Finish the testcase for extensions

Testing is still needed. Also only 'optional' extension fields
are supported now, 'repeated' fields are not yet supported.

.gitignore

@@ -0,0 +1,39 @@
+*.gcda
+*.gcno
+*.gcov
+*.o
+*.pb.c
+*.pb.h
+*.pb
+*~
+*.tar.gz
+julkaisu.txt
+docs/*.html
+docs/generator_flow.png
+example/client
+example/server
+example_avr_double/decode_double
+example_avr_double/encode_double
+example_avr_double/test_conversions
+example_unions/decode
+example_unions/encode
+generator/nanopb_pb2.pyc
+tests/decode_unittests
+tests/encode_unittests
+tests/test_compiles
+tests/test_decode1
+tests/test_decode2
+tests/test_decode3
+tests/test_decode3_buf
+tests/test_decode_callbacks
+tests/test_encode1
+tests/test_encode2
+tests/test_encode3
+tests/test_encode3_buf
+tests/test_encode_callbacks
+tests/test_missing_fields
+tests/test_multiple_files
+tests/bc_decode
+tests/bc_encode
+tests/breakpoints
+

CHANGELOG

@@ -0,0 +1,94 @@
+nanopb-0.2.1
+ NOTE: The default callback function signature has changed.
+       If you don't want to update your code, define PB_OLD_CALLBACK_STYLE.
+ 
+ Change the callback function to use void** (issue 69)
+ Add support for defining the nanopb options in a separate file (issue 12)
+ Add support for packed structs in IAR and MSVC (in addition to GCC) (issue 66)
+ Implement error message support for the encoder side (issue 7)
+ Handle unterminated strings when encoding (issue 68)
+ Fix bug with empty strings in repeated string callbacks (issue 73)
+ Fix regression in 0.2.0 with optional callback fields (issue 70)
+ Fix bugs with empty message types (issues 64, 65)
+ Fix some compiler warnings on clang (issue 67)
+ Some portability improvements (issues 60, 62)
+ Various new generator options
+ Improved tests
+
+nanopb-0.2.0
+ NOTE: This release requires you to regenerate all .pb.c
+       files. Files generated by older versions will not
+       compile anymore.
+
+ Reformat generated .pb.c files using macros (issue 58)
+ Rename PB_HTYPE_ARRAY -> PB_HTYPE_REPEATED
+ Separate PB_HTYPE to PB_ATYPE and PB_HTYPE
+ Move STATIC_ASSERTs to .pb.c file
+ Added CMake file (by Pavel Ilin)
+ Add option to give file extension to generator (by Michael Haberler)
+ Documentation updates
+
+nanopb-0.1.9
+ Fixed error message bugs (issues 52, 56)
+ Sanitize #ifndef filename (issue 50)
+ Performance improvements
+ Add compile-time option PB_BUFFER_ONLY
+ Add Java package name to nanopb.proto
+ Check for sizeof(double) == 8 (issue 54)
+ Added generator option to ignore some fields. (issue 51)
+ Added generator option to make message structs packed. (issue 49)
+ Add more test cases.
+
+nanopb-0.1.8
+ Fix bugs in the enum short names introduced in 0.1.7 (issues 42, 43)
+ Fix STATIC_ASSERT macro when using multiple .proto files. (issue 41)
+ Fix missing initialization of istream.errmsg
+ Make tests/Makefile work for non-gcc compilers (issue 40)
+
+nanopb-0.1.7
+ Remove "skip" mode from pb_istream_t callbacks. Example implementation had a bug. (issue 37)
+ Add option to use shorter names for enum values (issue 38)
+ Improve options support in generator (issues 12, 30)
+ Add nanopb version number to generated files (issue 36)
+ Add extern "C" to generated headers (issue 35)
+ Add names for structs to allow forward declaration (issue 39)
+ Add buffer size check in example (issue 34)
+ Fix build warnings on MS compilers (issue 33)
+
+nanopb-0.1.6
+ Reorganize the field decoder interface (issue 2)
+ Improve performance in submessage decoding (issue 28)
+ Implement error messages in the decoder side (issue 7)
+ Extended testcases (alltypes test is now complete).
+ Fix some compiler warnings (issues 25, 26, 27, 32).
+
+nanopb-0.1.5
+ Fix bug in decoder with packed arrays (issue 23).
+ Extended testcases.
+ Fix some compiler warnings.
+
+nanopb-0.1.4
+ Add compile-time options for easy-to-use >255 field support.
+ Improve the detection of missing required fields.
+ Added example on how to handle union messages.
+ Fix generator error with .proto without messages.
+ Fix problems that stopped the code from compiling with some compilers.
+ Fix some compiler warnings.
+
+nanopb-0.1.3
+ Refactor the field encoder interface.
+ Improve generator error messages (issue 5)
+ Add descriptor.proto into the #include exclusion list
+ Fix some compiler warnings.
+
+nanopb-0.1.2
+ Make the generator to generate include for other .proto files (issue 4).
+ Fixed generator not working on Windows (issue 3)
+
+nanopb-0.1.1
+ Fixed bug in encoder with 'bytes' fields (issue 1).
+ Fixed a bug in the generator that caused a compiler error on sfixed32 and sfixed64 fields.
+ Extended testcases.
+
+nanopb-0.1.0
+ First stable release.

README

@@ -5,5 +5,7 @@ Homepage: http://kapsi.fi/~jpa/nanopb/
 To compile the library, you'll need these libraries:
 protobuf-compiler python-protobuf libprotobuf-dev
 
+The only runtime dependencies are memset() and memcpy().
+
 To run the tests, run make under the tests folder.
 If it completes without error, everything is fine.

cmake/FindNanopb.cmake

@@ -0,0 +1,224 @@
+# Locate and configure the nanopb library.
+#
+# The following varialbes have to be set:
+#
+#   NANOPB_SRC_ROOT_FOLDER  - Path to nanopb source folder
+#
+# The following variables can be set and are optional:
+#
+#
+#   PROTOBUF_SRC_ROOT_FOLDER - When compiling with MSVC, if this cache variable is set
+#                              the protobuf-default VS project build locations
+#                              (vsprojects/Debug & vsprojects/Release) will be searched
+#                              for libraries and binaries.
+#
+#   NANOPB_IMPORT_DIRS       - List of additional directories to be searched for
+#                              imported .proto files.
+#
+#   NANOPB_GENERATE_CPP_APPEND_PATH - By default -I will be passed to protoc
+#                                     for each directory where a proto file is referenced.
+#                                     Set to FALSE if you want to disable this behaviour.
+#
+# Defines the following variables:
+#
+#   NANOPB_FOUND - Found the nanopb library (source&header files, generator tool, protoc compiler tool)
+#   NANOPB_INCLUDE_DIRS - Include directories for Google Protocol Buffers
+#
+# The following cache variables are also available to set or use:
+#   NANOPB_GENERATOR_EXECUTABLE - The nanopb generator
+#   PROTOBUF_PROTOC_EXECUTABLE - The protoc compiler
+#
+#  ====================================================================
+#
+# NANOPB_GENERATE_CPP (public function)
+#   SRCS = Variable to define with autogenerated
+#          source files
+#   HDRS = Variable to define with autogenerated
+#          header files
+#   ARGN = proto files
+#
+#  ====================================================================
+#  Example:
+#
+#   set(NANOPB_SRC_ROOT_FOLDER "/path/to/nanopb")
+#   set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ${NANOPB_SRC_ROOT_FOLDER}/cmake)
+#   find_package( Nanopb REQUIRED )
+#   include_directories(${NANOPB_INCLUDE_DIRS})
+#
+#   NANOPB_GENERATE_CPP(PROTO_SRCS PROTO_HDRS foo.proto)
+#
+#   include_directories(${CMAKE_CURRENT_BINARY_DIR})
+#   add_executable(bar bar.cc ${PROTO_SRCS} ${PROTO_HDRS})
+#
+#  ====================================================================
+
+#=============================================================================
+# Copyright 2009 Kitware, Inc.
+# Copyright 2009-2011 Philip Lowman <philip@yhbt.com>
+# Copyright 2008 Esben Mose Hansen, Ange Optimization ApS
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+#
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in the
+#   documentation and/or other materials provided with the distribution.
+#
+# * Neither the names of Kitware, Inc., the Insight Software Consortium,
+#   nor the names of their contributors may be used to endorse or promote
+#   products derived from this software without specific prior written
+#   permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+#
+#=============================================================================
+#
+# Changes
+# 2013.01.31 - Pavlo Ilin - used Modules/FindProtobuf.cmake from cmake 2.8.10 to
+#                           write FindNanopb.cmake
+#
+#=============================================================================
+
+
+function(NANOPB_GENERATE_CPP SRCS HDRS)
+  if(NOT ARGN)
+    return()
+  endif()
+
+  if(NANOPB_GENERATE_CPP_APPEND_PATH)
+    # Create an include path for each file specified
+    foreach(FIL ${ARGN})
+      get_filename_component(ABS_FIL ${FIL} ABSOLUTE)
+      get_filename_component(ABS_PATH ${ABS_FIL} PATH)
+
+      list(FIND _nanobp_include_path ${ABS_PATH} _contains_already)
+      if(${_contains_already} EQUAL -1)
+          list(APPEND _nanobp_include_path -I ${ABS_PATH})
+      endif()
+    endforeach()
+  else()
+    set(_nanobp_include_path -I ${CMAKE_CURRENT_SOURCE_DIR})
+  endif()
+
+  if(DEFINED NANOPB_IMPORT_DIRS)
+    foreach(DIR ${NANOPB_IMPORT_DIRS})
+      get_filename_component(ABS_PATH ${DIR} ABSOLUTE)
+      list(FIND _nanobp_include_path ${ABS_PATH} _contains_already)
+      if(${_contains_already} EQUAL -1)
+          list(APPEND _nanobp_include_path -I ${ABS_PATH})
+      endif()
+    endforeach()
+  endif()
+
+  set(${SRCS})
+  set(${HDRS})
+  get_filename_component(GENERATOR_PATH ${NANOPB_GENERATOR_EXECUTABLE} PATH)
+
+  foreach(FIL ${ARGN})
+    get_filename_component(ABS_FIL ${FIL} ABSOLUTE)
+    get_filename_component(FIL_WE ${FIL} NAME_WE)
+
+    list(APPEND ${SRCS} "${CMAKE_CURRENT_BINARY_DIR}/${FIL_WE}.pb.c")
+    list(APPEND ${HDRS} "${CMAKE_CURRENT_BINARY_DIR}/${FIL_WE}.pb.h")
+
+    add_custom_command(
+      OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/${FIL_WE}.pb"
+      COMMAND  ${PROTOBUF_PROTOC_EXECUTABLE}
+      ARGS -I${GENERATOR_PATH} -I${CMAKE_CURRENT_BINARY_DIR} ${_nanobp_include_path} -o${FIL_WE}.pb ${ABS_FIL}
+      DEPENDS ${ABS_FIL}
+      COMMENT "Running C++ protocol buffer compiler on ${FIL}"
+      VERBATIM )
+
+    add_custom_command(
+      OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/${FIL_WE}.pb.c"
+             "${CMAKE_CURRENT_BINARY_DIR}/${FIL_WE}.pb.h"
+      COMMAND python
+      ARGS ${NANOPB_GENERATOR_EXECUTABLE} ${FIL_WE}.pb
+      DEPENDS ${FIL_WE}.pb
+      COMMENT "Running nanopb generator on ${FIL_WE}.pb"
+      VERBATIM )
+  endforeach()
+
+  set_source_files_properties(${${SRCS}} ${${HDRS}} PROPERTIES GENERATED TRUE)
+  set(${SRCS} ${${SRCS}} ${NANOPB_SRCS} PARENT_SCOPE)
+  set(${HDRS} ${${HDRS}} ${NANOPB_HDRS} PARENT_SCOPE)
+
+endfunction()
+
+
+
+#
+# Main.
+#
+
+# By default have NANOPB_GENERATE_CPP macro pass -I to protoc
+# for each directory where a proto file is referenced.
+if(NOT DEFINED NANOPB_GENERATE_CPP_APPEND_PATH)
+  set(NANOPB_GENERATE_CPP_APPEND_PATH TRUE)
+endif()
+
+# Find the include directory
+find_path(NANOPB_INCLUDE_DIRS
+    pb.h
+    PATHS ${NANOPB_SRC_ROOT_FOLDER}
+)
+mark_as_advanced(NANOPB_INCLUDE_DIRS)
+
+# Find nanopb source files
+set(NANOPB_SRCS)
+set(NANOPB_HDRS)
+list(APPEND _nanopb_srcs pb_decode.c pb_encode.c)
+list(APPEND _nanopb_hdrs pb_decode.h pb_encode.h pb.h)
+
+foreach(FIL ${_nanopb_srcs})
+  find_file(${FIL}__nano_pb_file NAMES ${FIL} PATHS ${NANOPB_SRC_ROOT_FOLDER} ${NANOPB_INCLUDE_DIRS})
+  list(APPEND NANOPB_SRCS "${${FIL}__nano_pb_file}")
+  mark_as_advanced(${FIL}__nano_pb_file)
+endforeach()
+
+foreach(FIL ${_nanopb_hdrs})
+  find_file(${FIL}__nano_pb_file NAMES ${FIL} PATHS ${NANOPB_INCLUDE_DIRS})
+  mark_as_advanced(${FIL}__nano_pb_file)
+  list(APPEND NANOPB_HDRS "${${FIL}__nano_pb_file}")
+endforeach()
+
+# Find the protoc Executable
+find_program(PROTOBUF_PROTOC_EXECUTABLE
+    NAMES protoc
+    DOC "The Google Protocol Buffers Compiler"
+    PATHS
+    ${PROTOBUF_SRC_ROOT_FOLDER}/vsprojects/Release
+    ${PROTOBUF_SRC_ROOT_FOLDER}/vsprojects/Debug
+)
+mark_as_advanced(PROTOBUF_PROTOC_EXECUTABLE)
+
+# Find nanopb generator
+find_file(NANOPB_GENERATOR_EXECUTABLE
+    NAMES nanopb_generator.py
+    DOC "nanopb generator"
+    PATHS
+    ${NANOPB_SRC_ROOT_FOLDER}/generator
+)
+mark_as_advanced(NANOPB_GENERATOR_EXECUTABLE)
+
+include(FindPackageHandleStandardArgs)
+FIND_PACKAGE_HANDLE_STANDARD_ARGS(NANOPB DEFAULT_MSG
+  NANOPB_INCLUDE_DIRS
+  NANOPB_SRCS NANOPB_HDRS
+  NANOPB_GENERATOR_EXECUTABLE
+  PROTOBUF_PROTOC_EXECUTABLE
+  )

docs/concepts.rst

@@ -10,33 +10,40 @@ The things outlined here are the underlying concepts of the nanopb design.
 
 Proto files
 ===========
-All Protocol Buffers implementations use .proto files to describe the message format.
-The point of these files is to be a portable interface description language.
+All Protocol Buffers implementations use .proto files to describe the message
+format. The point of these files is to be a portable interface description
+language.
 
 Compiling .proto files for nanopb
 ---------------------------------
-Nanopb uses the Google's protoc compiler to parse the .proto file, and then a python script to generate the C header and source code from it::
+Nanopb uses the Google's protoc compiler to parse the .proto file, and then a
+python script to generate the C header and source code from it::
 
     user@host:~$ protoc -omessage.pb message.proto
     user@host:~$ python ../generator/nanopb_generator.py message.pb
     Writing to message.h and message.c
     user@host:~$
 
-Compiling .proto files with nanopb options
-------------------------------------------
-Nanopb defines two extensions for message fields described in .proto files: *max_size* and *max_count*.
-These are the maximum size of a string and maximum count of items in an array::
+Modifying generator behaviour
+-----------------------------
+Using generator options, you can set maximum sizes for fields in order to
+allocate them statically. The preferred way to do this is to create an .options
+file with the same name as your .proto file::
 
-    required string name = 1 [(nanopb).max_size = 40];
-    repeated PhoneNumber phone = 4 [(nanopb).max_count = 5];
+   # Foo.proto
+   message Foo {
+      required string name = 1;
+   }
 
-To use these extensions, you need to place an import statement in the beginning of the file::
+::
 
-    import "nanopb.proto";
+   # Foo.options
+   Foo.name max_size:16
 
-This file, in turn, requires the file *google/protobuf/descriptor.proto*. This is usually installed under */usr/include*. Therefore, to compile a .proto file which uses options, use a protoc command similar to::
+For more information on this, see the `Proto file options`_ section in the
+reference manual.
 
-    protoc -I/usr/include -Inanopb/generator -I. -omessage.pb message.proto
+.. _`Proto file options`: reference.html#proto-file-options
 
 Streams
 =======
@@ -50,6 +57,7 @@ There are a few generic rules for callback functions:
 #) Use state to store your own data, such as a file descriptor.
 #) *bytes_written* and *bytes_left* are updated by pb_write and pb_read.
 #) Your callback may be used with substreams. In this case *bytes_left*, *bytes_written* and *max_size* have smaller values than the original stream. Don't use these values to calculate pointers.
+#) Always read or write the full requested length of data. For example, POSIX *recv()* needs the *MSG_WAITALL* parameter to accomplish this.
 
 Output streams
 --------------
@@ -91,9 +99,8 @@ Writing to stdout::
 
 Input streams
 -------------
-For input streams, there are a few extra rules:
+For input streams, there is one extra rule:
 
-#) If buf is NULL, read from stream but don't store the data. This is used to skip unknown input.
 #) You don't need to know the length of the message in advance. After getting EOF error when reading, set bytes_left to 0 and return false. Pb_decode will detect this and if the EOF was in a proper position, it will return true.
 
 Here is the structure::
@@ -167,7 +174,9 @@ Field callbacks
 ===============
 When a field has dynamic length, nanopb cannot statically allocate storage for it. Instead, it allows you to handle the field in whatever way you want, using a callback function.
 
-The `pb_callback_t`_ structure contains a function pointer and a *void* pointer you can use for passing data to the callback. If the function pointer is NULL, the field will be skipped. The actual behavior of the callback function is different in encoding and decoding modes.
+The `pb_callback_t`_ structure contains a function pointer and a *void* pointer called *arg* you can use for passing data to the callback. If the function pointer is NULL, the field will be skipped. A pointer to the *arg* is passed to the function, so that it can modify it and retrieve the value.
+
+The actual behavior of the callback function is different in encoding and decoding modes. In encoding mode, the callback is called once and should write out everything, including field tags. In decoding mode, the callback is called repeatedly for every data item.
 
 .. _`pb_callback_t`: reference.html#pb-callback-t
 
@@ -175,7 +184,7 @@ Encoding callbacks
 ------------------
 ::
 
-    bool (*encode)(pb_ostream_t *stream, const pb_field_t *field, const void *arg);
+    bool (*encode)(pb_ostream_t *stream, const pb_field_t *field, void * const *arg);
 
 When encoding, the callback should write out complete fields, including the wire type and field number tag. It can write as many or as few fields as it likes. For example, if you want to write out an array as *repeated* field, you should do it all in a single call.
 
@@ -189,7 +198,7 @@ If the callback is used in a submessage, it will be called multiple times during
 
 This callback writes out a dynamically sized string::
 
-    bool write_string(pb_ostream_t *stream, const pb_field_t *field, const void *arg)
+    bool write_string(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
     {
         char *str = get_string_from_somewhere();
         if (!pb_encode_tag_for_field(stream, field))
@@ -202,7 +211,7 @@ Decoding callbacks
 ------------------
 ::
 
-    bool (*decode)(pb_istream_t *stream, const pb_field_t *field, void *arg);
+    bool (*decode)(pb_istream_t *stream, const pb_field_t *field, void **arg);
 
 When decoding, the callback receives a length-limited substring that reads the contents of a single field. The field tag has already been read. For *string* and *bytes*, the length value has already been parsed, and is available at *stream->bytes_left*.
 
@@ -212,7 +221,7 @@ The callback will be called multiple times for repeated fields. For packed field
 
 This callback reads multiple integers and prints them::
 
-    bool read_ints(pb_istream_t *stream, const pb_field_t *field, void *arg)
+    bool read_ints(pb_istream_t *stream, const pb_field_t *field, void **arg)
     {
         while (stream->bytes_left)
         {
@@ -241,16 +250,8 @@ For example this submessage in the Person.proto file::
 generates this field description array for the structure *Person_PhoneNumber*::
 
  const pb_field_t Person_PhoneNumber_fields[3] = {
-    {1, PB_HTYPE_REQUIRED | PB_LTYPE_STRING,
-    offsetof(Person_PhoneNumber, number), 0,
-    pb_membersize(Person_PhoneNumber, number), 0, 0},
-
-    {2, PB_HTYPE_OPTIONAL | PB_LTYPE_VARINT,
-    pb_delta(Person_PhoneNumber, type, number),
-    pb_delta(Person_PhoneNumber, has_type, type),
-    pb_membersize(Person_PhoneNumber, type), 0,
-    &Person_PhoneNumber_type_default},
-
+    PB_FIELD(  1, STRING  , REQUIRED, STATIC, Person_PhoneNumber, number, number, 0),
+    PB_FIELD(  2, ENUM    , OPTIONAL, STATIC, Person_PhoneNumber, type, number, &Person_PhoneNumber_type_default),
     PB_LAST_FIELD
  };
 
@@ -258,18 +259,14 @@ generates this field description array for the structure *Person_PhoneNumber*::
 Return values and error handling
 ================================
 
-Most functions in nanopb return bool: *true* means success, *false* means failure. If this is enough for you, skip this section.
+Most functions in nanopb return bool: *true* means success, *false* means failure. There is also some support for error messages for debugging purposes: the error messages go in *stream->errmsg*.
 
-For simplicity, nanopb doesn't define it's own error codes. This might be added if there is a compelling need for it. You can however deduce something about the error causes:
+The error messages help in guessing what is the underlying cause of the error. The most common error conditions are:
 
-1) Running out of memory. Because everything is allocated from the stack, nanopb can't detect this itself. Encoding or decoding the same type of a message always takes the same amount of stack space. Therefore, if it works once, it works always.
-2) Invalid field description. These are usually stored as constants, so if it works under the debugger, it always does.
-3) IO errors in your own stream callbacks. Because encoding/decoding stops at the first error, you can overwrite the *state* field in the struct and store your own error code there.
-4) Errors that happen in your callback functions. You can use the state field in the callback structure.
+1) Running out of memory, i.e. stack overflow.
+2) Invalid field descriptors (would usually mean a bug in the generator).
+3) IO errors in your own stream callbacks.
+4) Errors that happen in your callback functions.
 5) Exceeding the max_size or bytes_left of a stream.
 6) Exceeding the max_size of a string or array field
-7) Invalid protocol buffers binary message. It's not like you could recover from it anyway, so a simple failure should be enough.
-
-In my opinion, it is enough that 1. and 2. can be resolved using a debugger.
-
-However, you may be interested which of the remaining conditions caused the error. For 3. and 4., you can set and check the state. If you have to detect 5. and 6., you should convert the fields to callback type. Any remaining problem is of type 7.
+7) Invalid protocol buffers binary message.

docs/index.rst

@@ -36,23 +36,26 @@ Features and limitations
 **Features**
 
 #) Pure C runtime
-#) Small code size (2–10 kB depending on processor)
-#) Small ram usage (typically 200 bytes)
+#) Small code size (2–10 kB depending on processor, plus any message definitions)
+#) Small ram usage (typically ~300 bytes, plus any message structs)
 #) Allows specifying maximum size for strings and arrays, so that they can be allocated statically.
 #) No malloc needed: everything can be allocated statically or on the stack.
 #) You can use either encoder or decoder alone to cut the code size in half.
+#) Support for most protobuf features, including: all data types, nested submessages, default values, repeated and optional fields, packed arrays.
+#) Callback mechanism for handling messages larger than can fit in available RAM.
+#) Extensive set of tests.
 
 **Limitations**
 
 #) User must provide callbacks when decoding arrays or strings without maximum size. Malloc support could be added as a separate module.
-#) Some speed has been sacrificed for code size. For example varint calculations are always done in 64 bits.
+#) Some speed has been sacrificed for code size.
 #) Encoding is focused on writing to streams. For memory buffers only it could be made more efficient.
 #) The deprecated Protocol Buffers feature called "groups" is not supported.
 #) Fields in the generated structs are ordered by the tag number, instead of the natural ordering in .proto file.
 #) Unknown fields are not preserved when decoding and re-encoding a message.
 #) Reflection (runtime introspection) is not supported. E.g. you can't request a field by giving its name in a string.
 #) Numeric arrays are always encoded as packed, even if not marked as packed in .proto. This causes incompatibility with decoders that do not support packed format.
-#) Cyclic references between messages are not supported. They could be supported in callback-mode if there was an option in the generator to set the mode.
+#) Cyclic references between messages are supported only in callback mode.
 
 Getting started
 ===============
@@ -104,10 +107,3 @@ Debugging and testing
 =====================
 Extensive unittests are included under the *tests* folder. Just type *make* there to run the tests.
 
-This also generates a file called *breakpoints* which includes all lines returning *false* in nanopb. You can use this in gdb by typing *source breakpoints*, after which gdb will break on first nanopb error.
-
-Wishlist
-========
-#) A specialized encoder for encoding to a memory buffer. Should serialize in reverse order to avoid having to determine submessage size beforehand.
-#) A cleaner rewrite of the Python-based source generator.
-#) Better performance for 16- and 8-bit platforms: use smaller datatypes where possible.

docs/reference.rst

@@ -6,25 +6,182 @@ Nanopb: API reference
 
 .. contents ::
 
+
+
+
 Compilation options
 ===================
-The following options can be specified using -D switch given to the C compiler:
+The following options can be specified in one of two ways:
+
+1. Using the -D switch on the C compiler command line.
+2. By #defining them at the top of pb.h.
+
+You must have the same settings for the nanopb library and all code that
+includes pb.h.
+
+============================  ================================================
+__BIG_ENDIAN__                 Set this if your platform stores integers and
+                               floats in big-endian format. Mixed-endian
+                               systems (different layout for ints and floats)
+                               are currently not supported.
+NANOPB_INTERNALS               Set this to expose the field encoder functions
+                               that are hidden since nanopb-0.1.3.
+PB_MAX_REQUIRED_FIELDS         Maximum number of required fields to check for
+                               presence. Default value is 64. Increases stack
+                               usage 1 byte per every 8 fields. Compiler
+                               warning will tell if you need this.
+PB_FIELD_16BIT                 Add support for tag numbers > 255 and fields
+                               larger than 255 bytes or 255 array entries.
+                               Increases code size 3 bytes per each field.
+                               Compiler error will tell if you need this.
+PB_FIELD_32BIT                 Add support for tag numbers > 65535 and fields
+                               larger than 65535 bytes or 65535 array entries.
+                               Increases code size 9 bytes per each field.
+                               Compiler error will tell if you need this.
+PB_NO_ERRMSG                   Disables the support for error messages; only
+                               error information is the true/false return
+                               value. Decreases the code size by a few hundred
+                               bytes.
+PB_BUFFER_ONLY                 Disables the support for custom streams. Only
+                               supports encoding and decoding with memory
+                               buffers. Speeds up execution and decreases code
+                               size slightly.
+PB_OLD_CALLBACK_STYLE          Use the old function signature (void\* instead
+                               of void\*\*) for callback fields. This was the
+                               default until nanopb-0.2.1.
+============================  ================================================
+
+The PB_MAX_REQUIRED_FIELDS, PB_FIELD_16BIT and PB_FIELD_32BIT settings allow
+raising some datatype limits to suit larger messages. Their need is recognized
+automatically by C-preprocessor #if-directives in the generated .pb.h files.
+The default setting is to use the smallest datatypes (least resources used).
+
+
+
+
+Proto file options
+==================
+The generator behaviour can be adjusted using these options, defined in the
+'nanopb.proto' file in the generator folder:
+
+============================  ================================================
+max_size                       Allocated size for *bytes* and *string* fields.
+max_count                      Allocated number of entries in arrays
+                               (*repeated* fields).
+type                           Type of the generated field. Default value
+                               is *FT_DEFAULT*, which selects automatically.
+                               You can use *FT_CALLBACK*, *FT_STATIC* or
+                               *FT_IGNORE* to force a callback field, a static
+                               field or to completely ignore the field.
+long_names                     Prefix the enum name to the enum value in
+                               definitions, i.e. *EnumName_EnumValue*. Enabled
+                               by default.
+packed_struct                  Make the generated structures packed.
+                               NOTE: This cannot be used on CPUs that break
+                               on unaligned accesses to variables.
+============================  ================================================
+
+These options can be defined for the .proto files before they are converted
+using the nanopb-generatory.py. There are three ways to define the options:
+
+1. Using a separate .options file.
+   This is the preferred way as of nanopb-0.2.1, because it has the best
+   compatibility with other protobuf libraries.
+2. Defining the options on the command line of nanopb_generator.py.
+   This only makes sense for settings that apply to a whole file.
+3. Defining the options in the .proto file using the nanopb extensions.
+   This is the way used in nanopb-0.1, and will remain supported in the
+   future. It however sometimes causes trouble when using the .proto file
+   with other protobuf libraries.
+
+The effect of the options is the same no matter how they are given. The most
+common purpose is to define maximum size for string fields in order to
+statically allocate them.
+
+Defining the options in a .options file
+---------------------------------------
+The preferred way to define options is to have a separate file
+'myproto.options' in the same directory as the 'myproto.proto'. ::
+
+    # myproto.proto
+    message MyMessage {
+        required string name = 1;
+        repeated int32 ids = 4;
+    }
+
+::
+
+    # myproto.options
+    MyMessage.name         max_size:40
+    MyMessage.ids          max_count:5
+
+The generator will automatically search for this file and read the
+options from it. The file format is as follows:
+
+* Lines starting with '#' or '//' are regarded as comments.
+* Blank lines are ignored.
+* All other lines should start with a field name pattern, followed by one or
+  more options. For example: *"MyMessage.myfield max_size:5 max_count:10"*.
+* The field name pattern is matched against a string of form *'Message.field'*.
+  For nested messages, the string is *'Message.SubMessage.field'*.
+* The field name pattern may use the notation recognized by Python fnmatch():
+
+  - *\** matches any part of string, like 'Message.\*' for all fields
+  - *\?* matches any single character
+  - *[seq]* matches any of characters 's', 'e' and 'q'
+  - *[!seq]* matches any other character
+
+* The options are written as *'option_name:option_value'* and several options
+  can be defined on same line, separated by whitespace.
+* Options defined later in the file override the ones specified earlier, so
+  it makes sense to define wildcard options first in the file and more specific
+  ones later.
+  
+If preferred, the name of the options file can be set using the command line
+switch *-f* to nanopb_generator.py.
+
+Defining the options on command line
+------------------------------------
+The nanopb_generator.py has a simple command line option *-s OPTION:VALUE*.
+The setting applies to the whole file that is being processed.
+
+Defining the options in the .proto file
+---------------------------------------
+The .proto file format allows defining custom options for the fields.
+The nanopb library comes with *nanopb.proto* which does exactly that, allowing
+you do define the options directly in the .proto file::
+
+    import "nanopb.proto";
+    
+    message MyMessage {
+        required string name = 1 [(nanopb).max_size = 40];
+        repeated int32 ids = 4   [(nanopb).max_count = 5];
+    }
+
+A small complication is that you have to set the include path of protoc so that
+nanopb.proto can be found. This file, in turn, requires the file
+*google/protobuf/descriptor.proto*. This is usually installed under
+*/usr/include*. Therefore, to compile a .proto file which uses options, use a
+protoc command similar to::
+
+    protoc -I/usr/include -Inanopb/generator -I. -omessage.pb message.proto
+
+The options can be defined in file, message and field scopes::
+
+    option (nanopb_fileopt).max_size = 20; // File scope
+    message Message
+    {
+        option (nanopb_msgopt).max_size = 30; // Message scope
+        required string fieldsize = 1 [(nanopb).max_size = 40]; // Field scope
+    }
+
+
+
+
+
+
 
-============================  ================================================================================================
-__BIG_ENDIAN__                 Set this if your platform stores integers and floats in big-endian format.
-                               Mixed-endian systems (different layout for ints and floats) are currently not supported.
-NANOPB_INTERNALS               Set this to expose the field encoder functions that are hidden since nanopb-0.1.3.
-PB_MAX_REQUIRED_FIELDS         Maximum number of required fields to check for presence. Default value is 64. Increases stack
-                               usage 1 byte per every 8 fields. Compiler warning will tell if you need this.
-PB_FIELD_16BIT                 Add support for tag numbers > 255 and fields larger than 255 bytes or 255 array entries.
-                               Increases code size 3 bytes per each field. Compiler error will tell if you need this.
-PB_FIELD_32BIT                 Add support for tag numbers > 65535 and fields larger than 65535 bytes or 65535 array entries.
-                               Increases code size 9 bytes per each field. Compiler error will tell if you need this.
-============================  ================================================================================================
 
-The PB_MAX_REQUIRED_FIELDS, PB_FIELD_16BIT and PB_FIELD_32BIT settings allow raising some datatype limits to suit larger messages.
-Their need is recognized automatically by C-preprocessor #if-directives in the generated .pb.h files. The default setting is to use
-the smallest datatypes (least resources used).
 
 pb.h
 ====
@@ -33,22 +190,23 @@ pb_type_t
 ---------
 Defines the encoder/decoder behaviour that should be used for a field. ::
 
-    typedef enum { ... } pb_type_t;
+    typedef uint8_t pb_type_t;
 
-The low-order byte of the enumeration values defines the function that can be used for encoding and decoding the field data:
+The low-order nibble of the enumeration values defines the function that can be used for encoding and decoding the field data:
 
 ==================== ===== ================================================
 LTYPE identifier     Value Storage format
 ==================== ===== ================================================
 PB_LTYPE_VARINT      0x00  Integer.
 PB_LTYPE_SVARINT     0x01  Integer, zigzag encoded.
-PB_LTYPE_FIXED       0x02  Integer or floating point.
-PB_LTYPE_BYTES       0x03  Structure with *size_t* field and byte array.
-PB_LTYPE_STRING      0x04  Null-terminated string.
-PB_LTYPE_SUBMESSAGE  0x05  Submessage structure.
+PB_LTYPE_FIXED32     0x02  32-bit integer or floating point.
+PB_LTYPE_FIXED64     0x03  64-bit integer or floating point.
+PB_LTYPE_BYTES       0x04  Structure with *size_t* field and byte array.
+PB_LTYPE_STRING      0x05  Null-terminated string.
+PB_LTYPE_SUBMESSAGE  0x06  Submessage structure.
 ==================== ===== ================================================
 
-The high-order byte defines whether the field is required, optional, repeated or callback:
+The bits 4-5 define whether the field is required, optional or repeated:
 
 ==================== ===== ================================================
 HTYPE identifier     Value Field handling
@@ -56,13 +214,24 @@ HTYPE identifier     Value Field handling
 PB_HTYPE_REQUIRED    0x00  Verify that field exists in decoded message.
 PB_HTYPE_OPTIONAL    0x10  Use separate *has_<field>* boolean to specify
                            whether the field is present.
-PB_HTYPE_ARRAY       0x20  A repeated field with preallocated array.
+                           (Unless it is a callback)
+PB_HTYPE_REPEATED    0x20  A repeated field with preallocated array.
                            Separate *<field>_count* for number of items.
-PB_HTYPE_CALLBACK    0x30  A field with dynamic storage size, data is
-                           actually a pointer to a structure containing a
-                           callback function.
+                           (Unless it is a callback)
 ==================== ===== ================================================
 
+The bits 6-7 define the how the storage for the field is allocated:
+
+==================== ===== ================================================
+ATYPE identifier     Value Allocation method
+==================== ===== ================================================
+PB_ATYPE_STATIC      0x00  Statically allocated storage in the structure.
+PB_ATYPE_CALLBACK    0x40  A field with dynamic storage size. Struct field
+                           actually contains a pointer to a callback
+                           function.
+==================== ===== ================================================
+
+
 pb_field_t
 ----------
 Describes a single structure field with memory position in relation to others. The descriptions are usually autogenerated. ::
@@ -79,7 +248,7 @@ Describes a single structure field with memory position in relation to others. T
     } pb_packed;
 
 :tag:           Tag number of the field or 0 to terminate a list of fields.
-:type:          LTYPE and HTYPE of the field.
+:type:          LTYPE, HTYPE and ATYPE of the field.
 :data_offset:   Offset of field data, relative to the end of the previous field.
 :size_offset:   Offset of *bool* flag for optional fields or *size_t* count for arrays, relative to field data.
 :data_size:     Size of a single data entry, in bytes. For PB_LTYPE_BYTES, the size of the byte array inside the containing structure. For PB_HTYPE_CALLBACK, size of the C data type if known.
@@ -106,14 +275,16 @@ Part of a message structure, for fields with type PB_HTYPE_CALLBACK::
     typedef struct _pb_callback_t pb_callback_t;
     struct _pb_callback_t {
         union {
-            bool (*decode)(pb_istream_t *stream, const pb_field_t *field, void *arg);
-            bool (*encode)(pb_ostream_t *stream, const pb_field_t *field, const void *arg);
+            bool (*decode)(pb_istream_t *stream, const pb_field_t *field, void **arg);
+            bool (*encode)(pb_ostream_t *stream, const pb_field_t *field, void * const *arg);
         } funcs;
         
         void *arg;
     };
 
-The *arg* is passed to the callback when calling. It can be used to store any information that the callback might need.
+A pointer to the *arg* is passed to the callback when calling. It can be used to store any information that the callback might need.
+
+Previously the function received just the value of *arg* instead of a pointer to it. This old behaviour can be enabled by defining *PB_OLD_CALLBACK_STYLE*.
 
 When calling `pb_encode`_, *funcs.encode* is used, and similarly when calling `pb_decode`_, *funcs.decode* is used. The function pointers are stored in the same memory location but are of incompatible types. You can set the function pointer to NULL to skip the field.
 
@@ -128,6 +299,41 @@ Protocol Buffers wire types. These are used with `pb_encode_tag`_. ::
         PB_WT_32BIT  = 5
     } pb_wire_type_t;
 
+PB_GET_ERROR
+------------
+Get the current error message from a stream, or a placeholder string if
+there is no error message::
+
+    #define PB_GET_ERROR(stream) (string expression)
+
+This should be used for printing errors, for example::
+
+    if (!pb_decode(...))
+    {
+        printf("Decode failed: %s\n", PB_GET_ERROR(stream));
+    }
+
+The macro only returns pointers to constant strings (in code memory),
+so that there is no need to release the returned pointer.
+
+PB_RETURN_ERROR
+---------------
+Set the error message and return false::
+
+    #define PB_RETURN_ERROR(stream,msg) (sets error and returns false)
+
+This should be used to handle error conditions inside nanopb functions
+and user callback functions::
+
+    if (error_condition)
+    {
+        PB_RETURN_ERROR(stream, "something went wrong");
+    }
+
+The *msg* parameter must be a constant string.
+
+
+
 pb_encode.h
 ===========
 
@@ -277,6 +483,17 @@ In Protocol Buffers format, the submessage size must be written before the subme
 
 If the submessage contains callback fields, the callback function might misbehave and write out a different amount of data on the second call. This situation is recognized and *false* is returned, but garbage will be written to the output before the problem is detected.
 
+
+
+
+
+
+
+
+
+
+
+
 pb_decode.h
 ===========
 
@@ -320,15 +537,15 @@ In addition to EOF, the pb_decode implementation supports terminating a message
 
 For optional fields, this function applies the default value and sets *has_<field>* to false if the field is not present.
 
-pb_decode_varint
+pb_decode_noinit
 ----------------
-Read and decode a varint_ encoded integer. ::
+Same as `pb_decode`_, except does not apply the default values to fields. ::
 
-    bool pb_decode_varint(pb_istream_t *stream, uint64_t *dest);
+    bool pb_decode_noinit(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct);
 
-:stream:        Input stream to read from. 1-10 bytes will be read.
-:dest:          Storage for the decoded integer. Value is undefined on error.
-:returns:       True on success, false if value exceeds uint64_t range or an IO error happens.
+(parameters are the same as for `pb_decode`_.)
+
+The destination structure should be filled with zeros before calling this function. Doing a *memset* manually can be slightly faster than using `pb_decode`_ if you don't need any default values.
 
 pb_skip_varint
 --------------
@@ -373,48 +590,41 @@ Remove the data for a field from the stream, without actually decoding it::
 :wire_type:     Type of field to skip.
 :returns:       True on success, false on IO error.
 
-.. sidebar:: Field decoders
+.. sidebar:: Decoding fields manually
     
-    The functions with names beginning with *pb_dec_* are called field decoders. Each PB_LTYPE has an own field decoder, which handles translating from Protocol Buffers data to C data.
+    The functions with names beginning with *pb_decode_* are used when dealing with callback fields. The typical reason for using callbacks is to have an array of unlimited size. In that case, `pb_decode`_ will call your callback function repeatedly, which can then store the values into e.g. filesystem in the order received in.
 
-    Each field decoder reads and decodes a single value. For arrays, the decoder is called repeatedly.
+    For decoding numeric (including enumerated and boolean) values, use `pb_decode_varint`_, `pb_decode_svarint`_, `pb_decode_fixed32`_ and `pb_decode_fixed64`_. They take a pointer to a 32- or 64-bit C variable, which you may then cast to smaller datatype for storage.
 
-    You can use the decoders from your callbacks. Just be aware that the pb_field_t passed to the callback is not directly compatible 
-    with the *varint* field decoders. Instead, you must create a new pb_field_t structure and set the data_size according to the data type 
-    you pass to *dest*, e.g. *field.data_size = sizeof(int);*. Other fields in the *pb_field_t* don't matter.
+    For decoding strings and bytes fields, the length has already been decoded. You can therefore check the total length in *stream->bytes_left* and read the data using `pb_read`_.
 
-    The field decoder interface is a bit messy as a result of the interface required inside the nanopb library.
-    Eventually they may be replaced by separate wrapper functions with a more friendly interface.
+    Finally, for decoding submessages in a callback, simply use `pb_decode`_ and pass it the *SubMessage_fields* descriptor array.
 
-pb_dec_varint
--------------
-Field decoder for PB_LTYPE_VARINT. ::
+pb_decode_varint
+----------------
+Read and decode a varint_ encoded integer. ::
 
-    bool pb_dec_varint(pb_istream_t *stream, const pb_field_t *field, void *dest)
+    bool pb_decode_varint(pb_istream_t *stream, uint64_t *dest);
 
 :stream:        Input stream to read from. 1-10 bytes will be read.
-:field:         Field description structure. Only *field->data_size* matters.
-:dest:          Pointer to destination integer. Must have size of *field->data_size* bytes.
-:returns:       True on success, false on IO errors or if `pb_decode_varint`_ fails.
+:dest:          Storage for the decoded integer. Value is undefined on error.
+:returns:       True on success, false if value exceeds uint64_t range or an IO error happens.
 
-This function first calls `pb_decode_varint`_. It then copies the first bytes of the 64-bit result value to *dest*, or on big endian architectures, the last bytes.
+pb_decode_svarint
+-----------------
+Similar to `pb_decode_varint`_, except that it performs zigzag-decoding on the value. This corresponds to the Protocol Buffers *sint32* and *sint64* datatypes. ::
 
-pb_dec_svarint
---------------
-Field decoder for PB_LTYPE_SVARINT. Similar to `pb_dec_varint`_, except that it performs zigzag-decoding on the value. ::
+    bool pb_decode_svarint(pb_istream_t *stream, int64_t *dest);
 
-    bool pb_dec_svarint(pb_istream_t *stream, const pb_field_t *field, void *dest);
+(parameters are the same as `pb_decode_varint`_)
 
-(parameters are the same as `pb_dec_varint`_)
+pb_decode_fixed32
+-----------------
+Decode a *fixed32*, *sfixed32* or *float* value. ::
 
-pb_dec_fixed32
---------------
-Field decoder for PB_LTYPE_FIXED32. ::
-
-    bool pb_dec_fixed32(pb_istream_t *stream, const pb_field_t *field, void *dest);
+    bool pb_decode_fixed32(pb_istream_t *stream, void *dest);
 
 :stream:        Input stream to read from. 4 bytes will be read.
-:field:         Not used.
 :dest:          Pointer to destination *int32_t*, *uint32_t* or *float*.
 :returns:       True on success, false on IO errors.
 
@@ -422,9 +632,9 @@ This function reads 4 bytes from the input stream.
 On big endian architectures, it then reverses the order of the bytes.
 Finally, it writes the bytes to *dest*.
 
-pb_dec_fixed64
---------------
-Field decoder for PB_LTYPE_FIXED64. ::
+pb_decode_fixed64
+-----------------
+Decode a *fixed64*, *sfixed64* or *double* value. ::
 
     bool pb_dec_fixed(pb_istream_t *stream, const pb_field_t *field, void *dest);
 
@@ -433,53 +643,28 @@ Field decoder for PB_LTYPE_FIXED64. ::
 :dest:          Pointer to destination *int64_t*, *uint64_t* or *double*.
 :returns:       True on success, false on IO errors.
 
-Same as `pb_dec_fixed32`_, except this reads 8 bytes.
+Same as `pb_decode_fixed32`_, except this reads 8 bytes.
 
-pb_dec_bytes
-------------
-Field decoder for PB_LTYPE_BYTES. Reads a length-prefixed block of bytes. ::
+pb_make_string_substream
+------------------------
+Decode the length for a field with wire type *PB_WT_STRING* and create a substream for reading the data. ::
 
-    bool pb_dec_bytes(pb_istream_t *stream, const pb_field_t *field, void *dest);
+    bool pb_make_string_substream(pb_istream_t *stream, pb_istream_t *substream);
 
-**Note:** This is an internal function that is not useful in decoder callbacks. To read bytes fields in callbacks, use 
-*stream->bytes_left* and `pb_read`_.
+:stream:        Original input stream to read the length and data from.
+:substream:     New substream that has limited length. Filled in by the function.
+:returns:       True on success, false if reading the length fails.
 
-:stream:        Input stream to read from.
-:field:         Field description structure. Only *field->data_size* matters.
-:dest:          Pointer to a structure similar to pb_bytes_array_t.
-:returns:       True on success, false on IO error or if length exceeds the array size.
+This function uses `pb_decode_varint`_ to read an integer from the stream. This is interpreted as a number of bytes, and the substream is set up so that its `bytes_left` is initially the same as the length, and its callback function and state the same as the parent stream.
 
-This function expects a pointer to a structure with a *size_t* field at start, and a variable sized byte array after it. It will deduce the maximum size of the array from *field->data_size*.
+pb_close_string_substream
+-------------------------
+Close the substream created with `pb_make_string_substream`_. ::
 
-pb_dec_string
--------------
-Field decoder for PB_LTYPE_STRING. Reads a length-prefixed string. ::
+    void pb_close_string_substream(pb_istream_t *stream, pb_istream_t *substream);
 
-    bool pb_dec_string(pb_istream_t *stream, const pb_field_t *field, void *dest);
-
-**Note:** This is an internal function that is not useful in decoder callbacks. To read string fields in callbacks, use 
-*stream->bytes_left* and `pb_read`_.
-
-:stream:        Input stream to read from.
-:field:         Field description structure. Only *field->data_size* matters.
-:dest:          Pointer to a character array of size *field->data_size*.
-:returns:       True on success, false on IO error or if length exceeds the array size.
-
-This function null-terminates the string when successful. On error, the contents of the destination array is undefined.
-
-pb_dec_submessage
------------------
-Field decoder for PB_LTYPE_SUBMESSAGE. Calls `pb_decode`_ to perform the actual decoding. ::
-
-    bool pb_dec_submessage(pb_istream_t *stream, const pb_field_t *field, void *dest)
-
-**Note:** This is an internal function that is not useful in decoder callbacks. To read submessage fields in callbacks, use 
-`pb_decode`_ directly.
-
-:stream:        Input stream to read from.
-:field:         Field description structure. Only *field->ptr* matters.
-:dest:          Pointer to the destination structure.
-:returns:       True on success, false on IO error or if `pb_decode`_ fails.
-
-The *field->ptr* should be a pointer to *pb_field_t* array describing the submessage.
+:stream:        Original input stream to read the length and data from.
+:substream:     Substream to close
 
+This function copies back the state from the substream to the parent stream.
+It must be called after done with the substream.

example/client.c

@@ -23,7 +23,7 @@
 #include "fileproto.pb.h"
 #include "common.h"
 
-bool printfile_callback(pb_istream_t *stream, const pb_field_t *field, void *arg)
+bool printfile_callback(pb_istream_t *stream, const pb_field_t *field, void **arg)
 {
     FileInfo fileinfo;
     
@@ -72,7 +72,7 @@ bool listdir(int fd, char *path)
     
     if (!pb_decode(&input, ListFilesResponse_fields, &response))
     {
-        fprintf(stderr, "Decoding failed.\n");
+        fprintf(stderr, "Decode failed: %s\n", PB_GET_ERROR(&input));
         return false;
     }
     

example/common.c

@@ -19,15 +19,6 @@ static bool read_callback(pb_istream_t *stream, uint8_t *buf, size_t count)
     int fd = (intptr_t)stream->state;
     int result;
     
-    if (buf == NULL)
-    {
-        /* Well, this is a really inefficient way to skip input. */
-        /* It is only used when there are unknown fields. */
-        char dummy;
-        while (count-- && recv(fd, &dummy, 1, 0) == 1);
-        return count == 0;
-    }
-    
     result = recv(fd, buf, count, MSG_WAITALL);
     
     if (result == 0)

example/fileproto.options

@@ -0,0 +1,13 @@
+# This file defines the nanopb-specific options for the messages defined
+# in fileproto.proto.
+#
+# If you come from high-level programming background, the hardcoded
+# maximum lengths may disgust you. However, if your microcontroller only
+# has a few kB of ram to begin with, setting reasonable limits for
+# filenames is ok.
+#
+# On the other hand, using the callback interface, it is not necessary
+# to set a limit on the number of files in the response.
+
+ListFilesRequest.path 	max_size:128
+FileInfo.name 		max_size:128

example/fileproto.proto

@@ -1,22 +1,14 @@
-import "nanopb.proto";
-
 // This defines protocol for a simple server that lists files.
 //
-// If you come from high-level programming background, the hardcoded
-// maximum lengths may disgust you. However, if your microcontroller only
-// has a few kB of ram to begin with, setting reasonable limits for
-// filenames is ok.
-//
-// On the other hand, using the callback interface, it is not necessary
-// to set a limit on the number of files in the response.
+// See also the nanopb-specific options in fileproto.options.
 
 message ListFilesRequest {
-    optional string path = 1 [default = "/", (nanopb).max_size = 128];
+    optional string path = 1 [default = "/"];
 }
 
 message FileInfo {
     required uint64 inode = 1;
-    required string name = 2 [(nanopb).max_size = 128];
+    required string name = 2;
 }
 
 message ListFilesResponse {

example/server.c

@@ -23,9 +23,9 @@
 #include "fileproto.pb.h"
 #include "common.h"
 
-bool listdir_callback(pb_ostream_t *stream, const pb_field_t *field, const void *arg)
+bool listdir_callback(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
 {
-    DIR *dir = (DIR*) arg;
+    DIR *dir = (DIR*) *arg;
     struct dirent *file;
     FileInfo fileinfo;
     
@@ -55,7 +55,7 @@ void handle_connection(int connfd)
     
     if (!pb_decode(&input, ListFilesRequest_fields, &request))
     {
-        printf("Decoding failed.\n");
+        printf("Decode failed: %s\n", PB_GET_ERROR(&input));
         return;
     }
     

example_avr_double/Makefile

@@ -0,0 +1,22 @@
+CFLAGS=-Wall -Werror -I .. -g -O0
+DEPS=double_conversion.c ../pb_decode.c ../pb_decode.h ../pb_encode.c ../pb_encode.h ../pb.h
+
+all: run_tests
+
+clean:
+	rm -f test_conversions encode_double decode_double doubleproto.pb.c doubleproto.pb.h
+
+test_conversions: test_conversions.c double_conversion.c
+	$(CC) $(CFLAGS) -o $@ $^
+
+%: %.c $(DEPS) doubleproto.pb.h doubleproto.pb.c
+	$(CC) $(CFLAGS) -o $@ $< double_conversion.c ../pb_decode.c ../pb_encode.c doubleproto.pb.c
+
+doubleproto.pb.c doubleproto.pb.h: doubleproto.proto ../generator/nanopb_generator.py
+	protoc -I. -I../generator -I/usr/include -odoubleproto.pb $<
+	python ../generator/nanopb_generator.py doubleproto.pb
+
+run_tests: test_conversions encode_double decode_double
+	./test_conversions
+	./encode_double | ./decode_double
+    

example_avr_double/README.txt

@@ -0,0 +1,22 @@
+Some processors/compilers, such as AVR-GCC, do not support the double
+datatype. Instead, they have sizeof(double) == 4. Because protocol
+binary format uses the double encoding directly, this causes trouble
+if the protocol in .proto requires double fields.
+
+This directory contains a solution to this problem. It uses uint64_t
+to store the raw wire values, because its size is correct on all
+platforms. The file double_conversion.c provides functions that
+convert these values to/from floats, without relying on compiler
+support.
+
+To use this method, you need to make two modifications to your code:
+
+1) Change all 'double' fields into 'fixed64' in the .proto.
+
+2) Whenever writing to a 'double' field, use float_to_double().
+
+3) Whenever reading a 'double' field, use double_to_float().
+
+The conversion routines should be as accurate as the float datatype can
+be. Furthermore, they should handle all special values (NaN, inf, denormalized
+numbers) correctly. There are testcases in test_conversions.c.

example_avr_double/decode_double.c

@@ -0,0 +1,33 @@
+/* Decodes a double value into a float variable.
+ * Used to read double values with AVR code, which doesn't support double directly.
+ */
+
+#include <stdio.h>
+#include <pb_decode.h>
+#include "double_conversion.h"
+#include "doubleproto.pb.h"
+
+int main()
+{
+    uint8_t buffer[32];
+    size_t count = fread(buffer, 1, sizeof(buffer), stdin);
+    pb_istream_t stream = pb_istream_from_buffer(buffer, count);
+    
+    AVRDoubleMessage message;
+    pb_decode(&stream, AVRDoubleMessage_fields, &message);
+    
+    float v1 = double_to_float(message.field1);
+    float v2 = double_to_float(message.field2);
+
+    printf("Values: %f %f\n", v1, v2);
+    
+    if (v1 == 1234.5678f &&
+        v2 == 0.00001f)
+    {
+        return 0;
+    }
+    else
+    {
+        return 1;
+    }
+}

example_avr_double/double_conversion.c

@@ -0,0 +1,123 @@
+/* Conversion routines for platforms that do not support 'double' directly. */
+
+#include "double_conversion.h"
+#include <math.h>
+
+typedef union {
+    float f;
+    uint32_t i;
+} conversion_t;
+
+/* Note: IEE 754 standard specifies float formats as follows:
+ * Single precision: sign,  8-bit exp, 23-bit frac.
+ * Double precision: sign, 11-bit exp, 52-bit frac.
+ */
+
+uint64_t float_to_double(float value)
+{
+    conversion_t in;
+    in.f = value;
+    uint8_t sign;
+    int16_t exponent;
+    uint64_t mantissa;
+    
+    /* Decompose input value */
+    sign = (in.i >> 31) & 1;
+    exponent = ((in.i >> 23) & 0xFF) - 127;
+    mantissa = in.i & 0x7FFFFF;
+    
+    if (exponent == 128)
+    {
+        /* Special value (NaN etc.) */
+        exponent = 1024;
+    }
+    else if (exponent == -127)
+    {
+        if (!mantissa)
+        {
+            /* Zero */
+            exponent = -1023;
+        }
+        else
+        {
+            /* Denormalized */
+            mantissa <<= 1;
+            while (!(mantissa & 0x800000))
+            {
+                mantissa <<= 1;
+                exponent--;
+            }
+            mantissa &= 0x7FFFFF;
+        }
+    }
+    
+    /* Combine fields */
+    mantissa <<= 29;
+    mantissa |= (uint64_t)(exponent + 1023) << 52;
+    mantissa |= (uint64_t)sign << 63;
+    
+    return mantissa;
+}
+
+float double_to_float(uint64_t value)
+{
+    uint8_t sign;
+    int16_t exponent;
+    uint32_t mantissa;
+    conversion_t out;
+
+    /* Decompose input value */
+    sign = (value >> 63) & 1;
+    exponent = ((value >> 52) & 0x7FF) - 1023;
+    mantissa = (value >> 28) & 0xFFFFFF; /* Highest 24 bits */
+ 
+    /* Figure if value is in range representable by floats. */
+    if (exponent == 1024)
+    {
+        /* Special value */
+        exponent = 128;
+    }
+    else if (exponent > 127)
+    {
+        /* Too large */        
+        if (sign)
+            return -INFINITY;
+        else
+            return INFINITY;
+    }
+    else if (exponent < -150)
+    {
+        /* Too small */
+        if (sign)
+            return -0.0f;
+        else
+            return 0.0f;
+    }
+    else if (exponent < -126)
+    {
+        /* Denormalized */
+        mantissa |= 0x1000000;
+        mantissa >>= (-126 - exponent);
+        exponent = -127;
+    }
+ 
+    /* Round off mantissa */
+    mantissa = (mantissa + 1) >> 1;
+    
+    /* Check if mantissa went over 2.0 */
+    if (mantissa & 0x800000)
+    {
+        exponent += 1;
+        mantissa &= 0x7FFFFF;
+        mantissa >>= 1;
+    }
+    
+    /* Combine fields */
+    out.i = mantissa;
+    out.i |= (uint32_t)(exponent + 127) << 23;
+    out.i |= (uint32_t)sign << 31;
+    
+    return out.f;
+}
+
+

example_avr_double/double_conversion.h

@@ -0,0 +1,26 @@
+/* AVR-GCC does not have real double datatype. Instead its double
+ * is equal to float, i.e. 32 bit value. If you need to communicate
+ * with other systems that use double in their .proto files, you
+ * need to do some conversion.
+ *
+ * These functions use bitwise operations to mangle floats into doubles
+ * and then store them in uint64_t datatype.
+ */
+
+#ifndef DOUBLE_CONVERSION
+#define DOUBLE_CONVERSION
+
+#include <stdint.h>
+
+/* Convert native 4-byte float into a 8-byte double. */
+extern uint64_t float_to_double(float value);
+
+/* Convert 8-byte double into native 4-byte float.
+ * Values are rounded to nearest, 0.5 away from zero.
+ * Overflowing values are converted to Inf or -Inf.
+ */
+extern float double_to_float(uint64_t value);
+
+
+#endif
+

example_avr_double/doubleproto.proto

@@ -0,0 +1,13 @@
+// A message containing doubles, as used by other applications.
+message DoubleMessage {
+    required double field1 = 1;
+    required double field2 = 2;
+}
+
+// A message containing doubles, but redefined using uint64_t.
+// For use in AVR code.
+message AVRDoubleMessage {
+    required fixed64 field1 = 1;
+    required fixed64 field2 = 2;
+}
+

example_avr_double/encode_double.c

@@ -0,0 +1,25 @@
+/* Encodes a float value into a double on the wire.
+ * Used to emit doubles from AVR code, which doesn't support double directly.
+ */
+
+#include <stdio.h>
+#include <pb_encode.h>
+#include "double_conversion.h"
+#include "doubleproto.pb.h"
+
+int main()
+{
+    AVRDoubleMessage message = {
+        float_to_double(1234.5678f),
+        float_to_double(0.00001f)
+    };
+    
+    uint8_t buffer[32];
+    pb_ostream_t stream = pb_ostream_from_buffer(buffer, sizeof(buffer));
+    
+    pb_encode(&stream, AVRDoubleMessage_fields, &message);
+    fwrite(buffer, 1, stream.bytes_written, stdout);
+
+    return 0;
+}
+

example_avr_double/test_conversions.c

@@ -0,0 +1,56 @@
+#include "double_conversion.h"
+#include <math.h>
+#include <stdio.h>
+
+static const double testvalues[] = {
+           0.0,        -0.0,         0.1,         -0.1,
+          M_PI,       -M_PI,  123456.789,  -123456.789,
+      INFINITY,   -INFINITY,         NAN, INFINITY - INFINITY,
+          1e38,       -1e38,        1e39,        -1e39,
+         1e-38,      -1e-38,       1e-39,       -1e-39,
+   3.14159e-37,-3.14159e-37, 3.14159e-43, -3.14159e-43,
+         1e-60,      -1e-60,       1e-45,       -1e-45,
+    0.99999999999999, -0.99999999999999, 127.999999999999, -127.999999999999
+};
+
+#define TESTVALUES_COUNT (sizeof(testvalues)/sizeof(testvalues[0]))
+
+int main()
+{
+    int status = 0;
+    int i;
+    for (i = 0; i < TESTVALUES_COUNT; i++)
+    {
+        double orig = testvalues[i];
+        float expected_float = (float)orig;
+        double expected_double = (double)expected_float;
+        
+        float got_float = double_to_float(*(uint64_t*)&orig);
+        uint64_t got_double = float_to_double(got_float);
+        
+        uint32_t e1 = *(uint32_t*)&expected_float;
+        uint32_t g1 = *(uint32_t*)&got_float;
+        uint64_t e2 = *(uint64_t*)&expected_double;
+        uint64_t g2 = got_double;
+        
+        if (g1 != e1)
+        {
+            printf("%3d double_to_float fail: %08x != %08x\n", i, g1, e1);
+            status = 1;
+        }
+        
+        if (g2 != e2)
+        {
+            printf("%3d float_to_double fail: %016llx != %016llx\n", i,
+                (unsigned long long)g2,
+                (unsigned long long)e2);
+            status = 1;
+        }
+    }
+
+    return status;
+}
+
+    
+    
+

example_unions/decode.c

@@ -17,7 +17,7 @@
 const pb_field_t* decode_unionmessage_type(pb_istream_t *stream)
 {
     pb_wire_type_t wire_type;
-    int tag;
+    uint32_t tag;
     bool eof;
 
     while (pb_decode_tag(stream, &wire_type, &tag, &eof))
@@ -44,10 +44,14 @@ const pb_field_t* decode_unionmessage_type(pb_istream_t *stream)
 
 bool decode_unionmessage_contents(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct)
 {
-    pb_field_t field = {}; /* NB: Could get rid of this wrapper by fixing issue #2. */
-    field.ptr = fields;
+    pb_istream_t substream;
+    bool status;
+    if (!pb_make_string_substream(stream, &substream))
+        return false;
     
-    return pb_dec_submessage(stream, &field, dest_struct);
+    status = pb_decode(&substream, fields, dest_struct);
+    pb_close_string_substream(stream, &substream);
+    return status;
 }
 
 int main()
@@ -81,7 +85,7 @@ int main()
     
     if (!status)
     {
-        printf("Decoding failed.\n");
+        printf("Decode failed: %s\n", PB_GET_ERROR(&stream));
         return 1;
     }
     

generator/nanopb.proto

@@ -7,11 +7,40 @@
 
 import "google/protobuf/descriptor.proto";
 
-message NanoPBOptions {
-  optional int32 max_size = 1;
-  optional int32 max_count = 2;
+option java_package = "fi.kapsi.koti.jpa.nanopb";
+
+enum FieldType {
+    FT_DEFAULT = 0; // Automatically decide field type, generate static field if possible.
+    FT_CALLBACK = 1; // Always generate a callback field.
+    FT_STATIC = 2; // Generate a static field or raise an exception if not possible.
+    FT_IGNORE = 3; // Ignore the field completely.
 }
 
+// This is the inner options message, which basically defines options for
+// a field. When it is used in message or file scope, it applies to all
+// fields.
+message NanoPBOptions {
+  // Allocated size for 'bytes' and 'string' fields.
+  optional int32 max_size = 1;
+  
+  // Allocated number of entries in arrays ('repeated' fields)
+  optional int32 max_count = 2;
+  
+  // Force type of field (callback or static allocation)
+  optional FieldType type = 3 [default = FT_DEFAULT];
+  
+  // Use long names for enums, i.e. EnumName_EnumValue.
+  optional bool long_names = 4 [default = true];
+  
+  // Add 'packed' attribute to generated structs.
+  // Note: this cannot be used on CPUs that break on unaligned
+  // accesses to variables.
+  optional bool packed_struct = 5 [default = false];
+}
+
+// Extensions to protoc 'Descriptor' type in order to define options
+// inside a .proto file.
+//
 // Protocol Buffers extension number registry
 // --------------------------------
 // Project:  Nanopb
@@ -20,7 +49,20 @@ message NanoPBOptions {
 // Extensions: 1010 (all types)
 // --------------------------------
 
+extend google.protobuf.FileOptions {
+    optional NanoPBOptions nanopb_fileopt = 1010;
+}
+
+extend google.protobuf.MessageOptions {
+    optional NanoPBOptions nanopb_msgopt = 1010;
+}
+
+extend google.protobuf.EnumOptions {
+    optional NanoPBOptions nanopb_enumopt = 1010;
+}
+
 extend google.protobuf.FieldOptions {
     optional NanoPBOptions nanopb = 1010;
 }
 
+

generator/nanopb_generator.py

@@ -1,7 +1,9 @@
 '''Generate header file for nanopb from a ProtoBuf FileDescriptorSet.'''
+nanopb_version = "nanopb-0.2.2-dev"
 
 try:
     import google.protobuf.descriptor_pb2 as descriptor
+    import google.protobuf.text_format as text_format
 except:
     print
     print "*************************************************************"
@@ -22,30 +24,38 @@ except:
     print
     raise
 
+
+
+
+
+
+# ---------------------------------------------------------------------------
+#                     Generation of single fields
+# ---------------------------------------------------------------------------
+
+import time
 import os.path
 
-# Values are tuple (c type, pb ltype)
+# Values are tuple (c type, pb type)
 FieldD = descriptor.FieldDescriptorProto
 datatypes = {
-    FieldD.TYPE_BOOL: ('bool', 'PB_LTYPE_VARINT'),
-    FieldD.TYPE_DOUBLE: ('double', 'PB_LTYPE_FIXED64'),
-    FieldD.TYPE_FIXED32: ('uint32_t', 'PB_LTYPE_FIXED32'),
-    FieldD.TYPE_FIXED64: ('uint64_t', 'PB_LTYPE_FIXED64'),
-    FieldD.TYPE_FLOAT: ('float', 'PB_LTYPE_FIXED32'),
-    FieldD.TYPE_INT32: ('int32_t', 'PB_LTYPE_VARINT'),
-    FieldD.TYPE_INT64: ('int64_t', 'PB_LTYPE_VARINT'),
-    FieldD.TYPE_SFIXED32: ('int32_t', 'PB_LTYPE_FIXED32'),
-    FieldD.TYPE_SFIXED64: ('int64_t', 'PB_LTYPE_FIXED64'),
-    FieldD.TYPE_SINT32: ('int32_t', 'PB_LTYPE_SVARINT'),
-    FieldD.TYPE_SINT64: ('int64_t', 'PB_LTYPE_SVARINT'),
-    FieldD.TYPE_UINT32: ('uint32_t', 'PB_LTYPE_VARINT'),
-    FieldD.TYPE_UINT64: ('uint64_t', 'PB_LTYPE_VARINT')
+    FieldD.TYPE_BOOL:       ('bool',     'BOOL'),
+    FieldD.TYPE_DOUBLE:     ('double',   'DOUBLE'),
+    FieldD.TYPE_FIXED32:    ('uint32_t', 'FIXED32'),
+    FieldD.TYPE_FIXED64:    ('uint64_t', 'FIXED64'),
+    FieldD.TYPE_FLOAT:      ('float',    'FLOAT'),
+    FieldD.TYPE_INT32:      ('int32_t',  'INT32'),
+    FieldD.TYPE_INT64:      ('int64_t',  'INT64'),
+    FieldD.TYPE_SFIXED32:   ('int32_t',  'SFIXED32'),
+    FieldD.TYPE_SFIXED64:   ('int64_t',  'SFIXED64'),
+    FieldD.TYPE_SINT32:     ('int32_t',  'SINT32'),
+    FieldD.TYPE_SINT64:     ('int64_t',  'SINT64'),
+    FieldD.TYPE_UINT32:     ('uint32_t', 'UINT32'),
+    FieldD.TYPE_UINT64:     ('uint64_t', 'UINT64')
 }
 
 class Names:
-    '''Keeps a set of nested names and formats them to C identifier.
-    You can subclass this with your own implementation.
-    '''
+    '''Keeps a set of nested names and formats them to C identifier.'''
     def __init__(self, parts = ()):
         if isinstance(parts, Names):
             parts = parts.parts
@@ -62,6 +72,9 @@ class Names:
         else:
             raise ValueError("Name parts should be of type str")
     
+    def __eq__(self, other):
+        return isinstance(other, Names) and self.parts == other.parts
+    
 def names_from_type_name(type_name):
     '''Parse Names() from FieldDescriptorProto type_name'''
     if type_name[0] != '.':
@@ -69,19 +82,27 @@ def names_from_type_name(type_name):
     return Names(type_name[1:].split('.'))
 
 class Enum:
-    def __init__(self, names, desc):
+    def __init__(self, names, desc, enum_options):
         '''desc is EnumDescriptorProto'''
+        
+        self.options = enum_options
         self.names = names + desc.name
+        
+        if enum_options.long_names:
             self.values = [(self.names + x.name, x.number) for x in desc.value]            
+        else:
+            self.values = [(names + x.name, x.number) for x in desc.value] 
+        
+        self.value_longnames = [self.names + x.name for x in desc.value]
     
     def __str__(self):
-        result = 'typedef enum {\n'
+        result = 'typedef enum _%s {\n' % self.names
         result += ',\n'.join(["    %s = %d" % x for x in self.values])
         result += '\n} %s;' % self.names
         return result
 
 class Field:
-    def __init__(self, struct_name, desc):
+    def __init__(self, struct_name, desc, field_options):
         '''desc is FieldDescriptorProto'''
         self.tag = desc.number
         self.struct_name = struct_name
@@ -91,76 +112,83 @@ class Field:
         self.max_count = None
         self.array_decl = ""
         
-        # Parse nanopb-specific field options
-        if desc.options.HasExtension(nanopb_pb2.nanopb):
-            ext = desc.options.Extensions[nanopb_pb2.nanopb]
-            if ext.HasField("max_size"):
-                self.max_size = ext.max_size
-            if ext.HasField("max_count"):
-                self.max_count = ext.max_count
+        # Parse field options
+        if field_options.HasField("max_size"):
+            self.max_size = field_options.max_size
+        
+        if field_options.HasField("max_count"):
+            self.max_count = field_options.max_count
         
         if desc.HasField('default_value'):
             self.default = desc.default_value
            
-        # Decide HTYPE
-        # HTYPE is the high-order nibble of nanopb field description,
-        # defining whether value is required/optional/repeated.
-        is_callback = False
+        # Check field rules, i.e. required/optional/repeated.
+        can_be_static = True
         if desc.label == FieldD.LABEL_REQUIRED:
-            self.htype = 'PB_HTYPE_REQUIRED'
+            self.rules = 'REQUIRED'
         elif desc.label == FieldD.LABEL_OPTIONAL:
-            self.htype = 'PB_HTYPE_OPTIONAL'
+            self.rules = 'OPTIONAL'
         elif desc.label == FieldD.LABEL_REPEATED:
+            self.rules = 'REPEATED'
             if self.max_count is None:
-                is_callback = True
+                can_be_static = False
             else:
-                self.htype = 'PB_HTYPE_ARRAY'
                 self.array_decl = '[%d]' % self.max_count
         else:
             raise NotImplementedError(desc.label)
         
-        # Decide LTYPE and CTYPE
-        # LTYPE is the low-order nibble of nanopb field description,
-        # defining how to decode an individual value.
-        # CTYPE is the name of the c type to use in the struct.
+        # Decide the C data type to use in the struct.
         if datatypes.has_key(desc.type):
-            self.ctype, self.ltype = datatypes[desc.type]
+            self.ctype, self.pbtype = datatypes[desc.type]
         elif desc.type == FieldD.TYPE_ENUM:
-            self.ltype = 'PB_LTYPE_VARINT'
+            self.pbtype = 'ENUM'
             self.ctype = names_from_type_name(desc.type_name)
             if self.default is not None:
                 self.default = self.ctype + self.default
         elif desc.type == FieldD.TYPE_STRING:
-            self.ltype = 'PB_LTYPE_STRING'
+            self.pbtype = 'STRING'
             if self.max_size is None:
-                is_callback = True
+                can_be_static = False
             else:
                 self.ctype = 'char'
                 self.array_decl += '[%d]' % self.max_size
         elif desc.type == FieldD.TYPE_BYTES:
-            self.ltype = 'PB_LTYPE_BYTES'
+            self.pbtype = 'BYTES'
             if self.max_size is None:
-                is_callback = True
+                can_be_static = False
             else:
                 self.ctype = self.struct_name + self.name + 't'
         elif desc.type == FieldD.TYPE_MESSAGE:
-            self.ltype = 'PB_LTYPE_SUBMESSAGE'
+            self.pbtype = 'MESSAGE'
             self.ctype = self.submsgname = names_from_type_name(desc.type_name)
         else:
             raise NotImplementedError(desc.type)
         
-        if is_callback:
-            self.htype = 'PB_HTYPE_CALLBACK'
+        if field_options.type == nanopb_pb2.FT_DEFAULT:
+            if can_be_static:
+                field_options.type = nanopb_pb2.FT_STATIC
+            else:
+                field_options.type = nanopb_pb2.FT_CALLBACK
+        
+        if field_options.type == nanopb_pb2.FT_STATIC and not can_be_static:
+            raise Exception("Field %s is defined as static, but max_size or max_count is not given." % self.name)
+        
+        if field_options.type == nanopb_pb2.FT_STATIC:
+            self.allocation = 'STATIC'
+        elif field_options.type == nanopb_pb2.FT_CALLBACK:
+            self.allocation = 'CALLBACK'
             self.ctype = 'pb_callback_t'
             self.array_decl = ''
+        else:
+            raise NotImplementedError(field_options.type)
     
     def __cmp__(self, other):
         return cmp(self.tag, other.tag)
     
     def __str__(self):
-        if self.htype == 'PB_HTYPE_OPTIONAL':
+        if self.rules == 'OPTIONAL' and self.allocation == 'STATIC':
             result = '    bool has_' + self.name + ';\n'
-        elif self.htype == 'PB_HTYPE_ARRAY':
+        elif self.rules == 'REPEATED' and self.allocation == 'STATIC':
             result = '    size_t ' + self.name + '_count;\n'
         else:
             result = ''
@@ -169,7 +197,7 @@ class Field:
     
     def types(self):
         '''Return definitions for any special types this field might need.'''
-        if self.ltype == 'PB_LTYPE_BYTES' and self.max_size is not None:
+        if self.pbtype == 'BYTES' and self.allocation == 'STATIC':
             result = 'typedef struct {\n'
             result += '    size_t size;\n'
             result += '    uint8_t bytes[%d];\n' % self.max_size
@@ -183,80 +211,63 @@ class Field:
         if self.default is None:
             return None
 
-        if self.ltype == 'PB_LTYPE_STRING':
-            ctype = 'char'
-            if self.max_size is None:
+        ctype, default = self.ctype, self.default
+        array_decl = ''
+        
+        if self.pbtype == 'STRING':
+            if self.allocation != 'STATIC':
                 return None # Not implemented
-            else:
-                array_decl = '[%d]' % (self.max_size + 1)
+        
+            array_decl = '[%d]' % self.max_size
             default = str(self.default).encode('string_escape')
             default = default.replace('"', '\\"')
             default = '"' + default + '"'
-        elif self.ltype == 'PB_LTYPE_BYTES':
+        elif self.pbtype == 'BYTES':
+            if self.allocation != 'STATIC':
+                return None # Not implemented
+
             data = self.default.decode('string_escape')
             data = ['0x%02x' % ord(c) for c in data]
-            
-            if self.max_size is None:
-                return None # Not implemented
-            else:
-                ctype = self.ctype
-            
             default = '{%d, {%s}}' % (len(data), ','.join(data))
-            array_decl = ''
-        else:
-            ctype, default = self.ctype, self.default
-            array_decl = ''
         
         if declaration_only:
             return 'extern const %s %s_default%s;' % (ctype, self.struct_name + self.name, array_decl)
         else:
             return 'const %s %s_default%s = %s;' % (ctype, self.struct_name + self.name, array_decl, default)
     
+    def tags(self):
+        '''Return the #define for the tag number of this field.'''
+        identifier = '%s_%s_tag' % (self.struct_name, self.name)
+        return '#define %-40s %d\n' % (identifier, self.tag)
+    
     def pb_field_t(self, prev_field_name):
         '''Return the pb_field_t initializer to use in the constant array.
         prev_field_name is the name of the previous field or None.
         '''
-        result = '    {%d, ' % self.tag
-        result += '(pb_type_t) ((int) ' + self.htype
-        if self.ltype is not None:
-            result += ' | (int) ' + self.ltype
-        result += '),\n'
+        result  = '    PB_FIELD(%3d, ' % self.tag
+        result += '%-8s, ' % self.pbtype
+        result += '%s, ' % self.rules
+        result += '%s, ' % self.allocation
+        result += '%s, ' % self.struct_name
+        result += '%s, ' % self.name
+        result += '%s, ' % (prev_field_name or self.name)
         
-        if prev_field_name is None:
-            result += '    offsetof(%s, %s),' % (self.struct_name, self.name)
+        if self.pbtype == 'MESSAGE':
+            result += '&%s_fields)' % self.submsgname
+        elif self.default is None:
+            result += '0)'
+        elif self.pbtype in ['BYTES', 'STRING'] and self.allocation != 'STATIC':
+            result += '0)' # Arbitrary size default values not implemented
         else:
-            result += '    pb_delta_end(%s, %s, %s),' % (self.struct_name, self.name, prev_field_name)
-        
-        if self.htype == 'PB_HTYPE_OPTIONAL':
-            result += '\n    pb_delta(%s, has_%s, %s),' % (self.struct_name, self.name, self.name)
-        elif self.htype == 'PB_HTYPE_ARRAY':
-            result += '\n    pb_delta(%s, %s_count, %s),' % (self.struct_name, self.name, self.name)
-        else:
-            result += ' 0,'
-        
-        
-        if self.htype == 'PB_HTYPE_ARRAY':
-            result += '\n    pb_membersize(%s, %s[0]),' % (self.struct_name, self.name)
-            result += ('\n    pb_membersize(%s, %s) / pb_membersize(%s, %s[0]),'
-                       % (self.struct_name, self.name, self.struct_name, self.name))
-        else:
-            result += '\n    pb_membersize(%s, %s),' % (self.struct_name, self.name)
-            result += ' 0,'
-        
-        if self.ltype == 'PB_LTYPE_SUBMESSAGE':
-            result += '\n    &%s_fields}' % self.submsgname
-        elif self.default is None or self.htype == 'PB_HTYPE_CALLBACK':
-            result += ' 0}'
-        else:
-            result += '\n    &%s_default}' % (self.struct_name + self.name)
+            result += '&%s_default)' % (self.struct_name + self.name)
         
         return result
     
     def largest_field_value(self):
         '''Determine if this field needs 16bit or 32bit pb_field_t structure to compile properly.
         Returns numeric value or a C-expression for assert.'''
-        if self.ltype == 'PB_LTYPE_SUBMESSAGE':
-            if self.htype == 'PB_HTYPE_ARRAY':
+        if self.pbtype == 'MESSAGE':
+            if self.rules == 'REPEATED' and self.allocation == 'STATIC':
                 return 'pb_membersize(%s, %s[0])' % (self.struct_name, self.name)
             else:
                 return 'pb_membersize(%s, %s)' % (self.struct_name, self.name)
@@ -264,10 +275,88 @@ class Field:
         return max(self.tag, self.max_size, self.max_count)        
 
 
+class ExtensionRange(Field):
+    def __init__(self, struct_name, range_start, field_options):
+        '''Implements a special pb_extension_t* field in an extensible message
+        structure. The range_start signifies the index at which the extensions
+        start. Not necessarily all tags above this are extensions, it is merely
+        a speed optimization.
+        '''
+        self.tag = range_start
+        self.struct_name = struct_name
+        self.name = 'extensions'
+        self.pbtype = 'EXTENSION'
+        self.rules = 'OPTIONAL'
+        self.allocation = 'CALLBACK'
+        self.ctype = 'pb_extension_t'
+        self.array_decl = ''
+        self.default = None
+        self.max_size = 0
+        self.max_count = 0
+        
+    def __str__(self):
+        return '    pb_extension_t *extensions;'
+    
+    def types(self):
+        return None
+    
+    def tags(self):
+        return ''
+
+class ExtensionField(Field):
+    def __init__(self, struct_name, desc, field_options):
+        self.fullname = struct_name + desc.name
+        self.extendee_name = names_from_type_name(desc.extendee)
+        Field.__init__(self, self.fullname + 'struct', desc, field_options)
+        
+        if self.rules != 'OPTIONAL':
+            raise NotImplementedError("Only 'optional' is supported for extension fields. "
+               + "(%s.rules == %s)" % (self.fullname, self.rules))
+
+        self.rules = 'OPTEXT'
+
+    def extension_decl(self):
+        '''Declaration of the extension type in the .pb.h file'''
+        return 'extern const pb_extension_type_t %s;\n' % self.fullname
+
+    def extension_def(self):
+        '''Definition of the extension type in the .pb.c file'''
+
+        result  = 'typedef struct {\n'
+        result += str(self)
+        result += '\n} %s;\n\n' % self.struct_name
+        result += ('static const pb_field_t %s_field = \n  %s;\n\n' %
+                    (self.fullname, self.pb_field_t(None)))
+        result += 'const pb_extension_type_t %s = {\n' % self.fullname
+        result += '    NULL,\n'
+        result += '    NULL,\n'
+        result += '    &%s_field\n' % self.fullname
+        result += '};\n'
+        return result
+
+
+# ---------------------------------------------------------------------------
+#                   Generation of messages (structures)
+# ---------------------------------------------------------------------------
+
+
 class Message:
-    def __init__(self, names, desc):
+    def __init__(self, names, desc, message_options):
         self.name = names
-        self.fields = [Field(self.name, f) for f in desc.field]
+        self.fields = []
+        
+        for f in desc.field:
+            field_options = get_nanopb_suboptions(f, message_options, self.name + f.name)
+            if field_options.type != nanopb_pb2.FT_IGNORE:
+                self.fields.append(Field(self.name, f, field_options))
+        
+        if len(desc.extension_range) > 0:
+            field_options = get_nanopb_suboptions(desc, message_options, self.name + 'extensions')
+            range_start = min([r.start for r in desc.extension_range])
+            if field_options.type != nanopb_pb2.FT_IGNORE:
+                self.fields.append(ExtensionRange(self.name, range_start, field_options))
+        
+        self.packed = message_options.packed_struct
         self.ordered_fields = self.fields[:]
         self.ordered_fields.sort()
 
@@ -276,9 +365,25 @@ class Message:
         return [str(field.ctype) for field in self.fields]
     
     def __str__(self):
-        result = 'typedef struct {\n'
+        result = 'typedef struct _%s {\n' % self.name
+
+        if not self.ordered_fields:
+            # Empty structs are not allowed in C standard.
+            # Therefore add a dummy field if an empty message occurs.
+            result += '    uint8_t dummy_field;'
+
         result += '\n'.join([str(f) for f in self.ordered_fields])
-        result += '\n} %s;' % self.name
+        result += '\n}'
+        
+        if self.packed:
+            result += ' pb_packed'
+        
+        result += ' %s;' % self.name
+        
+        if self.packed:
+            result = 'PB_PACKED_STRUCT_START\n' + result
+            result += '\nPB_PACKED_STRUCT_END'
+        
         return result
     
     def types(self):
@@ -307,12 +412,19 @@ class Message:
         prev = None
         for field in self.ordered_fields:
             result += field.pb_field_t(prev)
-            result += ',\n\n'
+            result += ',\n'
             prev = field.name
         
         result += '    PB_LAST_FIELD\n};'
         return result
 
+
+
+# ---------------------------------------------------------------------------
+#                    Processing of entire .proto files
+# ---------------------------------------------------------------------------
+
+
 def iterate_messages(desc, names = Names()):
     '''Recursively find all messages. For each, yield name, DescriptorProto.'''
     if hasattr(desc, 'message_type'):
@@ -327,11 +439,23 @@ def iterate_messages(desc, names = Names()):
         for x in iterate_messages(submsg, sub_names):
             yield x
 
-def parse_file(fdesc):
-    '''Takes a FileDescriptorProto and returns tuple (enum, messages).'''
+def iterate_extensions(desc, names = Names()):
+    '''Recursively find all extensions.
+    For each, yield name, FieldDescriptorProto.
+    '''
+    for extension in desc.extension:
+        yield names, extension
+
+    for subname, subdesc in iterate_messages(desc, names):
+        for extension in subdesc.extension:
+            yield subname, extension
+
+def parse_file(fdesc, file_options):
+    '''Takes a FileDescriptorProto and returns tuple (enums, messages, extensions).'''
     
     enums = []
     messages = []
+    extensions = []
     
     if fdesc.package:
         base_name = Names(fdesc.package.split('.'))
@@ -339,14 +463,30 @@ def parse_file(fdesc):
         base_name = Names()
     
     for enum in fdesc.enum_type:
-        enums.append(Enum(base_name, enum))
+        enum_options = get_nanopb_suboptions(enum, file_options, base_name + enum.name)
+        enums.append(Enum(base_name, enum, enum_options))
     
     for names, message in iterate_messages(fdesc, base_name):
-        messages.append(Message(names, message))
+        message_options = get_nanopb_suboptions(message, file_options, names)
+        messages.append(Message(names, message, message_options))
         for enum in message.enum_type:
-            enums.append(Enum(names, enum))
+            enum_options = get_nanopb_suboptions(enum, message_options, names + enum.name)
+            enums.append(Enum(names, enum, enum_options))
     
-    return enums, messages
+    for names, extension in iterate_extensions(fdesc, base_name):
+        field_options = get_nanopb_suboptions(extension, file_options, names)
+        extensions.append(ExtensionField(names, extension, field_options))
+    
+    # Fix field default values where enum short names are used.
+    for enum in enums:
+        if not enum.options.long_names:
+            for message in messages:
+                for field in message.fields:
+                    if field.default in enum.value_longnames:
+                        idx = enum.value_longnames.index(field.default)
+                        field.default = enum.values[idx][0]
+    
+    return enums, messages, extensions
 
 def toposort2(data):
     '''Topological sort.
@@ -379,23 +519,43 @@ def sort_dependencies(messages):
         if msgname in message_by_name:
             yield message_by_name[msgname]
 
-def generate_header(dependencies, headername, enums, messages):
+def make_identifier(headername):
+    '''Make #ifndef identifier that contains uppercase A-Z and digits 0-9'''
+    result = ""
+    for c in headername.upper():
+        if c.isalnum():
+            result += c
+        else:
+            result += '_'
+    return result
+
+def generate_header(dependencies, headername, enums, messages, extensions, options):
     '''Generate content for a header file.
     Generates strings, which should be concatenated and stored to file.
     '''
     
     yield '/* Automatically generated nanopb header */\n'
+    yield '/* Generated by %s at %s. */\n\n' % (nanopb_version, time.asctime())
     
-    symbol = headername.replace('.', '_').upper()
+    symbol = make_identifier(headername)
     yield '#ifndef _PB_%s_\n' % symbol
     yield '#define _PB_%s_\n' % symbol
-    yield '#include <pb.h>\n\n'
+    try:
+        yield options.libformat % ('pb.h')
+    except TypeError:
+        # no %s specified - use whatever was passed in as options.libformat
+        yield options.libformat
+    yield '\n'
     
     for dependency in dependencies:
         noext = os.path.splitext(dependency)[0]
-        yield '#include "%s.pb.h"\n' % noext
+        yield options.genformat % (noext + '.' + options.extension + '.h')
         yield '\n'
 
+    yield '#ifdef __cplusplus\n'
+    yield 'extern "C" {\n'
+    yield '#endif\n\n'
+    
     yield '/* Enum definitions */\n'
     for enum in enums:
         yield str(enum) + '\n\n'
@@ -405,17 +565,56 @@ def generate_header(dependencies, headername, enums, messages):
         yield msg.types()
         yield str(msg) + '\n\n'
     
+    if extensions:
+        yield '/* Extensions */\n'
+        for extension in extensions:
+            yield extension.extension_decl()
+        yield '\n'
+        
     yield '/* Default values for struct fields */\n'
     for msg in messages:
         yield msg.default_decl(True)
     yield '\n'
     
+    yield '/* Field tags (for use in manual encoding/decoding) */\n'
+    for msg in sort_dependencies(messages):
+        for field in msg.fields:
+            yield field.tags()
+    yield '\n'
+    
     yield '/* Struct field encoding specification for nanopb */\n'
     for msg in messages:
         yield msg.fields_declaration() + '\n'
     
+    yield '\n#ifdef __cplusplus\n'
+    yield '} /* extern "C" */\n'
+    yield '#endif\n'
+    
+    # End of header
+    yield '\n#endif\n'
+
+def generate_source(headername, enums, messages, extensions):
+    '''Generate content for a source file.'''
+    
+    yield '/* Automatically generated nanopb constant definitions */\n'
+    yield '/* Generated by %s at %s. */\n\n' % (nanopb_version, time.asctime())
+    yield options.genformat % (headername)
+    yield '\n'
+    
+    for msg in messages:
+        yield msg.default_decl(False)
+    
+    yield '\n\n'
+    
+    for msg in messages:
+        yield msg.fields_definition() + '\n\n'
+    
+    for ext in extensions:
+        yield ext.extension_def() + '\n'
+        
+    # Add checks for numeric limits
     if messages:
-        count_required_fields = lambda m: len([f for f in msg.fields if f.htype == 'PB_HTYPE_REQUIRED'])
+        count_required_fields = lambda m: len([f for f in msg.fields if f.rules == 'REQUIRED'])
         largest_msg = max(messages, key = count_required_fields)
         largest_count = count_required_fields(largest_msg)
         if largest_count > 64:
@@ -428,7 +627,9 @@ def generate_header(dependencies, headername, enums, messages):
     worst = 0
     worst_field = ''
     checks = []
+    checks_msgnames = []
     for msg in messages:
+        checks_msgnames.append(msg.name)
         for field in msg.fields:
             status = field.largest_field_value()
             if isinstance(status, (str, unicode)):
@@ -446,7 +647,8 @@ def generate_header(dependencies, headername, enums, messages):
                 yield '#error Field descriptor for %s is too large. Define PB_FIELD_16BIT to fix this.\n' % worst_field
             else:
                 assertion = ' && '.join(str(c) + ' < 256' for c in checks)
-                yield 'STATIC_ASSERT((%s), YOU_MUST_DEFINE_PB_FIELD_16BIT)\n' % assertion
+                msgs = '_'.join(str(n) for n in checks_msgnames)
+                yield 'STATIC_ASSERT((%s), YOU_MUST_DEFINE_PB_FIELD_16BIT_FOR_MESSAGES_%s)\n'%(assertion,msgs)
             yield '#endif\n\n'
         
         if worst > 65535 or checks:
@@ -455,59 +657,187 @@ def generate_header(dependencies, headername, enums, messages):
                 yield '#error Field descriptor for %s is too large. Define PB_FIELD_32BIT to fix this.\n' % worst_field
             else:
                 assertion = ' && '.join(str(c) + ' < 65536' for c in checks)
-                yield 'STATIC_ASSERT((%s), YOU_MUST_DEFINE_PB_FIELD_32BIT)\n' % assertion
+                msgs = '_'.join(str(n) for n in checks_msgnames)
+                yield 'STATIC_ASSERT((%s), YOU_MUST_DEFINE_PB_FIELD_32BIT_FOR_MESSAGES_%s)\n'%(assertion,msgs)
             yield '#endif\n'
     
-    # End of header
-    yield '\n#endif\n'
-
-def generate_source(headername, enums, messages):
-    '''Generate content for a source file.'''
-    
-    yield '/* Automatically generated nanopb constant definitions */\n'
-    yield '#include "%s"\n\n' % headername
-    
+    # Add check for sizeof(double)
+    has_double = False
     for msg in messages:
-        yield msg.default_decl(False)
+        for field in msg.fields:
+            if field.ctype == 'double':
+                has_double = True
     
-    yield '\n\n'
+    if has_double:
+        yield '\n'
+        yield '/* On some platforms (such as AVR), double is really float.\n'
+        yield ' * These are not directly supported by nanopb, but see example_avr_double.\n'
+        yield ' * To get rid of this error, remove any double fields from your .proto.\n'
+        yield ' */\n'
+        yield 'STATIC_ASSERT(sizeof(double) == 8, DOUBLE_MUST_BE_8_BYTES)\n'
     
-    for msg in messages:
-        yield msg.fields_definition() + '\n\n'
+    yield '\n'
+
+# ---------------------------------------------------------------------------
+#                    Options parsing for the .proto files
+# ---------------------------------------------------------------------------
+
+from fnmatch import fnmatch
+
+def read_options_file(infile):
+    '''Parse a separate options file to list:
+        [(namemask, options), ...]
+    '''
+    results = []
+    for line in infile:
+        line = line.strip()
+        if not line or line.startswith('//') or line.startswith('#'):
+            continue
+        
+        parts = line.split(None, 1)
+        opts = nanopb_pb2.NanoPBOptions()
+        text_format.Merge(parts[1], opts)
+        results.append((parts[0], opts))
+
+    return results
+
+class Globals:
+    '''Ugly global variables, should find a good way to pass these.'''
+    verbose_options = False
+    separate_options = []
+
+def get_nanopb_suboptions(subdesc, options, name):
+    '''Get copy of options, and merge information from subdesc.'''
+    new_options = nanopb_pb2.NanoPBOptions()
+    new_options.CopyFrom(options)
+    
+    # Handle options defined in a separate file
+    dotname = '.'.join(name.parts)
+    for namemask, options in Globals.separate_options:
+        if fnmatch(dotname, namemask):
+            new_options.MergeFrom(options)
+    
+    # Handle options defined in .proto
+    if isinstance(subdesc.options, descriptor.FieldOptions):
+        ext_type = nanopb_pb2.nanopb
+    elif isinstance(subdesc.options, descriptor.FileOptions):
+        ext_type = nanopb_pb2.nanopb_fileopt
+    elif isinstance(subdesc.options, descriptor.MessageOptions):
+        ext_type = nanopb_pb2.nanopb_msgopt
+    elif isinstance(subdesc.options, descriptor.EnumOptions):
+        ext_type = nanopb_pb2.nanopb_enumopt
+    else:
+        raise Exception("Unknown options type")
+    
+    if subdesc.options.HasExtension(ext_type):
+        ext = subdesc.options.Extensions[ext_type]
+        new_options.MergeFrom(ext)
+    
+    if Globals.verbose_options:
+        print "Options for " + dotname + ":"
+        print text_format.MessageToString(new_options)
+    
+    return new_options
+
+
+# ---------------------------------------------------------------------------
+#                         Command line interface
+# ---------------------------------------------------------------------------
 
-if __name__ == '__main__':
 import sys
 import os.path    
+from optparse import OptionParser
 
-    if len(sys.argv) != 2:
-        print "Usage: " + sys.argv[0] + " file.pb"
-        print "where file.pb has been compiled from .proto by:"
-        print "protoc -ofile.pb file.proto"
-        print "Output fill be written to file.pb.h and file.pb.c"
-        sys.exit(1)
+optparser = OptionParser(
+    usage = "Usage: nanopb_generator.py [options] file.pb ...",
+    epilog = "Compile file.pb from file.proto by: 'protoc -ofile.pb file.proto'. " +
+             "Output will be written to file.pb.h and file.pb.c.")
+optparser.add_option("-x", dest="exclude", metavar="FILE", action="append", default=[],
+    help="Exclude file from generated #include list.")
+optparser.add_option("-e", "--extension", dest="extension", metavar="EXTENSION", default="pb",
+    help="Set extension to use instead of 'pb' for generated files. [default: %default]")
+optparser.add_option("-f", "--options-file", dest="options_file", metavar="FILE", default="%s.options",
+    help="Set name of a separate generator options file.")
+optparser.add_option("-Q", "--generated-include-format", dest="genformat",
+    metavar="FORMAT", default='#include "%s"\n',
+    help="Set format string to use for including other .pb.h files. [default: %default]")
+optparser.add_option("-L", "--library-include-format", dest="libformat",
+    metavar="FORMAT", default='#include <%s>\n',
+    help="Set format string to use for including the nanopb pb.h header. [default: %default]")
+optparser.add_option("-q", "--quiet", dest="quiet", action="store_true", default=False,
+    help="Don't print anything except errors.")
+optparser.add_option("-v", "--verbose", dest="verbose", action="store_true", default=False,
+    help="Print more information.")
+optparser.add_option("-s", dest="settings", metavar="OPTION:VALUE", action="append", default=[],
+    help="Set generator option (max_size, max_count etc.).")
 
-    data = open(sys.argv[1], 'rb').read()
+def process(filenames, options):
+    '''Process the files given on the command line.'''
+    
+    if not filenames:
+        optparser.print_help()
+        return False
+    
+    if options.quiet:
+        options.verbose = False
+
+    Globals.verbose_options = options.verbose
+    
+    toplevel_options = nanopb_pb2.NanoPBOptions()
+    for s in options.settings:
+        text_format.Merge(s, toplevel_options)
+    
+    for filename in filenames:
+        data = open(filename, 'rb').read()
         fdesc = descriptor.FileDescriptorSet.FromString(data)
-    enums, messages = parse_file(fdesc.file[0])
         
-    noext = os.path.splitext(sys.argv[1])[0]
-    headername = noext + '.pb.h'
-    sourcename = noext + '.pb.c'
+        # Check if any separate options are specified
+        try:
+            optfilename = options.options_file % os.path.splitext(filename)[0]
+        except TypeError:
+            # No %s specified, use the filename as-is
+            optfilename = options.options_file
+        
+        if options.verbose:
+            print 'Reading options from ' + optfilename
+        
+        if os.path.isfile(optfilename):
+            Globals.separate_options = read_options_file(open(optfilename, "rU"))
+        else:
+            Globals.separate_options = []
+        
+        # Parse the file
+        file_options = get_nanopb_suboptions(fdesc.file[0], toplevel_options, Names([filename]))
+        enums, messages, extensions = parse_file(fdesc.file[0], file_options)
+        
+        noext = os.path.splitext(filename)[0]
+        headername = noext + '.' + options.extension + '.h'
+        sourcename = noext + '.' + options.extension + '.c'
         headerbasename = os.path.basename(headername)
         
+        if not options.quiet:
             print "Writing to " + headername + " and " + sourcename
         
         # List of .proto files that should not be included in the C header file
         # even if they are mentioned in the source .proto.
-    excludes = ['nanopb.proto', 'google/protobuf/descriptor.proto']
+        excludes = ['nanopb.proto', 'google/protobuf/descriptor.proto'] + options.exclude
         dependencies = [d for d in fdesc.file[0].dependency if d not in excludes]
         
         header = open(headername, 'w')
-    for part in generate_header(dependencies, headerbasename, enums, messages):
+        for part in generate_header(dependencies, headerbasename, enums,
+                                    messages, extensions, options):
             header.write(part)
 
         source = open(sourcename, 'w')
-    for part in generate_source(headerbasename, enums, messages):
+        for part in generate_source(headerbasename, enums, messages, extensions):
             source.write(part)
 
+    return True
+
+if __name__ == '__main__':
+    options, filenames = optparser.parse_args()
+    status = process(filenames, options)
+    
+    if not status:
+        sys.exit(1)
     

generator/nanopb_pb2.py

@@ -7,15 +7,75 @@ from google.protobuf import descriptor_pb2
 # @@protoc_insertion_point(imports)
 
 
+import google.protobuf.descriptor_pb2
+
 DESCRIPTOR = descriptor.FileDescriptor(
   name='nanopb.proto',
   package='',
-  serialized_pb='\n\x0cnanopb.proto\x1a google/protobuf/descriptor.proto\"4\n\rNanoPBOptions\x12\x10\n\x08max_size\x18\x01 \x01(\x05\x12\x11\n\tmax_count\x18\x02 \x01(\x05:>\n\x06nanopb\x12\x1d.google.protobuf.FieldOptions\x18\xf2\x07 \x01(\x0b\x32\x0e.NanoPBOptions')
+  serialized_pb='\n\x0cnanopb.proto\x1a google/protobuf/descriptor.proto\"\x92\x01\n\rNanoPBOptions\x12\x10\n\x08max_size\x18\x01 \x01(\x05\x12\x11\n\tmax_count\x18\x02 \x01(\x05\x12$\n\x04type\x18\x03 \x01(\x0e\x32\n.FieldType:\nFT_DEFAULT\x12\x18\n\nlong_names\x18\x04 \x01(\x08:\x04true\x12\x1c\n\rpacked_struct\x18\x05 \x01(\x08:\x05\x66\x61lse*J\n\tFieldType\x12\x0e\n\nFT_DEFAULT\x10\x00\x12\x0f\n\x0b\x46T_CALLBACK\x10\x01\x12\r\n\tFT_STATIC\x10\x02\x12\r\n\tFT_IGNORE\x10\x03:E\n\x0enanopb_fileopt\x12\x1c.google.protobuf.FileOptions\x18\xf2\x07 \x01(\x0b\x32\x0e.NanoPBOptions:G\n\rnanopb_msgopt\x12\x1f.google.protobuf.MessageOptions\x18\xf2\x07 \x01(\x0b\x32\x0e.NanoPBOptions:E\n\x0enanopb_enumopt\x12\x1c.google.protobuf.EnumOptions\x18\xf2\x07 \x01(\x0b\x32\x0e.NanoPBOptions:>\n\x06nanopb\x12\x1d.google.protobuf.FieldOptions\x18\xf2\x07 \x01(\x0b\x32\x0e.NanoPBOptions')
+
+_FIELDTYPE = descriptor.EnumDescriptor(
+  name='FieldType',
+  full_name='FieldType',
+  filename=None,
+  file=DESCRIPTOR,
+  values=[
+    descriptor.EnumValueDescriptor(
+      name='FT_DEFAULT', index=0, number=0,
+      options=None,
+      type=None),
+    descriptor.EnumValueDescriptor(
+      name='FT_CALLBACK', index=1, number=1,
+      options=None,
+      type=None),
+    descriptor.EnumValueDescriptor(
+      name='FT_STATIC', index=2, number=2,
+      options=None,
+      type=None),
+    descriptor.EnumValueDescriptor(
+      name='FT_IGNORE', index=3, number=3,
+      options=None,
+      type=None),
+  ],
+  containing_type=None,
+  options=None,
+  serialized_start=199,
+  serialized_end=273,
+)
 
 
+FT_DEFAULT = 0
+FT_CALLBACK = 1
+FT_STATIC = 2
+FT_IGNORE = 3
+
+NANOPB_FILEOPT_FIELD_NUMBER = 1010
+nanopb_fileopt = descriptor.FieldDescriptor(
+  name='nanopb_fileopt', full_name='nanopb_fileopt', index=0,
+  number=1010, type=11, cpp_type=10, label=1,
+  has_default_value=False, default_value=None,
+  message_type=None, enum_type=None, containing_type=None,
+  is_extension=True, extension_scope=None,
+  options=None)
+NANOPB_MSGOPT_FIELD_NUMBER = 1010
+nanopb_msgopt = descriptor.FieldDescriptor(
+  name='nanopb_msgopt', full_name='nanopb_msgopt', index=1,
+  number=1010, type=11, cpp_type=10, label=1,
+  has_default_value=False, default_value=None,
+  message_type=None, enum_type=None, containing_type=None,
+  is_extension=True, extension_scope=None,
+  options=None)
+NANOPB_ENUMOPT_FIELD_NUMBER = 1010
+nanopb_enumopt = descriptor.FieldDescriptor(
+  name='nanopb_enumopt', full_name='nanopb_enumopt', index=2,
+  number=1010, type=11, cpp_type=10, label=1,
+  has_default_value=False, default_value=None,
+  message_type=None, enum_type=None, containing_type=None,
+  is_extension=True, extension_scope=None,
+  options=None)
 NANOPB_FIELD_NUMBER = 1010
 nanopb = descriptor.FieldDescriptor(
-  name='nanopb', full_name='nanopb', index=0,
+  name='nanopb', full_name='nanopb', index=3,
   number=1010, type=11, cpp_type=10, label=1,
   has_default_value=False, default_value=None,
   message_type=None, enum_type=None, containing_type=None,
@@ -44,6 +104,27 @@ _NANOPBOPTIONS = descriptor.Descriptor(
       message_type=None, enum_type=None, containing_type=None,
       is_extension=False, extension_scope=None,
       options=None),
+    descriptor.FieldDescriptor(
+      name='type', full_name='NanoPBOptions.type', index=2,
+      number=3, type=14, cpp_type=8, label=1,
+      has_default_value=True, default_value=0,
+      message_type=None, enum_type=None, containing_type=None,
+      is_extension=False, extension_scope=None,
+      options=None),
+    descriptor.FieldDescriptor(
+      name='long_names', full_name='NanoPBOptions.long_names', index=3,
+      number=4, type=8, cpp_type=7, label=1,
+      has_default_value=True, default_value=True,
+      message_type=None, enum_type=None, containing_type=None,
+      is_extension=False, extension_scope=None,
+      options=None),
+    descriptor.FieldDescriptor(
+      name='packed_struct', full_name='NanoPBOptions.packed_struct', index=4,
+      number=5, type=8, cpp_type=7, label=1,
+      has_default_value=True, default_value=False,
+      message_type=None, enum_type=None, containing_type=None,
+      is_extension=False, extension_scope=None,
+      options=None),
   ],
   extensions=[
   ],
@@ -53,12 +134,12 @@ _NANOPBOPTIONS = descriptor.Descriptor(
   options=None,
   is_extendable=False,
   extension_ranges=[],
-  serialized_start=50,
-  serialized_end=102,
+  serialized_start=51,
+  serialized_end=197,
 )
 
-import google.protobuf.descriptor_pb2
-
+_NANOPBOPTIONS.fields_by_name['type'].enum_type = _FIELDTYPE
+DESCRIPTOR.message_types_by_name['NanoPBOptions'] = _NANOPBOPTIONS
 
 class NanoPBOptions(message.Message):
   __metaclass__ = reflection.GeneratedProtocolMessageType
@@ -66,6 +147,12 @@ class NanoPBOptions(message.Message):
   
   # @@protoc_insertion_point(class_scope:NanoPBOptions)
 
+nanopb_fileopt.message_type = _NANOPBOPTIONS
+google.protobuf.descriptor_pb2.FileOptions.RegisterExtension(nanopb_fileopt)
+nanopb_msgopt.message_type = _NANOPBOPTIONS
+google.protobuf.descriptor_pb2.MessageOptions.RegisterExtension(nanopb_msgopt)
+nanopb_enumopt.message_type = _NANOPBOPTIONS
+google.protobuf.descriptor_pb2.EnumOptions.RegisterExtension(nanopb_enumopt)
 nanopb.message_type = _NANOPBOPTIONS
 google.protobuf.descriptor_pb2.FieldOptions.RegisterExtension(nanopb)
 # @@protoc_insertion_point(module_scope)

pb.h

@@ -1,19 +1,92 @@
+/* Common parts of the nanopb library. Most of these are quite low-level
+ * stuff. For the high-level interface, see pb_encode.h and pb_decode.h.
+ */
+
 #ifndef _PB_H_
 #define _PB_H_
 
-/* pb.h: Common parts for nanopb library.
- * Most of these are quite low-level stuff. For the high-level interface,
- * see pb_encode.h or pb_decode.h
- */
+/*****************************************************************
+ * Nanopb compilation time options. You can change these here by *
+ * uncommenting the lines, or on the compiler command line.      *
+ *****************************************************************/
 
+/* Define this if your CPU architecture is big endian, i.e. it
+ * stores the most-significant byte first. */
+/* #define __BIG_ENDIAN__ 1 */
+
+/* Increase the number of required fields that are tracked.
+ * A compiler warning will tell if you need this. */
+/* #define PB_MAX_REQUIRED_FIELDS 256 */
+
+/* Add support for tag numbers > 255 and fields larger than 255 bytes. */
+/* #define PB_FIELD_16BIT 1 */
+
+/* Add support for tag numbers > 65536 and fields larger than 65536 bytes. */
+/* #define PB_FIELD_32BIT 1 */
+
+/* Disable support for error messages in order to save some code space. */
+/* #define PB_NO_ERRMSG 1 */
+
+/* Disable support for custom streams (support only memory buffers). */
+/* #define PB_BUFFER_ONLY 1 */
+
+/* Switch back to the old-style callback function signature.
+ * This was the default until nanopb-0.2.1. */
+/* #define PB_OLD_CALLBACK_STYLE */
+
+
+/******************************************************************
+ * You usually don't need to change anything below this line.     *
+ * Feel free to look around and use the defined macros, though.   *
+ ******************************************************************/
+
+
+/* Version of the nanopb library. Just in case you want to check it in
+ * your own program. */
+#define NANOPB_VERSION nanopb-0.2.2-dev
+
+/* Include all the system headers needed by nanopb. You will need the
+ * definitions of the following:
+ * - strlen, memcpy, memset functions
+ * - [u]int8_t, [u]int16_t, [u]int32_t, [u]int64_t
+ * - size_t
+ * - bool
+ *
+ * If you don't have the standard header files, you can instead provide
+ * a custom header that defines or includes all this. In that case,
+ * define PB_SYSTEM_HEADER to the path of this file.
+ */
+#ifdef PB_SYSTEM_HEADER
+#include PB_SYSTEM_HEADER
+#else
 #include <stdint.h>
 #include <stddef.h>
 #include <stdbool.h>
+#include <string.h>
+#endif
 
-#ifdef __GNUC__
-/* This just reduces memory requirements, but is not required. */
+/* Macro for defining packed structures (compiler dependent).
+ * This just reduces memory requirements, but is not required.
+ */
+#if defined(__GNUC__) || defined(__clang__)
+    /* For GCC and clang */
+#   define PB_PACKED_STRUCT_START
+#   define PB_PACKED_STRUCT_END
 #   define pb_packed __attribute__((packed))
+#elif defined(__ICCARM__)
+    /* For IAR ARM compiler */
+#   define PB_PACKED_STRUCT_START _Pragma("pack(push, 1)")
+#   define PB_PACKED_STRUCT_END _Pragma("pack(pop)")
+#   define pb_packed
+#elif defined(_MSC_VER) && (_MSC_VER >= 1500)
+    /* For Microsoft Visual C++ */
+#   define PB_PACKED_STRUCT_START __pragma(pack(push, 1))
+#   define PB_PACKED_STRUCT_END __pragma(pack(pop))
+#   define pb_packed
 #else
+    /* Unknown compiler */
+#   define PB_PACKED_STRUCT_START
+#   define PB_PACKED_STRUCT_END
 #   define pb_packed
 #endif
 
@@ -22,13 +95,16 @@
 #define UNUSED(x) (void)(x)
 #endif
 
-/* Compile-time assertion, used for checking compatible compilation options. */
+/* Compile-time assertion, used for checking compatible compilation options.
+ * If this fails on your compiler for some reason, use #define STATIC_ASSERT
+ * to disable it. */
 #ifndef STATIC_ASSERT
-#define STATIC_ASSERT(COND,MSG) typedef char static_assertion_##MSG[(COND)?1:-1];
+#define STATIC_ASSERT(COND,MSG) typedef char STATIC_ASSERT_MSG(MSG, __LINE__, __COUNTER__)[(COND)?1:-1];
+#define STATIC_ASSERT_MSG(MSG, LINE, COUNTER) STATIC_ASSERT_MSG_(MSG, LINE, COUNTER)
+#define STATIC_ASSERT_MSG_(MSG, LINE, COUNTER) static_assertion_##MSG##LINE##COUNTER
 #endif
 
-/* Number of required fields to keep track of
- * (change here or on compiler command line). */
+/* Number of required fields to keep track of. */
 #ifndef PB_MAX_REQUIRED_FIELDS
 #define PB_MAX_REQUIRED_FIELDS 64
 #endif
@@ -47,59 +123,55 @@
  * SINT* is different, though, because it is zig-zag coded.
  */
 
-typedef enum {
-    /************************
-     * Field contents types *
-     ************************/
+typedef uint8_t pb_type_t;
+
+/**** Field data types ****/
 
 /* Numeric types */
-    PB_LTYPE_VARINT = 0x00, /* int32, uint32, int64, uint64, bool, enum */
-    PB_LTYPE_SVARINT = 0x01, /* sint32, sint64 */
-    PB_LTYPE_FIXED32 = 0x02, /* fixed32, sfixed32, float */
-    PB_LTYPE_FIXED64 = 0x03, /* fixed64, sfixed64, double */
+#define PB_LTYPE_VARINT 0x00 /* int32, uint32, int64, uint64, bool, enum */
+#define PB_LTYPE_SVARINT 0x01 /* sint32, sint64 */
+#define PB_LTYPE_FIXED32 0x02 /* fixed32, sfixed32, float */
+#define PB_LTYPE_FIXED64 0x03 /* fixed64, sfixed64, double */
 
 /* Marker for last packable field type. */
-    PB_LTYPE_LAST_PACKABLE = 0x03,
+#define PB_LTYPE_LAST_PACKABLE 0x03
 
 /* Byte array with pre-allocated buffer.
  * data_size is the length of the allocated PB_BYTES_ARRAY structure. */
-    PB_LTYPE_BYTES = 0x04,
+#define PB_LTYPE_BYTES 0x04
 
 /* String with pre-allocated buffer.
  * data_size is the maximum length. */
-    PB_LTYPE_STRING = 0x05,
+#define PB_LTYPE_STRING 0x05
 
 /* Submessage
  * submsg_fields is pointer to field descriptions */
-    PB_LTYPE_SUBMESSAGE = 0x06,
+#define PB_LTYPE_SUBMESSAGE 0x06
+
+/* Extension pseudo-field
+ * The field contains a pointer to pb_extension_t */
+#define PB_LTYPE_EXTENSION 0x07
 
 /* Number of declared LTYPES */
-    PB_LTYPES_COUNT = 7,
+#define PB_LTYPES_COUNT 8
+#define PB_LTYPE_MASK 0x0F
 
-    /******************
-     * Modifier flags *
-     ******************/
+/**** Field repetition rules ****/
 
-    /* Just the basic, write data at data_offset */
-    PB_HTYPE_REQUIRED = 0x00,
+#define PB_HTYPE_REQUIRED 0x00
+#define PB_HTYPE_OPTIONAL 0x10
+#define PB_HTYPE_REPEATED 0x20
+#define PB_HTYPE_MASK     0x30
 
-    /* Write true at size_offset */
-    PB_HTYPE_OPTIONAL = 0x10,
+/**** Field allocation types ****/
  
-    /* Read to pre-allocated array
-     * Maximum number of entries is array_size,
-     * actual number is stored at size_offset */
-    PB_HTYPE_ARRAY = 0x20,
+#define PB_ATYPE_STATIC   0x00
+#define PB_ATYPE_CALLBACK 0x40
+#define PB_ATYPE_MASK     0xC0
 
-    /* Works for all required/optional/repeated fields.
-     * data_offset points to pb_callback_t structure.
-     * LTYPE should be 0 (it is ignored, but sometimes
-     * used to speculatively index an array). */
-    PB_HTYPE_CALLBACK = 0x30
-} pb_packed pb_type_t;
-
-#define PB_HTYPE(x) ((x) & 0xF0)
-#define PB_LTYPE(x) ((x) & 0x0F)
+#define PB_ATYPE(x) ((x) & PB_ATYPE_MASK)
+#define PB_HTYPE(x) ((x) & PB_HTYPE_MASK)
+#define PB_LTYPE(x) ((x) & PB_LTYPE_MASK)
 
 /* This structure is used in auto-generated constants
  * to specify struct fields.
@@ -109,6 +181,7 @@ typedef enum {
  * structures. Fix that by defining PB_FIELD_16BIT or
  * PB_FIELD_32BIT.
  */
+PB_PACKED_STRUCT_START
 typedef struct _pb_field_t pb_field_t;
 struct _pb_field_t {
 
@@ -140,15 +213,18 @@ struct _pb_field_t {
      * If null, then field will zeroed. */
     const void *ptr;
 } pb_packed;
+PB_PACKED_STRUCT_END
 
 /* This structure is used for 'bytes' arrays.
  * It has the number of bytes in the beginning, and after that an array.
  * Note that actual structs used will have a different length of bytes array.
  */
-typedef struct {
+struct _pb_bytes_array_t {
     size_t size;
     uint8_t bytes[1];
-} pb_bytes_array_t;
+};
+
+typedef struct _pb_bytes_array_t pb_bytes_array_t;
 
 /* This structure is used for giving the callback function.
  * It is stored in the message structure and filled in by the method that
@@ -172,10 +248,19 @@ typedef struct _pb_istream_t pb_istream_t;
 typedef struct _pb_ostream_t pb_ostream_t;
 typedef struct _pb_callback_t pb_callback_t;
 struct _pb_callback_t {
+#ifdef PB_OLD_CALLBACK_STYLE
+    /* Deprecated since nanopb-0.2.1 */
     union {
         bool (*decode)(pb_istream_t *stream, const pb_field_t *field, void *arg);
         bool (*encode)(pb_ostream_t *stream, const pb_field_t *field, const void *arg);
     } funcs;
+#else
+    /* New function signature, which allows modifying arg contents in callback. */
+    union {
+        bool (*decode)(pb_istream_t *stream, const pb_field_t *field, void **arg);
+        bool (*encode)(pb_ostream_t *stream, const pb_field_t *field, void * const *arg);
+    } funcs;
+#endif    
     
     /* Free arg for use by callback */
     void *arg;
@@ -189,12 +274,163 @@ typedef enum {
     PB_WT_32BIT  = 5
 } pb_wire_type_t;
 
+/* Structure for defining the handling of unknown/extension fields.
+ * Usually the pb_extension_type_t structure is automatically generated,
+ * while the pb_extension_t structure is created by the user. However,
+ * if you want to catch all unknown fields, you can also create a custom
+ * pb_extension_type_t with your own callback.
+ */
+typedef struct _pb_extension_type_t pb_extension_type_t;
+typedef struct _pb_extension_t pb_extension_t;
+struct _pb_extension_type_t {
+    /* Called for each unknown field in the message.
+     * If you handle the field, read off all of its data and return true.
+     * If you do not handle the field, do not read anything and return true.
+     * If you run into an error, return false.
+     * Set to NULL for default handler.
+     */
+    bool (*decode)(pb_istream_t *stream, pb_extension_t *extension,
+                   uint32_t tag, pb_wire_type_t wire_type);
+    
+    /* Called once after all regular fields have been encoded.
+     * If you have something to write, do so and return true.
+     * If you do not have anything to write, just return true.
+     * If you run into an error, return false.
+     * Set to NULL for default handler.
+     */
+    bool (*encode)(pb_ostream_t *stream, const pb_extension_t *extension);
+    
+    /* Free field for use by the callback. */
+    const void *arg;
+};
+
+struct _pb_extension_t {
+    /* Type describing the extension field. Usually you'll initialize
+     * this to a pointer to the automatically generated structure. */
+    const pb_extension_type_t *type;
+    
+    /* Destination for the decoded data. This must match the datatype
+     * of the extension field. */
+    void *dest;
+    
+    /* Pointer to the next extension handler, or NULL.
+     * If this extension does not match a field, the next handler is
+     * automatically called. */
+    pb_extension_t *next;
+};
+
 /* These macros are used to declare pb_field_t's in the constant array. */
 #define pb_membersize(st, m) (sizeof ((st*)0)->m)
 #define pb_arraysize(st, m) (pb_membersize(st, m) / pb_membersize(st, m[0]))
 #define pb_delta(st, m1, m2) ((int)offsetof(st, m1) - (int)offsetof(st, m2))
-#define pb_delta_end(st, m1, m2) (offsetof(st, m1) - offsetof(st, m2) - pb_membersize(st, m2))
-#define PB_LAST_FIELD {0,(pb_type_t) 0,0,0}
+#define pb_delta_end(st, m1, m2) (int)(offsetof(st, m1) == offsetof(st, m2) \
+                                  ? offsetof(st, m1) \
+                                  : offsetof(st, m1) - offsetof(st, m2) - pb_membersize(st, m2))
+#define PB_LAST_FIELD {0,(pb_type_t) 0,0,0,0,0,0}
 
+/* Required fields are the simplest. They just have delta (padding) from
+ * previous field end, and the size of the field. Pointer is used for
+ * submessages and default values.
+ */
+#define PB_REQUIRED_STATIC(tag, st, m, pm, ltype, ptr) \
+    {tag, PB_ATYPE_STATIC | PB_HTYPE_REQUIRED | ltype, \
+    pb_delta_end(st, m, pm), 0, pb_membersize(st, m), 0, ptr}
+
+/* Optional fields add the delta to the has_ variable. */
+#define PB_OPTIONAL_STATIC(tag, st, m, pm, ltype, ptr) \
+    {tag, PB_ATYPE_STATIC | PB_HTYPE_OPTIONAL | ltype, \
+    pb_delta_end(st, m, pm), \
+    pb_delta(st, has_ ## m, m), \
+    pb_membersize(st, m), 0, ptr}
+
+/* Repeated fields have a _count field and also the maximum number of entries. */
+#define PB_REPEATED_STATIC(tag, st, m, pm, ltype, ptr) \
+    {tag, PB_ATYPE_STATIC | PB_HTYPE_REPEATED | ltype, \
+    pb_delta_end(st, m, pm), \
+    pb_delta(st, m ## _count, m), \
+    pb_membersize(st, m[0]), \
+    pb_arraysize(st, m), ptr}
+
+/* Callbacks are much like required fields except with special datatype. */
+#define PB_REQUIRED_CALLBACK(tag, st, m, pm, ltype, ptr) \
+    {tag, PB_ATYPE_CALLBACK | PB_HTYPE_REQUIRED | ltype, \
+    pb_delta_end(st, m, pm), 0, pb_membersize(st, m), 0, ptr}
+
+#define PB_OPTIONAL_CALLBACK(tag, st, m, pm, ltype, ptr) \
+    {tag, PB_ATYPE_CALLBACK | PB_HTYPE_OPTIONAL | ltype, \
+    pb_delta_end(st, m, pm), 0, pb_membersize(st, m), 0, ptr}
+    
+#define PB_REPEATED_CALLBACK(tag, st, m, pm, ltype, ptr) \
+    {tag, PB_ATYPE_CALLBACK | PB_HTYPE_REPEATED | ltype, \
+    pb_delta_end(st, m, pm), 0, pb_membersize(st, m), 0, ptr}
+
+/* Optional extensions don't have the has_ field, as that would be redundant. */
+#define PB_OPTEXT_STATIC(tag, st, m, pm, ltype, ptr) \
+    {tag, PB_ATYPE_STATIC | PB_HTYPE_OPTIONAL | ltype, \
+    0, \
+    0, \
+    pb_membersize(st, m), 0, ptr}
+
+#define PB_OPTEXT_CALLBACK(tag, st, m, pm, ltype, ptr) \
+    {tag, PB_ATYPE_CALLBACK | PB_HTYPE_OPTIONAL | ltype, \
+    0, 0, pb_membersize(st, m), 0, ptr}
+
+/* The mapping from protobuf types to LTYPEs is done using these macros. */
+#define PB_LTYPE_MAP_BOOL       PB_LTYPE_VARINT
+#define PB_LTYPE_MAP_BYTES      PB_LTYPE_BYTES
+#define PB_LTYPE_MAP_DOUBLE     PB_LTYPE_FIXED64
+#define PB_LTYPE_MAP_ENUM       PB_LTYPE_VARINT
+#define PB_LTYPE_MAP_FIXED32    PB_LTYPE_FIXED32
+#define PB_LTYPE_MAP_FIXED64    PB_LTYPE_FIXED64
+#define PB_LTYPE_MAP_FLOAT      PB_LTYPE_FIXED32
+#define PB_LTYPE_MAP_INT32      PB_LTYPE_VARINT
+#define PB_LTYPE_MAP_INT64      PB_LTYPE_VARINT
+#define PB_LTYPE_MAP_MESSAGE    PB_LTYPE_SUBMESSAGE
+#define PB_LTYPE_MAP_SFIXED32   PB_LTYPE_FIXED32
+#define PB_LTYPE_MAP_SFIXED64   PB_LTYPE_FIXED64
+#define PB_LTYPE_MAP_SINT32     PB_LTYPE_SVARINT
+#define PB_LTYPE_MAP_SINT64     PB_LTYPE_SVARINT
+#define PB_LTYPE_MAP_STRING     PB_LTYPE_STRING
+#define PB_LTYPE_MAP_UINT32     PB_LTYPE_VARINT
+#define PB_LTYPE_MAP_UINT64     PB_LTYPE_VARINT
+#define PB_LTYPE_MAP_EXTENSION  PB_LTYPE_EXTENSION
+
+/* This is the actual macro used in field descriptions.
+ * It takes these arguments:
+ * - Field tag number
+ * - Field type:   BOOL, BYTES, DOUBLE, ENUM, FIXED32, FIXED64,
+ *                 FLOAT, INT32, INT64, MESSAGE, SFIXED32, SFIXED64
+ *                 SINT32, SINT64, STRING, UINT32, UINT64 or EXTENSION
+ * - Field rules:  REQUIRED, OPTIONAL or REPEATED
+ * - Allocation:   STATIC or CALLBACK
+ * - Message name
+ * - Field name
+ * - Previous field name (or field name again for first field)
+ * - Pointer to default value or submsg fields.
+ */
+
+#define PB_FIELD(tag, type, rules, allocation, message, field, prevfield, ptr) \
+    PB_ ## rules ## _ ## allocation(tag, message, field, prevfield, \
+                                    PB_LTYPE_MAP_ ## type, ptr)
+
+
+/* These macros are used for giving out error messages.
+ * They are mostly a debugging aid; the main error information
+ * is the true/false return value from functions.
+ * Some code space can be saved by disabling the error
+ * messages if not used.
+ */
+#ifdef PB_NO_ERRMSG
+#define PB_RETURN_ERROR(stream,msg) return false
+#define PB_GET_ERROR(stream) "(errmsg disabled)"
+#else
+#define PB_RETURN_ERROR(stream,msg) \
+    do {\
+        if ((stream)->errmsg == NULL) \
+            (stream)->errmsg = (msg); \
+        return false; \
+    } while(0)
+#define PB_GET_ERROR(stream) ((stream)->errmsg ? (stream)->errmsg : "(none)")
+#endif
 
 #endif

pb_decode.c

@@ -3,17 +3,17 @@
  * 2011 Petteri Aimonen <jpa@kapsi.fi>
  */
 
-
-#ifdef __GNUC__
+/* The warn_unused_result attribute appeared first in gcc-3.4.0 */
+#if !defined(__GNUC__) || ( __GNUC__ < 3) || (__GNUC__ == 3 && __GNUC_MINOR__ < 4)
+    #define checkreturn
+#else
     /* Verify that we remember to check all return values for proper error propagation */
     #define checkreturn __attribute__((warn_unused_result))
-#else
-#define checkreturn
 #endif
 
+#define NANOPB_INTERNALS
 #include "pb.h"
 #include "pb_decode.h"
-#include <string.h>
 
 typedef bool (*pb_decoder_t)(pb_istream_t *stream, const pb_field_t *field, void *dest) checkreturn;
 
@@ -28,42 +28,75 @@ static const pb_decoder_t PB_DECODERS[PB_LTYPES_COUNT] = {
     
     &pb_dec_bytes,
     &pb_dec_string,
-    &pb_dec_submessage
+    &pb_dec_submessage,
+    NULL /* extensions */
 };
 
 /**************
  * pb_istream *
  **************/
 
-bool checkreturn pb_read(pb_istream_t *stream, uint8_t *buf, size_t count)
-{
-    if (stream->bytes_left < count)
-        return false;
-    
-    if (!stream->callback(stream, buf, count))
-        return false;
-    
-    stream->bytes_left -= count;
-    return true;
-}
-
 static bool checkreturn buf_read(pb_istream_t *stream, uint8_t *buf, size_t count)
 {
     uint8_t *source = (uint8_t*)stream->state;
+    stream->state = source + count;
     
     if (buf != NULL)
-        memcpy(buf, source, count);
+    {
+        while (count--)
+            *buf++ = *source++;
+    }
     
-    stream->state = source + count;
+    return true;
+}
+
+bool checkreturn pb_read(pb_istream_t *stream, uint8_t *buf, size_t count)
+{
+#ifndef PB_BUFFER_ONLY
+	if (buf == NULL && stream->callback != buf_read)
+	{
+		/* Skip input bytes */
+		uint8_t tmp[16];
+		while (count > 16)
+		{
+			if (!pb_read(stream, tmp, 16))
+				return false;
+			
+			count -= 16;
+		}
+		
+		return pb_read(stream, tmp, count);
+	}
+#endif
+
+    if (stream->bytes_left < count)
+        PB_RETURN_ERROR(stream, "end-of-stream");
+    
+#ifndef PB_BUFFER_ONLY
+    if (!stream->callback(stream, buf, count))
+        PB_RETURN_ERROR(stream, "io error");
+#else
+    if (!buf_read(stream, buf, count))
+        return false;
+#endif
+    
+    stream->bytes_left -= count;
     return true;
 }
 
 pb_istream_t pb_istream_from_buffer(uint8_t *buf, size_t bufsize)
 {
     pb_istream_t stream;
+#ifdef PB_BUFFER_ONLY
+    stream.callback = NULL;
+#else
     stream.callback = &buf_read;
+#endif
     stream.state = buf;
     stream.bytes_left = bufsize;
+#ifndef PB_NO_ERRMSG
+    stream.errmsg = NULL;
+#endif
     return stream;
 }
 
@@ -73,28 +106,60 @@ pb_istream_t pb_istream_from_buffer(uint8_t *buf, size_t bufsize)
 
 static bool checkreturn pb_decode_varint32(pb_istream_t *stream, uint32_t *dest)
 {
-    uint64_t temp;
-    bool status = pb_decode_varint(stream, &temp);
-    *dest = (uint32_t)temp;
-    return status;
+    uint8_t byte;
+    uint32_t result;
+    
+    if (!pb_read(stream, &byte, 1))
+        return false;
+    
+    if (!(byte & 0x80))
+    {
+        /* Quick case, 1 byte value */
+        result = byte;
+    }
+    else
+    {
+        /* Multibyte case */
+        uint8_t bitpos = 7;
+        result = byte & 0x7F;
+        
+        do
+        {
+            if (bitpos >= 32)
+                PB_RETURN_ERROR(stream, "varint overflow");
+            
+            if (!pb_read(stream, &byte, 1))
+                return false;
+            
+            result |= (uint32_t)(byte & 0x7F) << bitpos;
+            bitpos = (uint8_t)(bitpos + 7);
+        } while (byte & 0x80);
+   }
+   
+   *dest = result;
+   return true;
 }
 
 bool checkreturn pb_decode_varint(pb_istream_t *stream, uint64_t *dest)
 {
     uint8_t byte;
     uint8_t bitpos = 0;
-    *dest = 0;
+    uint64_t result = 0;
     
-    while (bitpos < 64 && pb_read(stream, &byte, 1))
+    do
     {
-        *dest |= (uint64_t)(byte & 0x7F) << bitpos;
-        bitpos += 7;
-        
-        if (!(byte & 0x80))
-            return true;
-    }
+        if (bitpos >= 64)
+            PB_RETURN_ERROR(stream, "varint overflow");
         
+        if (!pb_read(stream, &byte, 1))
             return false;
+
+        result |= (uint64_t)(byte & 0x7F) << bitpos;
+        bitpos = (uint8_t)(bitpos + 7);
+    } while (byte & 0x80);
+    
+    *dest = result;
+    return true;
 }
 
 bool checkreturn pb_skip_varint(pb_istream_t *stream)
@@ -117,11 +182,11 @@ bool checkreturn pb_skip_string(pb_istream_t *stream)
     return pb_read(stream, NULL, length);
 }
 
-bool checkreturn pb_decode_tag(pb_istream_t *stream, pb_wire_type_t *wire_type, int *tag, bool *eof)
+bool checkreturn pb_decode_tag(pb_istream_t *stream, pb_wire_type_t *wire_type, uint32_t *tag, bool *eof)
 {
     uint32_t temp;
     *eof = false;
-    *wire_type = 0;
+    *wire_type = (pb_wire_type_t) 0;
     *tag = 0;
     
     if (!pb_decode_varint32(stream, &temp))
@@ -151,7 +216,7 @@ bool checkreturn pb_skip_field(pb_istream_t *stream, pb_wire_type_t wire_type)
         case PB_WT_64BIT: return pb_read(stream, NULL, 8);
         case PB_WT_STRING: return pb_skip_string(stream);
         case PB_WT_32BIT: return pb_read(stream, NULL, 4);
-        default: return false;
+        default: PB_RETURN_ERROR(stream, "invalid wire_type");
     }
 }
 
@@ -181,15 +246,14 @@ static bool checkreturn read_raw_value(pb_istream_t *stream, pb_wire_type_t wire
             *size = 4;
             return pb_read(stream, buf, 4);
         
-        default: return false;
+        default: PB_RETURN_ERROR(stream, "invalid wire_type");
     }
 }
 
 /* Decode string length from stream and return a substream with limited length.
- * Before disposing the substream, remember to copy the substream->state back
- * to stream->state.
+ * Remember to close the substream using pb_close_string_substream().
  */
-static bool checkreturn make_string_substream(pb_istream_t *stream, pb_istream_t *substream)
+bool checkreturn pb_make_string_substream(pb_istream_t *stream, pb_istream_t *substream)
 {
     uint32_t size;
     if (!pb_decode_varint32(stream, &size))
@@ -197,19 +261,28 @@ static bool checkreturn make_string_substream(pb_istream_t *stream, pb_istream_t
     
     *substream = *stream;
     if (substream->bytes_left < size)
-        return false;
+        PB_RETURN_ERROR(stream, "parent stream too short");
     
     substream->bytes_left = size;
     stream->bytes_left -= size;
     return true;
 }
 
+void pb_close_string_substream(pb_istream_t *stream, pb_istream_t *substream)
+{
+    stream->state = substream->state;
+
+#ifndef PB_NO_ERRMSG
+    stream->errmsg = substream->errmsg;
+#endif
+}
+
 /* Iterator for pb_field_t list */
 typedef struct {
     const pb_field_t *start; /* Start of the pb_field_t array */
-    const pb_field_t *current; /* Current position of the iterator */
-    int field_index; /* Zero-based index of the field. */
-    int required_field_index; /* Zero-based index that counts only the required fields */
+    const pb_field_t *pos; /* Current position of the iterator */
+    unsigned field_index; /* Zero-based index of the field. */
+    unsigned required_field_index; /* Zero-based index that counts only the required fields */
     void *dest_struct; /* Pointer to the destination structure to decode to */
     void *pData; /* Pointer where to store current field value */
     void *pSize; /* Pointer where to store the size of current array field */
@@ -217,30 +290,36 @@ typedef struct {
 
 static void pb_field_init(pb_field_iterator_t *iter, const pb_field_t *fields, void *dest_struct)
 {
-    iter->start = iter->current = fields;
+    iter->start = iter->pos = fields;
     iter->field_index = 0;
     iter->required_field_index = 0;
-    iter->pData = (char*)dest_struct + iter->current->data_offset;
-    iter->pSize = (char*)iter->pData + iter->current->size_offset;
+    iter->pData = (char*)dest_struct + iter->pos->data_offset;
+    iter->pSize = (char*)iter->pData + iter->pos->size_offset;
     iter->dest_struct = dest_struct;
 }
 
 static bool pb_field_next(pb_field_iterator_t *iter)
 {
     bool notwrapped = true;
-    size_t prev_size = iter->current->data_size;
+    size_t prev_size = iter->pos->data_size;
     
-    if (PB_HTYPE(iter->current->type) == PB_HTYPE_ARRAY)
-        prev_size *= iter->current->array_size;
+    if (PB_ATYPE(iter->pos->type) == PB_ATYPE_STATIC &&
+        PB_HTYPE(iter->pos->type) == PB_HTYPE_REPEATED)
+    {
+        prev_size *= iter->pos->array_size;
+    }
     
-    if (PB_HTYPE(iter->current->type) == PB_HTYPE_REQUIRED)
+    if (iter->pos->tag == 0)
+        return false; /* Only happens with empty message types */
+    
+    if (PB_HTYPE(iter->pos->type) == PB_HTYPE_REQUIRED)
         iter->required_field_index++;
     
-    iter->current++;
+    iter->pos++;
     iter->field_index++;
-    if (iter->current->tag == 0)
+    if (iter->pos->tag == 0)
     {
-        iter->current = iter->start;
+        iter->pos = iter->start;
         iter->field_index = 0;
         iter->required_field_index = 0;
         iter->pData = iter->dest_struct;
@@ -248,18 +327,21 @@ static bool pb_field_next(pb_field_iterator_t *iter)
         notwrapped = false;
     }
     
-    iter->pData = (char*)iter->pData + prev_size + iter->current->data_offset;
-    iter->pSize = (char*)iter->pData + iter->current->size_offset;
+    iter->pData = (char*)iter->pData + prev_size + iter->pos->data_offset;
+    iter->pSize = (char*)iter->pData + iter->pos->size_offset;
     return notwrapped;
 }
 
-static bool checkreturn pb_field_find(pb_field_iterator_t *iter, int tag)
+static bool checkreturn pb_field_find(pb_field_iterator_t *iter, uint32_t tag)
 {
-    int start = iter->field_index;
+    unsigned start = iter->field_index;
     
     do {
-        if (iter->current->tag == tag)
+        if (iter->pos->tag == tag &&
+            PB_LTYPE(iter->pos->type) != PB_LTYPE_EXTENSION)
+        {
             return true;
+        }
         pb_field_next(iter);
     } while (iter->field_index != start);
     
@@ -270,54 +352,78 @@ static bool checkreturn pb_field_find(pb_field_iterator_t *iter, int tag)
  * Decode a single field *
  *************************/
 
-static bool checkreturn decode_field(pb_istream_t *stream, pb_wire_type_t wire_type, pb_field_iterator_t *iter)
+static bool checkreturn decode_static_field(pb_istream_t *stream, pb_wire_type_t wire_type, pb_field_iterator_t *iter)
 {
-    pb_decoder_t func = PB_DECODERS[PB_LTYPE(iter->current->type)];
+    pb_type_t type;
+    pb_decoder_t func;
     
-    switch (PB_HTYPE(iter->current->type))
+    type = iter->pos->type;
+    func = PB_DECODERS[PB_LTYPE(type)];
+
+    switch (PB_HTYPE(type))
     {
         case PB_HTYPE_REQUIRED:
-            return func(stream, iter->current, iter->pData);
+            return func(stream, iter->pos, iter->pData);
             
         case PB_HTYPE_OPTIONAL:
             *(bool*)iter->pSize = true;
-            return func(stream, iter->current, iter->pData);
+            return func(stream, iter->pos, iter->pData);
     
-        case PB_HTYPE_ARRAY:
+        case PB_HTYPE_REPEATED:
             if (wire_type == PB_WT_STRING
-                && PB_LTYPE(iter->current->type) <= PB_LTYPE_LAST_PACKABLE)
+                && PB_LTYPE(type) <= PB_LTYPE_LAST_PACKABLE)
             {
                 /* Packed array */
+                bool status = true;
                 size_t *size = (size_t*)iter->pSize;
                 pb_istream_t substream;
-                if (!make_string_substream(stream, &substream))
+                if (!pb_make_string_substream(stream, &substream))
                     return false;
                 
-                while (substream.bytes_left && *size < iter->current->array_size)
+                while (substream.bytes_left && *size < iter->pos->array_size)
                 {
-                    void *pItem = (uint8_t*)iter->pData + iter->current->data_size * (*size);
-                    if (!func(&substream, iter->current, pItem))
-                        return false;
+                    void *pItem = (uint8_t*)iter->pData + iter->pos->data_size * (*size);
+                    if (!func(&substream, iter->pos, pItem))
+                    {
+                        status = false;
+                        break;
+                    }
                     (*size)++;
                 }
-                return (substream.bytes_left == 0);
+                pb_close_string_substream(stream, &substream);
+                
+                if (substream.bytes_left != 0)
+                    PB_RETURN_ERROR(stream, "array overflow");
+                
+                return status;
             }
             else
             {
                 /* Repeated field */
                 size_t *size = (size_t*)iter->pSize;
-                void *pItem = (uint8_t*)iter->pData + iter->current->data_size * (*size);
-                if (*size >= iter->current->array_size)
-                    return false;
+                void *pItem = (uint8_t*)iter->pData + iter->pos->data_size * (*size);
+                if (*size >= iter->pos->array_size)
+                    PB_RETURN_ERROR(stream, "array overflow");
                 
                 (*size)++;
-                return func(stream, iter->current, pItem);
+                return func(stream, iter->pos, pItem);
             }
 
-        case PB_HTYPE_CALLBACK:
+        default:
+            PB_RETURN_ERROR(stream, "invalid field type");
+    }
+}
+
+static bool checkreturn decode_callback_field(pb_istream_t *stream, pb_wire_type_t wire_type, pb_field_iterator_t *iter)
 {
     pb_callback_t *pCallback = (pb_callback_t*)iter->pData;
     
+#ifdef PB_OLD_CALLBACK_STYLE
+    void *arg = pCallback->arg;
+#else
+    void **arg = &(pCallback->arg);
+#endif
+    
     if (pCallback->funcs.decode == NULL)
         return pb_skip_field(stream, wire_type);
     
@@ -325,16 +431,16 @@ static bool checkreturn decode_field(pb_istream_t *stream, pb_wire_type_t wire_t
     {
         pb_istream_t substream;
         
-                if (!make_string_substream(stream, &substream))
+        if (!pb_make_string_substream(stream, &substream))
             return false;
         
-                while (substream.bytes_left)
+        do
         {
-                    if (!pCallback->funcs.decode(&substream, iter->current, pCallback->arg))
-                        return false;
-                }
+            if (!pCallback->funcs.decode(&substream, iter->pos, arg))
+                PB_RETURN_ERROR(stream, "callback failed");
+        } while (substream.bytes_left);
         
-                stream->state = substream.state;
+        pb_close_string_substream(stream, &substream);
         return true;
     }
     else
@@ -351,15 +457,89 @@ static bool checkreturn decode_field(pb_istream_t *stream, pb_wire_type_t wire_t
             return false;
         substream = pb_istream_from_buffer(buffer, size);
         
-                return pCallback->funcs.decode(&substream, iter->current, pCallback->arg);
+        return pCallback->funcs.decode(&substream, iter->pos, arg);
     }
 }
 
+static bool checkreturn decode_field(pb_istream_t *stream, pb_wire_type_t wire_type, pb_field_iterator_t *iter)
+{
+    switch (PB_ATYPE(iter->pos->type))
+    {
+        case PB_ATYPE_STATIC:
+            return decode_static_field(stream, wire_type, iter);
+        
+        case PB_ATYPE_CALLBACK:
+            return decode_callback_field(stream, wire_type, iter);
+        
         default:
-            return false;
+            PB_RETURN_ERROR(stream, "invalid field type");
     }
 }
 
+/* Default handler for extension fields. Expects a pb_field_t structure
+ * in extension->type->arg. */
+static bool checkreturn default_extension_handler(pb_istream_t *stream,
+    pb_extension_t *extension, uint32_t tag, pb_wire_type_t wire_type)
+{
+    const pb_field_t *field = (const pb_field_t*)extension->type->arg;
+    pb_field_iterator_t iter;
+    bool dummy;
+    
+    if (field->tag != tag)
+        return true;
+    
+    iter.start = field;
+    iter.pos = field;
+    iter.field_index = 0;
+    iter.required_field_index = 0;
+    iter.dest_struct = extension->dest;
+    iter.pData = extension->dest;
+    iter.pSize = &dummy;
+    
+    return decode_field(stream, wire_type, &iter);
+}
+
+/* Try to decode an unknown field as an extension field. Tries each extension
+ * decoder in turn, until one of them handles the field or loop ends. */
+static bool checkreturn decode_extension(pb_istream_t *stream,
+    uint32_t tag, pb_wire_type_t wire_type, pb_field_iterator_t *iter)
+{
+    pb_extension_t *extension = *(pb_extension_t* const *)iter->pData;
+    size_t pos = stream->bytes_left;
+    
+    while (extension && pos == stream->bytes_left)
+    {
+        bool status;
+        if (extension->type->decode)
+            status = extension->type->decode(stream, extension, tag, wire_type);
+        else
+            status = default_extension_handler(stream, extension, tag, wire_type);
+
+        if (!status)
+            return false;
+        
+        extension = extension->next;
+    }
+    
+    return true;
+}
+
+/* Step through the iterator until an extension field is found or until all
+ * entries have been checked. There can be only one extension field per
+ * message. Returns false if no extension field is found. */
+static bool checkreturn find_extension_field(pb_field_iterator_t *iter)
+{
+    unsigned start = iter->field_index;
+    
+    do {
+        if (PB_LTYPE(iter->pos->type) == PB_LTYPE_EXTENSION)
+            return true;
+        pb_field_next(iter);
+    } while (iter->field_index != start);
+    
+    return false;
+}
+
 /* Initialize message fields to default values, recursively */
 static void pb_message_set_to_defaults(const pb_field_t fields[], void *dest_struct)
 {
@@ -369,36 +549,42 @@ static void pb_message_set_to_defaults(const pb_field_t fields[], void *dest_str
     /* Initialize size/has fields and apply default values */
     do
     {
-        if (iter.current->tag == 0)
+        pb_type_t type;
+        type = iter.pos->type;
+    
+        if (iter.pos->tag == 0)
             continue;
         
+        if (PB_ATYPE(type) == PB_ATYPE_STATIC)
+        {
             /* Initialize the size field for optional/repeated fields to 0. */
-        if (PB_HTYPE(iter.current->type) == PB_HTYPE_OPTIONAL)
+            if (PB_HTYPE(type) == PB_HTYPE_OPTIONAL)
             {
                 *(bool*)iter.pSize = false;
             }
-        else if (PB_HTYPE(iter.current->type) == PB_HTYPE_ARRAY)
+            else if (PB_HTYPE(type) == PB_HTYPE_REPEATED)
             {
                 *(size_t*)iter.pSize = 0;
                 continue; /* Array is empty, no need to initialize contents */
             }
             
             /* Initialize field contents to default value */
-        if (PB_HTYPE(iter.current->type) == PB_HTYPE_CALLBACK)
+            if (PB_LTYPE(iter.pos->type) == PB_LTYPE_SUBMESSAGE)
             {
-            continue; /* Don't overwrite callback */
+                pb_message_set_to_defaults((const pb_field_t *) iter.pos->ptr, iter.pData);
             }
-        else if (PB_LTYPE(iter.current->type) == PB_LTYPE_SUBMESSAGE)
+            else if (iter.pos->ptr != NULL)
             {
-            pb_message_set_to_defaults(iter.current->ptr, iter.pData);
-        }
-        else if (iter.current->ptr != NULL)
-        {
-            memcpy(iter.pData, iter.current->ptr, iter.current->data_size);
+                memcpy(iter.pData, iter.pos->ptr, iter.pos->data_size);
             }
             else
             {
-            memset(iter.pData, 0, iter.current->data_size);
+                memset(iter.pData, 0, iter.pos->data_size);
+            }
+        }
+        else if (PB_ATYPE(type) == PB_ATYPE_CALLBACK)
+        {
+            continue; /* Don't overwrite callback */
         }
     } while (pb_field_next(&iter));
 }
@@ -407,18 +593,17 @@ static void pb_message_set_to_defaults(const pb_field_t fields[], void *dest_str
  * Decode all fields *
  *********************/
 
-bool checkreturn pb_decode(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct)
+bool checkreturn pb_decode_noinit(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct)
 {
-    uint8_t fields_seen[(PB_MAX_REQUIRED_FIELDS + 7) / 8] = {}; /* Used to check for required fields */
+    uint8_t fields_seen[(PB_MAX_REQUIRED_FIELDS + 7) / 8] = {0}; /* Used to check for required fields */
+    uint32_t extension_range_start = 0;
     pb_field_iterator_t iter;
     
-    pb_message_set_to_defaults(fields, dest_struct);
-    
     pb_field_init(&iter, fields, dest_struct);
     
     while (stream->bytes_left)
     {
-        int tag;
+        uint32_t tag;
         pb_wire_type_t wire_type;
         bool eof;
         
@@ -432,16 +617,39 @@ bool checkreturn pb_decode(pb_istream_t *stream, const pb_field_t fields[], void
         
         if (!pb_field_find(&iter, tag))
         {
+            /* No match found, check if it matches an extension. */
+            if (tag >= extension_range_start)
+            {
+                if (!find_extension_field(&iter))
+                    extension_range_start = (uint32_t)-1;
+                else
+                    extension_range_start = iter.pos->tag;
+                
+                if (tag >= extension_range_start)
+                {
+                    size_t pos = stream->bytes_left;
+                
+                    if (!decode_extension(stream, tag, wire_type, &iter))
+                        return false;
+                    
+                    if (pos != stream->bytes_left)
+                    {
+                        /* The field was handled */
+                        continue;                    
+                    }
+                }
+            }
+        
             /* No match found, skip data */
             if (!pb_skip_field(stream, wire_type))
                 return false;
             continue;
         }
         
-        if (PB_HTYPE(iter.current->type) == PB_HTYPE_REQUIRED
+        if (PB_HTYPE(iter.pos->type) == PB_HTYPE_REQUIRED
             && iter.required_field_index < PB_MAX_REQUIRED_FIELDS)
         {
-            fields_seen[iter.required_field_index >> 3] |= 1 << (iter.required_field_index & 7);
+            fields_seen[iter.required_field_index >> 3] |= (uint8_t)(1 << (iter.required_field_index & 7));
         }
             
         if (!decode_field(stream, wire_type, &iter))
@@ -449,93 +657,159 @@ bool checkreturn pb_decode(pb_istream_t *stream, const pb_field_t fields[], void
     }
     
     /* Check that all required fields were present. */
-    pb_field_init(&iter, fields, dest_struct);
-    do {
-        if (PB_HTYPE(iter.current->type) == PB_HTYPE_REQUIRED &&
-            iter.required_field_index < PB_MAX_REQUIRED_FIELDS &&
-            !(fields_seen[iter.required_field_index >> 3] & (1 << (iter.required_field_index & 7))))
     {
-            return false;
-        }
+        /* First figure out the number of required fields by
+         * seeking to the end of the field array. Usually we
+         * are already close to end after decoding.
+         */
+        unsigned req_field_count;
+        pb_type_t last_type;
+        unsigned i;
+        do {
+            req_field_count = iter.required_field_index;
+            last_type = iter.pos->type;
         } while (pb_field_next(&iter));
         
+        /* Fixup if last field was also required. */
+        if (PB_HTYPE(last_type) == PB_HTYPE_REQUIRED && iter.pos->tag)
+            req_field_count++;
+        
+        /* Check the whole bytes */
+        for (i = 0; i < (req_field_count >> 3); i++)
+        {
+            if (fields_seen[i] != 0xFF)
+                PB_RETURN_ERROR(stream, "missing required field");
+        }
+        
+        /* Check the remaining bits */
+        if (fields_seen[req_field_count >> 3] != (0xFF >> (8 - (req_field_count & 7))))
+            PB_RETURN_ERROR(stream, "missing required field");
+    }
+    
     return true;
 }
 
+bool checkreturn pb_decode(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct)
+{
+    pb_message_set_to_defaults(fields, dest_struct);
+    return pb_decode_noinit(stream, fields, dest_struct);
+}
+
+bool pb_decode_delimited(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct)
+{
+    pb_istream_t substream;
+    bool status;
+    
+    if (!pb_make_string_substream(stream, &substream))
+        return false;
+    
+    status = pb_decode(&substream, fields, dest_struct);
+    pb_close_string_substream(stream, &substream);
+    return status;
+}
+
 /* Field decoders */
 
-/* Copy destsize bytes from src so that values are casted properly.
- * On little endian machine, copy first n bytes of src
- * On big endian machine, copy last n bytes of src
- * srcsize must always be larger than destsize
- */
-static void endian_copy(void *dest, void *src, size_t destsize, size_t srcsize)
+bool pb_decode_svarint(pb_istream_t *stream, int64_t *dest)
+{
+    uint64_t value;
+    if (!pb_decode_varint(stream, &value))
+        return false;
+    
+    if (value & 1)
+        *dest = (int64_t)(~(value >> 1));
+    else
+        *dest = (int64_t)(value >> 1);
+    
+    return true;
+}
+
+bool pb_decode_fixed32(pb_istream_t *stream, void *dest)
 {
     #ifdef __BIG_ENDIAN__
-    memcpy(dest, (char*)src + (srcsize - destsize), destsize);
+    uint8_t *bytes = (uint8_t*)dest;
+    uint8_t lebytes[4];
+    
+    if (!pb_read(stream, lebytes, 4))
+        return false;
+    
+    bytes[0] = lebytes[3];
+    bytes[1] = lebytes[2];
+    bytes[2] = lebytes[1];
+    bytes[3] = lebytes[0];
+    return true;
     #else
-    UNUSED(srcsize);
-    memcpy(dest, src, destsize);
+    return pb_read(stream, (uint8_t*)dest, 4);
+    #endif   
+}
+
+bool pb_decode_fixed64(pb_istream_t *stream, void *dest)
+{
+    #ifdef __BIG_ENDIAN__
+    uint8_t *bytes = (uint8_t*)dest;
+    uint8_t lebytes[8];
+    
+    if (!pb_read(stream, lebytes, 8))
+        return false;
+    
+    bytes[0] = lebytes[7];
+    bytes[1] = lebytes[6];
+    bytes[2] = lebytes[5];
+    bytes[3] = lebytes[4];
+    bytes[4] = lebytes[3];
+    bytes[5] = lebytes[2];
+    bytes[6] = lebytes[1];
+    bytes[7] = lebytes[0];
+    return true;
+    #else
+    return pb_read(stream, (uint8_t*)dest, 8);
     #endif   
 }
 
 bool checkreturn pb_dec_varint(pb_istream_t *stream, const pb_field_t *field, void *dest)
 {
-    uint64_t temp;
-    bool status = pb_decode_varint(stream, &temp);
-    endian_copy(dest, &temp, field->data_size, sizeof(temp));
-    return status;
+    uint64_t value;
+    if (!pb_decode_varint(stream, &value))
+        return false;
+    
+    switch (field->data_size)
+    {
+        case 1: *(uint8_t*)dest = (uint8_t)value; break;
+        case 2: *(uint16_t*)dest = (uint16_t)value; break;
+        case 4: *(uint32_t*)dest = (uint32_t)value; break;
+        case 8: *(uint64_t*)dest = value; break;
+        default: PB_RETURN_ERROR(stream, "invalid data_size");
+    }
+    
+    return true;
 }
 
 bool checkreturn pb_dec_svarint(pb_istream_t *stream, const pb_field_t *field, void *dest)
 {
-    uint64_t temp;
-    bool status = pb_decode_varint(stream, &temp);
-    temp = (temp >> 1) ^ -(int64_t)(temp & 1);
-    endian_copy(dest, &temp, field->data_size, sizeof(temp));
-    return status;
+    int64_t value;
+    if (!pb_decode_svarint(stream, &value))
+        return false;
+    
+    switch (field->data_size)
+    {
+        case 4: *(int32_t*)dest = (int32_t)value; break;
+        case 8: *(int64_t*)dest = value; break;
+        default: PB_RETURN_ERROR(stream, "invalid data_size");
+    }
+    
+    return true;
 }
 
 bool checkreturn pb_dec_fixed32(pb_istream_t *stream, const pb_field_t *field, void *dest)
 {
-#ifdef __BIG_ENDIAN__
-    uint8_t bytes[4] = {0};
-    bool status = pb_read(stream, bytes, 4);
-    if (status) {
-        uint8_t *d = (uint8_t*)dest;
-        d[0] = bytes[3];
-        d[1] = bytes[2];
-        d[2] = bytes[1];
-        d[3] = bytes[0];
-    }
-    return status;
-#else
     UNUSED(field);
-    return pb_read(stream, (uint8_t*)dest, 4);
-#endif
+    return pb_decode_fixed32(stream, dest);
 }
 
 bool checkreturn pb_dec_fixed64(pb_istream_t *stream, const pb_field_t *field, void *dest)
 {
-#ifdef __BIG_ENDIAN__
-    uint8_t bytes[8] = {0};
-    bool status = pb_read(stream, bytes, 8);
-    if (status) {
-        uint8_t *d = (uint8_t*)dest;
-        d[0] = bytes[7];
-        d[1] = bytes[6];
-        d[2] = bytes[5]; 
-        d[3] = bytes[4];
-        d[4] = bytes[3];
-        d[5] = bytes[2];
-        d[6] = bytes[1];
-        d[7] = bytes[0];
-    }
-    return status;
-#else
     UNUSED(field);
-    return pb_read(stream, (uint8_t*)dest, 8);
-#endif
+    return pb_decode_fixed64(stream, dest);
 }
 
 bool checkreturn pb_dec_bytes(pb_istream_t *stream, const pb_field_t *field, void *dest)
@@ -549,7 +823,7 @@ bool checkreturn pb_dec_bytes(pb_istream_t *stream, const pb_field_t *field, voi
     
     /* Check length, noting the space taken by the size_t header. */
     if (x->size > field->data_size - offsetof(pb_bytes_array_t, bytes))
-        return false;
+        PB_RETURN_ERROR(stream, "bytes overflow");
     
     return pb_read(stream, x->bytes, x->size);
 }
@@ -563,7 +837,7 @@ bool checkreturn pb_dec_string(pb_istream_t *stream, const pb_field_t *field, vo
     
     /* Check length, noting the null terminator */
     if (size + 1 > field->data_size)
-        return false;
+        PB_RETURN_ERROR(stream, "string overflow");
     
     status = pb_read(stream, (uint8_t*)dest, size);
     *((uint8_t*)dest + size) = 0;
@@ -574,14 +848,21 @@ bool checkreturn pb_dec_submessage(pb_istream_t *stream, const pb_field_t *field
 {
     bool status;
     pb_istream_t substream;
+    const pb_field_t* submsg_fields = (const pb_field_t*)field->ptr;
     
-    if (!make_string_substream(stream, &substream))
+    if (!pb_make_string_substream(stream, &substream))
         return false;
     
     if (field->ptr == NULL)
-        return false;
+        PB_RETURN_ERROR(stream, "invalid field descriptor");
     
-    status = pb_decode(&substream, (pb_field_t*)field->ptr, dest);
-    stream->state = substream.state;
+    /* New array entries need to be initialized, while required and optional
+     * submessages have already been initialized in the top-level pb_decode. */
+    if (PB_HTYPE(field->type) == PB_HTYPE_REPEATED)
+        status = pb_decode(&substream, submsg_fields, dest);
+    else
+        status = pb_decode_noinit(&substream, submsg_fields, dest);
+    
+    pb_close_string_substream(stream, &substream);
     return status;
 }

pb_decode.h

@@ -1,67 +1,142 @@
+/* pb_decode.h: Functions to decode protocol buffers. Depends on pb_decode.c.
+ * The main function is pb_decode. You also need an input stream, and the
+ * field descriptions created by nanopb_generator.py.
+ */
+
 #ifndef _PB_DECODE_H_
 #define _PB_DECODE_H_
 
-/* pb_decode.h: Functions to decode protocol buffers. Depends on pb_decode.c.
- * The main function is pb_decode. You will also need to create an input
- * stream, which is easiest to do with pb_istream_t.
- * 
- * You also need structures and their corresponding pb_field_t descriptions.
- * These are usually generated from .proto-files with a script.
- */
-
-#include <stdbool.h>
 #include "pb.h"
 
-/* Lightweight input stream.
- * You can provide a callback function for reading or use
- * pb_istream_from_buffer.
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/***************************
+ * Main decoding functions *
+ ***************************/
+ 
+/* Decode a single protocol buffers message from input stream into a C structure.
+ * Returns true on success, false on any failure.
+ * The actual struct pointed to by dest must match the description in fields.
+ * Callback fields of the destination structure must be initialized by caller.
+ * All other fields will be initialized by this function.
+ *
+ * Example usage:
+ *    MyMessage msg = {};
+ *    uint8_t buffer[64];
+ *    pb_istream_t stream;
+ *    
+ *    // ... read some data into buffer ...
+ *
+ *    stream = pb_istream_from_buffer(buffer, count);
+ *    pb_decode(&stream, MyMessage_fields, &msg);
+ */
+bool pb_decode(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct);
+
+/* Same as pb_decode, except does not initialize the destination structure
+ * to default values. This is slightly faster if you need no default values
+ * and just do memset(struct, 0, sizeof(struct)) yourself.
+ *
+ * This can also be used for 'merging' two messages, i.e. update only the
+ * fields that exist in the new message.
+ */
+bool pb_decode_noinit(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct);
+
+/* Same as pb_decode, except expects the stream to start with the message size
+ * encoded as varint. Corresponds to parseDelimitedFrom() in Google's
+ * protobuf API.
+ */
+bool pb_decode_delimited(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct);
+
+
+/**************************************
+ * Functions for manipulating streams *
+ **************************************/
+
+/* Create an input stream for reading from a memory buffer.
+ *
+ * Alternatively, you can use a custom stream that reads directly from e.g.
+ * a file or a network socket.
+ */
+pb_istream_t pb_istream_from_buffer(uint8_t *buf, size_t bufsize);
+
+/* Function to read from a pb_istream_t. You can use this if you need to
+ * read some custom header data, or to read data in field callbacks.
+ */
+bool pb_read(pb_istream_t *stream, uint8_t *buf, size_t count);
+
+/* Structure for defining custom input streams. You will need to provide
+ * a callback function to read the bytes from your storage, which can be
+ * for example a file or a network socket.
+ * 
+ * The callback must conform to these rules:
  *
- * Rules for callback:
  * 1) Return false on IO errors. This will cause decoding to abort.
- * 
- * 2) If buf is NULL, read but don't store bytes ("skip input").
- * 
- * 3) You can use state to store your own data (e.g. buffer pointer),
+ * 2) You can use state to store your own data (e.g. buffer pointer),
  *    and rely on pb_read to verify that no-body reads past bytes_left.
- * 
- * 4) Your callback may be used with substreams, in which case bytes_left
+ * 3) Your callback may be used with substreams, in which case bytes_left
  *    is different than from the main stream. Don't use bytes_left to compute
  *    any pointers.
  */
 struct _pb_istream_t
 {
+#ifdef PB_BUFFER_ONLY
+    /* Callback pointer is not used in buffer-only configuration.
+     * Having an int pointer here allows binary compatibility but
+     * gives an error if someone tries to assign callback function.
+     */
+    int *callback;
+#else
     bool (*callback)(pb_istream_t *stream, uint8_t *buf, size_t count);
+#endif
+
     void *state; /* Free field for use by callback implementation */
     size_t bytes_left;
+    
+#ifndef PB_NO_ERRMSG
+    const char *errmsg;
+#endif
 };
 
-pb_istream_t pb_istream_from_buffer(uint8_t *buf, size_t bufsize);
-bool pb_read(pb_istream_t *stream, uint8_t *buf, size_t count);
 
-/* Decode from stream to destination struct.
- * Returns true on success, false on any failure.
- * The actual struct pointed to by dest must match the description in fields.
- */
-bool pb_decode(pb_istream_t *stream, const pb_field_t fields[], void *dest_struct);
+/************************************************
+ * Helper functions for writing field callbacks *
+ ************************************************/
 
-/* --- Helper functions ---
- * You may want to use these from your caller or callbacks.
- */
+/* Decode the tag for the next field in the stream. Gives the wire type and
+ * field tag. At end of the message, returns false and sets eof to true. */
+bool pb_decode_tag(pb_istream_t *stream, pb_wire_type_t *wire_type, uint32_t *tag, bool *eof);
 
-bool pb_decode_tag(pb_istream_t *stream, pb_wire_type_t *wire_type, int *tag, bool *eof);
+/* Skip the field payload data, given the wire type. */
 bool pb_skip_field(pb_istream_t *stream, pb_wire_type_t wire_type);
 
+/* Decode an integer in the varint format. This works for bool, enum, int32,
+ * int64, uint32 and uint64 field types. */
 bool pb_decode_varint(pb_istream_t *stream, uint64_t *dest);
 
-bool pb_skip_varint(pb_istream_t *stream);
-bool pb_skip_string(pb_istream_t *stream);
+/* Decode an integer in the zig-zagged svarint format. This works for sint32
+ * and sint64. */
+bool pb_decode_svarint(pb_istream_t *stream, int64_t *dest);
 
-/* --- Field decoders ---
- * Each decoder takes stream and field description, and a pointer to the field
- * in the destination struct (dest = struct_addr + field->data_offset).
- * For arrays, these functions are called repeatedly.
- */
+/* Decode a fixed32, sfixed32 or float value. You need to pass a pointer to
+ * a 4-byte wide C variable. */
+bool pb_decode_fixed32(pb_istream_t *stream, void *dest);
 
+/* Decode a fixed64, sfixed64 or double value. You need to pass a pointer to
+ * a 8-byte wide C variable. */
+bool pb_decode_fixed64(pb_istream_t *stream, void *dest);
+
+/* Make a limited-length substream for reading a PB_WT_STRING field. */
+bool pb_make_string_substream(pb_istream_t *stream, pb_istream_t *substream);
+void pb_close_string_substream(pb_istream_t *stream, pb_istream_t *substream);
+
+
+/*******************************
+ * Internal / legacy functions *
+ *******************************/
+
+#ifdef NANOPB_INTERNALS
 bool pb_dec_varint(pb_istream_t *stream, const pb_field_t *field, void *dest);
 bool pb_dec_svarint(pb_istream_t *stream, const pb_field_t *field, void *dest);
 bool pb_dec_fixed32(pb_istream_t *stream, const pb_field_t *field, void *dest);
@@ -71,4 +146,12 @@ bool pb_dec_bytes(pb_istream_t *stream, const pb_field_t *field, void *dest);
 bool pb_dec_string(pb_istream_t *stream, const pb_field_t *field, void *dest);
 bool pb_dec_submessage(pb_istream_t *stream, const pb_field_t *field, void *dest);
 
+bool pb_skip_varint(pb_istream_t *stream);
+bool pb_skip_string(pb_istream_t *stream);
+#endif
+
+#ifdef __cplusplus
+} /* extern "C" */
+#endif
+
 #endif

pb_encode.c

@@ -6,16 +6,15 @@
 #define NANOPB_INTERNALS
 #include "pb.h"
 #include "pb_encode.h"
-#include <string.h>
 
-#ifdef __GNUC__
+/* The warn_unused_result attribute appeared first in gcc-3.4.0 */
+#if !defined(__GNUC__) || ( __GNUC__ < 3) || (__GNUC__ == 3 && __GNUC_MINOR__ < 4)
+    #define checkreturn
+#else
     /* Verify that we remember to check all return values for proper error propagation */
     #define checkreturn __attribute__((warn_unused_result))
-#else
-#define checkreturn
 #endif
 
-
 typedef bool (*pb_encoder_t)(pb_ostream_t *stream, const pb_field_t *field, const void *src) checkreturn;
 
 /* --- Function pointers to field encoders ---
@@ -29,7 +28,8 @@ static const pb_encoder_t PB_ENCODERS[PB_LTYPES_COUNT] = {
     
     &pb_enc_bytes,
     &pb_enc_string,
-    &pb_enc_submessage
+    &pb_enc_submessage,
+    NULL /* extensions */
 };
 
 /* pb_ostream_t implementation */
@@ -37,18 +37,28 @@ static const pb_encoder_t PB_ENCODERS[PB_LTYPES_COUNT] = {
 static bool checkreturn buf_write(pb_ostream_t *stream, const uint8_t *buf, size_t count)
 {
     uint8_t *dest = (uint8_t*)stream->state;
-    memcpy(dest, buf, count);
     stream->state = dest + count;
+    
+    while (count--)
+        *dest++ = *buf++;
+    
     return true;
 }
 
 pb_ostream_t pb_ostream_from_buffer(uint8_t *buf, size_t bufsize)
 {
     pb_ostream_t stream;
+#ifdef PB_BUFFER_ONLY
+    stream.callback = (void*)1; /* Just some marker value */
+#else
     stream.callback = &buf_write;
+#endif
     stream.state = buf;
     stream.max_size = bufsize;
     stream.bytes_written = 0;
+#ifndef PB_NO_ERRMSG
+    stream.errmsg = NULL;
+#endif
     return stream;
 }
 
@@ -57,10 +67,15 @@ bool checkreturn pb_write(pb_ostream_t *stream, const uint8_t *buf, size_t count
     if (stream->callback != NULL)
     {
         if (stream->bytes_written + count > stream->max_size)
-            return false;
+            PB_RETURN_ERROR(stream, "stream full");
 
+#ifdef PB_BUFFER_ONLY
+        if (!buf_write(stream, buf, count))
+            PB_RETURN_ERROR(stream, "io error");
+#else        
         if (!stream->callback(stream, buf, count))
-            return false;
+            PB_RETURN_ERROR(stream, "io error");
+#endif
     }
     
     stream->bytes_written += count;
@@ -69,10 +84,7 @@ bool checkreturn pb_write(pb_ostream_t *stream, const uint8_t *buf, size_t count
 
 /* Main encoding stuff */
 
-/* Callbacks don't need this function because they usually know the data type
- * without examining the field structure.
- * Therefore it is static for now.
- */
+/* Encode a static array. Handles the size calculations and possible packing. */
 static bool checkreturn encode_array(pb_ostream_t *stream, const pb_field_t *field,
                          const void *pData, size_t count, pb_encoder_t func)
 {
@@ -83,6 +95,7 @@ static bool checkreturn encode_array(pb_ostream_t *stream, const pb_field_t *fie
     if (count == 0)
         return true;
     
+    /* We always pack arrays if the datatype allows it. */
     if (PB_LTYPE(field->type) <= PB_LTYPE_LAST_PACKABLE)
     {
         if (!pb_encode_tag(stream, PB_WT_STRING, field->tag))
@@ -99,7 +112,7 @@ static bool checkreturn encode_array(pb_ostream_t *stream, const pb_field_t *fie
         }
         else
         { 
-            pb_ostream_t sizestream = {0};
+            pb_ostream_t sizestream = PB_OSTREAM_SIZING;
             p = pData;
             for (i = 0; i < count; i++)
             {
@@ -141,22 +154,21 @@ static bool checkreturn encode_array(pb_ostream_t *stream, const pb_field_t *fie
     return true;
 }
 
-bool checkreturn pb_encode(pb_ostream_t *stream, const pb_field_t fields[], const void *src_struct)
+/* Encode a field with static allocation, i.e. one whose data is stored
+ * in the structure itself. */
+static bool checkreturn encode_static_field(pb_ostream_t *stream,
+    const pb_field_t *field, const void *pData)
 {
-    const pb_field_t *field = fields;
-    const void *pData = src_struct;
+    pb_encoder_t func;
     const void *pSize;
-    size_t prev_size = 0;
+    bool dummy = true;
     
-    while (field->tag != 0)
-    {
-        pb_encoder_t func = PB_ENCODERS[PB_LTYPE(field->type)];
-        pData = (const char*)pData + prev_size + field->data_offset;
+    func = PB_ENCODERS[PB_LTYPE(field->type)];
+    
+    if (field->size_offset)
         pSize = (const char*)pData + field->size_offset;
-        
-        prev_size = field->data_size;
-        if (PB_HTYPE(field->type) == PB_HTYPE_ARRAY)
-            prev_size *= field->array_size;
+    else
+        pSize = &dummy;
     
     switch (PB_HTYPE(field->type))
     {
@@ -168,7 +180,7 @@ bool checkreturn pb_encode(pb_ostream_t *stream, const pb_field_t fields[], cons
             break;
         
         case PB_HTYPE_OPTIONAL:
-                if (*(bool*)pSize)
+            if (*(const bool*)pSize)
             {
                 if (!pb_encode_tag_for_field(stream, field))
                     return false;
@@ -178,21 +190,119 @@ bool checkreturn pb_encode(pb_ostream_t *stream, const pb_field_t fields[], cons
             }
             break;
         
-            case PB_HTYPE_ARRAY:
-                if (!encode_array(stream, field, pData, *(size_t*)pSize, func))
+        case PB_HTYPE_REPEATED:
+            if (!encode_array(stream, field, pData, *(const size_t*)pSize, func))
                 return false;
             break;
         
-            case PB_HTYPE_CALLBACK:
+        default:
+            PB_RETURN_ERROR(stream, "invalid field type");
+    }
+    
+    return true;
+}
+
+/* Encode a field with callback semantics. This means that a user function is
+ * called to provide and encode the actual data. */
+static bool checkreturn encode_callback_field(pb_ostream_t *stream,
+    const pb_field_t *field, const void *pData)
 {
-                pb_callback_t *callback = (pb_callback_t*)pData;
+    const pb_callback_t *callback = (const pb_callback_t*)pData;
+    
+#ifdef PB_OLD_CALLBACK_STYLE
+    const void *arg = callback->arg;
+#else
+    void * const *arg = &(callback->arg);
+#endif    
+    
     if (callback->funcs.encode != NULL)
     {
-                    if (!callback->funcs.encode(stream, field, callback->arg))
+        if (!callback->funcs.encode(stream, field, arg))
+            PB_RETURN_ERROR(stream, "callback error");
+    }
+    return true;
+}
+
+/* Encode a single field of any callback or static type. */
+static bool checkreturn encode_field(pb_ostream_t *stream,
+    const pb_field_t *field, const void *pData)
+{
+    switch (PB_ATYPE(field->type))
+    {
+        case PB_ATYPE_STATIC:
+            return encode_static_field(stream, field, pData);
+        
+        case PB_ATYPE_CALLBACK:
+            return encode_callback_field(stream, field, pData);
+        
+        default:
+            PB_RETURN_ERROR(stream, "invalid field type");
+    }
+}
+
+/* Default handler for extension fields. Expects to have a pb_field_t
+ * pointer in the extension->type->arg field. */
+static bool checkreturn default_extension_handler(pb_ostream_t *stream,
+    const pb_extension_t *extension)
+{
+    const pb_field_t *field = (const pb_field_t*)extension->type->arg;
+    return encode_field(stream, field, extension->dest);
+}
+
+/* Walk through all the registered extensions and give them a chance
+ * to encode themselves. */
+static bool checkreturn encode_extension_field(pb_ostream_t *stream,
+    const pb_field_t *field, const void *pData)
+{
+    const pb_extension_t *extension = *(const pb_extension_t* const *)pData;
+    UNUSED(field);
+    
+    while (extension)
+    {
+        bool status;
+        if (extension->type->encode)
+            status = extension->type->encode(stream, extension);
+        else
+            status = default_extension_handler(stream, extension);
+
+        if (!status)
+            return false;
+        
+        extension = extension->next;
+    }
+    
+    return true;
+}
+
+bool checkreturn pb_encode(pb_ostream_t *stream, const pb_field_t fields[], const void *src_struct)
+{
+    const pb_field_t *field = fields;
+    const void *pData = src_struct;
+    size_t prev_size = 0;
+    
+    while (field->tag != 0)
+    {
+        pData = (const char*)pData + prev_size + field->data_offset;
+        prev_size = field->data_size;
+        
+        /* Special case for static arrays */
+        if (PB_ATYPE(field->type) == PB_ATYPE_STATIC &&
+            PB_HTYPE(field->type) == PB_HTYPE_REPEATED)
+        {
+            prev_size *= field->array_size;
+        }
+        
+        if (PB_LTYPE(field->type) == PB_LTYPE_EXTENSION)
+        {
+            /* Special case for the extension field placeholder */
+            if (!encode_extension_field(stream, field, pData))
                 return false;
         }
-                break;
-            }
+        else
+        {
+            /* Regular field */
+            if (!encode_field(stream, field, pData))
+                return false;
         }
     
         field++;
@@ -201,11 +311,16 @@ bool checkreturn pb_encode(pb_ostream_t *stream, const pb_field_t fields[], cons
     return true;
 }
 
+bool pb_encode_delimited(pb_ostream_t *stream, const pb_field_t fields[], const void *src_struct)
+{
+    return pb_encode_submessage(stream, fields, src_struct);
+}
+
 /* Helper functions */
 bool checkreturn pb_encode_varint(pb_ostream_t *stream, uint64_t value)
 {
     uint8_t buffer[10];
-    int i = 0;
+    size_t i = 0;
     
     if (value == 0)
         return pb_write(stream, (uint8_t*)&value, 1);
@@ -225,9 +340,9 @@ bool checkreturn pb_encode_svarint(pb_ostream_t *stream, int64_t value)
 {
     uint64_t zigzagged;
     if (value < 0)
-        zigzagged = ~(value << 1);
+        zigzagged = (uint64_t)(~(value << 1));
     else
-        zigzagged = value << 1;
+        zigzagged = (uint64_t)(value << 1);
     
     return pb_encode_varint(stream, zigzagged);
 }
@@ -243,7 +358,7 @@ bool checkreturn pb_encode_fixed32(pb_ostream_t *stream, const void *value)
     lebytes[3] = bytes[0];
     return pb_write(stream, lebytes, 4);
     #else
-    return pb_write(stream, (uint8_t*)value, 4);
+    return pb_write(stream, (const uint8_t*)value, 4);
     #endif
 }
 
@@ -262,14 +377,14 @@ bool checkreturn pb_encode_fixed64(pb_ostream_t *stream, const void *value)
     lebytes[7] = bytes[0];
     return pb_write(stream, lebytes, 8);
     #else
-    return pb_write(stream, (uint8_t*)value, 8);
+    return pb_write(stream, (const uint8_t*)value, 8);
     #endif
 }
 
-bool checkreturn pb_encode_tag(pb_ostream_t *stream, pb_wire_type_t wiretype, int field_number)
+bool checkreturn pb_encode_tag(pb_ostream_t *stream, pb_wire_type_t wiretype, uint32_t field_number)
 {
-    int tag = wiretype | (field_number << 3);
-    return pb_encode_varint(stream, (uint64_t)tag);
+    uint64_t tag = wiretype | (field_number << 3);
+    return pb_encode_varint(stream, tag);
 }
 
 bool checkreturn pb_encode_tag_for_field(pb_ostream_t *stream, const pb_field_t *field)
@@ -297,7 +412,7 @@ bool checkreturn pb_encode_tag_for_field(pb_ostream_t *stream, const pb_field_t
             break;
         
         default:
-            return false;
+            PB_RETURN_ERROR(stream, "invalid field type");
     }
     
     return pb_encode_tag(stream, wiretype, field->tag);
@@ -314,7 +429,7 @@ bool checkreturn pb_encode_string(pb_ostream_t *stream, const uint8_t *buffer, s
 bool checkreturn pb_encode_submessage(pb_ostream_t *stream, const pb_field_t fields[], const void *src_struct)
 {
     /* First calculate the message size using a non-writing substream. */
-    pb_ostream_t substream = {0};
+    pb_ostream_t substream = PB_OSTREAM_SIZING;
     size_t size;
     bool status;
     
@@ -330,7 +445,7 @@ bool checkreturn pb_encode_submessage(pb_ostream_t *stream, const pb_field_t fie
         return pb_write(stream, NULL, size); /* Just sizing */
     
     if (stream->bytes_written + size > stream->max_size)
-        return false;
+        PB_RETURN_ERROR(stream, "stream full");
         
     /* Use a substream to verify that a callback doesn't write more than
      * what it did the first time. */
@@ -338,14 +453,20 @@ bool checkreturn pb_encode_submessage(pb_ostream_t *stream, const pb_field_t fie
     substream.state = stream->state;
     substream.max_size = size;
     substream.bytes_written = 0;
+#ifndef PB_NO_ERRMSG
+    substream.errmsg = NULL;
+#endif
     
     status = pb_encode(&substream, fields, src_struct);
     
     stream->bytes_written += substream.bytes_written;
     stream->state = substream.state;
+#ifndef PB_NO_ERRMSG
+    stream->errmsg = substream.errmsg;
+#endif
     
     if (substream.bytes_written != size)
-        return false;
+        PB_RETURN_ERROR(stream, "submsg size changed");
     
     return status;
 }
@@ -358,11 +479,11 @@ bool checkreturn pb_enc_varint(pb_ostream_t *stream, const pb_field_t *field, co
     
     switch (field->data_size)
     {
-        case 1: value = *(uint8_t*)src; break;
-        case 2: value = *(uint16_t*)src; break;
-        case 4: value = *(uint32_t*)src; break;
-        case 8: value = *(uint64_t*)src; break;
-        default: return false;
+        case 1: value = *(const uint8_t*)src; break;
+        case 2: value = *(const uint16_t*)src; break;
+        case 4: value = *(const uint32_t*)src; break;
+        case 8: value = *(const uint64_t*)src; break;
+        default: PB_RETURN_ERROR(stream, "invalid data_size");
     }
     
     return pb_encode_varint(stream, value);
@@ -370,13 +491,13 @@ bool checkreturn pb_enc_varint(pb_ostream_t *stream, const pb_field_t *field, co
 
 bool checkreturn pb_enc_svarint(pb_ostream_t *stream, const pb_field_t *field, const void *src)
 {
-    uint64_t value = 0;
+    int64_t value = 0;
     
     switch (field->data_size)
     {
-        case 4: value = *(int32_t*)src; break;
-        case 8: value = *(int64_t*)src; break;
-        default: return false;
+        case 4: value = *(const int32_t*)src; break;
+        case 8: value = *(const int64_t*)src; break;
+        default: PB_RETURN_ERROR(stream, "invalid data_size");
     }
     
     return pb_encode_svarint(stream, value);
@@ -396,22 +517,30 @@ bool checkreturn pb_enc_fixed32(pb_ostream_t *stream, const pb_field_t *field, c
 
 bool checkreturn pb_enc_bytes(pb_ostream_t *stream, const pb_field_t *field, const void *src)
 {
-    pb_bytes_array_t *bytes = (pb_bytes_array_t*)src;
+    const pb_bytes_array_t *bytes = (const pb_bytes_array_t*)src;
     UNUSED(field);
     return pb_encode_string(stream, bytes->bytes, bytes->size);
 }
 
 bool checkreturn pb_enc_string(pb_ostream_t *stream, const pb_field_t *field, const void *src)
 {
-    UNUSED(field);
-    return pb_encode_string(stream, (uint8_t*)src, strlen((char*)src));
+    /* strnlen() is not always available, so just use a for-loop */
+    size_t size = 0;
+    const char *p = (const char*)src;
+    while (size < field->data_size && *p != '\0')
+    {
+        size++;
+        p++;
+    }
+
+    return pb_encode_string(stream, (const uint8_t*)src, size);
 }
 
 bool checkreturn pb_enc_submessage(pb_ostream_t *stream, const pb_field_t *field, const void *src)
 {
     if (field->ptr == NULL)
-        return false;
+        PB_RETURN_ERROR(stream, "invalid field descriptor");
     
-    return pb_encode_submessage(stream, (pb_field_t*)field->ptr, src);
+    return pb_encode_submessage(stream, (const pb_field_t*)field->ptr, src);
 }
 

pb_encode.h

@@ -1,60 +1,121 @@
+/* pb_encode.h: Functions to encode protocol buffers. Depends on pb_encode.c.
+ * The main function is pb_encode. You also need an output stream, and the
+ * field descriptions created by nanopb_generator.py.
+ */
+
 #ifndef _PB_ENCODE_H_
 #define _PB_ENCODE_H_
 
-/* pb_encode.h: Functions to encode protocol buffers. Depends on pb_encode.c.
- * The main function is pb_encode. You also need an output stream, structures
- * and their field descriptions (just like with pb_decode).
- */
-
-#include <stdbool.h>
 #include "pb.h"
 
-/* Lightweight output stream.
- * You can provide callback for writing or use pb_ostream_from_buffer.
- * 
- * Alternatively, callback can be NULL in which case the stream will just
- * count the number of bytes that would have been written. In this case
- * max_size is not checked.
- *
- * Rules for callback:
- * 1) Return false on IO errors. This will cause encoding to abort.
- * 
- * 2) You can use state to store your own data (e.g. buffer pointer).
- * 
- * 3) pb_write will update bytes_written after your callback runs.
- * 
- * 4) Substreams will modify max_size and bytes_written. Don't use them to
- * calculate any pointers.
- */
-struct _pb_ostream_t
-{
-    bool (*callback)(pb_ostream_t *stream, const uint8_t *buf, size_t count);
-    void *state; /* Free field for use by callback implementation */
-    size_t max_size; /* Limit number of output bytes written (or use SIZE_MAX). */
-    size_t bytes_written;
-};
+#ifdef __cplusplus
+extern "C" {
+#endif
 
-pb_ostream_t pb_ostream_from_buffer(uint8_t *buf, size_t bufsize);
-bool pb_write(pb_ostream_t *stream, const uint8_t *buf, size_t count);
+/***************************
+ * Main encoding functions *
+ ***************************/
 
-/* Encode struct to given output stream.
+/* Encode a single protocol buffers message from C structure into a stream.
  * Returns true on success, false on any failure.
  * The actual struct pointed to by src_struct must match the description in fields.
  * All required fields in the struct are assumed to have been filled in.
+ *
+ * Example usage:
+ *    MyMessage msg = {};
+ *    uint8_t buffer[64];
+ *    pb_ostream_t stream;
+ *
+ *    msg.field1 = 42;
+ *    stream = pb_ostream_from_buffer(buffer, sizeof(buffer));
+ *    pb_encode(&stream, MyMessage_fields, &msg);
  */
 bool pb_encode(pb_ostream_t *stream, const pb_field_t fields[], const void *src_struct);
 
-/* --- Helper functions ---
- * You may want to use these from your caller or callbacks.
+/* Same as pb_encode, but prepends the length of the message as a varint.
+ * Corresponds to writeDelimitedTo() in Google's protobuf API.
  */
+bool pb_encode_delimited(pb_ostream_t *stream, const pb_field_t fields[], const void *src_struct);
 
-/* Encode field header based on LTYPE and field number defined in the field structure.
- * Call this from the callback before writing out field contents. */
+/**************************************
+ * Functions for manipulating streams *
+ **************************************/
+
+/* Create an output stream for writing into a memory buffer.
+ * The number of bytes written can be found in stream.bytes_written after
+ * encoding the message.
+ *
+ * Alternatively, you can use a custom stream that writes directly to e.g.
+ * a file or a network socket.
+ */
+pb_ostream_t pb_ostream_from_buffer(uint8_t *buf, size_t bufsize);
+
+/* Pseudo-stream for measuring the size of a message without actually storing
+ * the encoded data.
+ * 
+ * Example usage:
+ *    MyMessage msg = {};
+ *    pb_ostream_t stream = PB_OSTREAM_SIZING;
+ *    pb_encode(&stream, MyMessage_fields, &msg);
+ *    printf("Message size is %d\n", stream.bytes_written);
+ */
+#ifndef PB_NO_ERRMSG
+#define PB_OSTREAM_SIZING {0,0,0,0,0}
+#else
+#define PB_OSTREAM_SIZING {0,0,0,0}
+#endif
+
+/* Function to write into a pb_ostream_t stream. You can use this if you need
+ * to append or prepend some custom headers to the message.
+ */
+bool pb_write(pb_ostream_t *stream, const uint8_t *buf, size_t count);
+
+/* Structure for defining custom output streams. You will need to provide
+ * a callback function to write the bytes to your storage, which can be
+ * for example a file or a network socket.
+ *
+ * The callback must conform to these rules:
+ *
+ * 1) Return false on IO errors. This will cause encoding to abort.
+ * 2) You can use state to store your own data (e.g. buffer pointer).
+ * 3) pb_write will update bytes_written after your callback runs.
+ * 4) Substreams will modify max_size and bytes_written. Don't use them
+ *    to calculate any pointers.
+ */
+struct _pb_ostream_t
+{
+#ifdef PB_BUFFER_ONLY
+    /* Callback pointer is not used in buffer-only configuration.
+     * Having an int pointer here allows binary compatibility but
+     * gives an error if someone tries to assign callback function.
+     * Also, NULL pointer marks a 'sizing stream' that does not
+     * write anything.
+     */
+    int *callback;
+#else
+    bool (*callback)(pb_ostream_t *stream, const uint8_t *buf, size_t count);
+#endif
+    void *state;          /* Free field for use by callback implementation. */
+    size_t max_size;      /* Limit number of output bytes written (or use SIZE_MAX). */
+    size_t bytes_written; /* Number of bytes written so far. */
+    
+#ifndef PB_NO_ERRMSG
+    const char *errmsg;
+#endif
+};
+
+
+/************************************************
+ * Helper functions for writing field callbacks *
+ ************************************************/
+
+/* Encode field header based on type and field number defined in the field
+ * structure. Call this from the callback before writing out field contents. */
 bool pb_encode_tag_for_field(pb_ostream_t *stream, const pb_field_t *field);
 
-/* Encode field header by manually specifing wire type. You need to use this if
- * you want to write out packed arrays from a callback field. */
-bool pb_encode_tag(pb_ostream_t *stream, pb_wire_type_t wiretype, int field_number);
+/* Encode field header by manually specifing wire type. You need to use this
+ * if you want to write out packed arrays from a callback field. */
+bool pb_encode_tag(pb_ostream_t *stream, pb_wire_type_t wiretype, uint32_t field_number);
 
 /* Encode an integer in the varint format.
  * This works for bool, enum, int32, int64, uint32 and uint64 field types. */
@@ -76,15 +137,16 @@ bool pb_encode_fixed32(pb_ostream_t *stream, const void *value);
 bool pb_encode_fixed64(pb_ostream_t *stream, const void *value);
 
 /* Encode a submessage field.
- * You need to pass the pb_field_t array and pointer to struct, just like with pb_encode().
- * This internally encodes the submessage twice, first to calculate message size and then to actually write it out.
+ * You need to pass the pb_field_t array and pointer to struct, just like
+ * with pb_encode(). This internally encodes the submessage twice, first to
+ * calculate message size and then to actually write it out.
  */
 bool pb_encode_submessage(pb_ostream_t *stream, const pb_field_t fields[], const void *src_struct);
 
-/* --- Internal functions ---
- * These functions are not terribly useful for the average library user, but
- * are exported to make the unit testing and extending nanopb easier.
- */
+
+/*******************************
+ * Internal / legacy functions *
+ *******************************/
 
 #ifdef NANOPB_INTERNALS
 bool pb_enc_varint(pb_ostream_t *stream, const pb_field_t *field, const void *src);
@@ -99,5 +161,8 @@ bool pb_enc_string(pb_ostream_t *stream, const pb_field_t *field, const void *sr
  * instead, it has the same functionality with a less confusing interface. */
 bool pb_enc_submessage(pb_ostream_t *stream, const pb_field_t *field, const void *src);
 
+#ifdef __cplusplus
+} /* extern "C" */
+#endif
 
 #endif

tests/Makefile

@@ -1,34 +1,91 @@
-CFLAGS=-ansi -Wall -Werror -I .. -g -O0 --coverage
-LDFLAGS=--coverage
-DEPS=../pb_decode.h ../pb_encode.h ../pb.h person.pb.h callbacks.pb.h unittests.h unittestproto.pb.h alltypes.pb.h missing_fields.pb.h
-TESTS=test_decode1 test_encode1 decode_unittests encode_unittests test_no_messages
+CFLAGS=-ansi -Wall -Werror -I .. -g -O0
+DEPS=../pb_decode.h ../pb_encode.h ../pb.h person.pb.h \
+	callbacks2.pb.h callbacks.pb.h unittests.h unittestproto.pb.h \
+	alltypes.pb.h missing_fields.pb.h
+TESTS=  decode_unittests encode_unittests \
+	test_decode1 test_decode2 test_decode3 test_decode3_buf \
+	test_encode1 test_encode2 test_encode3 test_encode3_buf \
+	test_decode_callbacks test_encode_callbacks \
+	test_missing_fields test_no_messages test_funny_name \
+	test_multiple_files test_cxxcompile test_options \
+	bc_encode bc_decode test_encode_extensions test_decode_extensions
+
+# More strict checks for the core part of nanopb
+CC_VERSION=$(shell $(CC) -v 2>&1)
+CFLAGS_CORE=
+ifneq (,$(findstring gcc,$(CC_VERSION)))
+	CFLAGS_CORE=-pedantic -Wextra -Wcast-qual -Wlogical-op -Wconversion
+	CFLAGS+=--coverage -fstack-protector-all
+	LDFLAGS+=--coverage
+endif
+ifneq (,$(findstring clang,$(CC_VERSION)))
+	CFLAGS_CORE=-pedantic -Wextra -Wcast-qual -Wconversion
+endif
+
+# Also use mudflap if it is available
+# To enable, run with    make -B USE_MUDFLAP=y
+USE_MUDFLAP ?= n
+ifeq ($(USE_MUDFLAP),y)
+	CFLAGS += -fmudflap
+	LDFLAGS += -lmudflap -fmudflap
+endif
 
 all: breakpoints $(TESTS) run_unittests
 
 clean:
-	rm -f $(TESTS) person.pb* alltypes.pb* *.o *.gcda *.gcno
+	rm -f $(TESTS) person.pb* alltypes.pb* *.o *.gcda *.gcno *.pb.h *.pb.c
+
+%.pb.o: %.pb.c %.pb.h
+	$(CC) $(CFLAGS) $(CFLAGS_CORE) -c -o $@ $<
 
 %.o: %.c
 %.o: %.c $(DEPS)
 	$(CC) $(CFLAGS) -c -o $@ $<
 
 pb_encode.o: ../pb_encode.c $(DEPS)
-	$(CC) $(CFLAGS) -c -o $@ $<
+	$(CC) $(CFLAGS) $(CFLAGS_CORE) -c -o $@ $<
 pb_decode.o: ../pb_decode.c $(DEPS)
-	$(CC) $(CFLAGS) -c -o $@ $<
+	$(CC) $(CFLAGS) $(CFLAGS_CORE) -c -o $@ $<
 
+# Test for compilability with c++ compiler
+
+pb_encode.cxx.o: ../pb_encode.c $(DEPS)
+	$(CXX) $(CFLAGS) $(CFLAGS_CORE) -c -o $@ $<
+pb_decode.cxx.o: ../pb_decode.c $(DEPS)
+	$(CXX) $(CFLAGS) $(CFLAGS_CORE) -c -o $@ $<
+
+# Test for PB_BUF_ONLY compilation option
+
+pb_encode.buf.o: ../pb_encode.c $(DEPS)
+	$(CC) -DPB_BUFFER_ONLY $(CFLAGS) $(CFLAGS_CORE) -c -o $@ $<
+pb_decode.buf.o: ../pb_decode.c $(DEPS)
+	$(CC) -DPB_BUFFER_ONLY $(CFLAGS) $(CFLAGS_CORE) -c -o $@ $<
+%.buf.o: %.c $(DEPS)
+	$(CC) -DPB_BUFFER_ONLY $(CFLAGS) -c -o $@ $<	
+test_encode3_buf: test_encode3.buf.o pb_encode.buf.o alltypes.pb.o
+	$(CC) $(LDFLAGS) $^ -o $@
+test_decode3_buf: test_decode3.buf.o pb_decode.buf.o alltypes.pb.o
+	$(CC) $(LDFLAGS) $^ -o $@
+
+test_cxxcompile: pb_encode.cxx.o pb_decode.cxx.o
 test_decode1: test_decode1.o pb_decode.o person.pb.o
 test_decode2: test_decode2.o pb_decode.o person.pb.o
 test_decode3: test_decode3.o pb_decode.o alltypes.pb.o
 test_encode1: test_encode1.o pb_encode.o person.pb.o
 test_encode2: test_encode2.o pb_encode.o person.pb.o
 test_encode3: test_encode3.o pb_encode.o alltypes.pb.o
+test_multiple_files: test_multiple_files.o pb_encode.o callbacks2.pb.o callbacks.pb.o
 test_decode_callbacks: test_decode_callbacks.o pb_decode.o callbacks.pb.o
 test_encode_callbacks: test_encode_callbacks.o pb_encode.o callbacks.pb.o
 test_missing_fields: test_missing_fields.o pb_encode.o pb_decode.o missing_fields.pb.o
 decode_unittests: decode_unittests.o pb_decode.o unittestproto.pb.o
 encode_unittests: encode_unittests.o pb_encode.o unittestproto.pb.o
 test_no_messages: no_messages.pb.h no_messages.pb.c no_messages.pb.o
+test_funny_name: funny-proto+name.pb.h funny-proto+name.pb.o
+bc_encode: bc_alltypes.pb.o pb_encode.o bc_encode.o
+bc_decode: bc_alltypes.pb.o pb_decode.o bc_decode.o
+test_encode_extensions: test_encode_extensions.c pb_encode.o alltypes.pb.o extensions.pb.o
+test_decode_extensions: test_decode_extensions.c pb_decode.o alltypes.pb.o extensions.pb.o
 
 %.pb: %.proto
 	protoc -I. -I../generator -I/usr/include -o$@ $<
@@ -43,7 +100,7 @@ coverage: run_unittests
 	gcov pb_encode.gcda
 	gcov pb_decode.gcda
 
-run_unittests: decode_unittests encode_unittests test_encode1 test_encode2 test_encode3 test_decode1 test_decode2 test_decode3 test_encode_callbacks test_decode_callbacks test_missing_fields
+run_unittests: $(TESTS)
 	rm -f *.gcda
 	
 	./decode_unittests > /dev/null
@@ -58,13 +115,29 @@ run_unittests: decode_unittests encode_unittests test_encode1 test_encode2 test_
 	[ "`./test_encode2 | ./test_decode2`" = \
 	"`./test_encode2 | protoc --decode=Person -I. -I../generator -I/usr/include person.proto`" ]
 	
+	[ "`./test_decode2 < person_with_extra_field.pb`" = \
+	"`./test_encode2 | ./test_decode2`" ]
+	
 	[ "`./test_encode_callbacks | ./test_decode_callbacks`" = \
 	"`./test_encode_callbacks | protoc --decode=TestMessage callbacks.proto`" ]
 
 	./test_encode3 | ./test_decode3
-	./test_encode3 | protoc --decode=AllTypes -I. -I../generator -I/usr/include alltypes.proto >/dev/null
+	./test_encode3 1 | ./test_decode3 1
+	./test_encode3 1 | protoc --decode=AllTypes -I. -I../generator -I/usr/include alltypes.proto >/dev/null
+	./test_encode3_buf 1 | ./test_decode3_buf 1
+	./test_decode3 < alltypes_with_extra_fields.pb
+	./bc_encode | ./bc_decode
+	./test_encode_extensions | ./test_decode_extensions
 
 	./test_missing_fields
 
-run_fuzztest: test_decode2
-	bash -c 'I=1; while true; do cat /dev/urandom | ./test_decode2 > /dev/null; I=$$(($$I+1)); echo -en "\r$$I"; done'
+test_options: options.pb.h options.expected options.pb.o
+	cat options.expected | while read -r p; do \
+	    if ! grep -q "$$p" $<; then \
+	        echo Expected: "$$p"; \
+	        exit 1; \
+	    fi \
+	done
+
+run_fuzztest: test_decode3
+	bash -c 'ulimit -c unlimited; I=1; while true; do cat /dev/urandom | ./test_decode3 > /dev/null; I=$$(($$I+1)); echo -en "\r$$I"; done'

tests/alltypes.options

@@ -0,0 +1,3 @@
+* max_size:16
+* max_count:5
+

tests/alltypes.proto

@@ -1,11 +1,15 @@
-import "nanopb.proto";
-
 message SubMessage {
-    required string substuff1 = 1 [(nanopb).max_size = 16];
-    required int32 substuff2 = 2;
+    required string substuff1 = 1 [default = "1"];
+    required int32 substuff2 = 2 [default = 2];
+    optional fixed32 substuff3 = 3 [default = 3];
+}
+
+message EmptyMessage {
+
 }
 
 enum MyEnum {
+    Zero = 0;
     First = 1;
     Second = 2;
     Truth = 42;
@@ -28,13 +32,62 @@ message AllTypes {
     required sfixed64   req_sfixed64= 12;
     required double     req_double  = 13;
     
-    required string     req_string  = 14 [(nanopb).max_size = 16];
-    required bytes      req_bytes   = 15 [(nanopb).max_size = 16];
+    required string     req_string  = 14;
+    required bytes      req_bytes   = 15;
     required SubMessage req_submsg  = 16;
     required MyEnum     req_enum    = 17;
+    required EmptyMessage req_emptymsg = 18;
+    
+    
+    repeated int32      rep_int32   = 21;
+    repeated int64      rep_int64   = 22;
+    repeated uint32     rep_uint32  = 23;
+    repeated uint64     rep_uint64  = 24;
+    repeated sint32     rep_sint32  = 25;
+    repeated sint64     rep_sint64  = 26;
+    repeated bool       rep_bool    = 27;
+    
+    repeated fixed32    rep_fixed32 = 28;
+    repeated sfixed32   rep_sfixed32= 29;
+    repeated float      rep_float   = 30;
+    
+    repeated fixed64    rep_fixed64 = 31;
+    repeated sfixed64   rep_sfixed64= 32;
+    repeated double     rep_double  = 33;
+    
+    repeated string     rep_string  = 34;
+    repeated bytes      rep_bytes   = 35;
+    repeated SubMessage rep_submsg  = 36;
+    repeated MyEnum     rep_enum    = 37;
+    repeated EmptyMessage rep_emptymsg = 38;
+    
+    optional int32      opt_int32   = 41 [default = 4041];
+    optional int64      opt_int64   = 42 [default = 4042];
+    optional uint32     opt_uint32  = 43 [default = 4043];
+    optional uint64     opt_uint64  = 44 [default = 4044];
+    optional sint32     opt_sint32  = 45 [default = 4045];
+    optional sint64     opt_sint64  = 46 [default = 4046];
+    optional bool       opt_bool    = 47 [default = false];
+    
+    optional fixed32    opt_fixed32 = 48 [default = 4048];
+    optional sfixed32   opt_sfixed32= 49 [default = 4049];
+    optional float      opt_float   = 50 [default = 4050];
+    
+    optional fixed64    opt_fixed64 = 51 [default = 4051];
+    optional sfixed64   opt_sfixed64= 52 [default = 4052];
+    optional double     opt_double  = 53 [default = 4053];
+    
+    optional string     opt_string  = 54 [default = "4054"];
+    optional bytes      opt_bytes   = 55 [default = "4055"];
+    optional SubMessage opt_submsg  = 56;
+    optional MyEnum     opt_enum    = 57 [default = Second];
+    optional EmptyMessage opt_emptymsg = 58;
 
     // Just to make sure that the size of the fields has been calculated
     // properly, i.e. otherwise a bug in last field might not be detected.
     required int32      end = 99;
+
+
+    extensions 200 to 255;
 }
 

tests/bc_alltypes.pb.c

@@ -0,0 +1,93 @@
+/* Automatically generated nanopb constant definitions */
+/* Generated by 0.2.0-dev at Sun Feb 17 00:09:53 2013. */
+/* This is a file generated using nanopb-0.2.0-dev.
+ * It is used as a part of test suite in order to detect any
+ * incompatible changes made to the generator in future versions.
+ */
+
+#include "bc_alltypes.pb.h"
+
+const char SubMessage_substuff1_default[16] = "1";
+const int32_t SubMessage_substuff2_default = 2;
+const uint32_t SubMessage_substuff3_default = 3;
+const int32_t AllTypes_opt_int32_default = 4041;
+const int64_t AllTypes_opt_int64_default = 4042;
+const uint32_t AllTypes_opt_uint32_default = 4043;
+const uint64_t AllTypes_opt_uint64_default = 4044;
+const int32_t AllTypes_opt_sint32_default = 4045;
+const int64_t AllTypes_opt_sint64_default = 4046;
+const bool AllTypes_opt_bool_default = false;
+const uint32_t AllTypes_opt_fixed32_default = 4048;
+const int32_t AllTypes_opt_sfixed32_default = 4049;
+const float AllTypes_opt_float_default = 4050;
+const uint64_t AllTypes_opt_fixed64_default = 4051;
+const int64_t AllTypes_opt_sfixed64_default = 4052;
+const double AllTypes_opt_double_default = 4053;
+const char AllTypes_opt_string_default[16] = "4054";
+const AllTypes_opt_bytes_t AllTypes_opt_bytes_default = {4, {0x34,0x30,0x35,0x35}};
+const MyEnum AllTypes_opt_enum_default = MyEnum_Second;
+
+
+const pb_field_t SubMessage_fields[4] = {
+    PB_FIELD(  1, STRING  , REQUIRED, STATIC, SubMessage, substuff1, substuff1, &SubMessage_substuff1_default),
+    PB_FIELD(  2, INT32   , REQUIRED, STATIC, SubMessage, substuff2, substuff1, &SubMessage_substuff2_default),
+    PB_FIELD(  3, FIXED32 , OPTIONAL, STATIC, SubMessage, substuff3, substuff2, &SubMessage_substuff3_default),
+    PB_LAST_FIELD
+};
+
+const pb_field_t AllTypes_fields[53] = {
+    PB_FIELD(  1, INT32   , REQUIRED, STATIC, AllTypes, req_int32, req_int32, 0),
+    PB_FIELD(  2, INT64   , REQUIRED, STATIC, AllTypes, req_int64, req_int32, 0),
+    PB_FIELD(  3, UINT32  , REQUIRED, STATIC, AllTypes, req_uint32, req_int64, 0),
+    PB_FIELD(  4, UINT64  , REQUIRED, STATIC, AllTypes, req_uint64, req_uint32, 0),
+    PB_FIELD(  5, SINT32  , REQUIRED, STATIC, AllTypes, req_sint32, req_uint64, 0),
+    PB_FIELD(  6, SINT64  , REQUIRED, STATIC, AllTypes, req_sint64, req_sint32, 0),
+    PB_FIELD(  7, BOOL    , REQUIRED, STATIC, AllTypes, req_bool, req_sint64, 0),
+    PB_FIELD(  8, FIXED32 , REQUIRED, STATIC, AllTypes, req_fixed32, req_bool, 0),
+    PB_FIELD(  9, SFIXED32, REQUIRED, STATIC, AllTypes, req_sfixed32, req_fixed32, 0),
+    PB_FIELD( 10, FLOAT   , REQUIRED, STATIC, AllTypes, req_float, req_sfixed32, 0),
+    PB_FIELD( 11, FIXED64 , REQUIRED, STATIC, AllTypes, req_fixed64, req_float, 0),
+    PB_FIELD( 12, SFIXED64, REQUIRED, STATIC, AllTypes, req_sfixed64, req_fixed64, 0),
+    PB_FIELD( 13, DOUBLE  , REQUIRED, STATIC, AllTypes, req_double, req_sfixed64, 0),
+    PB_FIELD( 14, STRING  , REQUIRED, STATIC, AllTypes, req_string, req_double, 0),
+    PB_FIELD( 15, BYTES   , REQUIRED, STATIC, AllTypes, req_bytes, req_string, 0),
+    PB_FIELD( 16, MESSAGE , REQUIRED, STATIC, AllTypes, req_submsg, req_bytes, &SubMessage_fields),
+    PB_FIELD( 17, ENUM    , REQUIRED, STATIC, AllTypes, req_enum, req_submsg, 0),
+    PB_FIELD( 21, INT32   , REPEATED, STATIC, AllTypes, rep_int32, req_enum, 0),
+    PB_FIELD( 22, INT64   , REPEATED, STATIC, AllTypes, rep_int64, rep_int32, 0),
+    PB_FIELD( 23, UINT32  , REPEATED, STATIC, AllTypes, rep_uint32, rep_int64, 0),
+    PB_FIELD( 24, UINT64  , REPEATED, STATIC, AllTypes, rep_uint64, rep_uint32, 0),
+    PB_FIELD( 25, SINT32  , REPEATED, STATIC, AllTypes, rep_sint32, rep_uint64, 0),
+    PB_FIELD( 26, SINT64  , REPEATED, STATIC, AllTypes, rep_sint64, rep_sint32, 0),
+    PB_FIELD( 27, BOOL    , REPEATED, STATIC, AllTypes, rep_bool, rep_sint64, 0),
+    PB_FIELD( 28, FIXED32 , REPEATED, STATIC, AllTypes, rep_fixed32, rep_bool, 0),
+    PB_FIELD( 29, SFIXED32, REPEATED, STATIC, AllTypes, rep_sfixed32, rep_fixed32, 0),
+    PB_FIELD( 30, FLOAT   , REPEATED, STATIC, AllTypes, rep_float, rep_sfixed32, 0),
+    PB_FIELD( 31, FIXED64 , REPEATED, STATIC, AllTypes, rep_fixed64, rep_float, 0),
+    PB_FIELD( 32, SFIXED64, REPEATED, STATIC, AllTypes, rep_sfixed64, rep_fixed64, 0),
+    PB_FIELD( 33, DOUBLE  , REPEATED, STATIC, AllTypes, rep_double, rep_sfixed64, 0),
+    PB_FIELD( 34, STRING  , REPEATED, STATIC, AllTypes, rep_string, rep_double, 0),
+    PB_FIELD( 35, BYTES   , REPEATED, STATIC, AllTypes, rep_bytes, rep_string, 0),
+    PB_FIELD( 36, MESSAGE , REPEATED, STATIC, AllTypes, rep_submsg, rep_bytes, &SubMessage_fields),
+    PB_FIELD( 37, ENUM    , REPEATED, STATIC, AllTypes, rep_enum, rep_submsg, 0),
+    PB_FIELD( 41, INT32   , OPTIONAL, STATIC, AllTypes, opt_int32, rep_enum, &AllTypes_opt_int32_default),
+    PB_FIELD( 42, INT64   , OPTIONAL, STATIC, AllTypes, opt_int64, opt_int32, &AllTypes_opt_int64_default),
+    PB_FIELD( 43, UINT32  , OPTIONAL, STATIC, AllTypes, opt_uint32, opt_int64, &AllTypes_opt_uint32_default),
+    PB_FIELD( 44, UINT64  , OPTIONAL, STATIC, AllTypes, opt_uint64, opt_uint32, &AllTypes_opt_uint64_default),
+    PB_FIELD( 45, SINT32  , OPTIONAL, STATIC, AllTypes, opt_sint32, opt_uint64, &AllTypes_opt_sint32_default),
+    PB_FIELD( 46, SINT64  , OPTIONAL, STATIC, AllTypes, opt_sint64, opt_sint32, &AllTypes_opt_sint64_default),
+    PB_FIELD( 47, BOOL    , OPTIONAL, STATIC, AllTypes, opt_bool, opt_sint64, &AllTypes_opt_bool_default),
+    PB_FIELD( 48, FIXED32 , OPTIONAL, STATIC, AllTypes, opt_fixed32, opt_bool, &AllTypes_opt_fixed32_default),
+    PB_FIELD( 49, SFIXED32, OPTIONAL, STATIC, AllTypes, opt_sfixed32, opt_fixed32, &AllTypes_opt_sfixed32_default),
+    PB_FIELD( 50, FLOAT   , OPTIONAL, STATIC, AllTypes, opt_float, opt_sfixed32, &AllTypes_opt_float_default),
+    PB_FIELD( 51, FIXED64 , OPTIONAL, STATIC, AllTypes, opt_fixed64, opt_float, &AllTypes_opt_fixed64_default),
+    PB_FIELD( 52, SFIXED64, OPTIONAL, STATIC, AllTypes, opt_sfixed64, opt_fixed64, &AllTypes_opt_sfixed64_default),
+    PB_FIELD( 53, DOUBLE  , OPTIONAL, STATIC, AllTypes, opt_double, opt_sfixed64, &AllTypes_opt_double_default),
+    PB_FIELD( 54, STRING  , OPTIONAL, STATIC, AllTypes, opt_string, opt_double, &AllTypes_opt_string_default),
+    PB_FIELD( 55, BYTES   , OPTIONAL, STATIC, AllTypes, opt_bytes, opt_string, &AllTypes_opt_bytes_default),
+    PB_FIELD( 56, MESSAGE , OPTIONAL, STATIC, AllTypes, opt_submsg, opt_bytes, &SubMessage_fields),
+    PB_FIELD( 57, ENUM    , OPTIONAL, STATIC, AllTypes, opt_enum, opt_submsg, &AllTypes_opt_enum_default),
+    PB_FIELD( 99, INT32   , REQUIRED, STATIC, AllTypes, end, opt_enum, 0),
+    PB_LAST_FIELD
+};
+

tests/bc_alltypes.pb.h

@@ -0,0 +1,178 @@
+/* Automatically generated nanopb header */
+/* This is a file generated using nanopb-0.2.0-dev.
+ * It is used as a part of test suite in order to detect any
+ * incompatible changes made to the generator in future versions.
+ */
+
+#ifndef _PB_ALLTYPES_PB_H_
+#define _PB_ALLTYPES_PB_H_
+#include <pb.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* Enum definitions */
+typedef enum _MyEnum {
+    MyEnum_Zero = 0,
+    MyEnum_First = 1,
+    MyEnum_Second = 2,
+    MyEnum_Truth = 42
+} MyEnum;
+
+/* Struct definitions */
+typedef struct _SubMessage {
+    char substuff1[16];
+    int32_t substuff2;
+    bool has_substuff3;
+    uint32_t substuff3;
+} SubMessage;
+
+typedef struct {
+    size_t size;
+    uint8_t bytes[16];
+} AllTypes_req_bytes_t;
+
+typedef struct {
+    size_t size;
+    uint8_t bytes[16];
+} AllTypes_rep_bytes_t;
+
+typedef struct {
+    size_t size;
+    uint8_t bytes[16];
+} AllTypes_opt_bytes_t;
+
+typedef struct _AllTypes {
+    int32_t req_int32;
+    int64_t req_int64;
+    uint32_t req_uint32;
+    uint64_t req_uint64;
+    int32_t req_sint32;
+    int64_t req_sint64;
+    bool req_bool;
+    uint32_t req_fixed32;
+    int32_t req_sfixed32;
+    float req_float;
+    uint64_t req_fixed64;
+    int64_t req_sfixed64;
+    double req_double;
+    char req_string[16];
+    AllTypes_req_bytes_t req_bytes;
+    SubMessage req_submsg;
+    MyEnum req_enum;
+    size_t rep_int32_count;
+    int32_t rep_int32[5];
+    size_t rep_int64_count;
+    int64_t rep_int64[5];
+    size_t rep_uint32_count;
+    uint32_t rep_uint32[5];
+    size_t rep_uint64_count;
+    uint64_t rep_uint64[5];
+    size_t rep_sint32_count;
+    int32_t rep_sint32[5];
+    size_t rep_sint64_count;
+    int64_t rep_sint64[5];
+    size_t rep_bool_count;
+    bool rep_bool[5];
+    size_t rep_fixed32_count;
+    uint32_t rep_fixed32[5];
+    size_t rep_sfixed32_count;
+    int32_t rep_sfixed32[5];
+    size_t rep_float_count;
+    float rep_float[5];
+    size_t rep_fixed64_count;
+    uint64_t rep_fixed64[5];
+    size_t rep_sfixed64_count;
+    int64_t rep_sfixed64[5];
+    size_t rep_double_count;
+    double rep_double[5];
+    size_t rep_string_count;
+    char rep_string[5][16];
+    size_t rep_bytes_count;
+    AllTypes_rep_bytes_t rep_bytes[5];
+    size_t rep_submsg_count;
+    SubMessage rep_submsg[5];
+    size_t rep_enum_count;
+    MyEnum rep_enum[5];
+    bool has_opt_int32;
+    int32_t opt_int32;
+    bool has_opt_int64;
+    int64_t opt_int64;
+    bool has_opt_uint32;
+    uint32_t opt_uint32;
+    bool has_opt_uint64;
+    uint64_t opt_uint64;
+    bool has_opt_sint32;
+    int32_t opt_sint32;
+    bool has_opt_sint64;
+    int64_t opt_sint64;
+    bool has_opt_bool;
+    bool opt_bool;
+    bool has_opt_fixed32;
+    uint32_t opt_fixed32;
+    bool has_opt_sfixed32;
+    int32_t opt_sfixed32;
+    bool has_opt_float;
+    float opt_float;
+    bool has_opt_fixed64;
+    uint64_t opt_fixed64;
+    bool has_opt_sfixed64;
+    int64_t opt_sfixed64;
+    bool has_opt_double;
+    double opt_double;
+    bool has_opt_string;
+    char opt_string[16];
+    bool has_opt_bytes;
+    AllTypes_opt_bytes_t opt_bytes;
+    bool has_opt_submsg;
+    SubMessage opt_submsg;
+    bool has_opt_enum;
+    MyEnum opt_enum;
+    int32_t end;
+} AllTypes;
+
+/* Default values for struct fields */
+extern const char SubMessage_substuff1_default[16];
+extern const int32_t SubMessage_substuff2_default;
+extern const uint32_t SubMessage_substuff3_default;
+extern const int32_t AllTypes_opt_int32_default;
+extern const int64_t AllTypes_opt_int64_default;
+extern const uint32_t AllTypes_opt_uint32_default;
+extern const uint64_t AllTypes_opt_uint64_default;
+extern const int32_t AllTypes_opt_sint32_default;
+extern const int64_t AllTypes_opt_sint64_default;
+extern const bool AllTypes_opt_bool_default;
+extern const uint32_t AllTypes_opt_fixed32_default;
+extern const int32_t AllTypes_opt_sfixed32_default;
+extern const float AllTypes_opt_float_default;
+extern const uint64_t AllTypes_opt_fixed64_default;
+extern const int64_t AllTypes_opt_sfixed64_default;
+extern const double AllTypes_opt_double_default;
+extern const char AllTypes_opt_string_default[16];
+extern const AllTypes_opt_bytes_t AllTypes_opt_bytes_default;
+extern const MyEnum AllTypes_opt_enum_default;
+
+/* Struct field encoding specification for nanopb */
+extern const pb_field_t SubMessage_fields[4];
+extern const pb_field_t AllTypes_fields[53];
+
+/* Check that field information fits in pb_field_t */
+#if !defined(PB_FIELD_16BIT) && !defined(PB_FIELD_32BIT)
+STATIC_ASSERT((pb_membersize(AllTypes, req_submsg) < 256 && pb_membersize(AllTypes, rep_submsg[0]) < 256 && pb_membersize(AllTypes, opt_submsg) < 256), YOU_MUST_DEFINE_PB_FIELD_16BIT_FOR_MESSAGES_SubMessage_AllTypes)
+#endif
+
+#if !defined(PB_FIELD_32BIT)
+STATIC_ASSERT((pb_membersize(AllTypes, req_submsg) < 65536 && pb_membersize(AllTypes, rep_submsg[0]) < 65536 && pb_membersize(AllTypes, opt_submsg) < 65536), YOU_MUST_DEFINE_PB_FIELD_32BIT_FOR_MESSAGES_SubMessage_AllTypes)
+#endif
+
+/* On some platforms (such as AVR), double is really float.
+ * These are not directly supported by nanopb, but see example_avr_double.
+ */
+STATIC_ASSERT(sizeof(double) == 8, DOUBLE_MUST_BE_8_BYTES)
+
+#ifdef __cplusplus
+} /* extern "C" */
+#endif
+
+#endif

tests/bc_decode.c

@@ -0,0 +1,197 @@
+/* Tests the decoding of all types.
+ * This is a backwards-compatibility test, using bc_alltypes.pb.h.
+ * It is similar to test_decode3, but duplicated in order to allow
+ * test_decode3 to test any new features introduced later.
+ *
+ * Run e.g. ./bc_encode | ./bc_decode
+ */
+
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+#include <pb_decode.h>
+#include "bc_alltypes.pb.h"
+
+#define TEST(x) if (!(x)) { \
+    printf("Test " #x " failed.\n"); \
+    return false; \
+    }
+
+/* This function is called once from main(), it handles
+   the decoding and checks the fields. */
+bool check_alltypes(pb_istream_t *stream, int mode)
+{
+    AllTypes alltypes;
+    
+    /* Fill with garbage to better detect initialization errors */
+    memset(&alltypes, 0xAA, sizeof(alltypes));
+    
+    if (!pb_decode(stream, AllTypes_fields, &alltypes))
+        return false;
+    
+    TEST(alltypes.req_int32         == -1001);
+    TEST(alltypes.req_int64         == -1002);
+    TEST(alltypes.req_uint32        == 1003);
+    TEST(alltypes.req_uint64        == 1004);
+    TEST(alltypes.req_sint32        == -1005);
+    TEST(alltypes.req_sint64        == -1006);
+    TEST(alltypes.req_bool          == true);
+    
+    TEST(alltypes.req_fixed32       == 1008);
+    TEST(alltypes.req_sfixed32      == -1009);
+    TEST(alltypes.req_float         == 1010.0f);
+    
+    TEST(alltypes.req_fixed64       == 1011);
+    TEST(alltypes.req_sfixed64      == -1012);
+    TEST(alltypes.req_double        == 1013.0f);
+    
+    TEST(strcmp(alltypes.req_string, "1014") == 0);
+    TEST(alltypes.req_bytes.size == 4);
+    TEST(memcmp(alltypes.req_bytes.bytes, "1015", 4) == 0);
+    TEST(strcmp(alltypes.req_submsg.substuff1, "1016") == 0);
+    TEST(alltypes.req_submsg.substuff2 == 1016);
+    TEST(alltypes.req_submsg.substuff3 == 3);
+    TEST(alltypes.req_enum == MyEnum_Truth);
+    
+    TEST(alltypes.rep_int32_count == 5 && alltypes.rep_int32[4] == -2001 && alltypes.rep_int32[0] == 0);
+    TEST(alltypes.rep_int64_count == 5 && alltypes.rep_int64[4] == -2002 && alltypes.rep_int64[0] == 0);
+    TEST(alltypes.rep_uint32_count == 5 && alltypes.rep_uint32[4] == 2003 && alltypes.rep_uint32[0] == 0);
+    TEST(alltypes.rep_uint64_count == 5 && alltypes.rep_uint64[4] == 2004 && alltypes.rep_uint64[0] == 0);
+    TEST(alltypes.rep_sint32_count == 5 && alltypes.rep_sint32[4] == -2005 && alltypes.rep_sint32[0] == 0);
+    TEST(alltypes.rep_sint64_count == 5 && alltypes.rep_sint64[4] == -2006 && alltypes.rep_sint64[0] == 0);
+    TEST(alltypes.rep_bool_count == 5 && alltypes.rep_bool[4] == true && alltypes.rep_bool[0] == false);
+    
+    TEST(alltypes.rep_fixed32_count == 5 && alltypes.rep_fixed32[4] == 2008 && alltypes.rep_fixed32[0] == 0);
+    TEST(alltypes.rep_sfixed32_count == 5 && alltypes.rep_sfixed32[4] == -2009 && alltypes.rep_sfixed32[0] == 0);
+    TEST(alltypes.rep_float_count == 5 && alltypes.rep_float[4] == 2010.0f && alltypes.rep_float[0] == 0.0f);
+    
+    TEST(alltypes.rep_fixed64_count == 5 && alltypes.rep_fixed64[4] == 2011 && alltypes.rep_fixed64[0] == 0);
+    TEST(alltypes.rep_sfixed64_count == 5 && alltypes.rep_sfixed64[4] == -2012 && alltypes.rep_sfixed64[0] == 0);
+    TEST(alltypes.rep_double_count == 5 && alltypes.rep_double[4] == 2013.0 && alltypes.rep_double[0] == 0.0);
+    
+    TEST(alltypes.rep_string_count == 5 && strcmp(alltypes.rep_string[4], "2014") == 0 && alltypes.rep_string[0][0] == '\0');
+    TEST(alltypes.rep_bytes_count == 5 && alltypes.rep_bytes[4].size == 4 && alltypes.rep_bytes[0].size == 0);
+    TEST(memcmp(alltypes.rep_bytes[4].bytes, "2015", 4) == 0);
+
+    TEST(alltypes.rep_submsg_count == 5);
+    TEST(strcmp(alltypes.rep_submsg[4].substuff1, "2016") == 0 && alltypes.rep_submsg[0].substuff1[0] == '\0');
+    TEST(alltypes.rep_submsg[4].substuff2 == 2016 && alltypes.rep_submsg[0].substuff2 == 0);
+    TEST(alltypes.rep_submsg[4].substuff3 == 2016 && alltypes.rep_submsg[0].substuff3 == 3);
+    
+    TEST(alltypes.rep_enum_count == 5 && alltypes.rep_enum[4] == MyEnum_Truth && alltypes.rep_enum[0] == MyEnum_Zero);
+    
+    if (mode == 0)
+    {
+        /* Expect default values */
+        TEST(alltypes.has_opt_int32     == false);
+        TEST(alltypes.opt_int32         == 4041);
+        TEST(alltypes.has_opt_int64     == false);
+        TEST(alltypes.opt_int64         == 4042);
+        TEST(alltypes.has_opt_uint32    == false);
+        TEST(alltypes.opt_uint32        == 4043);
+        TEST(alltypes.has_opt_uint64    == false);
+        TEST(alltypes.opt_uint64        == 4044);
+        TEST(alltypes.has_opt_sint32    == false);
+        TEST(alltypes.opt_sint32        == 4045);
+        TEST(alltypes.has_opt_sint64    == false);
+        TEST(alltypes.opt_sint64        == 4046);
+        TEST(alltypes.has_opt_bool      == false);
+        TEST(alltypes.opt_bool          == false);
+        
+        TEST(alltypes.has_opt_fixed32   == false);
+        TEST(alltypes.opt_fixed32       == 4048);
+        TEST(alltypes.has_opt_sfixed32  == false);
+        TEST(alltypes.opt_sfixed32      == 4049);
+        TEST(alltypes.has_opt_float     == false);
+        TEST(alltypes.opt_float         == 4050.0f);
+        
+        TEST(alltypes.has_opt_fixed64   == false);
+        TEST(alltypes.opt_fixed64       == 4051);
+        TEST(alltypes.has_opt_sfixed64  == false);
+        TEST(alltypes.opt_sfixed64      == 4052);
+        TEST(alltypes.has_opt_double    == false);
+        TEST(alltypes.opt_double        == 4053.0);
+        
+        TEST(alltypes.has_opt_string    == false);
+        TEST(strcmp(alltypes.opt_string, "4054") == 0);
+        TEST(alltypes.has_opt_bytes     == false);
+        TEST(alltypes.opt_bytes.size == 4);
+        TEST(memcmp(alltypes.opt_bytes.bytes, "4055", 4) == 0);
+        TEST(alltypes.has_opt_submsg    == false);
+        TEST(strcmp(alltypes.opt_submsg.substuff1, "1") == 0);
+        TEST(alltypes.opt_submsg.substuff2 == 2);
+        TEST(alltypes.opt_submsg.substuff3 == 3);
+        TEST(alltypes.has_opt_enum     == false);
+        TEST(alltypes.opt_enum == MyEnum_Second);
+    }
+    else
+    {
+        /* Expect filled-in values */
+        TEST(alltypes.has_opt_int32     == true);
+        TEST(alltypes.opt_int32         == 3041);
+        TEST(alltypes.has_opt_int64     == true);
+        TEST(alltypes.opt_int64         == 3042);
+        TEST(alltypes.has_opt_uint32    == true);
+        TEST(alltypes.opt_uint32        == 3043);
+        TEST(alltypes.has_opt_uint64    == true);
+        TEST(alltypes.opt_uint64        == 3044);
+        TEST(alltypes.has_opt_sint32    == true);
+        TEST(alltypes.opt_sint32        == 3045);
+        TEST(alltypes.has_opt_sint64    == true);
+        TEST(alltypes.opt_sint64        == 3046);
+        TEST(alltypes.has_opt_bool      == true);
+        TEST(alltypes.opt_bool          == true);
+        
+        TEST(alltypes.has_opt_fixed32   == true);
+        TEST(alltypes.opt_fixed32       == 3048);
+        TEST(alltypes.has_opt_sfixed32  == true);
+        TEST(alltypes.opt_sfixed32      == 3049);
+        TEST(alltypes.has_opt_float     == true);
+        TEST(alltypes.opt_float         == 3050.0f);
+        
+        TEST(alltypes.has_opt_fixed64   == true);
+        TEST(alltypes.opt_fixed64       == 3051);
+        TEST(alltypes.has_opt_sfixed64  == true);
+        TEST(alltypes.opt_sfixed64      == 3052);
+        TEST(alltypes.has_opt_double    == true);
+        TEST(alltypes.opt_double        == 3053.0);
+        
+        TEST(alltypes.has_opt_string    == true);
+        TEST(strcmp(alltypes.opt_string, "3054") == 0);
+        TEST(alltypes.has_opt_bytes     == true);
+        TEST(alltypes.opt_bytes.size == 4);
+        TEST(memcmp(alltypes.opt_bytes.bytes, "3055", 4) == 0);
+        TEST(alltypes.has_opt_submsg    == true);
+        TEST(strcmp(alltypes.opt_submsg.substuff1, "3056") == 0);
+        TEST(alltypes.opt_submsg.substuff2 == 3056);
+        TEST(alltypes.opt_submsg.substuff3 == 3);
+        TEST(alltypes.has_opt_enum      == true);
+        TEST(alltypes.opt_enum == MyEnum_Truth);
+    }
+    
+    TEST(alltypes.end == 1099);
+    
+    return true;
+}
+
+int main(int argc, char **argv)
+{
+    /* Whether to expect the optional values or the default values. */
+    int mode = (argc > 1) ? atoi(argv[1]) : 0;
+    
+    /* Read the data into buffer */
+    uint8_t buffer[1024];
+    size_t count = fread(buffer, 1, sizeof(buffer), stdin);
+    
+    /* Construct a pb_istream_t for reading from the buffer */
+    pb_istream_t stream = pb_istream_from_buffer(buffer, count);
+    
+    /* Decode and print out the stuff */
+    if (!check_alltypes(&stream, mode))
+    {
+        printf("Parsing failed: %s\n", PB_GET_ERROR(&stream));
+        return 1;
+    } else {
+        return 0;
+    }
+}

tests/bc_encode.c

@@ -0,0 +1,131 @@
+/* Attempts to test all the datatypes supported by ProtoBuf.
+ * This is a backwards-compatibility test, using bc_alltypes.pb.h.
+ * It is similar to test_encode3, but duplicated in order to allow
+ * test_encode3 to test any new features introduced later.
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <pb_encode.h>
+#include "bc_alltypes.pb.h"
+
+int main(int argc, char **argv)
+{
+    int mode = (argc > 1) ? atoi(argv[1]) : 0;
+    
+    /* Initialize the structure with constants */
+    AllTypes alltypes = {0};
+    
+    alltypes.req_int32         = -1001;
+    alltypes.req_int64         = -1002;
+    alltypes.req_uint32        = 1003;
+    alltypes.req_uint64        = 1004;
+    alltypes.req_sint32        = -1005;
+    alltypes.req_sint64        = -1006;
+    alltypes.req_bool          = true;
+    
+    alltypes.req_fixed32       = 1008;
+    alltypes.req_sfixed32      = -1009;
+    alltypes.req_float         = 1010.0f;
+    
+    alltypes.req_fixed64       = 1011;
+    alltypes.req_sfixed64      = -1012;
+    alltypes.req_double        = 1013.0;
+    
+    strcpy(alltypes.req_string, "1014");
+    alltypes.req_bytes.size = 4;
+    memcpy(alltypes.req_bytes.bytes, "1015", 4);
+    strcpy(alltypes.req_submsg.substuff1, "1016");
+    alltypes.req_submsg.substuff2 = 1016;
+    alltypes.req_enum = MyEnum_Truth;
+    
+    alltypes.rep_int32_count = 5; alltypes.rep_int32[4] = -2001;
+    alltypes.rep_int64_count = 5; alltypes.rep_int64[4] = -2002;
+    alltypes.rep_uint32_count = 5; alltypes.rep_uint32[4] = 2003;
+    alltypes.rep_uint64_count = 5; alltypes.rep_uint64[4] = 2004;
+    alltypes.rep_sint32_count = 5; alltypes.rep_sint32[4] = -2005;
+    alltypes.rep_sint64_count = 5; alltypes.rep_sint64[4] = -2006;
+    alltypes.rep_bool_count = 5; alltypes.rep_bool[4] = true;
+    
+    alltypes.rep_fixed32_count = 5; alltypes.rep_fixed32[4] = 2008;
+    alltypes.rep_sfixed32_count = 5; alltypes.rep_sfixed32[4] = -2009;
+    alltypes.rep_float_count = 5; alltypes.rep_float[4] = 2010.0f;
+    
+    alltypes.rep_fixed64_count = 5; alltypes.rep_fixed64[4] = 2011;
+    alltypes.rep_sfixed64_count = 5; alltypes.rep_sfixed64[4] = -2012;
+    alltypes.rep_double_count = 5; alltypes.rep_double[4] = 2013.0;
+    
+    alltypes.rep_string_count = 5; strcpy(alltypes.rep_string[4], "2014");
+    alltypes.rep_bytes_count = 5; alltypes.rep_bytes[4].size = 4;
+    memcpy(alltypes.rep_bytes[4].bytes, "2015", 4);
+
+    alltypes.rep_submsg_count = 5;
+    strcpy(alltypes.rep_submsg[4].substuff1, "2016");
+    alltypes.rep_submsg[4].substuff2 = 2016;
+    alltypes.rep_submsg[4].has_substuff3 = true;
+    alltypes.rep_submsg[4].substuff3 = 2016;
+    
+    alltypes.rep_enum_count = 5; alltypes.rep_enum[4] = MyEnum_Truth;
+    
+    if (mode != 0)
+    {
+        /* Fill in values for optional fields */
+        alltypes.has_opt_int32 = true;
+        alltypes.opt_int32         = 3041;
+        alltypes.has_opt_int64 = true;
+        alltypes.opt_int64         = 3042;
+        alltypes.has_opt_uint32 = true;
+        alltypes.opt_uint32        = 3043;
+        alltypes.has_opt_uint64 = true;
+        alltypes.opt_uint64        = 3044;
+        alltypes.has_opt_sint32 = true;
+        alltypes.opt_sint32        = 3045;
+        alltypes.has_opt_sint64 = true;
+        alltypes.opt_sint64        = 3046;
+        alltypes.has_opt_bool = true;
+        alltypes.opt_bool          = true;
+        
+        alltypes.has_opt_fixed32 = true;
+        alltypes.opt_fixed32       = 3048;
+        alltypes.has_opt_sfixed32 = true;
+        alltypes.opt_sfixed32      = 3049;
+        alltypes.has_opt_float = true;
+        alltypes.opt_float         = 3050.0f;
+        
+        alltypes.has_opt_fixed64 = true;
+        alltypes.opt_fixed64       = 3051;
+        alltypes.has_opt_sfixed64 = true;
+        alltypes.opt_sfixed64      = 3052;
+        alltypes.has_opt_double = true;
+        alltypes.opt_double        = 3053.0;
+        
+        alltypes.has_opt_string = true;
+        strcpy(alltypes.opt_string, "3054");
+        alltypes.has_opt_bytes = true;
+        alltypes.opt_bytes.size = 4;
+        memcpy(alltypes.opt_bytes.bytes, "3055", 4);
+        alltypes.has_opt_submsg = true;
+        strcpy(alltypes.opt_submsg.substuff1, "3056");
+        alltypes.opt_submsg.substuff2 = 3056;
+        alltypes.has_opt_enum = true;
+        alltypes.opt_enum = MyEnum_Truth;
+    }
+    
+    alltypes.end = 1099;
+    
+    uint8_t buffer[1024];
+    pb_ostream_t stream = pb_ostream_from_buffer(buffer, sizeof(buffer));
+    
+    /* Now encode it and check if we succeeded. */
+    if (pb_encode(&stream, AllTypes_fields, &alltypes))
+    {
+        fwrite(buffer, 1, stream.bytes_written, stdout);
+        return 0; /* Success */
+    }
+    else
+    {
+        fprintf(stderr, "Encoding failed!\n");
+        return 1; /* Failure */
+    }
+}

tests/callbacks.proto

@@ -11,5 +11,6 @@ message TestMessage {
     repeated fixed32 fixed32value = 3;
     repeated fixed64 fixed64value = 4;
     optional SubMessage submsg = 5;
+    repeated string repeatedstring = 6;
 }
 

tests/callbacks2.proto

@@ -0,0 +1,9 @@
+// Test if including generated header file for this file + implicit include of
+// callbacks.pb.h still compiles. Used with test_compiles.c.
+import "callbacks.proto";
+
+message Callback2Message {
+    required TestMessage tstmsg = 1;
+    required SubMessage submsg = 2;
+}
+

tests/decode_unittests.c

@@ -1,3 +1,5 @@
+#define NANOPB_INTERNALS
+
 #include <stdio.h>
 #include <string.h>
 #include "pb_decode.h"
@@ -17,11 +19,11 @@ bool stream_callback(pb_istream_t *stream, uint8_t *buf, size_t count)
 }
 
 /* Verifies that the stream passed to callback matches the byte array pointed to by arg. */
-bool callback_check(pb_istream_t *stream, const pb_field_t *field, void *arg)
+bool callback_check(pb_istream_t *stream, const pb_field_t *field, void **arg)
 {
     int i;
     uint8_t byte;
-    pb_bytes_array_t *ref = (pb_bytes_array_t*) arg;
+    pb_bytes_array_t *ref = (pb_bytes_array_t*) *arg;
     
     for (i = 0; i < ref->size; i++)
     {
@@ -287,6 +289,16 @@ int main()
         TEST((s = S("\x08"), !pb_decode(&s, IntegerArray_fields, &dest)))
     }
     
+    {
+        pb_istream_t s;
+        IntegerContainer dest = {};
+        
+        COMMENT("Testing pb_decode_delimited")
+        TEST((s = S("\x09\x0A\x07\x0A\x05\x01\x02\x03\x04\x05"),
+              pb_decode_delimited(&s, IntegerContainer_fields, &dest)) &&
+              dest.submsg.data_count == 5)
+    }
+    
     if (status != 0)
         fprintf(stdout, "\n\nSome tests FAILED!\n");
     

tests/encode_unittests.c

@@ -17,7 +17,7 @@ bool streamcallback(pb_ostream_t *stream, const uint8_t *buf, size_t count)
     return true;
 }
 
-bool fieldcallback(pb_ostream_t *stream, const pb_field_t *field, const void *arg)
+bool fieldcallback(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
 {
     int value = 0x55;
     if (!pb_encode_tag_for_field(stream, field))
@@ -25,7 +25,7 @@ bool fieldcallback(pb_ostream_t *stream, const pb_field_t *field, const void *ar
     return pb_encode_varint(stream, value);
 }
 
-bool crazyfieldcallback(pb_ostream_t *stream, const pb_field_t *field, const void *arg)
+bool crazyfieldcallback(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
 {
     /* This callback writes different amount of data the second time. */
     uint32_t *state = (uint32_t*)arg;
@@ -180,12 +180,14 @@ int main()
     {
         uint8_t buffer[30];
         pb_ostream_t s;
-        char value[] = "xyzzy";
+        char value[30] = "xyzzy";
         
         COMMENT("Test pb_enc_string")
-        TEST(WRITES(pb_enc_string(&s, NULL, &value), "\x05xyzzy"))
+        TEST(WRITES(pb_enc_string(&s, &StringMessage_fields[0], &value), "\x05xyzzy"))
         value[0] = '\0';
-        TEST(WRITES(pb_enc_string(&s, NULL, &value), "\x00"))
+        TEST(WRITES(pb_enc_string(&s, &StringMessage_fields[0], &value), "\x00"))
+        memset(value, 'x', 30);
+        TEST(WRITES(pb_enc_string(&s, &StringMessage_fields[0], &value), "\x0Axxxxxxxxxx"))
     }
     
     {
@@ -242,6 +244,16 @@ int main()
                     "\x0A\x07\x0A\x05\x01\x02\x03\x04\x05"))
     }
     
+    {
+        uint8_t buffer[20];
+        pb_ostream_t s;
+        IntegerContainer msg = {{5, {1,2,3,4,5}}};
+        
+        COMMENT("Test pb_encode_delimited.")
+        TEST(WRITES(pb_encode_delimited(&s, IntegerContainer_fields, &msg),
+                    "\x09\x0A\x07\x0A\x05\x01\x02\x03\x04\x05"))
+    }
+    
     {
         uint8_t buffer[10];
         pb_ostream_t s;

tests/extensions.options

@@ -0,0 +1 @@
+* max_size:16

tests/extensions.proto

@@ -0,0 +1,15 @@
+import 'alltypes.proto';
+
+extend AllTypes {
+    optional int32 AllTypes_extensionfield1 = 255;
+}
+
+message ExtensionMessage {
+    extend AllTypes {
+        optional ExtensionMessage AllTypes_extensionfield2 = 254;
+    }
+    
+    required string test1 = 1;
+    required int32 test2 = 2;
+}
+

tests/options.expected

@@ -0,0 +1,7 @@
+char filesize\[20\];
+char msgsize\[30\];
+char fieldsize\[40\];
+pb_callback_t int32_callback;
+\sEnumValue1 = 1
+Message5_EnumValue1
+} pb_packed my_packed_struct;

tests/options.proto

@@ -0,0 +1,73 @@
+/* Test nanopb option parsing.
+ * options.expected lists the patterns that are searched for in the output.
+ */
+
+import "nanopb.proto";
+
+// File level options
+option (nanopb_fileopt).max_size = 20;
+
+message Message1
+{
+    required string filesize = 1;
+}
+
+// Message level options
+message Message2
+{
+    option (nanopb_msgopt).max_size = 30;
+    required string msgsize = 1;
+}
+
+// Field level options
+message Message3
+{
+    required string fieldsize = 1 [(nanopb).max_size = 40];
+}
+
+// Forced callback field
+message Message4
+{
+    required int32 int32_callback = 1 [(nanopb).type = FT_CALLBACK];
+}
+
+// Short enum names
+enum Enum1
+{
+    option (nanopb_enumopt).long_names = false;
+    EnumValue1 = 1;
+    EnumValue2 = 2;
+}
+
+message EnumTest
+{
+    required Enum1 field = 1 [default = EnumValue2];
+}
+
+// Short enum names inside message
+message Message5
+{
+    enum Enum2
+    {
+       option (nanopb_enumopt).long_names = false;
+       EnumValue1 = 1;
+    }
+    required Enum2 field = 1 [default = EnumValue1];
+}
+
+// Packed structure
+message my_packed_struct
+{
+    option (nanopb_msgopt).packed_struct = true;
+    optional int32 myfield = 1;
+}
+
+// Message with ignored field
+// Note: doesn't really test if the field is missing in the output,
+// but atleast tests that the output compiles.
+message Message6
+{
+    required int32 field1 = 1;
+    optional int32 field2 = 2 [(nanopb).type = FT_IGNORE];
+}
+

tests/test_decode1.c

@@ -63,13 +63,19 @@ int main()
     uint8_t buffer[512];
     size_t count = fread(buffer, 1, sizeof(buffer), stdin);
     
+    if (!feof(stdin))
+    {
+    	printf("Message does not fit in buffer\n");
+    	return 1;
+    }
+    
     /* Construct a pb_istream_t for reading from the buffer */
     pb_istream_t stream = pb_istream_from_buffer(buffer, count);
     
     /* Decode and print out the stuff */
     if (!print_person(&stream))
     {
-        printf("Parsing failed.\n");
+        printf("Parsing failed: %s\n", PB_GET_ERROR(&stream));
         return 1;
     } else {
         return 0;

tests/test_decode2.c

@@ -59,13 +59,6 @@ bool callback(pb_istream_t *stream, uint8_t *buf, size_t count)
     FILE *file = (FILE*)stream->state;
     bool status;
     
-    if (buf == NULL)
-    {
-       /* Skipping data */
-        while (count-- && fgetc(file) != EOF);
-        return count == 0;
-    }
-    
     status = (fread(buf, 1, count, file) == count);
     
     if (feof(file))
@@ -82,7 +75,7 @@ int main()
     pb_istream_t stream = {&callback, stdin, 10000};
     if (!print_person(&stream))
     {
-        printf("Parsing failed.\n");
+        printf("Parsing failed: %s\n", PB_GET_ERROR(&stream));
         return 1;
     } else {
         return 0;

tests/test_decode3.c

@@ -1,10 +1,11 @@
-/* Tests the decoding of all types. Currently only in the 'required' variety.
+/* Tests the decoding of all types.
  * This is the counterpart of test_encode3.
  * Run e.g. ./test_encode3 | ./test_decode3
  */
 
 #include <stdio.h>
 #include <string.h>
+#include <stdlib.h>
 #include <pb_decode.h>
 #include "alltypes.pb.h"
 
@@ -15,27 +16,30 @@
 
 /* This function is called once from main(), it handles
    the decoding and checks the fields. */
-bool check_alltypes(pb_istream_t *stream)
+bool check_alltypes(pb_istream_t *stream, int mode)
 {
-    AllTypes alltypes = {};
+    AllTypes alltypes;
+    
+    /* Fill with garbage to better detect initialization errors */
+    memset(&alltypes, 0xAA, sizeof(alltypes));
     
     if (!pb_decode(stream, AllTypes_fields, &alltypes))
         return false;
     
-    TEST(alltypes.req_int32         == 1001);
-    TEST(alltypes.req_int64         == 1002);
+    TEST(alltypes.req_int32         == -1001);
+    TEST(alltypes.req_int64         == -1002);
     TEST(alltypes.req_uint32        == 1003);
     TEST(alltypes.req_uint64        == 1004);
-    TEST(alltypes.req_sint32        == 1005);
-    TEST(alltypes.req_sint64        == 1006);
+    TEST(alltypes.req_sint32        == -1005);
+    TEST(alltypes.req_sint64        == -1006);
     TEST(alltypes.req_bool          == true);
     
     TEST(alltypes.req_fixed32       == 1008);
-    TEST(alltypes.req_sfixed32      == 1009);
+    TEST(alltypes.req_sfixed32      == -1009);
     TEST(alltypes.req_float         == 1010.0f);
     
     TEST(alltypes.req_fixed64       == 1011);
-    TEST(alltypes.req_sfixed64      == 1012);
+    TEST(alltypes.req_sfixed64      == -1012);
     TEST(alltypes.req_double        == 1013.0f);
     
     TEST(strcmp(alltypes.req_string, "1014") == 0);
@@ -43,26 +47,149 @@ bool check_alltypes(pb_istream_t *stream)
     TEST(memcmp(alltypes.req_bytes.bytes, "1015", 4) == 0);
     TEST(strcmp(alltypes.req_submsg.substuff1, "1016") == 0);
     TEST(alltypes.req_submsg.substuff2 == 1016);
+    TEST(alltypes.req_submsg.substuff3 == 3);
     TEST(alltypes.req_enum == MyEnum_Truth);
     
+    TEST(alltypes.rep_int32_count == 5 && alltypes.rep_int32[4] == -2001 && alltypes.rep_int32[0] == 0);
+    TEST(alltypes.rep_int64_count == 5 && alltypes.rep_int64[4] == -2002 && alltypes.rep_int64[0] == 0);
+    TEST(alltypes.rep_uint32_count == 5 && alltypes.rep_uint32[4] == 2003 && alltypes.rep_uint32[0] == 0);
+    TEST(alltypes.rep_uint64_count == 5 && alltypes.rep_uint64[4] == 2004 && alltypes.rep_uint64[0] == 0);
+    TEST(alltypes.rep_sint32_count == 5 && alltypes.rep_sint32[4] == -2005 && alltypes.rep_sint32[0] == 0);
+    TEST(alltypes.rep_sint64_count == 5 && alltypes.rep_sint64[4] == -2006 && alltypes.rep_sint64[0] == 0);
+    TEST(alltypes.rep_bool_count == 5 && alltypes.rep_bool[4] == true && alltypes.rep_bool[0] == false);
+    
+    TEST(alltypes.rep_fixed32_count == 5 && alltypes.rep_fixed32[4] == 2008 && alltypes.rep_fixed32[0] == 0);
+    TEST(alltypes.rep_sfixed32_count == 5 && alltypes.rep_sfixed32[4] == -2009 && alltypes.rep_sfixed32[0] == 0);
+    TEST(alltypes.rep_float_count == 5 && alltypes.rep_float[4] == 2010.0f && alltypes.rep_float[0] == 0.0f);
+    
+    TEST(alltypes.rep_fixed64_count == 5 && alltypes.rep_fixed64[4] == 2011 && alltypes.rep_fixed64[0] == 0);
+    TEST(alltypes.rep_sfixed64_count == 5 && alltypes.rep_sfixed64[4] == -2012 && alltypes.rep_sfixed64[0] == 0);
+    TEST(alltypes.rep_double_count == 5 && alltypes.rep_double[4] == 2013.0 && alltypes.rep_double[0] == 0.0);
+    
+    TEST(alltypes.rep_string_count == 5 && strcmp(alltypes.rep_string[4], "2014") == 0 && alltypes.rep_string[0][0] == '\0');
+    TEST(alltypes.rep_bytes_count == 5 && alltypes.rep_bytes[4].size == 4 && alltypes.rep_bytes[0].size == 0);
+    TEST(memcmp(alltypes.rep_bytes[4].bytes, "2015", 4) == 0);
+
+    TEST(alltypes.rep_submsg_count == 5);
+    TEST(strcmp(alltypes.rep_submsg[4].substuff1, "2016") == 0 && alltypes.rep_submsg[0].substuff1[0] == '\0');
+    TEST(alltypes.rep_submsg[4].substuff2 == 2016 && alltypes.rep_submsg[0].substuff2 == 0);
+    TEST(alltypes.rep_submsg[4].substuff3 == 2016 && alltypes.rep_submsg[0].substuff3 == 3);
+    
+    TEST(alltypes.rep_enum_count == 5 && alltypes.rep_enum[4] == MyEnum_Truth && alltypes.rep_enum[0] == MyEnum_Zero);
+    TEST(alltypes.rep_emptymsg_count == 5);
+    
+    if (mode == 0)
+    {
+        /* Expect default values */
+        TEST(alltypes.has_opt_int32     == false);
+        TEST(alltypes.opt_int32         == 4041);
+        TEST(alltypes.has_opt_int64     == false);
+        TEST(alltypes.opt_int64         == 4042);
+        TEST(alltypes.has_opt_uint32    == false);
+        TEST(alltypes.opt_uint32        == 4043);
+        TEST(alltypes.has_opt_uint64    == false);
+        TEST(alltypes.opt_uint64        == 4044);
+        TEST(alltypes.has_opt_sint32    == false);
+        TEST(alltypes.opt_sint32        == 4045);
+        TEST(alltypes.has_opt_sint64    == false);
+        TEST(alltypes.opt_sint64        == 4046);
+        TEST(alltypes.has_opt_bool      == false);
+        TEST(alltypes.opt_bool          == false);
+        
+        TEST(alltypes.has_opt_fixed32   == false);
+        TEST(alltypes.opt_fixed32       == 4048);
+        TEST(alltypes.has_opt_sfixed32  == false);
+        TEST(alltypes.opt_sfixed32      == 4049);
+        TEST(alltypes.has_opt_float     == false);
+        TEST(alltypes.opt_float         == 4050.0f);
+        
+        TEST(alltypes.has_opt_fixed64   == false);
+        TEST(alltypes.opt_fixed64       == 4051);
+        TEST(alltypes.has_opt_sfixed64  == false);
+        TEST(alltypes.opt_sfixed64      == 4052);
+        TEST(alltypes.has_opt_double    == false);
+        TEST(alltypes.opt_double        == 4053.0);
+        
+        TEST(alltypes.has_opt_string    == false);
+        TEST(strcmp(alltypes.opt_string, "4054") == 0);
+        TEST(alltypes.has_opt_bytes     == false);
+        TEST(alltypes.opt_bytes.size == 4);
+        TEST(memcmp(alltypes.opt_bytes.bytes, "4055", 4) == 0);
+        TEST(alltypes.has_opt_submsg    == false);
+        TEST(strcmp(alltypes.opt_submsg.substuff1, "1") == 0);
+        TEST(alltypes.opt_submsg.substuff2 == 2);
+        TEST(alltypes.opt_submsg.substuff3 == 3);
+        TEST(alltypes.has_opt_enum     == false);
+        TEST(alltypes.opt_enum == MyEnum_Second);
+        TEST(alltypes.has_opt_emptymsg == false);
+    }
+    else
+    {
+        /* Expect filled-in values */
+        TEST(alltypes.has_opt_int32     == true);
+        TEST(alltypes.opt_int32         == 3041);
+        TEST(alltypes.has_opt_int64     == true);
+        TEST(alltypes.opt_int64         == 3042);
+        TEST(alltypes.has_opt_uint32    == true);
+        TEST(alltypes.opt_uint32        == 3043);
+        TEST(alltypes.has_opt_uint64    == true);
+        TEST(alltypes.opt_uint64        == 3044);
+        TEST(alltypes.has_opt_sint32    == true);
+        TEST(alltypes.opt_sint32        == 3045);
+        TEST(alltypes.has_opt_sint64    == true);
+        TEST(alltypes.opt_sint64        == 3046);
+        TEST(alltypes.has_opt_bool      == true);
+        TEST(alltypes.opt_bool          == true);
+        
+        TEST(alltypes.has_opt_fixed32   == true);
+        TEST(alltypes.opt_fixed32       == 3048);
+        TEST(alltypes.has_opt_sfixed32  == true);
+        TEST(alltypes.opt_sfixed32      == 3049);
+        TEST(alltypes.has_opt_float     == true);
+        TEST(alltypes.opt_float         == 3050.0f);
+        
+        TEST(alltypes.has_opt_fixed64   == true);
+        TEST(alltypes.opt_fixed64       == 3051);
+        TEST(alltypes.has_opt_sfixed64  == true);
+        TEST(alltypes.opt_sfixed64      == 3052);
+        TEST(alltypes.has_opt_double    == true);
+        TEST(alltypes.opt_double        == 3053.0);
+        
+        TEST(alltypes.has_opt_string    == true);
+        TEST(strcmp(alltypes.opt_string, "3054") == 0);
+        TEST(alltypes.has_opt_bytes     == true);
+        TEST(alltypes.opt_bytes.size == 4);
+        TEST(memcmp(alltypes.opt_bytes.bytes, "3055", 4) == 0);
+        TEST(alltypes.has_opt_submsg    == true);
+        TEST(strcmp(alltypes.opt_submsg.substuff1, "3056") == 0);
+        TEST(alltypes.opt_submsg.substuff2 == 3056);
+        TEST(alltypes.opt_submsg.substuff3 == 3);
+        TEST(alltypes.has_opt_enum      == true);
+        TEST(alltypes.opt_enum == MyEnum_Truth);
+        TEST(alltypes.has_opt_emptymsg  == true);
+    }
+    
     TEST(alltypes.end == 1099);
     
     return true;
 }
 
-int main()
+int main(int argc, char **argv)
 {
+    /* Whether to expect the optional values or the default values. */
+    int mode = (argc > 1) ? atoi(argv[1]) : 0;
+    
     /* Read the data into buffer */
-    uint8_t buffer[512];
+    uint8_t buffer[1024];
     size_t count = fread(buffer, 1, sizeof(buffer), stdin);
     
     /* Construct a pb_istream_t for reading from the buffer */
     pb_istream_t stream = pb_istream_from_buffer(buffer, count);
     
     /* Decode and print out the stuff */
-    if (!check_alltypes(&stream))
+    if (!check_alltypes(&stream, mode))
     {
-        printf("Parsing failed.\n");
+        printf("Parsing failed: %s\n", PB_GET_ERROR(&stream));
         return 1;
     } else {
         return 0;

tests/test_decode_callbacks.c

@@ -6,7 +6,7 @@
 #include <pb_decode.h>
 #include "callbacks.pb.h"
 
-bool print_string(pb_istream_t *stream, const pb_field_t *field, void *arg)
+bool print_string(pb_istream_t *stream, const pb_field_t *field, void **arg)
 {
     uint8_t buffer[1024] = {0};
     
@@ -20,37 +20,37 @@ bool print_string(pb_istream_t *stream, const pb_field_t *field, void *arg)
     /* Print the string, in format comparable with protoc --decode.
      * Format comes from the arg defined in main().
      */
-    printf((char*)arg, buffer);
+    printf((char*)*arg, buffer);
     return true;
 }
 
-bool print_int32(pb_istream_t *stream, const pb_field_t *field, void *arg)
+bool print_int32(pb_istream_t *stream, const pb_field_t *field, void **arg)
 {
     uint64_t value;
     if (!pb_decode_varint(stream, &value))
         return false;
     
-    printf((char*)arg, (long)value);
+    printf((char*)*arg, (long)value);
     return true;
 }
 
-bool print_fixed32(pb_istream_t *stream, const pb_field_t *field, void *arg)
+bool print_fixed32(pb_istream_t *stream, const pb_field_t *field, void **arg)
 {
     uint32_t value;
-    if (!pb_dec_fixed32(stream, NULL, &value))
+    if (!pb_decode_fixed32(stream, &value))
         return false;
     
-    printf((char*)arg, (long)value);
+    printf((char*)*arg, (long)value);
     return true;
 }
 
-bool print_fixed64(pb_istream_t *stream, const pb_field_t *field, void *arg)
+bool print_fixed64(pb_istream_t *stream, const pb_field_t *field, void **arg)
 {
     uint64_t value;
-    if (!pb_dec_fixed64(stream, NULL, &value))
+    if (!pb_decode_fixed64(stream, &value))
         return false;
     
-    printf((char*)arg, (long long)value);
+    printf((char*)*arg, (long long)value);
     return true;
 }
 
@@ -83,6 +83,8 @@ int main()
     testmessage.fixed32value.arg = "fixed32value: %ld\n";
     testmessage.fixed64value.funcs.decode = &print_fixed64;
     testmessage.fixed64value.arg = "fixed64value: %lld\n";
+    testmessage.repeatedstring.funcs.decode = &print_string;
+    testmessage.repeatedstring.arg = "repeatedstring: \"%s\"\n";
     
     if (!pb_decode(&stream, TestMessage_fields, &testmessage))
         return 1;

tests/test_decode_extensions.c

@@ -0,0 +1,43 @@
+/* Test decoding of extension fields. */
+
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+#include <pb_decode.h>
+#include "alltypes.pb.h"
+#include "extensions.pb.h"
+
+#define TEST(x) if (!(x)) { \
+    printf("Test " #x " failed.\n"); \
+    return 2; \
+    }
+
+int main(int argc, char **argv)
+{
+    uint8_t buffer[1024];
+    size_t count = fread(buffer, 1, sizeof(buffer), stdin);
+    pb_istream_t stream = pb_istream_from_buffer(buffer, count);
+    
+    AllTypes alltypes = {};
+    
+    int32_t extensionfield1;
+    pb_extension_t ext1 = {&AllTypes_extensionfield1, &extensionfield1, NULL};
+    alltypes.extensions = &ext1;
+        
+    ExtensionMessage extensionfield2 = {};
+    pb_extension_t ext2 = {&ExtensionMessage_AllTypes_extensionfield2, &extensionfield2, NULL};
+    ext1.next = &ext2;
+    
+    if (!pb_decode(&stream, AllTypes_fields, &alltypes))
+    {
+        printf("Parsing failed: %s\n", PB_GET_ERROR(&stream));
+        return 1;
+    }
+
+    TEST(extensionfield1 == 12345)
+    TEST(strcmp(extensionfield2.test1, "test") == 0)
+    TEST(extensionfield2.test2 == 54321)
+    
+    return 0;
+}
+

tests/test_encode1.c

@@ -27,6 +27,7 @@ int main()
     }
     else
     {
+        fprintf(stderr, "Encoding failed: %s\n", PB_GET_ERROR(&stream));
         return 1; /* Failure */
     }
 }

tests/test_encode2.c

@@ -26,7 +26,12 @@ int main()
     
     /* Now encode it and check if we succeeded. */
     if (pb_encode(&stream, Person_fields, &person))
+    {
         return 0; /* Success */
+    }
     else
+    {
+        fprintf(stderr, "Encoding failed: %s\n", PB_GET_ERROR(&stream));
         return 1; /* Failure */
     }
+}

tests/test_encode3.c

@@ -1,40 +1,119 @@
 /* Attempts to test all the datatypes supported by ProtoBuf.
- * Currently only tests the 'required' variety.
  */
 
 #include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
 #include <pb_encode.h>
 #include "alltypes.pb.h"
 
-int main()
+int main(int argc, char **argv)
 {
+    int mode = (argc > 1) ? atoi(argv[1]) : 0;
+    
     /* Initialize the structure with constants */
-    AllTypes alltypes = {
-        1001,
-        1002,
-        1003,
-        1004,
-        1005,
-        1006,
-        true,
+    AllTypes alltypes = {0};
     
-        1008,
-        1009,
-        1010.0f,
+    alltypes.req_int32         = -1001;
+    alltypes.req_int64         = -1002;
+    alltypes.req_uint32        = 1003;
+    alltypes.req_uint64        = 1004;
+    alltypes.req_sint32        = -1005;
+    alltypes.req_sint64        = -1006;
+    alltypes.req_bool          = true;
     
-        1011,
-        1012,
-        1013.0,
+    alltypes.req_fixed32       = 1008;
+    alltypes.req_sfixed32      = -1009;
+    alltypes.req_float         = 1010.0f;
     
-        "1014",
-        {4, "1015"},
-        {"1016", 1016},
-        MyEnum_Truth,
+    alltypes.req_fixed64       = 1011;
+    alltypes.req_sfixed64      = -1012;
+    alltypes.req_double        = 1013.0;
     
-        1099
-    };
+    strcpy(alltypes.req_string, "1014");
+    alltypes.req_bytes.size = 4;
+    memcpy(alltypes.req_bytes.bytes, "1015", 4);
+    strcpy(alltypes.req_submsg.substuff1, "1016");
+    alltypes.req_submsg.substuff2 = 1016;
+    alltypes.req_enum = MyEnum_Truth;
     
-    uint8_t buffer[512];
+    alltypes.rep_int32_count = 5; alltypes.rep_int32[4] = -2001;
+    alltypes.rep_int64_count = 5; alltypes.rep_int64[4] = -2002;
+    alltypes.rep_uint32_count = 5; alltypes.rep_uint32[4] = 2003;
+    alltypes.rep_uint64_count = 5; alltypes.rep_uint64[4] = 2004;
+    alltypes.rep_sint32_count = 5; alltypes.rep_sint32[4] = -2005;
+    alltypes.rep_sint64_count = 5; alltypes.rep_sint64[4] = -2006;
+    alltypes.rep_bool_count = 5; alltypes.rep_bool[4] = true;
+    
+    alltypes.rep_fixed32_count = 5; alltypes.rep_fixed32[4] = 2008;
+    alltypes.rep_sfixed32_count = 5; alltypes.rep_sfixed32[4] = -2009;
+    alltypes.rep_float_count = 5; alltypes.rep_float[4] = 2010.0f;
+    
+    alltypes.rep_fixed64_count = 5; alltypes.rep_fixed64[4] = 2011;
+    alltypes.rep_sfixed64_count = 5; alltypes.rep_sfixed64[4] = -2012;
+    alltypes.rep_double_count = 5; alltypes.rep_double[4] = 2013.0;
+    
+    alltypes.rep_string_count = 5; strcpy(alltypes.rep_string[4], "2014");
+    alltypes.rep_bytes_count = 5; alltypes.rep_bytes[4].size = 4;
+    memcpy(alltypes.rep_bytes[4].bytes, "2015", 4);
+
+    alltypes.rep_submsg_count = 5;
+    strcpy(alltypes.rep_submsg[4].substuff1, "2016");
+    alltypes.rep_submsg[4].substuff2 = 2016;
+    alltypes.rep_submsg[4].has_substuff3 = true;
+    alltypes.rep_submsg[4].substuff3 = 2016;
+    
+    alltypes.rep_enum_count = 5; alltypes.rep_enum[4] = MyEnum_Truth;
+    alltypes.rep_emptymsg_count = 5;
+    
+    if (mode != 0)
+    {
+        /* Fill in values for optional fields */
+        alltypes.has_opt_int32 = true;
+        alltypes.opt_int32         = 3041;
+        alltypes.has_opt_int64 = true;
+        alltypes.opt_int64         = 3042;
+        alltypes.has_opt_uint32 = true;
+        alltypes.opt_uint32        = 3043;
+        alltypes.has_opt_uint64 = true;
+        alltypes.opt_uint64        = 3044;
+        alltypes.has_opt_sint32 = true;
+        alltypes.opt_sint32        = 3045;
+        alltypes.has_opt_sint64 = true;
+        alltypes.opt_sint64        = 3046;
+        alltypes.has_opt_bool = true;
+        alltypes.opt_bool          = true;
+        
+        alltypes.has_opt_fixed32 = true;
+        alltypes.opt_fixed32       = 3048;
+        alltypes.has_opt_sfixed32 = true;
+        alltypes.opt_sfixed32      = 3049;
+        alltypes.has_opt_float = true;
+        alltypes.opt_float         = 3050.0f;
+        
+        alltypes.has_opt_fixed64 = true;
+        alltypes.opt_fixed64       = 3051;
+        alltypes.has_opt_sfixed64 = true;
+        alltypes.opt_sfixed64      = 3052;
+        alltypes.has_opt_double = true;
+        alltypes.opt_double        = 3053.0;
+        
+        alltypes.has_opt_string = true;
+        strcpy(alltypes.opt_string, "3054");
+        alltypes.has_opt_bytes = true;
+        alltypes.opt_bytes.size = 4;
+        memcpy(alltypes.opt_bytes.bytes, "3055", 4);
+        alltypes.has_opt_submsg = true;
+        strcpy(alltypes.opt_submsg.substuff1, "3056");
+        alltypes.opt_submsg.substuff2 = 3056;
+        alltypes.has_opt_enum = true;
+        alltypes.opt_enum = MyEnum_Truth;
+        alltypes.has_opt_emptymsg = true;
+    }
+    
+    alltypes.end = 1099;
+    
+    uint8_t buffer[1024];
     pb_ostream_t stream = pb_ostream_from_buffer(buffer, sizeof(buffer));
     
     /* Now encode it and check if we succeeded. */
@@ -45,6 +124,7 @@ int main()
     }
     else
     {
+        fprintf(stderr, "Encoding failed: %s\n", PB_GET_ERROR(&stream));
         return 1; /* Failure */
     }
 }

tests/test_encode_callbacks.c

@@ -5,7 +5,7 @@
 #include <pb_encode.h>
 #include "callbacks.pb.h"
 
-bool encode_string(pb_ostream_t *stream, const pb_field_t *field, const void *arg)
+bool encode_string(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
 {
     char *str = "Hello world!";
     
@@ -15,7 +15,7 @@ bool encode_string(pb_ostream_t *stream, const pb_field_t *field, const void *ar
     return pb_encode_string(stream, (uint8_t*)str, strlen(str));
 }
 
-bool encode_int32(pb_ostream_t *stream, const pb_field_t *field, const void *arg)
+bool encode_int32(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
 {
     if (!pb_encode_tag_for_field(stream, field))
         return false;
@@ -23,7 +23,7 @@ bool encode_int32(pb_ostream_t *stream, const pb_field_t *field, const void *arg
     return pb_encode_varint(stream, 42);
 }
 
-bool encode_fixed32(pb_ostream_t *stream, const pb_field_t *field, const void *arg)
+bool encode_fixed32(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
 {
     if (!pb_encode_tag_for_field(stream, field))
         return false;
@@ -32,7 +32,7 @@ bool encode_fixed32(pb_ostream_t *stream, const pb_field_t *field, const void *a
     return pb_encode_fixed32(stream, &value);
 }
 
-bool encode_fixed64(pb_ostream_t *stream, const pb_field_t *field, const void *arg)
+bool encode_fixed64(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
 {
     if (!pb_encode_tag_for_field(stream, field))
         return false;
@@ -41,6 +41,22 @@ bool encode_fixed64(pb_ostream_t *stream, const pb_field_t *field, const void *a
     return pb_encode_fixed64(stream, &value);
 }
 
+bool encode_repeatedstring(pb_ostream_t *stream, const pb_field_t *field, void * const *arg)
+{
+    char *str[4] = {"Hello world!", "", "Test", "Test2"};
+    int i;
+    
+    for (i = 0; i < 4; i++)
+    {
+        if (!pb_encode_tag_for_field(stream, field))
+            return false;
+        
+        if (!pb_encode_string(stream, (uint8_t*)str[i], strlen(str[i])))
+            return false;
+    }
+    return true;
+}
+
 int main()
 {
     uint8_t buffer[1024];
@@ -58,6 +74,8 @@ int main()
     testmessage.submsg.fixed32value.funcs.encode = &encode_fixed32;
     testmessage.submsg.fixed64value.funcs.encode = &encode_fixed64;
 
+    testmessage.repeatedstring.funcs.encode = &encode_repeatedstring;
+    
     if (!pb_encode(&stream, TestMessage_fields, &testmessage))
         return 1;
     

tests/test_encode_extensions.c

@@ -0,0 +1,38 @@
+/* Tests extension fields.
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <pb_encode.h>
+#include "alltypes.pb.h"
+#include "extensions.pb.h"
+
+int main(int argc, char **argv)
+{
+    AllTypes alltypes = {};
+
+    int32_t extensionfield1 = 12345;
+    pb_extension_t ext1 = {&AllTypes_extensionfield1, &extensionfield1, NULL};
+    alltypes.extensions = &ext1;
+
+    ExtensionMessage extensionfield2 = {"test", 54321};
+    pb_extension_t ext2 = {&ExtensionMessage_AllTypes_extensionfield2, &extensionfield2, NULL};
+    ext1.next = &ext2;
+    
+    uint8_t buffer[1024];
+    pb_ostream_t stream = pb_ostream_from_buffer(buffer, sizeof(buffer));
+    
+    /* Now encode it and check if we succeeded. */
+    if (pb_encode(&stream, AllTypes_fields, &alltypes))
+    {
+        fwrite(buffer, 1, stream.bytes_written, stdout);
+        return 0; /* Success */
+    }
+    else
+    {
+        fprintf(stderr, "Encoding failed: %s\n", PB_GET_ERROR(&stream));
+        return 1; /* Failure */
+    }
+}
+

tests/test_missing_fields.c

@@ -27,7 +27,7 @@ int main()
         
         if (!pb_decode(&stream, MissingField_fields, &msg))
         {
-            printf("Decode failed.\n");
+            printf("Decode failed: %s\n", PB_GET_ERROR(&stream));
             return 2;
         }
     }

tests/test_multiple_files.c

@@ -0,0 +1,13 @@
+/*
+ * Tests if still compile if typedefs are redfefined in STATIC_ASSERTS when
+ * proto file includes another poto file
+ */
+
+#include <stdio.h>
+#include <pb_encode.h>
+#include "callbacks2.pb.h"
+
+int main()
+{
+	return 0;
+}

tests/unittestproto.proto

@@ -8,6 +8,10 @@ message FloatArray {
     repeated float data = 1 [(nanopb).max_count = 10];
 }
 
+message StringMessage {
+    required string data = 1 [(nanopb).max_size = 10];
+}
+
 message CallbackArray {
     // We cheat a bit and use this message for testing other types, too.
     // Nanopb does not care about the actual defined data type for callback

tools/set_version.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+
+# Run this from the top directory of nanopb tree.
+# e.g. user@localhost:~/nanopb$ tools/set_version.sh nanopb-0.1.9-dev
+# It sets the version number in pb.h and generator/nanopb_generator.py.
+
+sed -i -e 's/nanopb_version\s*=\s*"[^"]*"/nanopb_version = "'$1'"/' generator/nanopb_generator.py
+sed -i -e 's/#define\s*NANOPB_VERSION\s*.*/#define NANOPB_VERSION '$1'/' pb.h
+
+