Commit graph

52 commits

Author SHA1 Message Date
Charlie Gordon
d210ba864b charset: improve eol and binary detection
- accept extra CR in badly encoded DOS files
- ignore ^Z in DOS files, do not detect them as binary
2017-04-25 16:56:26 +00:00
Charlie Gordon
d329669530 buffer: new APIs
- int eb_insert_uchars(EditBuffer *b, int offset, int c, int n);
- int eb_replace_uchar(EditBuffer *b, int offset, int c);
- inline eb_insert_spaces() using eb_insert_uchars() instead
- add minp(), maxp() and clampp() with side effect on first argument
- publish ucs2 and ucs4 charsets
2017-04-25 16:46:41 +00:00
Charlie Gordon
3580b53452 syntax: improve minibuffer and popup handling
- remove static state for minibuffer handling, use mode specific state attached
  to the buffer.
- remove static pointer for the popup last active window.  Use new
  target_window member in EditState.
- remove EditState.minibuf, replaced by WF_MINIBUF flag in EditState.flags
- rename minibuffer commands for consistency
- remove unused global variable debug_flags
- prevent search from minibuffer
- fix toggle-full-screen refresh bug. C-x f is bound to toggle-full-screen
  and could be typed by mistake for C-x C-f, switching to full screen mode
  but with a refresh delay.  This caused confusing behavior interpreted as a
  layout bug.
- make charset_ucs2le, charset_ucs2be, charset_ucs4le, charset_ucs4be local
  in charset.c
- make edit_detach() and edit_attach() local in qe.c
- redraw status and diag messages in edit_display(). This caused display
  bugs in x11 as asynchronous messages were erased by minibuffer window
  refresh.
- display status messages in diag area (bottom right screen corner) if
  minibuffer is active.
- force put_status message update if message starts with !
- force complete_refresh if window was resized or layout was changed
2017-04-17 18:14:15 +00:00
Charlie Gordon
e08867b692 basic: reduce global variable clutter
- make more registered objects static
- document some non static global objects
- remove some global variables
2017-04-16 21:51:19 +00:00
Charlie Gordon
2d59865702 charset: default to mac-roman for non utf-8 files with old style mac eol 2017-03-22 17:13:10 +00:00
Charlie Gordon
26fd39d9e0 unicode: improve support for wide and combining code points
- qe_isaccent() recognises all combining code points (to be refined)
- add computer generated width table unicode_width.h
- display difference in hex in compare-windows to distinguish
  space and nbsp, normal and combined form for accented letters.
- display multiple combining code points in what-cursor-position (C-x =)
- attach combining code points to preceeding space in display_char_bidir
- handle accents specificaaly in flush_fragment
- handle combining code-points in screen backstore via indirection table
  this transient table can be inspected via describe-screen
2017-03-16 09:16:31 +00:00
Charlie Gordon
ec07d5aec8 basic: charset improvements
- add QECharset.encode_table as a default for CharsetDecodeState.table
- make CharsetDecodeState.table const
- use static initializers for utf8 tables
- add al32utf8 charset name (used in Oracle for true utf-8 encoding)
- use direct pointers in unicode_glyph_range_index
- initialize unicode_glyph_range_index at run time
- rename unicode_glyph_tty_width as unicode_tty_glyph_width
- changed unihex display to space out single width characters
2017-03-15 23:42:23 +00:00
Charlie Gordon
252e2848ab basic: fix potentially invalid shifts
- use unsigned arithmetics to avoid implementation-defined behavior
  on (1 << 31)
2017-03-15 23:17:58 +00:00
Charlie Gordon
b962e33ade updated copyrights 2016-03-06 19:53:06 +00:00
Charlie Gordon
056db10c78 charset: fixed scanning bug for DOS eol modes
- goto_char had infinite loops for EOL_DOS in 8, 16 and 32 bits modes.
- this bug caused qemacs to hang on searches on DOS/Windows files.
2016-03-06 17:31:44 +00:00
Charlie Gordon
1f5a77a784 portability: stop defining symbols starting with __ 2015-09-16 22:18:22 +00:00
Charlie Gordon
f0aa3a4bd8 cosmetics: untabify, remove trailing spaces 2015-08-27 23:03:39 +00:00
Charlie Gordon
490a858c22 general: sanity checks
- make QECharset const by default, use struct QECharset for mutable references.
- use OWNED attribute to flag unique pointers to allocated memory.
- add typedefs for InputMethod, ISearchState
- use typedef for ModeDef
2015-08-19 16:22:30 +00:00
Charlie Gordon
a66cf5a68e minibuffer completion: add fuzzy matching
- add extra group argument in functions add_string() and set_string()
- group matches in completion_sort_func()
- add const char *strmem(const char *str, const void *mem, int size);
- add variable fuzzy-search
- add minibuffer-insert command to handle insert triggers
- display variable status after set-variable
- make TAB a regular char in incremental search
- display control chars as ^x in incremental search prompt
2015-08-11 17:45:30 +00:00
Charlie Gordon
aa2b5cc910 fix scoring of 4 byte charsets 2014-05-31 00:26:26 +00:00
Charlie Gordon
ebc8f89a82 prevent strict warnings 2014-05-27 22:59:48 +00:00
Charlie Gordon
e99211a8d1 fix automatic eol-type test for 16 bit charsets 2014-04-07 20:48:49 +00:00
Charlie Gordon
b0523a2dc9 improve utf8 and ucsx probing functions 2014-03-24 09:33:49 +00:00
Charlie Gordon
6d4d7c0e36 fix charset_detect
* use charset probing functions to detect ucs2 and ucs4 charsets in files
  without a BOM (byte order mark)
* add test files for multibyte charsets in tests directory
2014-03-15 14:55:05 +00:00
Charlie Gordon
de97ca1fe8 Add charset probing functions
* add QECharset method probe_func() to test charset likelihood
* probe function is not used for automatic charset detection yet
* fix kmaptoqe.c to use stable sorting and make kmaps reproducible
* fix ligtoqe.c to use stable sorting and make ligatures reproducible
2014-03-01 22:37:26 +00:00
Charlie Gordon
05ab3dc386 Improve charset detection and handling
* add charset_raw for binary files
* improve charset detection for ambiguous cases
* add do_show_coding_system()
* add do_set_auto_coding() to (re)select the best coding system
* handle BOM mark: display as \ufeff and ignore for syntax coloring
2014-02-10 20:29:26 +00:00
Charlie Gordon
6dc603ae27 fix some problems with dos eol_type.
* skip \n if EOL_DOS in get_pos and goto_line charset methods
* detect eol_type automatically for non 8 bit charsets
2014-02-06 00:19:39 +00:00
Charlie Gordon
a7853b3dff add support for end of line types
* support 3 types of end of line: Unix, Dos and old style Mac
* detect eol type automatically in detect_charset
* set buffer eol_type together with charset
* transparently convert eol sequence to \n upon reading buffers
  characters in eb_nextc().
* transparently convert \n to eol sequence in eb_encode_uchar()
* handle eol types in convert-file-buffer-coding-system: either preserve
  current value or force it via charset suffix -unix, -doc and -mac
* change read_charset() to accept eol_type suffix
* add eol_type and eol_char to CharsetDecodeState
* change charset methods to take CharsetDecodeState instead of charset pointers
* display eol_type in mode line if not unix
2014-02-05 00:56:48 +00:00
Charlie Gordon
d0c2e1d3e6 replace unicode_to_charset with eb_encode_uchar
* prepare support for eol_type support by making char encoding buffer specific
* fix small bugs and unsafe cases.
* fix bug in eb_goto_pos for EBCDIC charsets
2014-02-04 22:47:31 +00:00
Charlie Gordon
7037156cf6 fix charset bugs, caused infinite loop in usc4 2014-01-11 18:58:08 +00:00
Charlie Gordon
deb6972ca3 better charset handling
* make eb_insert_buffer return the number of bytes inserted
* reset page bits upon charset change
* add eb_delete_nextc: delete one encoded character at offset
* add eb_insert_uchar: insert a unicode character with encoding
* add eb_insert_utf8_buf: insert a utf8 encoded buffer with conversion
* make eb_printf insert utf8 bytes with conversion
* add eb_insert_buffer_convert: copy chars between buffers with charset
  conversion
* handle charset conversion correctly in various places
* insert wide caracters from hex mode
* create most buffers as BF_UTF8 by default
* set kill-buffer charset to that of kill source
* convert charsets when yanking
* handle charsets in search/replace system
2014-01-06 09:40:20 +00:00
Charlie Gordon
ee91b6c61f register ucs4 charsets 2014-01-06 04:30:09 +00:00
Charlie Gordon
02dec8bf1a added generic utility function unicode_glyph_tty_width in charset.c
* moved and improved code from tty_term_glyph_width
2014-01-02 11:28:30 +00:00
Charlie Gordon
5ee19ed240 default ucs2/ucs4 to big endian, remark on iso-2220-jp 2008-05-21 06:42:32 +00:00
Charlie Gordon
23cd92e91c fixed find-charset to use strxfind, implement strxfind correctly 2008-05-04 15:54:39 +00:00
Charlie Gordon
9cf6a3cb87 simplified buffer decoding methods: pass ptr in CharsetDecodeState 2008-05-02 17:05:38 +00:00
Charlie Gordon
e493a2e99c fixed many problems with charset encodings
added charset methods for buffer line and position computations
added QECharset fields for char_size and variable_size indicator
use charset's eol_char to count lines in buffers: this should fix EBCDIC support
added support for ucs2 and ucs4 little and big endian encodings
charset's decode_func call no longer be NULL
eb_nextc and eb_prev now require non NULL third parameter
2008-05-02 16:04:09 +00:00
Charlie Gordon
f58f2920de accept NULL in find_charset() 2008-03-29 17:41:46 +00:00
Charlie Gordon
28a0bd2c99 simplifications 2008-03-27 07:27:07 +00:00
Charlie Gordon
a5eae3ed92 bumped copyrights, some cosmetics 2008-01-08 16:37:54 +00:00
Charlie Gordon
98939ddcac changed API for completion functions:
- use CompleteState struct
- use complete_test for each potential match with simple prefix comparison
added new smart string compare function qe_collate to sort strings in
  lexicographical order but sort numeric portions in numerical order
use qe_collate in popup lists and dired
removed redundant alias "iso8859-2"
added strequal(a,b) as an alias for !strcmp(a,b)
  this function makes string collation with strcmp more explicit
  this even fixed a spurious bug in do_set_emulation
2008-01-08 03:42:53 +00:00
Charlie Gordon
ae163f897b fixed crash on charset_completion
renamed css_strtolower -> qe_strtolower
use offsetof for SAVED_DATA_SIZE macro (this stuff must go)
cosmetics in css_get_enum
remarks
fixed text-mode colorization bug: because of saved_data stuff,
  must reset colorization function after mode restore.  This saved data
  stuff is bogus and should be rewritten
smaller version of roman numeral converter
fixed alpha converter
smaller implementation of css_counter_str
fixed list_mode name: no longer generates extra mode completion entry
2008-01-07 16:25:22 +00:00
Charlie Gordon
a71b9efbb4 renamed do_set_mode -> edit_set_mode
use match_extension in mode probe functions
pass move indicator to do_return, removed do_open_line
renamed do_cmd_set_mode -> do_set_mode for consistency
separated strfind and strcasefind, optimized implementation
published register_input_method
unpublished input_methods
optimized KeyDef structure layout
changed eb_search arguments
used eb_delete_range
comments, remarks
better tests on fread failures
set buffer to read-only upon loading non writable file
do_insert_file tests for failure
  set mark at beginning of inserted block and offset at the end
fixed search for word boundary case
made raw_load_buffer1 return number of bytes read, of -1 on error
2008-01-03 18:31:24 +00:00
Charlie Gordon
4743fa1308 added remove_slash and append_slash path utilities
renamed canonize to canonicalize
added crude support for ~user -> /home/user
added win32 specific stuff in basename
added C-/ as an alias for C-_
some cosmetics, moved code
create files with mode 0644 in raw_save_buffer
untabified, removed redundant spaces
modified code generators to the same effect
2008-01-03 09:51:29 +00:00
Charlie Gordon
c83a481c6e remark 2007-12-18 12:19:32 +00:00
Charlie Gordon
5768f42cd8 no longer use ctype.h macros and functions
- ctype.h has strong constraints on arguments
- ctype.h implements bogus irrelevant inefficient locale specific behaviour
- use replacement css_xxx functions
- all take int arguments: can be any value, yet only test ASCII subset
- all css_isxxx return true boolean (1 or 0)
- removed utoupper(int) -> css_toupper(int) has same functionality

resolved ffmpeg clash on string functions from cutils.c module
- added #defines to make them local
- renamed some strixxx functions to strxxxx (extended behaviour)
- moved case insensitive stuff to util.c

fixed bug in strtokey: C-M-x combinations were not recognized
improved eb_changecase
fixed C comment parsing in css parser
2007-12-15 16:02:47 +00:00
Charlie Gordon
26ea3f69a0 added wrappers and utilities for safer memory allocation:
type *qe_malloc(type);
type *qe_mallocz(type);
type *qe_malloc_array(type, count);
type *qe_mallocz_array(type, count);
type *qe_malloc_hack(type, extra_bytes);
type *qe_mallocz_hack(type, extra_bytes);
void qe_free(&ptr);
void *qe_realloc(&ptr, new_size);
char *qe_strdup(const char *str);
void *qe_malloc_bytes(size_t size);
void *qe_mallocz_bytes(size_t size);
void *qe_malloc_dup(const void *src, size_t size);

pass pointer address to bmp_free()
2007-12-12 11:49:02 +00:00
Charlie Gordon
0507ffacc3 fixed function key mapping issues in X11 and Wndz
fixed C lang colorizer for some multiline stuff, wide char/strings
use countof() macro
cosmetics
2007-12-10 13:28:07 +00:00
Charlie Gordon
0446a52d21 added more charsets
simplified charset alias system
reorganized charset functions
groomed completion functions
2007-12-06 15:28:29 +00:00
Charlie Gordon
134ced6963 removed charset_table.c
added charsetjis.c and charsetjis.def
moved code from charsetmore.c to charset.c and charsetjis.c
improved cp parser cptoqe.c, make it generate charsetmore.c
small changes in QECharset structure: added eol_char and made aliases const
removed redundant QECharset declarations from qe.h
separate JIS table generation into jistoqe.c, make it generate charsetjis.def
updated Makefile, .cvsignore
improved fbftoqe.c
improved kmaptoqe.c, added --dump option
added argc, argv to QEmacsState
improved set_user_option: put runtime and user directories in front of res_path
added some missing file headings
2007-12-03 16:15:01 +00:00
Charlie Gordon
494e9f7866 include cflags.mk for extra gcc warnings if present
utf8_encode now returns the number of characters copied,
  simplified code accordingly
made local functions static
tagged unused function arguments
fixed shadowing issues
added support for GCC attributes for unused arguments and format strings
use snprintf() instead of sprintf()
moved scattered declarations to qe.h
removed duplicate declarations in header files
eb_write() payload argument is now const
fixed crash bug: reset trace_buffer upon deletion
improved trace_buffer readability: trace all pseudo tty streams separately
added missing structure initializers
removed buffered string handling from cutils.h
added new buffered string API for statically allocated buffers (should unify)
small improvements in X11 initialization
improved consistency in tty.c (function names...)
  added STDIN and STDOUT streams in QEditScreen to stop using stdin and stdout
  use unlocked stream I/O if available (faster)
  changed and documented terminal setup and exit sequences
  improved charset handling
mode_line functions now return the character count (like snprintf)
2006-12-31 19:34:00 +00:00
Charlie Gordon
3fa9f0ab84 moved charset_complete() to charset.c
made charset names case insensitive (also ignore spaces, dashes, underscores)
2006-12-11 23:17:03 +00:00
Charlie Gordon
ade489eba5 msg 2005-07-08 09:30:43 +00:00
Charlie Gordon
c52d77ea70 cosmetics only (spacing, untabify) 2005-05-11 15:12:14 +00:00
Charlie Gordon
421067af0b added fake charset vt100 for shell buffer
use simpler utf8 terminal test, ie: compatible with shell buffer.
added terminal dependant escape sequences,
  with limited support for vt100 variants.
more precise escape sequence parser in shell buffer.
trace shell output in trace_buffer.
improved file_completion :
  fix buffer overflows,
  ignore . and .. and backup files
  ignore file_completion_ignore_extensions,
2005-05-09 06:21:49 +00:00