cssed-devel Mailing List for cssed (Page 12)
Brought to you by:
iagorubio
You can subscribe to this list here.
2003 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(1) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2004 |
Jan
|
Feb
|
Mar
(73) |
Apr
(46) |
May
(16) |
Jun
(14) |
Jul
(3) |
Aug
(4) |
Sep
(185) |
Oct
(17) |
Nov
|
Dec
(2) |
2005 |
Jan
(3) |
Feb
(83) |
Mar
(8) |
Apr
|
May
|
Jun
(2) |
Jul
|
Aug
|
Sep
(3) |
Oct
|
Nov
(5) |
Dec
(11) |
2006 |
Jan
(3) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2007 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2008 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(2) |
Jun
|
Jul
|
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
|
2009 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2015 |
Jan
|
Feb
(6) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: <mic...@ea...> - 2004-09-12 10:27:52
|
When changing the base directory for search, the selection is not=20 refresh in the base directory dialog (the field at the bottom), though=20= the right directory is selected. In the same dialog, the popupmenu does not reflect the last base=20 directory. If the permission is denied to enter a given directory, there is no=20 error message shown to the user, but instead a dialog window shows as=20 if you can enter in it, though the selection does not change. I think=20 first it's very confusing to the user, second it gives a huge=20 impression of insecurity, not good at all for cssed. Here we should=20 have an error message=3D "You have not the required permissions to = access=20 this directory". When searching if no match at all is found, there should be a warning=20 for the user, so that it is clear the search has been performed. When using the popup menu in the editor window to search for terms and=20= if no term is selected in the current file, there is no refresh of the=20= base directory in the plugin. This only occurs when a search term is=20 selected in the active file. A bit confusing for me here. When the search process encounters a special file, the search takes=20 ages to end and there is a you can have the same error message as in=20 filebrowser plugin, just to reassure the user. More on that later. Mich=E8le <http://micmacfr.homeunix.org>= |
From: <mic...@ea...> - 2004-09-12 00:54:10
|
Since the name has changed, removing find_in_files_plugin |
From: <mic...@ea...> - 2004-09-12 00:32:13
|
Silly error in email address in line Last translator. |
From: <mic...@ea...> - 2004-09-11 23:59:40
|
Sorry, forgot to add the diff: |
From: <mic...@ea...> - 2004-09-11 23:57:58
|
Ok, if we consider that now strings are frozen, I've made the last=20 minor changes (misspelling, punctuation marks, first letter=20 capitalized) and updated all po files accordingly and created new pot=20 file. So now English, Spanish and French are OK. German, Italian, and=20 Galician are outdated. Plus the revert thing works fine. Mich=E8le <http://micmacfr.homeunix.org>= |
From: Iago R. <iag...@hi...> - 2004-09-11 18:59:18
|
I fixed a bug in getline() function. The line got was not null terminated so there was garbage at the en of the string. This bug was only present in MacOsX and BSD. Fixed in CVS, patch attached. -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-11 15:57:55
|
Some changes have been added to cssed, mostly a new menu item to revert changes in one document. From now we are in strings freeze for translation, and code freeze for testing. I need a week or two for packaging and plugins' tests so there's no hurry with translations. The changes are in CVS, patch attached. -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-11 14:16:18
|
On Fri, 2004-09-10 at 19:52, Mich=C3=A8le Garoche wrote: > Le 10 sept. 2004, =C3=A0 18:12, Iago Rubio a =C3=A9crit : >=20 [snip] > > Michelle any idea ?? > > http://fink.sourceforge.net/pdb/package.php/exuberant-ctags > > > > It's maintained by pogma, can you ask him how we can know the ctags > > binary is exuberant ctags or not ? > On mac, you can have exuberant-ctags or ctags embedded into emacs and=20 > xemacs, so for me the binary should be exuberant ctags, as I don't see=20 > a way to use the ctags embedded into xemacs or emacs. The alternative=20 > allows to list all packages which uses ctags and defines a priority.=20 > Then a symlink is created to the package which has the higher priority.=20 > The user can change it as he wants. >=20 > Example: > update-alternatives --display ctags > ctags - status is manual. > link currently points to /sw/bin/ctags-exuberant > /sw/bin/ctags.xemacs - priority 60 > slave ctags.1: /sw/share/man/man1/ctags.1.xemacs > /sw/bin/ctags.emacs21 - priority 40 > slave ctags.1: /sw/share/man/man1/ctags.1.emacs21 > /sw/bin/ctags-exuberant - priority 70 > slave ctags.1: /sw/share/man/man1/ctags.1-exuberant > Current `best' version is /sw/bin/ctags-exuberant. Ok, I see. [snip] > > Using just the ctags program we can not search anything but C source > > code, that's not really useful for cssed. [snip] > > Nothing apropiated to use in cssed. > You can maybe search toutdoux and dtags, dtags is an interface for=20 > using ctags for xsltproc within toutdoux. It can give you hints. Will test it. > I've patched the line in configure this way, then it works: >=20 > dnl This plugin needs the ctags program > dnl AC_CHECK_PROG([CTAGS], [ctags], [no-ctags]) >=20 > dnl Exuberant ctags > AC_MSG_CHECKING(for Exuberant Ctags) > if ctags-exuberant --version > /dev/null 2>&1; then > AC_MSG_RESULT(yes)=09 > ctags=3Dyes > else > AC_MSG_ERROR(ctags-exuberant software not found) > ctags=3Dno Ok, thanks. Some changes in CVS. Please test it. Mostly: dnl ///////////////////////////////////////////////////// case "${OSTYPE}" in LINUX)=20 CTAGS=3Dctags ;; DARWIN) CTAGS=3Dctags-exuberant ;;=20 BSD) CTAGS=3Dexctags ;; *)=20 CTAGS=3Dctags ;;=20 esac dnl Exuberant ctags AC_MSG_CHECKING(for Exuberant Ctags) if $CTAGS --version > /dev/null 2>&1; then AC_MSG_RESULT(yes)=09 ctags=3Dyes else AC_MSG_ERROR(ctags-exuberant software not found) ctags=3Dno fi dnl ///////////////////////////////////////////////////// I should write some code, to use the $CTAGS in the command line invocation.=20 I will wait for your test, to see if it works on OsX. --=20 Iago Rubio =20 - Home page * http://www.iagorubio.com=20 - Last project * http://cssed.sourceforge.net =20 - GPG Keyserv * pgp.rediris.es id=3D0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-11 14:08:43
|
On Sat, 2004-09-11 at 10:12, Michèle Garoche wrote: > > Can you propose any change to make it build ? > > I'm a little lost here :) > What I have in configure.in: Ok, nice. > > ______________________________________________________________________ > It's based on what was on your site, not in cvs. I made some fixes to let it build also on Linux and BSD. Changes in cvs, patch attached. > Now I should make DVD for my parents, so I'll not be available till > tomorrow; have a nice day. Ok, have fun :-) -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-11 13:45:53
|
On Sat, 2004-09-11 at 06:25, Mich=C3=A8le Garoche wrote: > Le 10 sept. 2004, =C3=A0 18:47, Iago Rubio a =C3=A9crit : >=20 > > On Fri, 2004-09-10 at 18:09, Mich=C3=A8le Garoche wrote: > [snip] > but=20 > >> the > >> name in configure in ctags, then configure does not find it. > > > > Ok, so it's a configure problem then. > > > > In BSD it builts and works fine (my local patched version). > > > > Can you propose any change to make it build ? > > > > I'm a little lost here :) > Ok, I'll send something later, now I'm working on findinfiles plugin.=20 > Let me some minutes and I'll send changes to integrate first the=20 > changes in name (findinfiles instead of find_in_files) and the changes=20 > for DARWIN (triple conditional in src/Makefile.am), as well as the=20 > special (maxdepth=3D1 forced in wrc/callbacks.c). This way, it works fine= =20 > with multiple lines. I've just commited some changes, and sent the patch to the list. I hope we can advance and fix all those problems. > Just a glitch you may want to investigate later, when the changes are=20 > committed: > ** (cssed:6876): WARNING **: Invalid UTF8 string passed to=20 > pango_layout_set_text() > but it can be something specific to Mac. I will convert all output to UTF8 to avoid those error messages. --=20 Iago Rubio =20 - Home page * http://www.iagorubio.com=20 - Last project * http://cssed.sourceforge.net =20 - GPG Keyserv * pgp.rediris.es id=3D0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-11 13:27:34
|
There was a bug in quicksearch plugin build files that install the plugin in a wrong location. Fixed in cvs, patch attached. -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-11 13:19:20
|
On Sat, 2004-09-11 at 10:04, Michèle Garoche wrote: > I've made changes to make it work on Mac, + changes in strings and > update of po files. > > Works fine under Mac, please update your copy. > > Here is the diff: Great, thanks. > > ______________________________________________________________________ > Of course, you will have to regenerate the configure and Makefile.in, > as I did not touch them (since otherwise, they have changed also in > cvs, just configure.in and Makefile.am). Ok > A better (cleaner) change would be to block the interface for Mac with > the depth always activated and the minimum depth to 1. I've just bocked the "Depth" spin button at minimun value of 1, so there's no need for the test if it's < 1. Changes in CVS, patch attached. > I've noticed some glitches when using depth > 1, the search does not > end. Is it the same on other systems? Not at all. I'm still thinking there's something bad in getline(). The command line works fine in the shell, so the failure must be in the code. > But it works perfectly with depth =1; so maybe we could block the depth > on Mac to 1, then the user changes the base directory, not ideal, but > at least it works, could it a quick workaround in case you want to > release very quickly. Ok, will hide in OsX the "Depth" settings if we doesn't find other way, but it's still in my head to completly fix it on OsX. I'll also fix the encoding warnings converting all text to UTF8 before passing it to the pango layout. -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-11 11:38:14
|
Compilation was breaking due the use of some non ASCII characters in cssed's source files (interface.c). I commited some changes in Makefile.in.in to tell xgettext to expect UTF-8 characters in source files. Changes in CVS, patch attached. -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: <mic...@ea...> - 2004-09-11 08:12:38
|
Le 10 sept. 2004, à 18:47, Iago Rubio a écrit : > On Fri, 2004-09-10 at 18:09, Michèle Garoche wrote: >> Le 10 sept. 2004, à 17:56, Iago Rubio a écrit : >> >>> The tags plugin must include two interfaces one por ctags (BSD, OsX) >>> and >>> other for exuberant ctags (the one working now). >> No, the other way around, on Mac OS X, this is exuberant ctags, but >> the >> name in configure in ctags, then configure does not find it. > > Ok, so it's a configure problem then. > > In BSD it builts and works fine (my local patched version). > > Can you propose any change to make it build ? > > I'm a little lost here :) What I have in configure.in: |
From: <mic...@ea...> - 2004-09-11 08:04:33
|
I've made changes to make it work on Mac, + changes in strings and update of po files. Works fine under Mac, please update your copy. Here is the diff: |
From: Iago R. <iag...@hi...> - 2004-09-11 04:48:29
|
I commited some changes in document.c to recognize UTF8 text, some minor changes highlighting. Patch attached. -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: <mic...@ea...> - 2004-09-11 04:25:17
|
Le 10 sept. 2004, =E0 18:47, Iago Rubio a =E9crit : > On Fri, 2004-09-10 at 18:09, Mich=E8le Garoche wrote: >> Le 10 sept. 2004, =E0 17:56, Iago Rubio a =E9crit : >> >>> The tags plugin must include two interfaces one por ctags (BSD, OsX) >>> and >>> other for exuberant ctags (the one working now). >> No, the other way around, on Mac OS X, this is exuberant ctags, but=20= >> the >> name in configure in ctags, then configure does not find it. > > Ok, so it's a configure problem then. > > In BSD it builts and works fine (my local patched version). > > Can you propose any change to make it build ? > > I'm a little lost here :) Ok, I'll send something later, now I'm working on findinfiles plugin.=20 Let me some minutes and I'll send changes to integrate first the=20 changes in name (findinfiles instead of find_in_files) and the changes=20= for DARWIN (triple conditional in src/Makefile.am), as well as the=20 special (maxdepth=3D1 forced in wrc/callbacks.c). This way, it works = fine=20 with multiple lines. Just a glitch you may want to investigate later, when the changes are=20 committed: ** (cssed:6876): WARNING **: Invalid UTF8 string passed to=20 pango_layout_set_text() but it can be something specific to Mac. Mich=E8le <http://micmacfr.homeunix.org> |
From: Iago R. <iag...@hi...> - 2004-09-11 04:14:56
|
On Fri, 2004-09-10 at 18:09, Mich=C3=A8le Garoche wrote: > Le 10 sept. 2004, =C3=A0 17:56, Iago Rubio a =C3=A9crit : >=20 > > The tags plugin must include two interfaces one por ctags (BSD, OsX)=20 > > and > > other for exuberant ctags (the one working now). > No, the other way around, on Mac OS X, this is exuberant ctags, but the=20 > name in configure in ctags, then configure does not find it. Ok, so it's a configure problem then.=20 In BSD it builts and works fine (my local patched version). Can you propose any change to make it build ? I'm a little lost here :) --=20 Iago Rubio =20 - Home page * http://www.iagorubio.com=20 - Last project * http://cssed.sourceforge.net =20 - GPG Keyserv * pgp.rediris.es id=3D0x909BD4DD |
From: <mic...@ea...> - 2004-09-11 02:56:56
|
As for the problem of only one line printed, you could check how it is=20= handled by my script findininfo (in $home/bin) and on the website=20 <http://micmacfr.homeunix.org/tcpcliservdet/wrapsockcref.shtml.fr#a19>,=20= Readline, mg_readline, etc... I think the problem is the end of line is never sent to the pipe, so=20 the buffer is not dumped on to the screen, I can see it, because the=20 grep command is always pending. It may gives you hints. Mich=E8le <http://micmacfr.homeunix.org> |
From: Iago R. <iag...@hi...> - 2004-09-11 02:01:13
|
On Fri, 2004-09-10 at 03:09, Mich=C3=A8le Garoche wrote: > Le 10 sept. 2004, =C3=A0 1:59, Mich=C3=A8le Garoche a =C3=A9crit : >=20 > > Second it does not compile on Mac OS X, since it has a _getline symbol=20 > > undefined. This is a known bug, this symbol is not exported. Could you=20 > > find another way to do it? > I've disabled the comments around #ifdef BSD and the corresponding=20 > #endif in callback.c (i.e. I've removed the lines) and it loads. So,=20 > probably a #ifdef DARWIN would be appropriated here. Upsh ! It's untested right now. It was there to change it when tested in BSD. It can easily fail. Will check it now. --=20 Iago Rubio =20 - Home page * http://www.iagorubio.com=20 - Last project * http://cssed.sourceforge.net =20 - GPG Keyserv * pgp.rediris.es id=3D0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-10 22:55:41
|
Due a bug in autoconf ( it's what autoconf itself reported ) I made some - small - changes in cssed's build scripts to let it configure on FreeBSD. Commited to cvs, patch attached. -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-10 22:49:43
|
On Fri, 2004-09-10 at 12:18, Iago Rubio wrote: > On Fri, 2004-09-10 at 01:59, Michèle Garoche wrote: > > First, I've changed the plugin name to cssed-findinfiles-plugin to be > > consistent with other plugins. Hence changes in configure.in, > > src/Makefile.am. > > ok. > > > Second it does not compile on Mac OS X, since it has a _getline symbol > > undefined. This is a known bug, this symbol is not exported. > > The same apply in BSD, as it have no getline. > > > Could you > > find another way to do it? > > Sure, I will change it if fails on OsX. Well, some changes in cvs. Now it works in FreeBSD. Patch attached. -- Iago Rubio - Home page * http://www.iagorubio.com - Last project * http://cssed.sourceforge.net - GPG Keyserv * pgp.rediris.es id=0x909BD4DD |
From: Iago R. <iag...@hi...> - 2004-09-10 21:57:08
|
On Fri, 2004-09-10 at 04:21, Mich=C3=A8le Garoche wrote: > What is it supposed to do? I only get one line when I search for a=20 > term. It's a bug. > I would have thought I got as many lines as the term occurs in=20 > the files found recursively (at the given depth) in the given base=20 > directory. It's the intended behaviour. > Maybe it is related to the fact I've changed getline definition? Almost sure. > Other than that, it seems to work. I will fix it right now. --=20 Iago Rubio =20 - Home page * http://www.iagorubio.com=20 - Last project * http://cssed.sourceforge.net =20 - GPG Keyserv * pgp.rediris.es id=3D0x909BD4DD |
From: <mic...@ea...> - 2004-09-10 21:51:38
|
Le 10 sept. 2004, =E0 18:50, Iago Rubio a =E9crit : > On Fri, 2004-09-10 at 18:08, Mich=E8le Garoche wrote: >> Le 10 sept. 2004, =E0 17:56, Iago Rubio a =E9crit : >> >>> The find in files plugin is being fixed and I expect a working = update >>> tomorrow. >> I'm waiting for it :-) >> >> =46rom my investigation, it works with the getline you have = commented >> out, then if you use: >> >> find /Users/toto -maxdepth 1 type -f | xargs grep -n "foo" >> >> it should work (at least it works on the command line) >> >> The problem is a permission problem and a type problem and a depth >> problem. >> 1 - You could not search at depth 0 >> 2 - You could not search files that are not regular files (I mean >> printer drivers, symbolic links, directories) with this syntax. > > Oh! Great help, lots of thanks. > > I was fiddling in the getline function and was unable to fix it. > >> See if it works on BSD with this syntax (maxdepth >=3D1 and = obligatory), >> type -f (silently added). > > Will do it right now. It always prints only one line, that's something to correct, maybe use=20= -print at the end. This is the logic, but correct it to be right C code cmd_line =3D g_strdup_printf ("find %s ", basedir_str); if( = gtk_toggle_button_get_active(GTK_TOGGLE_BUTTON(checkbox_depth)) ){ depth_str =3D g_strdup_printf( "-maxdepth %d ",=20 gtk_spin_button_get_value_as_int(GTK_SPIN_BUTTON(spinbutton_depth)) ); // In case the malicious user changes it to 0 #ifdef DARWIN or #ifdef BSD if depth_Str < 1 then %d =3D=3D 1 #endif old_ptr =3D cmd_line; // Silently add -type f for DARWIN and BSD #ifdef DARWIN or #idfef BSD cmd_line =3D g_strdup_printf("%s %s -type f", old_ptr, = depth_str ); #elif=09 cmd_line =3D g_strdup_printf("%s %s", old_ptr, depth_str = ); #endif g_free(old_ptr); g_free(depth_str); else { // Force to 1 for BSD and DARWIN #ifdef DARWIN or #ifdef BSD old_ptr =3D cmd_line; cmd_line =3D g_strdup_printf("%s %s -type f", old_ptr, = depth_str ); g_free(old_ptr); #endif }=09 old_ptr =3D cmd_line; cmd_line =3D g_strdup_printf("%s | xargs grep -n %s", old_ptr,=20= searchterm_str ); g_free( old_ptr ); Mich=E8le <http://micmacfr.homeunix.org>#ifdef DARWIN or #ifdef BSD #ifdef DARWIN or #ifdef BSD |
From: <mic...@ea...> - 2004-09-10 17:52:32
|
Le 10 sept. 2004, =E0 18:12, Iago Rubio a =E9crit : > On Fri, 2004-09-10 at 17:56, Iago Rubio wrote: >> I'm even thinking about two plugins one for ctags and other for >> exuberant ctags. > > Ok, I think it will be the approach. > > I installed exuberant ctags on BSD and the plugin worked fine with a > small patch to rename ctags to exctags in BSD systems, to avoid = clashes > between both programs. > > I was wondering how it works in OsX, as it does not changes the = package > name but uses something called "update-alternatives". > > Michelle any idea ?? > http://fink.sourceforge.net/pdb/package.php/exuberant-ctags > > It's maintained by pogma, can you ask him how we can know the ctags > binary is exuberant ctags or not ? On mac, you can have exuberant-ctags or ctags embedded into emacs and=20 xemacs, so for me the binary should be exuberant ctags, as I don't see=20= a way to use the ctags embedded into xemacs or emacs. The alternative=20 allows to list all packages which uses ctags and defines a priority.=20 Then a symlink is created to the package which has the higher priority.=20= The user can change it as he wants. Example: update-alternatives --display ctags ctags - status is manual. link currently points to /sw/bin/ctags-exuberant /sw/bin/ctags.xemacs - priority 60 slave ctags.1: /sw/share/man/man1/ctags.1.xemacs /sw/bin/ctags.emacs21 - priority 40 slave ctags.1: /sw/share/man/man1/ctags.1.emacs21 /sw/bin/ctags-exuberant - priority 70 slave ctags.1: /sw/share/man/man1/ctags.1-exuberant Current `best' version is /sw/bin/ctags-exuberant. > > The ctags plugin is to come, the current one will be only for = exuberant > ctags and will be - surely - renamed. > > The idea on this plugin, was to use the exuberant ctags support for = web > languages as PHP, HTML and much others (as Java, Javascript, Perl, sh, > python, Makefiles ... ) in cssed. > > ITOH I'm writing support for CSS tags on exuberant ctags, but can not > bind it to ctags. > > Using just the ctags program we can not search anything but C source > code, that's not really useful for cssed. > > Am I wrong with this ? > > Does the ctags program find any other tag than C tags ? > > I've got in mind it can search throught Lisp, Fortran,Pascal, C, lex=20= > and > yacc sources. > > Nothing apropiated to use in cssed. You can maybe search toutdoux and dtags, dtags is an interface for=20 using ctags for xsltproc within toutdoux. It can give you hints. I've patched the line in configure this way, then it works: dnl This plugin needs the ctags program dnl AC_CHECK_PROG([CTAGS], [ctags], [no-ctags]) dnl Exuberant ctags AC_MSG_CHECKING(for Exuberant Ctags) if ctags-exuberant --version > /dev/null 2>&1; then AC_MSG_RESULT(yes)=09 ctags=3Dyes else AC_MSG_ERROR(ctags-exuberant software not found) ctags=3Dno Mich=E8le <http://micmacfr.homeunix.org> |