+2017-06-08 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * char-ucs.h (MAX_LEADING_BYTE_PRIVATE): Use 1024 instead of 512.
+
+2016-11-17 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * chartab.h (USE_CONCORD_OBJECT_SYSTEM): New macro; use it instead
+ of `HAVE_LIBCHISE' to detect to use Concord Object System.
+ (USE_CONCORD_OBJECT_SYSTEM_TO_COMPOSE): New macro.
+
+ * text-coding.c: Use `USE_CONCORD_OBJECT_SYSTEM_TO_COMPOSE'
+ instead of `HAVE_LIBCHISE' to detect to use Concord Object System
+ for character composing.
+
+ * mule-charset.c (charset_code_point): Use
+ `USE_CONCORD_OBJECT_SYSTEM' instead of `HAVE_LIBCHISE' to detect
+ to use Concord Object System.
+
+2016-11-16 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * chartab.c: Use `USE_CONCORD_OBJECT_SYSTEM' instead of
+ `HAVE_LIBCHISE' to detect to use Concord Object System.
+
+2016-07-23 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * text-coding.c (Fmake_coding_system): Accept property
+ `charset-g3' for CODESYS_UTF8.
+ (char_encode_utf8): Refer `charset-g3'.
+
+2015-11-25 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * chartab.c (Fchar_feature): When argument ATTRIBUTE is a CCS and
+ whose name is =>FOO, don't try to find =FOO; fix problem when
+ argument ATTRIBUTE is ==>FOO.
+
+2013-05-19 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * text-coding.c (COMPOSE_ADD_CHAR): Use
+ `concord_object_get_attribute' instead of
+ `concord_object_get_feature_value'.
+
+ * chartab.c (char_table_get_db_cos): Use
+ `concord_object_get_attribute' instead of
+ `concord_object_get_feature_value'.
+
+2013-05-02 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * chartab.h (get_char_id_table_ce): Use
+ `load_char_attribute_maybe_cos' instead of
+ `load_char_attribute_maybe'.
+
+ * chartab.c (load_char_attribute_maybe_cos): Return NULL instead
+ of Qunbound as the default value.
+
+2013-05-01 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * chartab.h (load_char_attribute_maybe_cos): New prototype when
+ HAVE_LIBCHISE is defined.
+ (get_char_id_table_ce): New inline function when HAVE_LIBCHISE is
+ defined.
+
+ * mule-charset.c (charset_code_point): Use `get_char_id_table_ce'
+ instead of `get_char_id_table' when HAVE_LIBCHISE is defined.
+ (load_char_decoding_entry_maybe): Don't use
+ `decoding_table_put_char' to avoid GC.
+
+ * text-coding.c (concord_setup_env_maybe): Abolished.
+ (COMPOSE_ADD_CHAR): Use `open_chise_data_source_maybe' instead of
+ `concord_setup_env_maybe'.
+
+2013-05-01 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * chartab.c (open_chise_data_source_maybe): New implementation;
+ don't use `build_string' and `Fexpand_file_name' to avoid GC.
+ (char_table_get_db_cos): New function when HAVE_LIBCHISE is
+ defined.
+ (load_char_attribute_maybe_cos): Likewise.
+
+2013-04-23 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * text-coding.c (struct decoding_stream): Use COS_object instead
+ of Lisp_Object for `combining_table' if HAVE_LIBCHISE is defined.
+ (reset_decoding_stream): Likewise.
+ (COMPOSE_FLUSH_CHARS): Likewise.
+ (concord_setup_env_maybe): New function.
+ (COMPOSE_ADD_CHAR): Use COS API for `str->combining_table' if
+ HAVE_LIBCHISE is defined.
+
+2013-02-16 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * chartab.c (find_char_feature_in_family): New implementation;
+ don't support multiple parents; support non-list mother.
+
+2012-06-11 MORIOKA Tomohiko <tomo.git@chise.org>
+
+ * text-coding.c (Qenable_decomposition): New variable in XEmacs
+ CHISE.
+ (Fmake_coding_system): Add new property `enable-decomposition' in
+ XEmacs CHISE.
+ (char_encode_utf8): Don't decompose character if
+ CODING_SYSTEM_ENABLE_DECOMPOSITION (str->codesys) is false.
+ (syms_of_file_coding): Add new symbol `enable-decomposition' in
+ XEmacs CHISE.
+ (complex_vars_of_file_coding): Define new coding-system-property
+ `enable-decomposition' in XEmacs CHISE.
+
+ * file-coding.h (struct Lisp_Coding_System): Add
+ `enable_decomposition' in XEmacs CHISE.
+ (CODING_SYSTEM_ENABLE_DECOMPOSITION): New macro in XEmacs CHISE.
+ (XCODING_SYSTEM_ENABLE_DECOMPOSITION): New macro in XEmacs CHISE.
+
+2012-02-17 MORIOKA Tomohiko <tomo@zinbun.kyoto-u.ac.jp>
+
+ * text-coding.c (DECODE_ADD_UCS_CHAR): Add `static'.
+
+2012-02-11 MORIOKA Tomohiko <tomo@zinbun.kyoto-u.ac.jp>
+
+ * text-coding.c (DECODE_ADD_UCS_CHAR): Don't use `INLINE_HEADER'.
+
+ * chartab.c (UINT8_DECODE): Don't use `INLINE_HEADER'.
+ (UINT16_DECODE): Ditto.
+
+ * glyphs-eimage.c (png_instantiate):
+ - Use `png_get_image_height (png_ptr, info_ptr)' instead of
+ `info_ptr->height'.
+ - Use `png_get_image_width (png_ptr, info_ptr)' instead of
+ `info_ptr->width'.
+ - Use `png_get_color_type(png_ptr, info_ptr)' instead of
+ `info_ptr->color_type'.
+ - Use `png_get_bit_depth(png_ptr, info_ptr)' instead of
+ `info_ptr->bit_depth'.
+
2011-07-20 MORIOKA Tomohiko <tomo@zinbun.kyoto-u.ac.jp>
* mule-charset.c (charset_code_point): If nil is specified as a