* doxygenize and update API documentation
* fix spelling
* add "fuzz" corpus for java decoder to improve coverage
* use upper-case-snake names for dictionary constant definitions
* use `LDFLAGS` in conventional `Makefile`
* Declare `BUILD_SHARED_LIBS` option for CMake
* Define `${LIB}_SHARED_COMPILATION` when compiling shared library
* Define and use BROTLI_xxx_API
* Fix remaining unprefixed defines in port.h
* move `common/port.h` to `includes/port.h`
* replace magic more magic numbers with constants
* artificially limit window size to 2^18 for quality 0 and 1
* use fixed shifts for quality 0 and 1 hashes
* removed `BrotliEncoderWriteMetadata`
* added `BROTLI_OPERATION_EMIT_METADATA` instead
* deprecated low-level API
* fixed MSVC warnings
* add Brotli*TakeOutput methods
* * flushing now doesn't require additional call
* add Brotli*Version methods
* moved public headers to 'public' directory
* removed C++ API
* do not assume STDC_VERSION is defined
* condense printf in port.h; use BROTLI_LOG everywhere
* mark non-exported functions with BROTLI_INTERNAL
* use BROTLI_DUMP instead of (void)(BROTLI_FAILURE())
* fix problems with CustomDictionary
* make decode.h independent of state.h
* fix "double-new-lines"
* fix some strict compilation warnings
* fix bro.cc compilation for MSVS
* added compressor.h as a replacement for encode.h + streams.h
While there, add -Wmissing-prototypes and -Wmissing-declarations
to shared.mk in order to catch similar errors in the future.
Signed-off-by: Piotr Sikora <piotrsikora@google.com>
Instead of returning false, if it is called with no
new input for a non-last block, just check if it has
any already processed data to flush, and if not,
return true with empty output.
The new modes process the input data in independent blocks,
using backward references only from within an input block.
The new modes can be used by specifying quality 0 or quality 1,
the old quality 1 and quality 2 modes are renamed quality 2 and
quality 3, respectively, and the old quality 3 mode is removed.
In quality 1, use static Huffman codes for distance
and command histograms with <= 128 symbols and dynamic
Huffman codes with static code length codes for the other
histograms.
