http://blog.gmane.org/gmane.comp.lib.ncurses.bugs hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4861 <pre>... in passing when reading the last 1000 articles here I come across this question. thought it was rather interesting and amusing... as curses was motivated by rogue and then rog-o-matic came along. :) now we have someone wanting to redo rog-o-matic and tinkering with the innards of curses. myself am working on rog-o-matic itself and rogue... songbird </pre> songbird 2012-05-17T16:05:14 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4860 <pre> hello and thank you so much folks for your work on curses/ncurses. Thomas, Sven, many others too. my question is if there is a flag for compiling that would tell ncurses to output the token code, capability name string, Cap name or Tcap code instead of the characters? preferably in an easily parsed format, i use "{" and "}" around them, but it can be any enclosure. this would save me a lot of time and let me write code that is more portable when capturing/scanning/processing output from a ncurses using source code project of any kind. and it would seem that there must be some way to generate a token name for purposes of debugging ncurses, but i've not dug into the source code yet. my current use is in processing the output of Rogue and Rog-O-Matic, but I can see the wider use case for having the tokens returned instead of the characters. and yes, i do know about trace mode, but i suspect that puts out too much. it would mean i wouldn't have to write a scanner for all of these obscure character strings i don't understand well and could support things much more easily. writing a scanner instead that parses based upon the names or codes would be much easier. currently i'm writing scanners for xterm just to see what Rogue/Rog-O-Matic produce in the course of tracking down bugs. my scanners output the codes recognized enclosed in braces and anything not seen is passed through as text. works well but is by no means as complete or correct as what would actually come out of ncurses itself. as an aside, it is funny how some things come around, ages ago i enjoyed playing rogue and always wanted to see rog-o- matic run. now i have time to tinker and figure some of this out so i get to see how it all works together. cheers and thanks again, songbird </pre> songbird 2012-05-17T05:05:17 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4859 <pre>I'll be out of town over the weekend. </pre> Thomas Dickey 2012-05-14T10:57:00 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4858 <pre> ncurses 5.9 - patch 20120512 - Thomas E. Dickey ------------------------------------------------------------------------------ Ncurses 5.9 is at ftp.gnu.org:/pub/gnu Patches for ncurses 5.9 are in the subdirectory ftp://invisible-island.net/ncurses/5.9 ------------------------------------------------------------------------------ ftp://invisible-island.net/ncurses/5.9/ncurses-5.9-20120512.patch.gz patch by Thomas E. Dickey &lt;dickey&lt; at &gt;invisible-island.net&gt; created Sun May 13 00:35:57 UTC 2012 ------------------------------------------------------------------------------ NEWS | 11 ++++++++- dist.mk | 4 +-- misc/terminfo.src | 37 ++++++------------------------ ncurses/tty/tty_update.c | 13 +++++----- test/background.c | 54 ++++++++++++++++++++++++++++++++++++++++----- 5 files changed, 75 insertions(+), 44 deletions(-) ------------------------------------------------------------------------------ 20120512 + corrected 'op' for bterm (report by Samuel Thibault) -TD + modify test/background.c to demonstrate a background character holding a colored ACS_HLINE. The behavior differs from SVr4 due to the thick- and double-line extension (cf: 20091003). + modify handling of acs characters in PutAttrChar to avoid mapping an unmapped character to a space with A_ALTCHARSET set. + rewrite vt520 entry based on vt420 -TD </pre> Thomas Dickey 2012-05-13T00:37:02 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4857 <pre> There's some discussion from April 2011 about an unofficial git repository, that you may find useful. </pre> Thomas Dickey 2012-05-07T00:59:55 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4856 <pre>Are there any plans on moving to some VCS? Maybe some DVCS? _______________________________________________ Bug-ncurses mailing list Bug-ncurses&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-ncurses </pre> Andrius Bentkus 2012-05-06T16:13:52 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4855 <pre> ncurses 5.9 - patch 20120505 - Thomas E. Dickey ------------------------------------------------------------------------------ Ncurses 5.9 is at ftp.gnu.org:/pub/gnu Patches for ncurses 5.9 are in the subdirectory ftp://invisible-island.net/ncurses/5.9 ------------------------------------------------------------------------------ ftp://invisible-island.net/ncurses/5.9/ncurses-5.9-20120505.patch.gz patch by Thomas E. Dickey &lt;dickey&lt; at &gt;invisible-island.net&gt; created Sat May 5 23:38:59 UTC 2012 ------------------------------------------------------------------------------ Ada95/configure | 324 ++++----- Ada95/configure.in | 5 NEWS | 19 configure | 1474 ++++++++++++++++++++--------------------- configure.in | 5 dist.mk | 4 misc/terminfo.src | 52 - ncurses/tinfo/entries.c | 5 ncurses/tinfo/read_termcap.c | 5 9 files changed, 959 insertions(+), 934 deletions(-) ------------------------------------------------------------------------------ 20120505 + remove p6 (bold) from opus3n1+ for consistency -TD + remove acs stuff from env230 per clues in Ingres termcap -TD + modify env230 sgr/sgr0 to match other capabilities -TD + modify smacs/rmacs in bq300-8 to match sgr/sgr0 -TD + make sgr for dku7202 agree with other caps -TD + make sgr for ibmpc agree with other caps -TD + make sgr for tek4107 agree with other caps -TD + make sgr for ndr9500 agree with other caps -TD + make sgr for sco-ansi agree with other caps -TD + make sgr for d410 agree with other caps -TD + make sgr for d210 agree with other caps -TD + make sgr for d470c, d470c-7b agree with other caps -TD + remove redundant AC_DEFINE for NDEBUG versus Makefile definition. + fix a back-link in _nc_delink_entry(), which is needed if ncurses is configured with --enable-termcap and --disable-getcap. </pre> Thomas Dickey 2012-05-05T23:43:47 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4854 <pre> ...a little slow (I've been working on xterm and vttest, and came back to tack to add to the checks for statusline, and was reminded of this). There's a new version of tack for your consideration ;-) </pre> Thomas Dickey 2012-04-30T00:53:38 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4853 <pre> ncurses 5.9 - patch 20120428 - Thomas E. Dickey ------------------------------------------------------------------------------ Ncurses 5.9 is at ftp.gnu.org:/pub/gnu Patches for ncurses 5.9 are in the subdirectory ftp://invisible-island.net/ncurses/5.9 ------------------------------------------------------------------------------ ftp://invisible-island.net/ncurses/5.9/ncurses-5.9-20120428.patch.gz patch by Thomas E. Dickey &lt;dickey&lt; at &gt;invisible-island.net&gt; created Sun Apr 29 00:54:13 UTC 2012 ------------------------------------------------------------------------------ NEWS | 17 +++++- dist.mk | 4 - doc/html/ncurses-intro.html | 5 + doc/ncurses-intro.doc | 1 man/curs_inopts.3x | 47 ++++++++++++++++- misc/terminfo.src | 114 ++++++++++++++++++------------------------ ncurses/tinfo/comp_scan.c | 4 - ncurses/trace/lib_trace.c | 4 - progs/tic.c | 91 ++++++++++++++++++++++++++------- 9 files changed, 192 insertions(+), 95 deletions(-) ------------------------------------------------------------------------------ 20120428 + fix some inconsistencies between vt320/vt420, e.g., cnorm/civis -TD + add eslok flag to dec+sl -TD + dec+sl applies to vt320 and up -TD + drop wsl width from xterm+sl -TD + reuse xterm+sl in putty and nsca-m -TD + add ansi+tabs to vt520 -TD + add ansi+enq to vt220-vt520 -TD + fix a compiler warning in example in ncurses-intro.doc (Paul Waring). + added paragraph in keyname manpage telling how extended capabilities are interpreted as key definitions. + modify tic's check of conflicting key definitions to include extended capability strings in addition to the existing check on predefined keys. </pre> Thomas Dickey 2012-04-29T00:58:15 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4852 <pre>I tried building the sample program supplied in the ncurses documentation and got the following warnings: $ gcc intro.c -lncurses intro.c: In function ‘finish’: intro.c:53:3: warning: incompatible implicit declaration of built-in function ‘exit’ [enabled by default] [gcc version 4.6.1 (Ubuntu/Linaro 4.6.1-9ubuntu3)] This warning can be fixed by including stdlib.h, which declares the exit() function. I've attached a short patch which updates the documentation. Thanks, Paul </pre> Paul Waring 2012-04-26T16:42:44 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4851 <pre> The wgetch-events feature has been neglected since there's no example of its use. If you have some ideas for a simple program that we could add to ncurses's test-directory, that would be helpful ;-) (the bug report is helpful too) </pre> Thomas Dickey 2012-04-23T09:52:06 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4850 <pre>I've been experimenting with a headless system and ncurses has been nice quick and easy way to create some of my programs. One of the things I need is the ability include my own input events into ncurses event loop. Given that I'm using a file descriptor to fetch my input the wgetch_events function looked like it had a lot of potential. In the end it gave me what I wanted but I had to change, dare I say fix, the code. I pulled the 5.9 release, configured it with "configure --enable-wgetch-events", and built it. When I ran my test program wgetch_events didn't catch inputs from the file descriptor I passed to it. After some debugging I found that the following code at line 391 in lib_twait.c was the problem. #ifdef NCURSES_WGETCH_EVENTS     if (evl) {         evl-&gt;result_flags = 0;         for (n = 0; n &lt; evl-&gt;count; ++n) {             _nc_event *ev = evl-&gt;events[n];             if (ev-&gt;type == _NC_EVENT_TIMEOUT_MSEC) {                 long diff = (returntime - starttime);                 if (ev-&gt;data.timeout_msec &lt;= diff)                     ev-&gt;data.timeout_msec = 0;                 else                     ev-&gt;data.timeout_msec -= diff;             }         }     } #endif If evl is set, which it is if you're sending in your own file descriptors, then evl-&gt;result_flags is always set to 0. That wipes out how it may have been set earlier. In my case it was being set to _NC_EVENT_FILE_READABLE earlier in the function. When it hits this code that gets wiped out. And then down stream when the program gets back to my code it isn't notified that input actually occurred. Looking at the code I concluded that lib_twait.c should be setting evl-&gt;result_flags only if one of the "ev's" is of type _NC_EVENT_TIMEOUT_MSEC. I think the code should look something like this. #ifdef NCURSES_WGETCH_EVENTS     if (evl) {         for (n = 0; n &lt; evl-&gt;count; ++n) {             _nc_event *ev = evl-&gt;events[n];             if (ev-&gt;type == _NC_EVENT_TIMEOUT_MSEC) {                 long diff = (returntime - starttime);                 if (ev-&gt;data.timeout_msec &lt;= diff)                     ev-&gt;data.timeout_msec = 0;                 else                     ev-&gt;data.timeout_msec -= diff;                 evl-&gt;result_flags |= _NC_EVENT_TIMEOUT_MSEC;             }         }     } #endif This code only sets evl-&gt;result_flags if in fact a "ev" represents a time out. And when set the value OR's it into the variable. My code now runs as expected and I'm able to process input from the devices associated with the file descriptors I pass into wgetch_events. Having made this change I wanted to know if it's something that should be put into the main code line. If so, please let me know and I'll post a diff. ... bruce </pre> bwgz 2012-04-23T04:42:44 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4849 <pre> ncurses 5.9 - patch 20120421 - Thomas E. Dickey ------------------------------------------------------------------------------ Ncurses 5.9 is at ftp.gnu.org:/pub/gnu Patches for ncurses 5.9 are in the subdirectory ftp://invisible-island.net/ncurses/5.9 ------------------------------------------------------------------------------ ftp://invisible-island.net/ncurses/5.9/ncurses-5.9-20120421.patch.gz patch by Thomas E. Dickey &lt;dickey&lt; at &gt;invisible-island.net&gt; created Sat Apr 21 21:46:50 UTC 2012 ------------------------------------------------------------------------------ NEWS | 29 +++++-- dist.mk | 4 - misc/terminfo.src | 195 +++++++++++++++++++++++++++++++++------------------- progs/tic.c | 13 +-- 4 files changed, 155 insertions(+), 86 deletions(-) ------------------------------------------------------------------------------ 20120421 + improve cleanup of temporary files in tic using atexit(). + add msgr to vt420, similar DEC vtXXX entries -TD + add several missing vt420 capabilities from vt220 -TD + factor out ansi+pp from several entries -TD + change xterm+sl and xterm+sl-twm to include only the status-line capabilities and not "use=xterm", making them more generally useful as building-blocks -TD + add dec+sl building block, as example -TD </pre> Thomas Dickey 2012-04-21T21:49:35 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4848 <pre> ncurses 5.9 - patch 20120414 - Thomas E. Dickey ------------------------------------------------------------------------------ Ncurses 5.9 is at ftp.gnu.org:/pub/gnu Patches for ncurses 5.9 are in the subdirectory ftp://invisible-island.net/ncurses/5.9 ------------------------------------------------------------------------------ ftp://invisible-island.net/ncurses/5.9/ncurses-5.9-20120414.patch.gz patch by Thomas E. Dickey &lt;dickey&lt; at &gt;invisible-island.net&gt; created Sat Apr 14 23:45:59 UTC 2012 ------------------------------------------------------------------------------ NEWS | 7 ++ dist.mk | 4 - misc/terminfo.src | 127 +++++++++++++++++++++++++++++++++++++++---- ncurses/tinfo/comp_parse.c | 21 ++++++- progs/tic.c | 71 +++++++++++++++++++++++- 5 files changed, 213 insertions(+), 17 deletions(-) ------------------------------------------------------------------------------ 20120414 + add consistency check in tic for screen's "XT" capability. + add section in terminfo.src summarizing the user-defined capabilities used in that file -TD </pre> Thomas Dickey 2012-04-14T23:50:05 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4847 <pre>Hi, I found an error at the compilation of ncurses 5.8 library in mingw&amp;msys, ms windows 32bits. I found it at the compilation of Ada Library.  It occurs at the compilation process in the stage: gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-imgllu.adb gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-imguns.adb gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-valllu.adb gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-powtab.ads gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-wchjis.adb gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/a-finali.adb gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-crtrun.ads gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-restri.adb gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-win32.ads gcc -gnatpg -c -I.\ -I. -I../src -I./../src -O3 -gnatpn -I. -I- C:/MinGW/lib/gcc /mingw32/4.6.2/adainclude/s-winext.ads gnatbind -aO. -aO../src -aO./../src -x tour.ali gnatlink tour.ali -L../../lib -lform -lmenu -lpanel -lncurses -L../lib -lAdaCurs es Here, it appears many lines of errors. Furthermore: collect2: ld returned 1 exit status gnatlink: error when calling C:\MinGW\bin\gcc.exe gnatmake: *** link failed. make[2]: *** [tour.exe] Error 4 make[2]: Leaving directory `/d/ncurses-5.8/Ada95/samples' make[1]: *** [all] Error 2 make[1]: Leaving directory `/d/ncurses-5.8/Ada95' make: *** [all] Error 2 End of description of the error. I suppose that this error is related with the configuration of Ada compiler for ncurses 5.8 library. I think it because I found a warning at the compilation process of ncurses 5.0 library. It is at "configure process" of  ncurses 5.0 library, in mingw&amp;msys, ms windows 32bits and it occurs at: checking if GNAT works... yes configure: warning: Ada95 applications will not link properly with static libraries Furthermore, the "configure process" does not append any rules for normal and debug models for all "Ada95 make process". Best Regards._______________________________________________ Bug-ncurses mailing list Bug-ncurses&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-ncurses </pre> daniel silva ferreira bruno 2012-04-08T15:35:52 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4846 <pre> sorry (I was in too much of a hurry early this morning). I see that the problem with last week's patch was that I had edited the file, but not checked it in. So I made a patch with just the id-line modified as you first suggested (and it applies correctly here). that's on my ftp area now... </pre> Thomas Dickey 2012-04-08T15:29:55 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4845 <pre> $ gunzip -c ../../patches/ncurses-5.9-20120407.patch.gz | git apply &lt;stdin&gt;:376: trailing whitespace. * a) capabilities listed past the "STOP-HERE" comment in the Caps file. error: patch failed: NEWS:45 error: NEWS: patch does not apply Now both this week's and last week's patches add: 20120331 + + update Ada95/configure to use CF_DISABLE_ECHO (cf: 20120317). -Alejandro _______________________________________________ Bug-ncurses mailing list Bug-ncurses&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-ncurses </pre> Alejandro R. Sedeño 2012-04-08T14:45:52 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4844 <pre> I uploaded an amended patch. Looking at the history, it shows that I made a build-fix last week and did not notice that I'd forgotten to regenerate my staging directory (used for uploads, not for builds). </pre> Thomas Dickey 2012-04-08T10:00:16 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4843 <pre> ncurses 5.9 - patch 20120407 - Thomas E. Dickey ------------------------------------------------------------------------------ Ncurses 5.9 is at ftp.gnu.org:/pub/gnu Patches for ncurses 5.9 are in the subdirectory ftp://invisible-island.net/ncurses/5.9 ------------------------------------------------------------------------------ ftp://invisible-island.net/ncurses/5.9/ncurses-5.9-20120407.patch.gz patch by Thomas E. Dickey &lt;dickey&lt; at &gt;invisible-island.net&gt; created Sun Apr 8 00:45:28 UTC 2012 ------------------------------------------------------------------------------ NEWS | 18 ++++++++- dist.mk | 4 +- misc/gen_edit.sh | 69 ++++++++++++++++++++++++----------- misc/terminfo.src | 11 ++++- ncurses/tinfo/comp_scan.c | 85 ++++++++++++++++++++++++++++++++++++-------- progs/infocmp.c | 20 +++++++++- 6 files changed, 165 insertions(+), 42 deletions(-) ------------------------------------------------------------------------------ 20120407 + fix an inconsistency between tic/infocmp "-x" option; tic omits all non-standard capabilities, while infocmp was ignoring only the user definable capabilities. + improve special case in tic parsing of description to allow it to be followed by terminfo capabilities. Previously the description had to be the last field on an input line to allow tic to distinguish between termcap and terminfo format while still allowing commas to be embedded in the description. + correct variable name in gen_edit.sh which broke configurability of the --with-xterm-kbs option. + revert 2011-07-16 change to "linux" alias, return to "linux2.2" -TD + further amend 20110910 change, providing for configure-script override of the "linux" terminfo entry to install and changing the default for that to "linux2.2" (Debian #665959). </pre> Thomas Dickey 2012-04-08T00:57:20 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4842 <pre> ncurses 5.9 - patch 20120331 - Thomas E. Dickey ------------------------------------------------------------------------------ Ncurses 5.9 is at ftp.gnu.org:/pub/gnu Patches for ncurses 5.9 are in the subdirectory ftp://invisible-island.net/ncurses/5.9 ------------------------------------------------------------------------------ ftp://invisible-island.net/ncurses/5.9/ncurses-5.9-20120331.patch.gz patch by Thomas E. Dickey &lt;dickey&lt; at &gt;invisible-island.net&gt; created Sun Apr 1 00:16:56 UTC 2012 ------------------------------------------------------------------------------ Ada95/aclocal.m4 | 96 + Ada95/configure | 2225 +++++++++++++---------------- Ada95/configure.in | 19 NEWS | 12 aclocal.m4 | 30 configure | 3819 ++++++++++++++++++++++++--------------------------- configure.in | 34 dist.mk | 4 misc/terminfo.src | 9 test/aclocal.m4 | 11 test/configure | 2202 ++++++++++++++--------------- 11 files changed, 4122 insertions(+), 4339 deletions(-) ------------------------------------------------------------------------------ 20120331 + update Ada95/configure to use CF_DISABLE_ECHO (cf: 20120317). + correct order of use-clauses in st-256color -TD + modify configure script to look for gnatgcc if the Ada95 binding is built, in preference to the default gcc/cc (suggested by Nicolas Boulenguez). + modify configure script to ensure that the same -On option used for the C compiler in CFLAGS is used for ADAFLAGS rather than simply using -O3 (suggested by Nicolas Boulenguez) </pre> Thomas Dickey 2012-04-01T01:03:58 http://permalink.gmane.org/gmane.comp.lib.ncurses.bugs/4841 <pre> ncurses 5.9 - patch 20120324 - Thomas E. Dickey ------------------------------------------------------------------------------ Ncurses 5.9 is at ftp.gnu.org:/pub/gnu Patches for ncurses 5.9 are in the subdirectory ftp://invisible-island.net/ncurses/5.9 ------------------------------------------------------------------------------ ftp://invisible-island.net/ncurses/5.9/ncurses-5.9-20120324.patch.gz patch by Thomas E. Dickey &lt;dickey&lt; at &gt;invisible-island.net&gt; created Sat Mar 24 23:37:46 UTC 2012 ------------------------------------------------------------------------------ NEWS | 10 ++- dist.mk | 4 - man/tic.1m | 47 ++++++++------ ncurses/tinfo/comp_expand.c | 4 - ncurses/tinfo/comp_scan.c | 5 - ncurses/widechar/lib_cchar.c | 4 - progs/tic.c | 131 ++++++++++++++++++++++++++++++----------- 7 files changed, 143 insertions(+), 62 deletions(-) ------------------------------------------------------------------------------ 20120324 + amend an old fix so that next_char() exits properly for empty files, e.g., from reading /dev/null (cf: 20080804). + modify tic so that it can read from the standard input, or from a character device. Because tic uses seek's, this requires writing the data to a temporary file first (prompted by remark by Sven Joachim) (cf: 20000923). </pre> Thomas Dickey 2012-03-24T23:43:16 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.lib.ncurses.bugs