VARCem/PDCurses
1
0
Fork 0
mirror of https://github.com/VARCem/PDCurses.git synced 2026-09-24 07:45:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • 0539bbfe62 Redundant declarations. William McBrine 2006-07-18 12:25:51 +00:00
  • ef8be8e68f c_save_trm: unused. William McBrine 2006-07-18 11:30:28 +00:00
  • 3085219215 PDC_newline() can be a little more private. William McBrine 2006-07-18 11:25:47 +00:00
  • 89936d5381 CURSES_CLEAR: no longer used. William McBrine 2006-07-17 22:08:20 +00:00
  • 1a437a4d22 PDC_gotoxy() was named wrong. Like other curses functions, it actually takes the y parameter first. William McBrine 2006-07-17 21:45:16 +00:00
  • f3ce1371d8 Oops... pointless non-optimization. William McBrine 2006-07-17 21:05:18 +00:00
  • f47dc31e66 Moved CURSES_REFRESH call from doupdate() to PDC_transform_line(). This is very slightly less efficient, but allows for better separation of the code. William McBrine 2006-07-17 20:26:22 +00:00
  • eb04c7c604 PDC_chg_attrs() -- a nice function that's never been used. (And incorrectly prototyped until 2.8.) Similar functionality is now available in the standard wchgat(). William McBrine 2006-07-17 07:02:30 +00:00
  • 76c9c885ee Comments. William McBrine 2006-07-17 02:20:48 +00:00
  • 14475c9591 Removed the XCURSES ifdefs from Xinitscr() and delscreen(), which now also calls XCursesExit(). William McBrine 2006-07-16 19:56:08 +00:00
  • 7c3c6bbccf The display_name parameter of XCursesInitscr() and XCursesSetupX() was never used. William McBrine 2006-07-16 18:41:53 +00:00
  • c8aa43edef Common initialized values from PDC_scr_open() to initscr(). To Do: Some of these aren't really used. William McBrine 2006-07-16 18:08:16 +00:00
  • c53fda8dd3 Second parameter to PDC_scr_open() was always 0. William McBrine 2006-07-16 16:42:31 +00:00
  • b161602300 tahead is already initialized in OS/2's PDC_scr_open(), and the call to v_init() should be moved there. William McBrine 2006-07-16 16:12:45 +00:00
  • f05e223ee0 Moved PDC_fix_cursor() prototype to pdcdisp.c. William McBrine 2006-07-16 02:00:40 +00:00
  • c4a6115e78 Moved PDC_slk_calc() prototype to slk.c. William McBrine 2006-07-16 01:26:23 +00:00
  • de4000af6f Moved XCursesInstruct() prototype to x11/pdcx11.h. William McBrine 2006-07-16 00:45:15 +00:00
  • c0e7bc6e99 Meh. William McBrine 2006-07-15 21:03:38 +00:00
  • d6de930a5e Added test for flash(). William McBrine 2006-07-15 21:03:26 +00:00
  • a8d7ac67ed flash() is off the to-do list. William McBrine 2006-07-15 20:57:59 +00:00
  • 39b10e2ac2 flash() was not implemented for Win32 or X (despite much code to that effect). Here's a nice portable implementation, which also allows me to remove PDC_scroll() (this is the last place it was used). To Do: Should the delay be longer? William McBrine 2006-07-15 20:46:24 +00:00
  • e9402b29c1 Removed empty implementations of PDC_fix_cursor(). To Do: Complete removal? William McBrine 2006-07-15 16:07:06 +00:00
  • 34f6b6d2ae Useless comments. William McBrine 2006-07-15 15:38:24 +00:00
  • ab0f0daa75 Comments. William McBrine 2006-07-15 15:15:37 +00:00
  • f8c27befff Dropped PDC_sysname(), and with it, the reporting of the adapter type in DOS and OS/2. Now the system name is defined in curspriv.h. William McBrine 2006-07-15 15:13:40 +00:00
  • c39157fcb0 Changed PDC_set_80x25() to DOS-only. William McBrine 2006-07-15 15:00:57 +00:00
  • ff1a077d8c PDC_get_cur_col() and _row(): unused and redundant. William McBrine 2006-07-15 13:32:52 +00:00
  • 965927e955 Removed meaningless PDC_sanity_check() in win32; removed it from curspriv.h and intro.man. William McBrine 2006-07-15 13:13:46 +00:00
  • 7eb0ff9010 Took out the word "independent". After comparing "pcurses" with PDCurses, I can't really say that. William McBrine 2006-07-15 12:49:15 +00:00
  • b5c5f06c4a Removed the "buggy compiler fix" for NDP. I'm tempted to remove all support for NDP. William McBrine 2006-07-14 22:59:38 +00:00
  • 700763ad3a Moved PDC_get_key_modifiers() into pdckey.c. William McBrine 2006-07-14 19:38:30 +00:00
  • 0557513a11 Moved PDC_rawgetch() and PDC_sysgetch() into pdckey.c. They were identical between ports, _except_ for X. So, XCurses_rawgetch() became PDC_get_bios_key(). Also elminated the XCURSES ifdef in getch.c. William McBrine 2006-07-14 19:22:02 +00:00
  • 53c80a30bb Updated version number. William McBrine 2006-07-14 18:11:56 +00:00
  • 545c0bea05 Undefine PDC_WIDE if not CHTYPE_LONG, or not Win32 or X11. William McBrine 2006-07-14 17:50:15 +00:00
  • 9c00e45c43 _ENDLINE, _FULLWIN and _SCROLLWIN properties were initialized, but never used. William McBrine 2006-07-14 17:03:27 +00:00
  • 1dd8451478 _BCHAR -- used only in 2.1. _GOCHAR, _STOPCHAR, _PRINTCHAR -- never used. William McBrine 2006-07-14 16:52:10 +00:00
  • 3358d6f979 The internal mouse macros don't need to be in curspriv.h, and many aren't needed at all. William McBrine 2006-07-14 16:43:38 +00:00
  • a69639c297 Dependency correction. William McBrine 2006-07-14 16:10:13 +00:00
  • d395e8bbe5 Broke out the DOS-only material from curspriv.h into dos/pdcdos.h. William McBrine 2006-07-14 16:00:52 +00:00
  • 54cd3da469 Removed all the EMALLOC stuff. It was more or less broken, since straight malloc calls were used elsewhere; it was undocumented outside of comments in curspriv.h; and there are better ways to use a substitute malloc(). This is another thing that I doubt anyone but Frotz ever used. William McBrine 2006-07-14 06:27:32 +00:00
  • 3e52d8385d Replaced most of the embedded externs with curspriv.h declarations. I think this is sufficiently isolated, and a lot less redundant. William McBrine 2006-07-14 03:21:16 +00:00
  • 3844191f4b key_name() -- this isn't quite right, but then, from the description in the spec, it sounds completely useless anyway. So maybe I'm missing something... William McBrine 2006-07-13 21:13:51 +00:00
  • 852a642053 X/Open specifies "UNKNOWN KEY" instead of "NO KEY NAME". To Do: Solaris curses and ncurses never use UNKNOWN KEY, returning NULL instead; while NetBSD curses behaves like PDCurses. The standard is ambiguous -- both behaviors are mentioned, with no clear reason to prefer one or the other. What's the right thing here? William McBrine 2006-07-13 20:55:43 +00:00
  • 8231fdd198 Use the single define PDC_WIDE to ifdef all wide-character stuff. My attempt to separate it from the UNICODE ifdef was quixotic... users will just have to define PDC_WIDE for any wide-character program. The configure parameter becomes "--enable-widec", as in ncurses. William McBrine 2006-07-13 00:20:27 +00:00
  • ac5a012ca1 USE_WIDE -> PDC_WIDE... let's minimize our "namespace pollution". William McBrine 2006-07-12 16:19:57 +00:00
  • c16da2ea97 The last wide character functions (the ones for input) present a problem: They use wint_t rather than wchar_t. wchar_t is defined in C89, but wint_t is only in Addendum 1, from '95. I could typedef wint_t myself, but that seems like a bad idea for many reasons. So: Create a new ifdef, USE_WIDE, and define it only for Win32 and X. That also saves the overhead in DOS and OS/2, but it means that the base build reqs for Win32 and X are even more recent. I doubt this is the final solution. William McBrine 2006-07-12 05:19:32 +00:00
  • 4fa6ff28d8 One more thing: define _XOPEN_SOURCE_EXTENDED so ncurses (and others?) will include wide-character support. William McBrine 2006-07-11 20:14:55 +00:00
  • ca5e1d9c95 OK, this should do the trick -- check for UNICODE only under PDCurses, WACS_S1 otherwise. Also, comments. William McBrine 2006-07-11 20:05:56 +00:00
  • 5eb36eed08 OK, these are the *right* Unicode characters for ACS_Sn. Of course, they don't work even in Lucida Console in Win32. <sigh> But they look great in X. William McBrine 2006-07-11 19:57:12 +00:00
  • b907636d49 Two pages for the ACS test with wide characters (ACS_*, and WACS_*); test for WACS_S1 rather than UNICODE; test for ACS_S3 before including the "extra" ACS_* (only defined in ncurses and PDCurses). To Do: Hmm, testing for WACS_S1 isn't quite right, either. William McBrine 2006-07-11 19:43:31 +00:00
  • 10a41919a3 At first, I thought that X/Open was saying WACS_* should be cchar_t constants; but everyone else implements them as (cchar_t *) constants. And it does make more sense. William McBrine 2006-07-11 19:14:41 +00:00
  • 895d80374b OK, Thai was a mistake, because it requires combining marks, which PDCurses doesn't support yet. I'm looking for something outside the WGL4 for the fourth test language, but due to our current limitations, it has to 1) be written left to right, 2) use single-width characters, and 3) not use combining marks. I was also hoping for something non-European, but Georgian will do for now. William McBrine 2006-07-11 17:56:36 +00:00
  • ccd2af366d One more change to getcchar(): Add the trailing zero only if the text of the cchar_t isn't already zero (i.e., write at most one zero to the destination wchar_t string). William McBrine 2006-07-11 17:21:23 +00:00
  • 667ad0c1f2 ncurses tests didn't like my getcchar(). This is better. William McBrine 2006-07-10 21:20:51 +00:00
  • 0eb35af700 Return ASCII and control key names from keyname() (problem revealed by ncurses' movewindow test). William McBrine 2006-07-10 18:40:51 +00:00
  • 18bf5f50a4 cchar_t typedef and functions, and WACS_* defines. This is a simplistic implementation, with exactly one wchar_t per cchar_t; the only characters it handles properly are those that are one column wide. However, the X/Open spec is vague enough that this may actually be a conforming implementation. To Do: Document; test; and (long term) make it work the way it's really supposed to. William McBrine 2006-07-09 22:48:16 +00:00
  • f38b39c785 ...and took the comment out again, since there's really no need to qualify it if it's standard practice. (Is there?) William McBrine 2006-07-09 03:58:29 +00:00
  • ba90fef8d8 Reading X/Open, I just realized that it specifies ACS_* as char constants, not chtypes. Kinda casts doubt on my "| ALTCHARSET" solution, not to mention the Unicode variants; but they've always been chtypes in PDCurses (since 2.0; they weren't there before then). Let's see... OK, all the other curses implementations I have here also use chtype. Minus one point for the spec. Anyway, clarified the comments... William McBrine 2006-07-09 03:56:15 +00:00
  • acdb6e3681 Things that don't need to be in pdcx11.h. To Do: Most of the rest are only shared between pdcx11.c and process/x11.c. William McBrine 2006-07-08 00:21:41 +00:00
  • ae229e94cd Minor reorg. William McBrine 2006-07-07 16:20:28 +00:00
  • 642bd047f4 XCursesEndwin() doesn't need to be declared in curspriv.h, nor does it need to return an int. William McBrine 2006-07-07 13:19:31 +00:00
  • 54d51fee3d Removed XCurses_refresh_scrollbar(). William McBrine 2006-07-07 13:11:43 +00:00
  • 74ef01d775 FARKeyword has not been used since 2.3. William McBrine 2006-07-07 12:49:41 +00:00
  • a62cab6e55 Oops -- config.h is not used in DOS. William McBrine 2006-07-07 05:56:18 +00:00
  • df027b6f13 Comments. William McBrine 2006-07-07 05:37:58 +00:00
  • 57aa0de7c9 OK, for now, napms() goes back in pdcurses/kernel.c, and the others become PDC_napms(). I didn't want it to be the _only_ curses-level function defined on the platform level... and I just noticed that curs_set() was already the same way, right above napms(). William McBrine 2006-07-07 05:34:05 +00:00
  • eb776b28dd Moved "Regs regs" from pdcurses/initscr.c to dos/pdcscrn.c. To Do: Seems like it should be more local... William McBrine 2006-07-07 05:14:37 +00:00
  • 524b6ef51c Moved the core of longname() to the platform directories. To Do: Do we really need the card info? Everything's VGA now; it's not accurate in an OS/2 window; it's never been available under what are now the main platforms; and it's the kind of thing that curses is supposed to shield you from anyway. Does anyone _use_ this info? William McBrine 2006-07-07 04:27:44 +00:00
  • e2c261a48f Updated build number. William McBrine 2006-07-07 03:11:51 +00:00
  • 4d3a1486ca "x11basedir" -> "osdir" to match the other platforms' makefiles. (Yeah, I know, X11 isn't an OS.) William McBrine 2006-07-07 03:10:33 +00:00
  • 22e982f0c1 Moved PDC_beep() and napms() to platform directories; moved pdcutil.c to platform directories, and created pdckey.c in the pdcurses directory (to hold PDC_breakout(), formerly from pdcutil.c). To Do: I'm still not happy with this structure. Also, it's an enormous pain to add, remove or rename any files, since every makefile has to be changed. I need to try to do something about this. William McBrine 2006-07-07 02:53:08 +00:00
  • 50279b31f1 Comments. William McBrine 2006-07-07 00:00:53 +00:00
  • 487d553f38 INT_MAX is good enough as the only return from baudrate(). (Previously it was used everywhere except in BIOS mode in DOS. This allows me to remove the ifdef.) William McBrine 2006-07-06 23:57:27 +00:00
  • 3b83bb5eea Replaced core of flushinp() with platform-specific PDC_flushinp(), instead of a bunch of ifdefs. It's cleaner this way. To Do: I'm working toward complete separation -- i.e., no platform ifdefs in the pdcurses directory. William McBrine 2006-07-06 23:50:13 +00:00
  • e04e0abc0a whline(), wvline(), and PDC_lineattr() used an incorrect (off by one) interpretation of _maxx and _maxy. If values of n greater than the max were specified, these functions could access unallocated memory. To Do: Check for more of this. William McBrine 2006-07-06 18:58:17 +00:00
  • 750691debd wchgat() needs to update firstch/lastch and call PDC_sync() (doh). William McBrine 2006-07-06 18:42:58 +00:00
  • a91035487a malloc() etc. are prototyped in stdlib.h; no need for the declarations here. William McBrine 2006-07-06 15:51:55 +00:00
  • 7745c6d713 Updated to correspond to the change in aclocal.m4 (search lib64). William McBrine 2006-07-06 13:43:36 +00:00
  • c062fb2a38 Include searching for X11 libraries in lib64 directories on 64 bit machines Mark Hessling 2006-07-06 10:37:32 +00:00
  • 06480e3ff4 Typo. William McBrine 2006-07-06 06:19:26 +00:00
  • 9362905762 Removed the const and size_t checks, since ANSI C is now a base requirement. William McBrine 2006-07-06 05:52:59 +00:00
  • 333183fdee config.h is not needed in these files. William McBrine 2006-07-05 20:20:49 +00:00
  • 63dca8b406 Moved x11 demos build to x11 -- it's now comparable to the other platforms, building everything in one directory. To Do: Makefile.aix.in? And, what about the top level Makefile.in and tools/Makefile.in? William McBrine 2006-07-05 06:36:57 +00:00
  • d9bdac5de1 Bringing it back... William McBrine 2006-07-04 03:51:38 +00:00
  • a4524bff67 Let's see if _this_ will reset the permissions... William McBrine 2006-07-04 03:51:04 +00:00
  • 1e2e1287c9 OK, let's try it this way... this is how it worked (_if_ it worked) before. But I can't test it. William McBrine 2006-07-04 03:08:34 +00:00
  • 1da24b4931 MISTAKE William McBrine 2006-07-04 03:07:17 +00:00
  • d6f325c9af Incomplete update of AIX makefile (doesn't build libpanel.a). However... William McBrine 2006-07-04 02:53:13 +00:00
  • 8087fd26ca Rearrangement -- build X11 port in x11/ dir, like other ports, instead of in pdcurses/ and panel/. To Do: demos; more. William McBrine 2006-07-04 00:23:05 +00:00
  • 94e83a71c1 Unused code. William McBrine 2006-07-02 22:15:59 +00:00
  • 21d84f0163 The wide-character functions shouldn't really be dependent on the UNICODE define (though they're unlikely to do anything useful without it). However, they're still dependent on CHTYPE_LONG, since without it, A_CHARTEXT is only 8 bits. To Do: Hmm... William McBrine 2006-07-02 22:03:02 +00:00
  • 2a0ef7f166 Formatting. William McBrine 2006-07-02 20:07:01 +00:00
  • 67058cad9b Don't try to archive x11/process/*.h. William McBrine 2006-07-02 19:12:30 +00:00
  • 132de9d3b7 Clarification of PD status. William McBrine 2006-07-02 19:04:44 +00:00
  • 75f38d4ff3 Minor reorganization of X11 port -- since there are no longer differing copyright terms on the files, there's less reason to keep the "PDCurses" and "XCurses" stuff separated that way; and since the thread port is dead, there's less need for the separation embodied in the "process" directory, which isn't clean anyway. Of course, the various ports and the "portable" stuff still need to be separated, but that's actually easier now -- because, for example, many of the XCurses* defines from curspriv.h are gone. To Do: Much more. William McBrine 2006-07-02 19:03:59 +00:00
  • e9ac0e4b11 Redundant error messages. William McBrine 2006-07-02 08:17:21 +00:00
  • 92a4d9bee0 Removed pointless memcpy() calls; eliminated some redundant code. William McBrine 2006-07-02 07:08:28 +00:00
  • 76d246e893 mvprintw() -> mvaddstr(). William McBrine 2006-07-02 04:59:41 +00:00
  • affb63b74d A better selection implementation -- this one uses XA_STRING appropriately even in Unicode, by not converting until XCursesConvertProc(). To Do: refinement; other modes? William McBrine 2006-07-01 18:06:43 +00:00
  • 59412f3dd4 Cut down redundancy in the UNICODE ifdefs. William McBrine 2006-06-22 13:20:36 +00:00