1 # utf8proc release history #
7 - New functions `utf8proc_map_custom` and `utf8proc_decompose_custom`
8 to allow user-supplied transformations of codepoints, in conjunction
9 with other transformations ([#89]).
11 - New function `utf8proc_normalize_utf32` to apply normalizations
12 directly to UTF-32 data (not just UTF-8) ([#88]).
14 - Fixed stack overflow that could occur due to incorrect definition
15 of `UINT16_MAX` with some compilers ([#84]).
17 - Fixed conflict with `stdbool.h` in Visual Studio ([#90]).
19 - Updated font metrics to use Unifont 9.0.04.
25 - Move `-Wmissing-prototypes` warning flag from `Makefile` to `.travis.yml`
26 since MSVC does not understand this flag and it is occasionally useful to
27 build using MSVC through the `Makefile` ([#79]).
29 - Use a different variable name for a nested loop in `bench/bench.c`, and
30 declare it in a C89 way rather than inside the `for` to avoid "error:
31 'for' loop initial declarations are only allowed in C99 mode" ([#80]).
37 - Bug fix in `utf8proc_grapheme_break_stateful` ([#77]).
39 - Tests now use versioned Unicode files, so they will no longer
40 break when a new version of Unicode is released ([#78]).
46 - Updated for Unicode 9.0 ([#70]).
48 - New `utf8proc_grapheme_break_stateful` to handle the complicated
49 grapheme-breaking rules in Unicode 9. The old `utf8proc_grapheme_break`
50 is still provided, but may incorrectly identify grapheme breaks
51 in some Unicode-9 sequences.
53 - Smaller Unicode tables ([#62], [#68]). This required changes
54 in the `utf8proc_property_t` structure, which breaks backward
55 compatibility if you access this `struct` directly. The
56 functions in the API remain backward-compatible, however.
58 - Buffer overrun fix ([#66]).
64 - Do not export symbol for internal function `unsafe_encode_char()` ([#55]).
66 - Install relative symbolic links for shared libraries ([#58]).
68 - Enable and fix compiler warnings ([#55], [#58]).
70 - Add missing files to `make clean` ([#58]).
76 - Updated for Unicode 8.0 ([#45]).
78 - New `utf8proc_tolower` and `utf8proc_toupper` functions, portable
79 replacements for `towlower` and `towupper` in the C library ([#40]).
81 - Don't treat Unicode "non-characters" as invalid, and improved
82 validity checking in general ([#35]).
84 - Prefix all typedefs with `utf8proc_`, e.g. `utf8proc_int32_t`,
85 to avoid collisions with other libraries ([#32]).
87 - Rename `DLLEXPORT` to `UTF8PROC_DLLEXPORT` to prevent collisions.
89 - Fix build breakage in the benchmark routines.
91 - More fine-grained Makefile variables (`PICFLAG` etcetera), so that
92 compilation flags can be selectively overridden, and in particular
93 so that `CFLAGS` can be changed without accidentally eliminating
94 necessary flags like `-fPIC` and `-std=c99` ([#43]).
96 - Updated character-width tables based on Unifont 8.0.01 ([#51]) and
97 the Unicode 8 character categories ([#47]).
103 - Updated for Unicode 7.0 ([#6]).
105 - New function `utf8proc_grapheme_break(c1,c2)` that returns whether
106 there is a grapheme break between `c1` and `c2` ([#20]).
108 - New function `utf8proc_charwidth(c)` that returns the number of
109 column-positions that should be required for `c`; essentially a
110 portable replacment for `wcwidth(c)` ([#27]).
112 - New function `utf8proc_category(c)` that returns the Unicode
113 category of `c` (as one of the constants `UTF8PROC_CATEGORY_xx`).
114 Also, a function `utf8proc_category_string(c)` that returns the Unicode
115 category of `c` as a two-character string.
117 - `cmake` script `CMakeLists.txt`, in addition to `Makefile`, for
118 easier compilation on Windows ([#28]).
120 - Various `Makefile` improvements: a `make check` target to perform
121 tests ([#13]), `make install`, a rule to automate updating the Unicode
124 - The shared library is now versioned (e.g. has a soname on GNU/Linux) ([#24]).
126 - C++/MSVC compatibility ([#17]).
128 - Most `#defined` constants are now `enums` ([#29]).
130 - New preprocessor constants `UTF8PROC_VERSION_MAJOR`,
131 `UTF8PROC_VERSION_MINOR`, and `UTF8PROC_VERSION_PATCH` for compile-time
132 detection of the API version.
134 - Doxygen-formatted documentation ([#29]).
136 - The Ruby and PostgreSQL plugins have been removed due to lack of testing ([#22]).
142 - PostgreSQL 9.2 and 9.3 compatibility (lowercase `c` language name)
148 - Use `RSTRING_PTR()` and `RSTRING_LEN()` instead of `RSTRING()->ptr` and
149 `RSTRING()->len` for ruby1.9 compatibility (and `#define` them, if not
154 - Patches for compatibility with Microsoft Visual Studio
158 - Fixes to make utf8proc usable in C++ programs
166 - replaced C++ style comments for compatibility reasons
167 - added typecasts to suppress compiler warnings
168 - removed redundant source files for ruby-gemfile generation
172 - Changed copyright notice for Public Software Group e. V.
173 - Minor changes in the `README` file
179 - Added a function `utf8proc_version` returning a string containing the version
180 number of the library.
181 - Included a target `libutf8proc.dylib` for MacOSX.
184 - PostgreSQL 8.3 compatibility (use of `SET_VARSIZE` macro)
190 - Fixed a serious bug in the data file generator, which caused characters
191 being treated incorrectly, when stripping default ignorable characters or
192 calculating grapheme cluster boundaries.
198 - Added a new PostgreSQL function `unistrip`, which behaves like `unifold`,
199 but also removes all character marks (e.g. accents).
203 - Changed license from BSD to MIT style.
204 - Added a new function `utf8proc_codepoint_valid` to the C library.
205 - Changed compiler flags in `Makefile` from `-g -O0` to `-O2`
206 - The ruby script, which was used to build the `utf8proc_data.c` file, is now
207 included in the distribution.
213 - Fixed a bug in the ruby library, which caused an error, when splitting an
214 empty string at grapheme cluster boundaries (method `String#utf8chars`).
220 - included a check in `Integer#utf8`, which raises an exception, if the given
221 code-point is invalid because of being too high (this was missing yet)
225 - added support for PostgreSQL version 8.2
231 - included a gem file for the ruby version of the library
233 Release of version 1.0.1
239 - added the `LUMP` option, which lumps certain characters together (see `lump.md`) (also used for the PostgreSQL `unifold` function)
240 - added the `STRIPMARK` option, which strips marking characters (or marks of composed characters)
241 - deprecated ruby method `String#char_ary` in favour of `String#utf8chars`
247 - changed normalization from NFC to NFKC for postgresql unifold function
251 - added support to mark the beginning of a grapheme cluster with 0xFF (option: `CHARBOUND`)
252 - added the ruby method `String#chars`, which is returning an array of UTF-8 encoded grapheme clusters
253 - added `NLF2LF` transformation in postgresql `unifold` function
254 - added the `DECOMPOSE` option, if you neither use `COMPOSE` or `DECOMPOSE`, no normalization will be performed (different from previous versions)
255 - using integer constants rather than C-strings for character properties
256 - fixed (hopefully) a problem with the ruby library on Mac OS X, which occurred when compiler optimization was switched on
262 - changed behaviour of PostgreSQL function to return NULL in case of invalid input, rather than raising an exceptional condition
263 - improved efficiency of PostgreSQL function (no transformation to C string is done)
267 - added -fpic compiler flag in Makefile
268 - fixed bug in the C code for the ruby library (usage of non-existent function)
272 2006-06-02: initial release of version 0.1
274 [#6]: https://github.com/JuliaLang/utf8proc/issues/6
275 [#13]: https://github.com/JuliaLang/utf8proc/issues/13
276 [#17]: https://github.com/JuliaLang/utf8proc/issues/17
277 [#20]: https://github.com/JuliaLang/utf8proc/issues/20
278 [#22]: https://github.com/JuliaLang/utf8proc/issues/22
279 [#24]: https://github.com/JuliaLang/utf8proc/issues/24
280 [#27]: https://github.com/JuliaLang/utf8proc/issues/27
281 [#28]: https://github.com/JuliaLang/utf8proc/issues/28
282 [#29]: https://github.com/JuliaLang/utf8proc/issues/29
283 [#32]: https://github.com/JuliaLang/utf8proc/issues/32
284 [#35]: https://github.com/JuliaLang/utf8proc/issues/35
285 [#40]: https://github.com/JuliaLang/utf8proc/issues/40
286 [#43]: https://github.com/JuliaLang/utf8proc/issues/43
287 [#45]: https://github.com/JuliaLang/utf8proc/issues/45
288 [#47]: https://github.com/JuliaLang/utf8proc/issues/47
289 [#51]: https://github.com/JuliaLang/utf8proc/issues/51
290 [#55]: https://github.com/JuliaLang/utf8proc/issues/55
291 [#58]: https://github.com/JuliaLang/utf8proc/issues/58
292 [#62]: https://github.com/JuliaLang/utf8proc/issues/62
293 [#66]: https://github.com/JuliaLang/utf8proc/issues/66
294 [#68]: https://github.com/JuliaLang/utf8proc/issues/68
295 [#70]: https://github.com/JuliaLang/utf8proc/issues/70
296 [#77]: https://github.com/JuliaLang/utf8proc/issues/77
297 [#78]: https://github.com/JuliaLang/utf8proc/issues/78
298 [#79]: https://github.com/JuliaLang/utf8proc/issues/79
299 [#80]: https://github.com/JuliaLang/utf8proc/issues/80
300 [#84]: https://github.com/JuliaLang/utf8proc/pull/84
301 [#88]: https://github.com/JuliaLang/utf8proc/pull/88
302 [#89]: https://github.com/JuliaLang/utf8proc/pull/89
303 [#90]: https://github.com/JuliaLang/utf8proc/issues/90