Hi,
Vladimir Nadvornik has made available a few patches to complete / fix the utf-8 patches for the midnight commander at http://www.suse.de/~nadvornik/mc.html . The largest of these (utf8-input patch) does not entirely patch against FC1's mc, but please have a look anyway. It's probably not too hard to fix these rejects.
Vladimir also helped me fix one and a half issue (hotkey binding / ampersand badness) with a very simple patch which is inlined in http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=120735 .
How about the Red Hat patches for mc? Will these be merged with upstream before mc-4.6.1? Maybe it's a good idea to join efforts. Vladimir will work on the patches in a couple of weeks.
Finally a remark on a couple of other mc issues in bugzilla. The issues below have attached patches and/or are easy fixes and IMO should be included in the next update of mc for FC1 and FC2. Also see http://www.ottolander.nl/bughunt/index.php?package=mc for reference.
Two times \0 inserted out of range: http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=110771 INFO/OBSOLETES always empty in #rpm view: http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=67341 out dated file in the rpm package: http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=85073
Leonard.
Hi all,
I wrote:
Vladimir Nadvornik has made available a few patches to complete / fix the utf-8 patches for the midnight commander at http://www.suse.de/~nadvornik/mc.html .
On my request Vladimir was so kind to fix some conflicts I saw with the jumbo patch and the edit-replace patch. The edit-replace patch is now totally integrated in Vladimir's utf8-input patch. He added a small fix and couple of checks as well (*p1++ = 'l' if (*p == 's') and a couple of *p < 128 tests before calling strchr).
He also integrated code from 6 hunks from the jumbo patch into his utf8-input patch. This means these 6 hunks (all 5 hunks regarding edit/editcmd.c and hunk "@@ -1290,10 +1291,11 @@ delete_region" for src/widget.c) can be dropped from the current jumbo patch when applying these patches.
The current state of the patches seems to fix all the issues I was seeing when using charsets other than UTF-8 (a.o. wrong key binding (&), missing key highlights). You can find a test src rpm for FC 1 from http://www.ottolander.nl/opensource/srpms/fc1/mc-4.6.0-14.10n.src.rpm .
Leonard.
On 2004-06-18 (Friday) 03:35, Leonard den Ottolander wrote:
Hi all,
I wrote:
Vladimir Nadvornik has made available a few patches to complete / fix the utf-8 patches for the midnight commander at http://www.suse.de/~nadvornik/mc.html .
On my request Vladimir was so kind to fix some conflicts I saw with the jumbo patch and the edit-replace patch. The edit-replace patch is now totally integrated in Vladimir's utf8-input patch. He added a small fix and couple of checks as well (*p1++ = 'l' if (*p == 's') and a couple of *p < 128 tests before calling strchr).
He also integrated code from 6 hunks from the jumbo patch into his utf8-input patch. This means these 6 hunks (all 5 hunks regarding edit/editcmd.c and hunk "@@ -1290,10 +1291,11 @@ delete_region" for src/widget.c) can be dropped from the current jumbo patch when applying these patches.
The current state of the patches seems to fix all the issues I was seeing when using charsets other than UTF-8 (a.o. wrong key binding (&), missing key highlights). You can find a test src rpm for FC 1 from http://www.ottolander.nl/opensource/srpms/fc1/mc-4.6.0-14.10n.src.rpm .
Leonard.
Yes, it works great now (I use this version). With LANG=bg_BG.UTF-8 I have no key binding highlights in Bulgarian, but this can be translation issue. A real problem is if you enter a directory with <CR><LF> in the name with mc - it freezes there. Ex: D=$(echo -e 'test\015\012test') mkdir "$D" cd "$D" mc At this point trying to get out of this dir or switch panels with <TAB> does nothing. Can someone check if this also happends with 'normal' mc?