Re: still can't import any latex tabulars
> "José" == José Matos <[EMAIL PROTECTED]> writes: José> I think that this may be related. In the machine I'm testing I José> don't have lyx installed and I don't see also the option to José> import the tex file, This is to be expected, you know :) JMarc
Re: Using the Qt port for native Macosx/Win32 ports?
> However, I think asking all 76 people in lib/CREDITS should be ok. Not > that I believe this is practicable... maybe it's worthwhile to do this. 10 people probably account for 90 percent of the code anyway. if the big contributors agree then we could easily remove code from people who complain and replace it by new code i think that it would be a great advantage to being able to legally distribute a windows executable. i also think that having a windows version of lyx will *greatly* increase it's popularity for me personally it means that i will be able to exchange documents with my coauthors, not a minor thing and something i've been looking forward to for over two years now consider this email as an ok from me to change the license to allow qt on windows for whatever i've contributed Ed.
Re: still can't import any latex tabulars
On Wednesday 08 January 2003 08:55, Jean-Marc Lasgouttes wrote: Good morning :-) > This is to be expected, you know :) Why? It used to. I made some work work to enable reLyX to run from the devel directory. Also this is inconsistent lyx2lyx runs from the devel directory but reLyX not. It doesn't make sense. :-) > JMarc -- José Abílio
Re: Remove trailing \r when reading...
> > Not with Win32! Have a look into src/support/os_win32.C. There > > os::read:mode() does return "rb" for Win32! > > I think this has nothing to do with the \r in LaTeX.C. There an ordinary > ifstream is used without giving the std::ios_base::binary flag, so this > should read in text mode. > > What the * is this os::read::mode good for anyway? Oops! You're right. This couldn't be the reason for seeing \r in Lyx ;-( I thought os::read_mode() is still used more often in Lyx. Ruurd has introduced the class os to handle different os depending things and os::read_mode was included for all f/popen calls. Now that method is only used in filetools.C (do_popen()). So what's the problem? I think its due to this: Installing Cygwin you have to choose which drives/directories you will mount in which mode (either binary or text mode). Choosing binary you'll have problems opening text files created with Cygwin in most Windows programms (missing \r). Choosing text mode you'll get problems with Lyx: At the very beginning of my Lyx port I've realized that opening localization files in intl/loadmsgcat.c needs to use binary mode. Its the same when copying a file (src/support/copy.C) as well as using any file where \n and \r are handled as different characters (i.e. the old tabulars). I'm using binary mode for all mounts in Cygwin. Therefore reading Win32/DOS-text files (like Miktex log files) will bring up those \r's which will be shown as new lines in an xforms window and therefore making error messages unreadable. Lyx should run independent from the mode a drive is mounted. Therefore we need binary mode in some cases and text mode in some other cases whereas in most cases it doesn't matter. Of course it could be done easier with ifstream ifs(tmp.c_str(), ios::in | ios::text); Maybe we should include that instead ;-) And we can rewrite all calls to open a file for both reading and writing. But: is os::text/binary supported on all systems? My experiences with SunOS as well as with AIX and HP/UX some years ago have shown that their compilers don't knew these flags! So introducing them in more places then necessary may lead into problems on other systems. Claus
1.3.0 stoppers
Is there anything else than 8, 436, 622, 638, 725, 763, 791 which I get with http://bugzilla.lyx.org/buglist.cgi?short_desc_type=allwordssubstr&short_desc=&target_milestone=1.3.0&long_desc_type=allwordssubstr&long_desc=&bug_file_loc_type=allwordssubstr&bug_file_loc=&keywords_type=anywords&keywords=&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&emailtype1=substring&email1=&emailtype2=substring&email2=&bugidtype=include&bug_id=&votes=&changedin=&chfieldfrom=&chfieldto=Now&chfieldvalue=&cmdtype=doit&order=Reuse+same+sort+as+last+time&field0-0-0=noop&type0-0-0=noop&value0-0-0= ? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: still can't import any latex tabulars
> "José" == José Matos <[EMAIL PROTECTED]> writes: José> On Wednesday 08 January 2003 08:55, Jean-Marc Lasgouttes wrote: José> Good morning :-) Good morning José! >> This is to be expected, you know :) José> Why? It used to. I made some work work to enable reLyX to run José> from the devel directory. Also this is inconsistent lyx2lyx runs José> from the devel directory but reLyX not. It doesn't make sense. José> :-) You wrote that, since you do not have lyx installed, you do not see this and that in the export menu. I say: this is not a surprise. I hope you will get the joke this time because it was already mildly funny on the first round, and it is going to become worse and worse everytime I have to spell it out further... JMarc
Re: Remove trailing \r when reading...
On Wed, Jan 08, 2003 at 09:55:18AM +0100, Claus Hentschel wrote: > Lyx should run independent from the mode a drive is mounted. Therefore we > need binary mode in some cases and text mode in some other cases whereas in > most cases it doesn't matter. > > Of course it could be done easier with > ifstream ifs(tmp.c_str(), ios::in | ios::text); > Maybe we should include that instead ;-) And we can rewrite all calls to > open a file for both reading and writing. Does that work? > But: is os::text/binary supported on all systems? My experiences with SunOS > as well as with AIX and HP/UX some years ago have shown that their compilers > don't knew these flags! Because there is no such distinction on Unix, support for it is somewhat of a luxury. Conformant compilers must provide that luxury, though. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
[Bug 763] lyx2lyx adds gibberish to title
http://bugzilla.lyx.org/show_bug.cgi?id=763 [EMAIL PROTECTED] changed: What|Removed |Added CC||[EMAIL PROTECTED] AssignedTo|[EMAIL PROTECTED] |[EMAIL PROTECTED] Component|lyx2lyx |lyxlex --- Additional Comments From [EMAIL PROTECTED] 2003-01-08 22:34 --- This is not a lyx2lyx bug. The output seems to be correct there. It appears to be the lyx parser who is responsible for the extra garbage. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is.
Re: Remove trailing \r when reading...
Andre Poenitz <[EMAIL PROTECTED]> writes: | On Wed, Jan 08, 2003 at 09:55:18AM +0100, Claus Hentschel wrote: | > Lyx should run independent from the mode a drive is mounted. Therefore we | > need binary mode in some cases and text mode in some other cases whereas in | > most cases it doesn't matter. | > | > Of course it could be done easier with | > ifstream ifs(tmp.c_str(), ios::in | ios::text); | > Maybe we should include that instead ;-) And we can rewrite all calls to | > open a file for both reading and writing. | | Does that work? ios::text? No. It is a non-standard flag. | > But: is os::text/binary supported on all systems? My experiences with SunOS | > as well as with AIX and HP/UX some years ago have shown that their compilers | > don't knew these flags! | | Because there is no such distinction on Unix, support for it is somewhat of | a luxury. Conformant compilers must provide that luxury, though. ios::binary otoh is a standard flag and should work everywhere. -- Lgb
[Bug 763] lyx2lyx adds gibberish to title
http://bugzilla.lyx.org/show_bug.cgi?id=763 [EMAIL PROTECTED] changed: What|Removed |Added Status|NEW |ASSIGNED --- Additional Comments From [EMAIL PROTECTED] 2003-01-08 22:52 --- The problem is the second space in e.g. \leftmargin 1 in Which LyX version produces such a thing? Andre' --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is.
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wednesday 08 January 2003 22:52, [EMAIL PROTECTED] wrote: > > Which LyX version produces such a thing? > > > Andre' The file header says: #This file was created by Thu Nov 20 17:46:00 1997 #LyX 0.11 (C) 1995-1997 Matthias Ettrich and the LyX Team One possible solution would be in the 215 -> 216 lyx2lyx step to search for such cases and remove the extra space. What do you think? Notice that 0.11 are not yet in lyx2lyx range, that will be for 1.4 but if this is simple enough we could fix it now. -- José Abílio
lyx2lyx style question
Currently nesting there is 1. level: 4 spaces 2. level: 1 tab 3. level: 1 tab + 4 spaces etc. which is a mess when tab != 8 spaces. Is this mandatory for python or can't we simply use the same rules as in the rest of LyX? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Remove trailing \r when reading...
> | > Of course it could be done easier with > | > ifstream ifs(tmp.c_str(), ios::in | ios::text); > | > Maybe we should include that instead ;-) And we can rewrite all calls to > | > open a file for both reading and writing. > | > | Does that work? No, not with gcc, because gcc is conform with ANSI. > ios::text? No. It is a non-standard flag. > ... > ios::binary otoh is a standard flag and should work everywhere. So there seems not to be another way then removong those trailing \r's the way I'v done. We do need that in all cases where text read from a (binary) opened file is presented as-is in a window. Claus
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wed, Jan 08, 2003 at 10:02:48AM +, José Matos wrote: > #This file was created by Thu Nov 20 17:46:00 1997 > #LyX 0.11 (C) 1995-1997 Matthias Ettrich and the LyX Team > > One possible solution would be in the 215 -> 216 lyx2lyx step to search for > such cases and remove the extra space. > > What do you think? The proper way would to determine up to which version LyX could read this... As lyx2lyx is new I'd even put that in the 218 -> 220 script. > Notice that 0.11 are not yet in lyx2lyx range, that will be for 1.4 but if > this is simple enough we could fix it now. So the target milestone should be changed. I got aware of it when asking for thing with 1.3.0 target. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Remove trailing \r when reading...
"Claus Hentschel" <[EMAIL PROTECTED]> writes: | > | > Of course it could be done easier with | > | > ifstream ifs(tmp.c_str(), ios::in | ios::text); | > | > Maybe we should include that instead ;-) And we can rewrite all calls | to | > | > open a file for both reading and writing. | > | | > | Does that work? | | No, not with gcc, because gcc is conform with ANSI. | | > ios::text? No. It is a non-standard flag. | > ... | > ios::binary otoh is a standard flag and should work everywhere. | | So there seems not to be another way then removong those trailing \r's the | way I'v done. We do need that in all cases where text read from a (binary) | opened file is presented as-is in a window. Then we should not open those files in binary mode. -- Lgb
Re: lyx2lyx style question
On Wednesday 08 January 2003 10:00, Andre Poenitz wrote: > Currently nesting there is > > 1. level: 4 spaces > 2. level: 1 tab > 3. level: 1 tab + 4 spaces > > etc. which is a mess when tab != 8 spaces. > > Is this mandatory for python or can't we simply use the same rules as in > the rest of LyX? For python the indentation only needs to be consistent locally, you could have different styles in different locations inside the same file. Example: def t(x): if x>0: return x else: return -x def f(x): if x>0: return x*x else: return -x*x This is valid although I have never seen such style (for obvious reasons ;-)). The cruel truth is that we use that style since it is emacs python mode default. > Andre' -- José Abílio
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wednesday 08 January 2003 10:05, Andre Poenitz wrote: > > The proper way would to determine up to which version LyX could read > this... > > As lyx2lyx is new I'd even put that in the 218 -> 220 script. If 0.11.x was the last version that used this style I propose to put the convertion code there and not later. > Andre' -- José Abílio
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wed, Jan 08, 2003 at 10:02:48AM +, José Matos wrote: > What do you think? Aehm... how does LyX currently lengths like '1 true in'? Simply removing spaces would not help here... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: [Bug 763] lyx2lyx adds gibberish to title
Andre Poenitz <[EMAIL PROTECTED]> writes: | > Notice that 0.11 are not yet in lyx2lyx range, that will be for 1.4 but if | > this is simple enough we could fix it now. | | So the target milestone should be changed. I got aware of it when asking | for thing with 1.3.0 target. I changed the milestone to make sure that all these (lyx2lyx) cases where looked at before 1.3.0. For this case, as it uses a file from 0.11, it is not really important anymore. My goals/expectations with lyx2lyx have been met if files from 1.0 and upwards in converted correctly. Everything beyond this is a bonus. One thing though: lyx2lyx should not try to do the best thing for formats it does not have, it should rather punt completely. A half converted format can be worse than having non at all... -- Lgb
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wed, Jan 08, 2003 at 10:20:00AM +, José Matos wrote: > If 0.11.x was the last version that used this style I propose to put the > convertion code there and not later. Do as you like. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wed, Jan 08, 2003 at 11:18:40AM +0100, Lars Gullik Bjønnes wrote: > One thing though: lyx2lyx should not try to do the best thing for > formats it does not have, it should rather punt completely. A half > converted format can be worse than having non at all... I agree if called from LyX. For calling lyx2lyx manually I really like to have a 'do the best you can' mode... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: [Bug 763] lyx2lyx adds gibberish to title
> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes: Lars> Andre Poenitz <[EMAIL PROTECTED]> writes: | > Notice that 0.11 are Lars> not yet in lyx2lyx range, that will be for 1.4 but if | > this Lars> is simple enough we could fix it now. Lars> | Lars> | So the target milestone should be changed. I got aware of it Lars> when asking | for thing with 1.3.0 target. Lars> I changed the milestone to make sure that all these (lyx2lyx) Lars> cases where looked at before 1.3.0. Lars> For this case, as it uses a file from 0.11, it is not really Lars> important anymore. My goals/expectations with lyx2lyx have been Lars> met if files from 1.0 and upwards in converted correctly. Lars> Everything beyond this is a bonus. It would be nice to cover 0.12, since this is basically KLyX format. JMarc
Re: [Bug 763] lyx2lyx adds gibberish to title
Andre Poenitz <[EMAIL PROTECTED]> writes: | On Wed, Jan 08, 2003 at 11:18:40AM +0100, Lars Gullik Bjønnes wrote: | > One thing though: lyx2lyx should not try to do the best thing for | > formats it does not have, it should rather punt completely. A half | > converted format can be worse than having non at all... | | I agree if called from LyX. For calling lyx2lyx manually I really like to | have a 'do the best you can' mode... --force kind of thing? -- Lgb
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wednesday 08 January 2003 10:18, Lars Gullik Bjønnes wrote: > > One thing though: lyx2lyx should not try to do the best thing for > formats it does not have, it should rather punt completely. A half > converted format can be worse than having non at all... The problem here is that 2.15 is not really a file format but a loose colection of slightly different formats and there isn't any simple way (even for a human) to distinguish between them. For now lyx2lyx follows the same lines you stated. -- José Abílio
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wednesday 08 January 2003 10:23, Jean-Marc Lasgouttes wrote: > > It would be nice to cover 0.12, since this is basically KLyX format. I have tried to do it. I expect it to convert it without flaws, so I would appreciate to have users testing and reporting problems. > JMarc -- José Abílio
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wednesday 08 January 2003 10:18, Andre Poenitz wrote: > > Aehm... how does LyX currently lengths like '1 true in'? > > Simply removing spaces would not help here... Give me a general pattern and I will lift the Earth (or is it the other way) ;-) > Andre' -- José Abílio
Re: lyx-1.2.2: ./configure deletes /dev/null
> "Dominik" == Dominik Vogt <[EMAIL PROTECTED]> writes: Dominik> When I run the configure script that comes with lyx-1.2.2, it Dominik> deletes the /dev/null device entry. Here is a patch to fix this in 1.3.0cvs. I intend to apply the same patch to 1.2.3cvs. Lars? JMarc ? config.h.in Index: config/ChangeLog === RCS file: /usr/local/lyx/cvsroot/lyx-devel/config/ChangeLog,v retrieving revision 1.83 diff -u -p -r1.83 ChangeLog --- config/ChangeLog 18 Dec 2002 16:09:11 - 1.83 +++ config/ChangeLog 8 Jan 2003 10:37:20 - @@ -1,3 +1,10 @@ +2003-01-08 Jean-Marc Lasgouttes <[EMAIL PROTECTED]> + + * lyxinclude.m4 (LYX_CHECK_ERRORS): do not delete $cache_file if + there has been an error in configuration. With autoconf 2.5x, + $cache_file is often set to /dev/null, and deleting that is not a + good idea... + 2002-12-18 Lars Gullik Bjønnes <[EMAIL PROTECTED]> * configure.ac (AC_CONFIG_FILES): add sourcedoc/Makefile Index: config/lyxinclude.m4 === RCS file: /usr/local/lyx/cvsroot/lyx-devel/config/lyxinclude.m4,v retrieving revision 1.79 diff -u -p -r1.79 lyxinclude.m4 --- config/lyxinclude.m4 25 Nov 2002 16:02:21 - 1.79 +++ config/lyxinclude.m4 8 Jan 2003 10:37:20 - @@ -81,9 +81,7 @@ cat <
Re: [Bug 763] lyx2lyx adds gibberish to title
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: | > "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes: | | Lars> Andre Poenitz <[EMAIL PROTECTED]> writes: | > Notice that 0.11 are | Lars> not yet in lyx2lyx range, that will be for 1.4 but if | > this | Lars> is simple enough we could fix it now. | Lars> | | Lars> | So the target milestone should be changed. I got aware of it | Lars> when asking | for thing with 1.3.0 target. | | Lars> I changed the milestone to make sure that all these (lyx2lyx) | Lars> cases where looked at before 1.3.0. | | Lars> For this case, as it uses a file from 0.11, it is not really | Lars> important anymore. My goals/expectations with lyx2lyx have been | Lars> met if files from 1.0 and upwards in converted correctly. | Lars> Everything beyond this is a bonus. | | It would be nice to cover 0.12, since this is basically KLyX format. Sure, and afaik lyx2lyx covers this. -- Lgb
Re: lyx2lyx style question
On Wednesday 08 January 2003 10:32, Andre Poenitz wrote: > > So I'd rather have that in 'LyX style' with which emacs users _and_ I seem > to be happy. Moreover, it would make the whole of LyX more consistent. Lars, any clue how to force emacs to do this? > Andre' -- José Abílio
Re: LyX for Win32 - Test release
> Issues that remain are: > -finding a suitable replacement for a unix style home directory > -fix some external scripts -ok/apply of dialogs crashes lyx -lot of flicker on menu change (e.g. file -> edit -> insert ) -symbols don't show in math panel -lyx starts with console window which remains open and: > -probably a lot more. Ed.
Bugs in ReLyX
Hi folks, I wanted to convert a TeX document of a colleague of mine to LyX (1.3.0cvs) but I failed for several reasons: 1. reLyX fails if there is a subdirectory with the same (base) name as a TeX file 2. reLyX fails if the full path starting with the main document directory is specified in a document in a subdirectory (astonishingly, LaTeX accepts this) 3. reLyX creates a "makeindex" entry in the LyX file header which is not accepted by LyX 4. reLyX produces an old file format; when converting a fig inset into the LyX 1.3 file format, the Python converter script fails Enclosed please find a minimized test case which covers all those bugs. It would be nice if at least the third and forth bug can be fixed. Michael -- === Michael Schmitt Telefon: +49 651 97551-40 Institut für TelematikTelefax: +49 651 97551-12 Bahnhofstrasse 30-32 WWW: http://www.ti.fhg.de D-54292 Trier E-Mail: mailto:[EMAIL PROTECTED] === relyx-fails.tar Description: Unix tar archive
Re: lyx-1.2.2: ./configure deletes /dev/null
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: | > "Dominik" == Dominik Vogt <[EMAIL PROTECTED]> writes: | | Dominik> When I run the configure script that comes with lyx-1.2.2, it | Dominik> deletes the /dev/null device entry. | | Here is a patch to fix this in 1.3.0cvs. I intend to apply the same | patch to 1.2.3cvs. | | Lars? Yes. This should absolutely go in. It is not really our task to decide how autoconf should handle its cache file. -- Lgb
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wed, Jan 08, 2003 at 10:41:35AM +, José Matos wrote: > On Wednesday 08 January 2003 10:18, Andre Poenitz wrote: > > > > Aehm... how does LyX currently lengths like '1 true in'? > > > > Simply removing spaces would not help here... > > Give me a general pattern and I will lift the Earth (or is it the other way) LyX itself can't cope with it anymore, as lyxlex::getString() doesn't read spaces... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: 1.3.0 stoppers
How about bugs #770, #795 ? Michael -- === Michael Schmitt Telefon: +49 651 97551-40 Institut für TelematikTelefax: +49 651 97551-12 Bahnhofstrasse 30-32 WWW: http://www.ti.fhg.de D-54292 Trier E-Mail: mailto:[EMAIL PROTECTED] ===
Re: lyx2lyx style question
On Wednesday 08 January 2003 10:58 am, José Matos wrote: > On Wednesday 08 January 2003 10:32, Andre Poenitz wrote: > > So I'd rather have that in 'LyX style' with which emacs users _and_ I > > seem to be happy. Moreover, it would make the whole of LyX more > > consistent. > > Lars, any clue how to force emacs to do this? You need the equivalent of this in your .emacs file: ;; C++ mode hook (defun my-c++-mode-hook () ;; Style parameters (setq c-basic-offset 8) ) (add-hook 'c++-mode-hook 'my-c++-mode-hook) I guess that you should replace the 'c' and 'c++' in the above with 'python' and you're there. I can't test here because my emacs knows no python. Like me really ;-) -- Angus
Re: 1.3.0 stoppers
On Wed, Jan 08, 2003 at 11:55:09AM +0100, Michael Schmitt wrote: > How about bugs #770, #795 ? They have no target milestone. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
[WARNING] Bad bug in lyx 1.2.2 configure script
There is a very nasty bug in LyX 1.2.2 configure script: if you configure as root and an error occurs during the configure step (e.g. xforms not found), the the script will delete the /dev/null device. This undoubtly leads to very bad effect, since /dev/null is a very basic device in any unix system. If you have been victim of this problem, please complain here and I'll try to find out how to reconstruct it (some magic command that I do not know yet). I plan to release a 1.2.3 version that does not have this problem very soon. The same problem occurs in 1.3.0cvs (and 1.3.0pre1 I guess) but is fixed now in cvs. A bit of background: there has been for a long time some code in LyX configure script that removed the config.cache file (that keep tracks of tests results) when an error has occured. This allows to re-run the script once you have made correction without seeing the old results. However, starting with autoconf 2.5x, the cache file now defaults to /dev/null (i.e. no cache). Since LyX 1.2.2 was the first version distributed with an autoconf script generated with autoconf 2.52, this triggered this awful behaviour. Sorry for the problems that this may cause to your systems... JMarc
python: length of list
len(foo) doesn't work:
(My first shot at python!)
def change_preamble(lines):
for line in lines:
words = string.split(line)
if len(words) > 2:
if [ "\\oddsidemargin", "\\evensidemargin", "\\marginparwidth",
"\\textwidth", "\\topmargin",
"\\textheight"].count(words[0]) != 0:
len = words[1] + words[2]
words[1:3] = [len]
line = string.join(words)
yields:
Traceback (most recent call last):
File "./lyx2lyx", line 183, in ?
main(sys.argv)
File "./lyx2lyx", line 176, in main
__import__("lyxconvert_" + fmt).convert(header,body)
File "./lyxconvert_220.py", line 102, in convert
change_preamble(header)
File "./lyxconvert_220.py", line 92, in change_preamble
l = len(words)
UnboundLocalError: local variable 'len' referenced before assignment
What does that mean?
Andre'
--
Those who desire to give up Freedom in order to gain Security,
will not have, nor do they deserve, either one. (T. Jefferson)
Re: lyx2lyx style question
José Matos <[EMAIL PROTECTED]> writes: | On Wednesday 08 January 2003 10:32, Andre Poenitz wrote: | > | > So I'd rather have that in 'LyX style' with which emacs users _and_ I seem | > to be happy. Moreover, it would make the whole of LyX more consistent. | | Lars, any clue how to force emacs to do this? (setq some-basic-offset 8) (perhaps 't) -- Lgb
Re: [Bug 763] lyx2lyx adds gibberish to title
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes: Andre> On Wed, Jan 08, 2003 at 10:41:35AM +, José Matos wrote: >> On Wednesday 08 January 2003 10:18, Andre Poenitz wrote: >> > >> > Aehm... how does LyX currently lengths like '1 true in'? >> > >> > Simply removing spaces would not help here... >> >> Give me a general pattern and I will lift the Earth (or is it the >> other way) Andre> LyX itself can't cope with it anymore, as lyxlex::getString() Andre> doesn't read spaces... The good fix in this case is to enclose the length inside double quotes (and make sure that lyxlex is asked to honor these quotes). What are the lyx tags affected? JMarc
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wed, Jan 08, 2003 at 12:40:02PM +0100, Jean-Marc Lasgouttes wrote: > The good fix in this case is to enclose the length inside double > quotes (and make sure that lyxlex is asked to honor these quotes). Probably. Another fix might be as simple as "use rest of line"... > What are the lyx tags affected? In this file: \oddsidemargin -10 pt \evensidemargin 10i pt \marginparwidth 1 in \oddsidemargin 0 in \evensidemargin 0 in \marginparwidth 0.75 in \textwidth 6.375 true in \topmargin -0.25 in \textheight 8.5 true in Note that three of them appear twice. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: python: length of list
On Wed, Jan 08, 2003 at 12:34:51PM +0100, Andre Poenitz wrote: > > len(foo) doesn't work: > > (My first shot at python!) > > def change_preamble(lines): > for line in lines: > words = string.split(line) > if len(words) > 2: > if [ "\\oddsidemargin", "\\evensidemargin", "\\marginparwidth", > "\\textwidth", "\\topmargin", >"\\textheight"].count(words[0]) != 0: > len = words[1] + words[2] If you assign a value to the builtin len function, bad things will happen... Also, the line if [ "\\oddsidemargin", ...].count(words[0]) != 0: can be replace by if words[0] in ["\\oddsidemargin", ...] However, it is probably better to use search&replace of regular expressions.
Re: [Bug 763] lyx2lyx adds gibberish to title
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes: Andre> On Wed, Jan 08, 2003 at 12:40:02PM +0100, Jean-Marc Lasgouttes Andre> wrote: >> The good fix in this case is to enclose the length inside double >> quotes (and make sure that lyxlex is asked to honor these quotes). Andre> Probably. Another fix might be as simple as "use rest of Andre> line"... >> What are the lyx tags affected? Andre> In this file: Andre> \oddsidemargin -10 pt \evensidemargin 10i pt \marginparwidth 1 Andre> in \oddsidemargin 0 in \evensidemargin 0 in \marginparwidth Andre> 0.75 in \textwidth 6.375 true in \topmargin -0.25 in Andre> \textheight 8.5 true in Of these, only \topmargin still exist in 1.3.0cvs. Do you mean that the others are kept unchanged by lyx2lyx? This would be very wrong. But indeed the code in buffer.C around line 765 could use lex.eatLine() instead of lex.next(). JMarc
Re: lyx2lyx style question
[EMAIL PROTECTED] (Lars Gullik Bjønnes) writes: | José Matos <[EMAIL PROTECTED]> writes: | | | On Wednesday 08 January 2003 10:32, Andre Poenitz wrote: | | > | | > So I'd rather have that in 'LyX style' with which emacs users _and_ I seem | | > to be happy. Moreover, it would make the whole of LyX more consistent. | | | | Lars, any clue how to force emacs to do this? | | (setq some-basic-offset 8) I checked... (setq py-indent-offset 8) should do the trick. (M-x set-variable RET py-indent-offset RET 8 RET) -- Lgb
Re: python: length of list
On Wed, Jan 08, 2003 at 01:52:04PM +0200, Dekel Tsur wrote: > > if [ "\\oddsidemargin", "\\evensidemargin", "\\marginparwidth", > > "\\textwidth", "\\topmargin", >"\\textheight"].count(words[0]) != 0: > > len = words[1] + words[2] > > If you assign a value to the builtin len function, bad things will happen... As I said: > > (My first shot at python!) > However, it is probably better to use search&replace of regular expressions. I doubt so. Or do you have a regexp that matches TeX lengths? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: [WARNING] Bad bug in lyx 1.2.2 configure script
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: | If you have been victim of this problem, please complain here and I'll | try to find out how to reconstruct it (some magic command that I do | not know yet). On linux: mknod /dev/null c 1 3 (also use the -m switch to set permissons correctly "rw-rw-rw-", or chmod the char device after creation "chmod a=rw /dev/null") -- Lgb
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wed, Jan 08, 2003 at 12:57:29PM +0100, Jean-Marc Lasgouttes wrote: > Of these, only \topmargin still exist in 1.3.0cvs. Do you mean that > the others are kept unchanged by lyx2lyx? This would be very wrong. LyX ignores when reading the file... > But indeed the code in buffer.C around line 765 could use > lex.eatLine() instead of lex.next(). Would be wrong to read a number and drop the unit Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: python: length of list
On Wed, Jan 08, 2003 at 01:52:04PM +0200, Dekel Tsur wrote: > However, it is probably better to use search&replace of regular expressions. def change_preamble(lines): for i in xrange(0, len(lines)): words = string.split(lines[i]) if len(words) > 2 and words[0] in [ "\\oddsidemargin", "\\evensidemargin", "\\marginparwidth", "\\textwidth", "\\topmargin", "\\textheight", "\\rightmargin", "\leftmargin", "\\bottommargin"] : words[1:3] = [ words[1] + words[2] ] lines[i] = string.join(words) in lyxconvert_220.py fixes the reported case. I can't build a patch as baywatch seems to be down? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Bugs in ReLyX
On Wednesday 08 January 2003 10:45, Michael Schmitt wrote: > Hi folks, > > I wanted to convert a TeX document of a colleague of mine to LyX > (1.3.0cvs) but I failed for several reasons: [...] > 4. reLyX produces an old file format; when converting a fig inset into > the LyX 1.3 > file format, the Python converter script fails > > Enclosed please find a minimized test case which covers all those bugs. > It would be nice if at least the third and forth bug can be fixed. I think that 4) is a know bug, but I will give it a try later. > Michael -- José Abílio
Re: 1.3.0 stoppers
Andre Poenitz wrote: On Wed, Jan 08, 2003 at 11:55:09AM +0100, Michael Schmitt wrote: How about bugs #770, #795 ? They have no target milestone. #795 has a target milestone now :-) 135 bugzilla bug reports in total including unconfirmed and trivial bugs! Not to bad. I will check my personal bug report list this evening; I guess I can contribute about 5-10 new entries. Michael -- === Michael Schmitt Telefon: +49 651 97551-40 Institut für TelematikTelefax: +49 651 97551-12 Bahnhofstrasse 30-32 WWW: http://www.ti.fhg.de D-54292 Trier E-Mail: mailto:[EMAIL PROTECTED] ===
Re: lyx2lyx style question
On Wednesday 08 January 2003 11:58, Lars Gullik Bjønnes wrote: > > (setq py-indent-offset 8) > > should do the trick. > > (M-x set-variable RET py-indent-offset RET 8 RET) Ok, I already knew that (trial and error), but there is a small (not) problem. It really inserts 8 spaces and not a tab. André will not like it. And ^Q + TAB is not a reasonable solution. -- José Abílio
Re: lyx2lyx style question
On Wed, Jan 08, 2003 at 12:27:53PM +, José Matos wrote: > Ok, I already knew that (trial and error), but there is a small (not) > problem. It really inserts 8 spaces and not a tab. André will not like it. I like it better than the mixed version... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: lyx2lyx style question
On Wed, Jan 08, 2003 at 12:27:53PM +, Jos? Matos wrote: > On Wednesday 08 January 2003 11:58, Lars Gullik Bj?nnes wrote: > > > > (setq py-indent-offset 8) > > > > should do the trick. > > > > (M-x set-variable RET py-indent-offset RET 8 RET) > > Ok, I already knew that (trial and error), but there is a small (not) > problem. It really inserts 8 spaces and not a tab. Andr? will not like it. > > And ^Q + TAB is not a reasonable solution. You need to add (setq-default indent-tabs-mode t)
Re: [WARNING] Bad bug in lyx 1.2.2 configure script
Am Mittwoch, 8. Januar 2003 12:32 schrieb Jean-Marc Lasgouttes: > There is a very nasty bug in LyX 1.2.2 configure script: if you > configure as root and an error occurs during the configure step > (e.g. xforms not found), the the script will delete the /dev/null > device. This undoubtly leads to very bad effect, since /dev/null is a > very basic device in any unix system. > > If you have been victim of this problem, please complain here and I'll > try to find out how to reconstruct it (some magic command that I do > not know yet). mknod -m666 /dev/null c 1 3 as root, of course -- ronny
Re: [WARNING] Bad bug in lyx 1.2.2 configure script
For anyone who needs to recover, On my RedHat 8.0 machine /dev/null is a character device with major mode 1, minor mode 3 and mode a+rw - use mknod to make a new one. On Solaris, it's a soft link to /devices/pseudo/mm@0:null - use ln to recreate. Bill. On 8 Jan 2003, Jean-Marc Lasgouttes wrote: > > There is a very nasty bug in LyX 1.2.2 configure script: if you > configure as root and an error occurs during the configure step > (e.g. xforms not found), the the script will delete the /dev/null > device. This undoubtly leads to very bad effect, since /dev/null is a > very basic device in any unix system. > > If you have been victim of this problem, please complain here and I'll > try to find out how to reconstruct it (some magic command that I do > not know yet). > > I plan to release a 1.2.3 version that does not have this problem very > soon. The same problem occurs in 1.3.0cvs (and 1.3.0pre1 I guess) but > is fixed now in cvs. > > A bit of background: there has been for a long time some code in > LyX configure script that removed the config.cache file (that keep > tracks of tests results) when an error has occured. This allows to > re-run the script once you have made correction without seeing the old > results. However, starting with autoconf 2.5x, the cache file now > defaults to /dev/null (i.e. no cache). Since LyX 1.2.2 was the first > version distributed with an autoconf script generated with autoconf > 2.52, this triggered this awful behaviour. > > Sorry for the problems that this may cause to your systems... > > JMarc >
Re: [Bug 763] lyx2lyx adds gibberish to title
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes: Andre> On Wed, Jan 08, 2003 at 12:57:29PM +0100, Jean-Marc Lasgouttes Andre> wrote: >> Of these, only \topmargin still exist in 1.3.0cvs. Do you mean that >> the others are kept unchanged by lyx2lyx? This would be very wrong. Andre> LyX ignores when reading the file... But it spits an error, doesn't it? Anyway they should be removed by lyx2lyx. >> But indeed the code in buffer.C around line 765 could use >> lex.eatLine() instead of lex.next(). Andre> Would be wrong to read a number and drop the unit No, eatLine reads the full line from cursor position, so it gets all the length. JMarc
Towards LyX 1.2.3 [status update #1]
Hello, It seems that we need to release LyX 1.2.3 right now. Please find appended the list of changes that are in cvs currently. If there is a patch that you really think should be added, please tell me about it. My current plan is to release what we have maybe tomorrow. I anticipate that LyX 1.2.3 will be the last 1.2.x release. JMarc -*- text -*- This file describes what has been done in the preparation of LyX 1.2.3. All comments are welcome. You can find a list of bugs pending for 1.2.3 at URL http://bugzilla.lyx.org/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&&target_milestone=1.2.3 I'd be glad if some of you could take the time to check it out (or fix a bug or two if you are feeling adventurous). Let me recall that all these fixes have been checked in the branch BRANCH-1_2_X, which you can get with the command cvs checkout -d lyx-1_2_x -r BRANCH-1_2_X lyx-devel JMarc What's new == ** Updates - updated italian User Guide ** Bug fixes - fix bug where the configure script would try to delete the /dev/null device when there has been an error in configuration step - fix bug where autosaving would leave a lot of zombie processes after a long time - fix problem where it was not possible to force some values to 0 in custom margins - fix bug where horizontal alignment would be reset automatically to "left" in the tabular dialog. - fix functions word-find-forward and word-find-backward, which meaning was inverted - when the lyx binary was actually a symbolic link, LyX printed annoying debug messages on console; they are debug messages now - fix an installation issue with solaris
Re: Towards LyX 1.2.3 [status update #1]
On Wednesday 08 January 2003 1:02 pm, Jean-Marc Lasgouttes wrote: > - when the lyx binary was actually a symbolic link, LyX printed > annoying debug messages on console; they are debug messages now s/annoying debug/annoying/ or this reads as 'no change' ;-) -- Angus
Re: [Bug 763] lyx2lyx adds gibberish to title
On Wed, Jan 08, 2003 at 01:58:55PM +0100, Jean-Marc Lasgouttes wrote: > Andre> LyX ignores when reading the file... > > But it spits an error, doesn't it? I did not see any. > Anyway they should be removed by lyx2lyx. Probably. How do we learn which option went at what time? > Andre> Would be wrong to read a number and drop the unit > > No, eatLine reads the full line from cursor position, so it gets all > the length. Ah, I thought you meant getString() followed be eatLine() ... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: [Bug 763] lyx2lyx adds gibberish to title
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes: Andre> Probably. Andre> How do we learn which option went at what time? In OLD_CHANGES, I find: since 0.11.24 - changed from vmargin.sty to geometry.sty (Jug) - changed also the configure-stuff from vmargin to geometry (Jug) [...] since 0.11.19 - Inserted handling for Vmargin.sty with new Layout-Page-Menu (Jug) So I guess the old syntax was removed in 0.11.20 and maybe updated in 0.11.24. Reading the lyx 0.12.0 source code will probably reveal how old syntax can be converted. JMarc
Re: lyx2lyx style question
On Wednesday 08 January 2003 12:37, Dekel Tsur wrote: > > > > And ^Q + TAB is not a reasonable solution. > > You need to add > (setq-default indent-tabs-mode t) That did the trick, thanks. -- José Abílio
Re: Remove trailing \r when reading...
> | > ios::text? No. It is a non-standard flag. > | > ... > | > ios::binary otoh is a standard flag and should work everywhere. > | > | So there seems not to be another way then removong those trailing \r's the > | way I'v done. We do need that in all cases where text read from a (binary) > | opened file is presented as-is in a window. > > Then we should not open those files in binary mode. That isn't possible if the drive/directory is mounted in binary mode!!! Mounting *all* drives/dirs in text mode does produce other problems because then (not only with Lyx) reading any (although non-text) file *always* causes a byte translation \n -> \n\r! Claus
Re: Towards LyX 1.2.3 [status update #1]
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
Andre> On Wed, Jan 08, 2003 at 02:38:19PM +0100, Jean-Marc Lasgouttes
Andre> wrote:
>> > "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
>>
Andre> On Wed, Jan 08, 2003 at 02:02:36PM +0100, Jean-Marc Lasgouttes
Andre> wrote:
>> >> - fix functions word-find-forward and word-find-backward, which
>> >> meaning was inverted
>>
Andre> Is 'which' the proper word here?
>> Would it be clearer if I wrote 'which meanings were inverted'?
Andre> No, that was just a language question. In German one would use
Andre> 'deren', which is often translated to 'whose' in English. But
Andre> then, rules in English are different ('whose' only for persons?
Andre> - I really can't remember...)
I think `which' is right here, but my english is far from perfect. But
I also think that the plural is needed. I add lyx-devel in cc to get
advice from real english people.
JMarc
Re: still can't import any latex tabulars
On Tue, Jan 07, 2003 at 06:14:14PM +0100, Juergen Spitzmueller wrote: > Dr. Richard E. Hawkins wrote: > > Yes; I can't even reproduce the order or pattern of blank/nonblank. All > > I have so far is that it seems to be affected by a failed import, abut > > also by using anything on the menu. > Here's what I figured out: > Start LyX: all is ok (5 import options) > if I: > a.) choose a menu command without opening a doc (e.g. TeX information), the > whole "Import" item is greyed out. > b.) open a doc, all but ASCII (as lines, as pars) disappears > > Having done b.), I get all 5 import options back if I move on the menue bar to > the right (from file to help) and back. If I chose a menue command now (open > a dialog), they disappear again. > This is with qt. I've found that if I open a new file, and close it, the 5 options come back. While any file is open, only the ascii options seem available. (though maybe I"m just looking during the wrong part of the menu dance . . . -- Richard E. Hawkins, Asst. Prof. of Economics/"\ ASCII ribbon campaign [EMAIL PROTECTED] Smeal 178 (814) 375-4700 \ / against HTML mail These opinions will not be those of Xand postings. Penn State until it pays my retainer. / \
Re: [PATCH] Re: Bug in tabular dialog
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes: Juergen> Jean-Marc Lasgouttes wrote: The major (multicolumn) bug was a Juergen> new regression in 1.3 Bug 572 is a rather minor annoyance. Juergen> It's not urgent, but I could backport it if requested. >> So we can live without it, then. However, if you do it, I'll apply >> it. Juergen> Here it comes. Please apply the the second patch to 1.3, it Juergen> fixes a stupid typo introduced by /me (with fix for bug 572). I applied both. JMarc
Re: lyx2lyx style question
José Matos <[EMAIL PROTECTED]> writes: | On Wednesday 08 January 2003 11:58, Lars Gullik Bjønnes wrote: | > | > (setq py-indent-offset 8) | > | > should do the trick. | > | > (M-x set-variable RET py-indent-offset RET 8 RET) | | Ok, I already knew that (trial and error), but there is a small (not) | problem. It really inserts 8 spaces and not a tab. André will not like it. M-x set-variable RET indent-tabs-mode RET t RET -- Lgb
Re: Remove trailing \r when reading...
"Claus Hentschel" <[EMAIL PROTECTED]> writes: | > | > ios::text? No. It is a non-standard flag. | > | > ... | > | > ios::binary otoh is a standard flag and should work everywhere. | > | | > | So there seems not to be another way then removong those trailing \r's | the | > | way I'v done. We do need that in all cases where text read from a | (binary) | > | opened file is presented as-is in a window. | > | > Then we should not open those files in binary mode. | | That isn't possible if the drive/directory is mounted in binary mode!!! I'd like to put this down as a configuration problem of cygwin/win drives/dirs etc. If you manage to setup the sustemt the stdlibc++ does not work, we should not cater for it. IMHO of course. | Mounting *all* drives/dirs in text mode does produce other problems because | then (not only with Lyx) reading any (although non-text) file *always* | causes a byte translation \n -> \n\r! Only on binaries that open their filen in the wrong mode. If you want to be portable to win you have to use the ios::binary when not reading a text file (and I know that lyx does not do this as it should) -- Lgb
Re: 1.3.0 stoppers
On Wed, Jan 08, 2003 at 11:55:09AM +0100, Michael Schmitt wrote: > How about bugs #770, #795 ? > Why must they be fixed before 1.3.0 ? Juergen is going to apply a fix for 770. Why must 795 be fixed ? I dont understand why we've added a load of bugs as important now, just as we're getting close to a release. It doesn't suddenly generate fixes especially bug 8 ! regards john -- "CUT IT OUT FACEHEAD" - jeffk
Re: lyx2lyx style question
On Wed, Jan 08, 2003 at 11:32:10AM +0100, Andre Poenitz wrote: > I don't use emacs. This 'default' is unreadable for me unless I set > tabstop==8 (I usually have *shriek* 2), moreover it is a hassle to edit Andre, Andre ... *shakes head* /me runs john -- "CUT IT OUT FACEHEAD" - jeffk
Re: Towards LyX 1.2.3 [status update #1]
On Wed, Jan 08, 2003 at 02:50:06PM +0100, Jean-Marc Lasgouttes wrote: > >> >> - fix functions word-find-forward and word-find-backward, which > >> >> meaning was inverted > > I think `which' is right here, but my english is far from perfect. But > I also think that the plural is needed. I add lyx-devel in cc to get > advice from real english people. "whose" reads better, but I don't have a justification. regards john -- "CUT IT OUT FACEHEAD" - jeffk
Re: Towards LyX 1.2.3 [status update #1]
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> On Wed, Jan 08, 2003 at 02:50:06PM +0100, Jean-Marc Lasgouttes John> wrote: >> >> >> - fix functions word-find-forward and word-find-backward, >> which >> >> meaning was inverted >> >> I think `which' is right here, but my english is far from perfect. >> But I also think that the plural is needed. I add lyx-devel in cc >> to get advice from real english people. John> "whose" reads better, but I don't have a justification. That's a good enough justification. And should I use a plural for `meaning'? JMarc
Re: Towards LyX 1.2.3 [status update #1]
On Wednesday 08 January 2003 2:24 pm, Jean-Marc Lasgouttes wrote: > > "John" == John Levon <[EMAIL PROTECTED]> writes: > > John> On Wed, Jan 08, 2003 at 02:50:06PM +0100, Jean-Marc Lasgouttes > > John> wrote: > >> >> >> - fix functions word-find-forward and word-find-backward, > >> > >> which >> >> meaning was inverted > >> > >> I think `which' is right here, but my english is far from perfect. > >> But I also think that the plural is needed. I add lyx-devel in cc > >> to get advice from real english people. > > John> "whose" reads better, but I don't have a justification. > > That's a good enough justification. And should I use a plural for > `meaning'? Yes. 'whose' refers to both functions so you need fix functions word-find-forward and word-find-backward whose meanings were inverted. or perhaps more elegantly fix functions word-find-forward and word-find-backward, the meanings of which were inverted. -- Angus
Re: Towards LyX 1.2.3 [status update #1]
On Wed, Jan 08, 2003 at 03:24:48PM +0100, Jean-Marc Lasgouttes wrote: > John> "whose" reads better, but I don't have a justification. > > That's a good enough justification. And should I use a plural for > `meaning'? It would be singular in German if that helps ;-] Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Bugzilla Bug 436
Hi Dekel, this bug is trigered by this piece \begin_inset Figure size 288 178 file graphs/gridNtree-confReal2.eps subcaption flags 9 \end_inset and you search it as subcaptionText = get_value(lines, "subcaption", i+1, j) the problem is that get_value always expect to have a value to catch a value there def get_value(lines, token, start, end = 0): i = find_token2(lines, token, start, end) if i == -1: return "" return string.split(lines[i])[1] Notice the [1] we are assuming that there are always 2 member. Is it ok to test for the presence of the second and if not present to return the empty string? I think that this is the right thing to do in get_value, at least. Closing this bug there aren't anymore lyx2lyx open bugs. :-) -- José Abílio
Re: [WARNING] Bad bug in lyx 1.2.2 configure script
On most Linux-distros you won't be able to properly boot since the root filesystem ist usually mounted read-only. Remounting it as read-write will fail when /dev/null is not present (at least in SuSE). So start you system in single user mode, log in as root on the console, do a 'mount -o remount /' an then recreate /dev/null. then you can reboot or use init to go to your preferred runlevel. This bug bit me couple of times when I set up my new Laptop. I never suspected the configure-script and chased the bug for a week. Michael [EMAIL PROTECTED] wrote: For anyone who needs to recover, On my RedHat 8.0 machine /dev/null is a character device with major mode 1, minor mode 3 and mode a+rw - use mknod to make a new one. On Solaris, it's a soft link to /devices/pseudo/mm@0:null - use ln to recreate. Bill. -- Michael Abshoff - MRB - Universität Dortmund - Telefon 755-3463 (intern) Where do you want to RTFM today?
Re: [WARNING] Bad bug in lyx 1.2.2 configure script
On Wed, Jan 08, 2003 at 12:32:24PM +0100, Jean-Marc Lasgouttes wrote: > > There is a very nasty bug in LyX 1.2.2 configure script: if you > configure as root and an error occurs during the configure step > (e.g. xforms not found), the the script will delete the /dev/null > device. This undoubtly leads to very bad effect, since /dev/null is a > very basic device in any unix system. Oh, /dev/null isn't much of a loss; I've never been able to get anything out of it, anyway :) hawk -- Richard E. Hawkins, Asst. Prof. of Economics/"\ ASCII ribbon campaign [EMAIL PROTECTED] Smeal 178 (814) 375-4700 \ / against HTML mail These opinions will not be those of Xand postings. Penn State until it pays my retainer. / \
Re: Towards LyX 1.2.3 [status update #1]
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes: Andre> On Wed, Jan 08, 2003 at 03:24:48PM +0100, Jean-Marc Lasgouttes Andre> wrote: John> "whose" reads better, but I don't have a justification. >> That's a good enough justification. And should I use a plural for >> `meaning'? Andre> It would be singular in German if that helps ;-] Usually, German does not help me much ;) JMarc
Re: Towards LyX 1.2.3 [status update #1]
On Wed, Jan 08, 2003 at 03:38:39PM +0100, Jean-Marc Lasgouttes wrote: > Andre> It would be singular in German if that helps ;-] > > Usually, German does not help me much ;) Seems to be a local thing... Actually, when I think about the example I believe I was wrong and plural is correct... Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Bugzilla Bug 436
On Wed, Jan 08, 2003 at 02:40:18PM +, Jos? Matos wrote: > the problem is that get_value always expect to have a value to catch a value > there > > def get_value(lines, token, start, end = 0): > i = find_token2(lines, token, start, end) > if i == -1: > return "" > return string.split(lines[i])[1] > > Notice the [1] we are assuming that there are always 2 member. Is it ok to > test for the presence of the second and if not present to return the empty > string? Yes, I think it would be fine. > > Closing this bug there aren't anymore lyx2lyx open bugs. :-) Did you fix the code to work with nested tabulars (as we discussed some time ago) ?
Make error in sourcedoc?
When I make lyx, it ends with make[1]: Leaving directory `/local/lasgoutt/qtbuild/src' Making all in sourcedoc make[1]: Entering directory `/local/lasgoutt/qtbuild/sourcedoc' make[1]: *** No rule to make target `all'. Stop. make[1]: Leaving directory `/local/lasgoutt/qtbuild/sourcedoc' make: *** [all-recursive] Error 1 There is indeed no Makefile in sourcedoc. Why is that? JMarc
Re: Make error in sourcedoc?
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: | When I make lyx, it ends with | | make[1]: Leaving directory `/local/lasgoutt/qtbuild/src' | Making all in sourcedoc | make[1]: Entering directory `/local/lasgoutt/qtbuild/sourcedoc' | make[1]: *** No rule to make target `all'. Stop. | make[1]: Leaving directory `/local/lasgoutt/qtbuild/sourcedoc' | make: *** [all-recursive] Error 1 | | There is indeed no Makefile in sourcedoc. Why is that? srcdir != builddir (or ==) issue? configure should certainly create a makefile. -- Lgb
Re: 1.3.0 stoppers
John Levon <[EMAIL PROTECTED]> writes: | On Wed, Jan 08, 2003 at 11:55:09AM +0100, Michael Schmitt wrote: | | > How about bugs #770, #795 ? | > | | Why must they be fixed before 1.3.0 ? Juergen is going to apply a fix | for 770. Why must 795 be fixed ? | | I dont understand why we've added a load of bugs as important now, just | as we're getting close to a release. It doesn't suddenly generate fixes | | | especially bug 8 ! It seems that I use not use "target milestone" in the same way you do. I use it both as a way to say that "this must be fixed", but that it must be fixed is also communicated with the severity and the priority. The target milestone also says when bugs to look into before release, if it is decided that this problem is minor etc. The target milestone can be removed. -- Lgb
Re: Make error in sourcedoc?
> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes: Lars> Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: | Lars> When I make lyx, it ends with Lars> | Lars> | make[1]: Leaving directory `/local/lasgoutt/qtbuild/src' | Lars> Making all in sourcedoc | make[1]: Entering directory Lars> `/local/lasgoutt/qtbuild/sourcedoc' | make[1]: *** No rule to Lars> make target `all'. Stop. | make[1]: Leaving directory Lars> `/local/lasgoutt/qtbuild/sourcedoc' | make: *** [all-recursive] Lars> Error 1 Lars> | Lars> | There is indeed no Makefile in sourcedoc. Why is that? Lars> srcdir != builddir (or ==) issue? I have srcdir != builddir Lars> configure should certainly create a makefile. Yes, it seems it should... JMarc
Re: Bugzilla Bug 436
On Wednesday 08 January 2003 14:46, Dekel Tsur wrote: > > Notice the [1] we are assuming that there are always 2 member. Is it ok > > to test for the presence of the second and if not present to return the > > empty string? > > Yes, I think it would be fine. > > > Closing this bug there aren't anymore lyx2lyx open bugs. :-) Just to be in the safe side: * in version 218 we have: begin_inset Figure size 285 396 file figures/article-fig5.ps width 3 48 flags 9 \end_inset * in version 220 we have: \begin_inset Graphics FormatVersion 1 filename figures/article-fig5.ps display monochrome size_type 1 width 48text% rotateOrigin leftBaseline lyxsize_type 1 lyxwidth 285pt lyxheight 396pt \end_inset * in version 221 we have: \begin_inset Graphics filename figures/article-fig5.ps display monochrome width 48text% \end_inset Is this ok? If yes, I will commit the fix and close the bug. > Did you fix the code to work with nested tabulars (as we discussed some > time ago) ? Oops, it looks like I forgot it. I will look into it again. -- José Abílio
Re: 1.3.0 stoppers
On Wed, Jan 08, 2003 at 04:08:11PM +0100, Lars Gullik Bj?nnes wrote: > | especially bug 8 ! > > The target milestone also says when bugs to look into before release, > if it is decided that this problem is minor etc. The target milestone > can be removed. OK fine. Then I would say that bug 8 is very old, reasonably obscure, and not dangerous much. So it can have the target milestone removed :) regards john -- "CUT IT OUT FACEHEAD" - jeffk
Re: Towards LyX 1.2.3 [status update #1]
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes: Angus> On Wednesday 08 January 2003 1:02 pm, Jean-Marc Lasgouttes Angus> wrote: >> - when the lyx binary was actually a symbolic link, LyX printed >> annoying debug messages on console; they are debug messages now Angus> s/annoying debug/annoying/ or this reads as 'no change' ;-) -- Well spotted. JMarc
Re: Bugzilla Bug 436
On Wed, Jan 08, 2003 at 03:24:55PM +, Jos? Matos wrote: > * in version 218 we have: > * in version 220 we have: > * in version 221 we have: > > Is this ok? If yes, I will commit the fix and close the bug. I don't really remember the exact differences between version. The change in get_value seems safe, so you can commit the fix.
Re: Towards LyX 1.2.3 [status update #1]
On Wednesday 08 January 2003 3:37 pm, Jean-Marc Lasgouttes wrote: > > "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes: > > Angus> On Wednesday 08 January 2003 1:02 pm, Jean-Marc Lasgouttes > > Angus> wrote: > >> - when the lyx binary was actually a symbolic link, LyX printed > >> annoying debug messages on console; they are debug messages now > > Angus> s/annoying debug/annoying/ or this reads as 'no change' ;-) -- > > Well spotted. > > JMarc Since we're being finicky: - when the lyx binary was invoked through a symbolic link, LyX would print annoying messages on the console; they are now output only when debugging. -- Angus
[PATCH] Re: still can't import any latex tabulars
> "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes:
Juergen> Dr. Richard E. Hawkins wrote:
>> Yes; I can't even reproduce the order or pattern of blank/nonblank.
>> All I have so far is that it seems to be affected by a failed
>> import, abut also by using anything on the menu.
Juergen> Here's what I figured out:
Juergen> Start LyX: all is ok (5 import options) if I: a.) choose a
Juergen> menu command without opening a doc (e.g. TeX information),
Juergen> the whole "Import" item is greyed out. b.) open a doc, all
Juergen> but ASCII (as lines, as pars) disappears
Juergen> Having done b.), I get all 5 import options back if I move on
Juergen> the menue bar to the right (from file to help) and back. If I
Juergen> chose a menue command now (open a dialog), they disappear
Juergen> again. This is with qt.
This happens with xforms too. Actually it is not needed to open some
dialog, just moving around in menus is needed.
After some investigation, it turns out that the following patch cures
the problem. I am not sure why this code caused these particular
symptoms, but it was broken anyway.
Lars, since the "(3)" is your doing, feel free to replace that with any
C++ally correct construct that may make you feel good.
JMarc
? config.h.in
Index: src/ChangeLog
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/ChangeLog,v
retrieving revision 1.1007
diff -u -p -r1.1007 ChangeLog
--- src/ChangeLog 8 Jan 2003 09:54:18 - 1.1007
+++ src/ChangeLog 8 Jan 2003 15:55:51 -
@@ -1,3 +1,8 @@
+2003-01-08 Jean-Marc Lasgouttes <[EMAIL PROTECTED]>
+
+ * importer.C (Loaders): do not preallocated 3 elements in the
+ vector, since one ends up with 6 elements otherwise
+
2002-12-20 Jean-Marc Lasgouttes <[EMAIL PROTECTED]>
* DepTable.C (write): write the file name as last element of the
Index: src/importer.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/importer.C,v
retrieving revision 1.21
diff -u -p -r1.21 importer.C
--- src/importer.C 24 Nov 2002 15:19:54 - 1.21
+++ src/importer.C 8 Jan 2003 15:55:51 -
@@ -116,7 +116,7 @@ vector const Importer::G
vector const Importer::Loaders()
{
- vector v(3);
+ vector v;
v.push_back("lyx");
v.push_back("text");
v.push_back("textparagraph");
[patch] remove unconditional error message
Ok?
Andre'
Index: formulabase.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/mathed/formulabase.C,v
retrieving revision 1.230
diff -u -p -r1.230 formulabase.C
--- formulabase.C 12 Dec 2002 09:42:35 - 1.230
+++ formulabase.C 8 Jan 2003 16:09:55 -
@@ -234,7 +234,7 @@ void InsetFormulaBase::getCursorPos(Buff
void InsetFormulaBase::toggleInsetCursor(BufferView * bv)
{
if (!mathcursor) {
- lyxerr << "toggleInsetCursor impossible" << endl;
+ lyxerr[Debug::MATHED] << "toggleInsetCursor impossible" << endl;
return;
}
//lyxerr << "toggleInsetCursor: " << isCursorVisible() << endl;
@@ -356,7 +356,7 @@ Inset::RESULT InsetFormulaBase::lfunMous
//lyxerr << "lfunMousePress: buttons: " << cmd.button() << endl;
if (!mathcursor || mathcursor->formula() != this) {
- lyxerr << "re-create cursor" << endl;
+ lyxerr[Debug::MATHED] << "re-create cursor" << endl;
releaseMathCursor(bv);
mathcursor = new MathCursor(this, cmd.x == 0);
metrics(bv);
alt-r doesn't replace in spellcheck dialog
edit the replacement line, hit alt-r, and nothing happens yesterday's cvs. hawk -- Richard E. Hawkins, Asst. Prof. of Economics/"\ ASCII ribbon campaign [EMAIL PROTECTED] Smeal 178 (814) 375-4700 \ / against HTML mail These opinions will not be those of Xand postings. Penn State until it pays my retainer. / \
Re: Compiler warnings
> "Michael" == Michael Schmitt <[EMAIL PROTECTED]> writes: Michael> Hi, please find the current list of compiler warnings below Michael> (gcc-3.2, xforms frontend). I think a few of them are easily Michael> fixable. Michael> xscreen.C: In member function Michael> `virtual void XScreen::draw(LyXText*, BufferView*, unsigned Michael> int)': xscreen.C:217: warning: comparison between signed and Michael> unsigned integer expressions xscreen.C:217: warning: Michael> comparison between signed and unsigned integer expressions This one is weird: the code reads // is any optimization possible? if ((y - old_first) < owner_.workHeight() && (old_first - y) < owner_.workHeight()) is there a reason why the test is not // is any optimization possible? if (std::abs(y - old_first) < owner_.workHeight()) Or maybe the test is not what it intended to be. Michael> MenuBackend.C: In function `void Michael> ::expandToc(Menu&, const Buffer*)': Michael> MenuBackend.C:489: warning: unused variable `const Michael> FloatList&floats' Fixed. Michael> paragraph_pimpl.C: In member function `void Michael> Paragraph::Pimpl::simpleTeXBlanks(std::ostream&, TexRow&, Michael> int, int&, const LyXFont&, const LyXLayout&)': Michael> paragraph_pimpl.C:219: warning: comparison between signed and Michael> unsigned integer expressions Fixed. Michael> tabular.C: In member function `int LyXTabular::latex(const Michael> Buffer*, std::ostream&, bool, bool) const': tabular.C:2259: Michael> warning: enumeration value `LYX_ALIGN_NONE' not handled in Michael> switch tabular.C:2259: warning: enumeration value Michael> `LYX_ALIGN_BLOCK' not handled in switch tabular.C:2259: Michael> warning: enumeration value `LYX_ALIGN_LAYOUT' not handled in Michael> switch tabular.C:2259: warning: enumeration value Michael> `LYX_ALIGN_SPECIAL' not handled in switch Fixed. Note that most of the #warnings are not as scary as they seem... (but of course it would be nice to resolve a few of them). JMarc
Re: [PATCH] Re: still can't import any latex tabulars
On Wednesday 08 January 2003 4:00 pm, Jean-Marc Lasgouttes wrote:
> Lars, since the "(3)" is your doing, feel free to replace that with any
> C++ally correct construct that may make you feel good.
>
> JMarc
Why not either of:
#include
#include
using std::vector;
using std::string;
void foo()
{
char const * const cstr[] = { "lyx", "text", "textparagraph" };
size_t const csize = sizeof(cstr) / sizeof(char*);
vector const v(cstr, cstr + csize);
}
void bar()
{
vector v(3);
v[0] = "lyx";
v[1] = "text";
v[2] = "textparagraph";
}
--
Angus
Re: Compiler warnings
On Wed, Jan 08, 2003 at 05:26:48PM +0100, Jean-Marc Lasgouttes wrote: > This one is weird: the code reads > > // is any optimization possible? > if ((y - old_first) < owner_.workHeight() > && (old_first - y) < owner_.workHeight()) > > is there a reason why the test is not > > // is any optimization possible? > if (std::abs(y - old_first) < owner_.workHeight()) > > Or maybe the test is not what it intended to be. Please be very careful. My attempt to fix this in qscreen.C led to several problems. I didn't try very hard though. regards john -- "CUT IT OUT FACEHEAD" - jeffk
Re: Towards LyX 1.2.3 [status update #1]
Am Mittwoch, 8. Januar 2003 14:02 schrieb Jean-Marc Lasgouttes:
> Hello,
>
> It seems that we need to release LyX 1.2.3 right now. Please find
> appended the list of changes that are in cvs currently. If there is a
> patch that you really think should be added, please tell me about it.
i have one attached
it does the following:
- add a desktop file
- correct the spec file to use License: instead of Copyright: (rpm-4.x)
- and use a commonly used (in RH) and better fitting Group:
- and let rpm compress the man pages
- extend and fix dependencies, i.e. Requires: for libraries are not needed
- and depend on xforms >= 1.0
- update lyx.rpm.README to reflect the changes
- use rpmbuild for make rpmdist (rpm 4.x)
so it should work for rh62 and above
--
ronny
--- development/lyx.desktop.in Wed Jan 8 14:46:19 2003
+++ development/lyx.desktop.in Wed Jan 8 14:46:19 2003
@@ -0,0 +1,8 @@
+[Desktop Entry]
+Name=LyX
+Comment=A WYSIWYM document processor
+Icon=@datadir@/lyx/images/lyx.xpm
+Exec=lyx
+Terminal=0
+Type=Application
+EOF
Index: Makefile.am
===
RCS file: /cvs/lyx/lyx-devel/Makefile.am,v
retrieving revision 1.37
diff -u -r1.37 Makefile.am
--- Makefile.am 2002/04/26 10:32:22 1.37
+++ Makefile.am 2003/01/08 14:19:19
@@ -5,6 +5,7 @@
MAINTAINERCLEANFILES += $(srcdir)/aclocal.m4 \
$(srcdir)/configure $(srcdir)/development/lyx.spec \
+ $(srcdir)/development/lyx.desktop \
$(srcdir)/acinclude.m4
SUBDIRS = config intl po sigc++ boost src lib
@@ -32,7 +33,7 @@
rpmdist: dist
ln -s lib/images/lyx.xpm . ; \
- rpm -ta $(PACKAGE)-$(VERSION).tar.gz ; saved_status=$$?; \
+ rpmbuild -ta $(PACKAGE)-$(VERSION).tar.gz ; saved_status=$$?; \
rm lyx.xpm; exit $$saved_status
bindist:
Index: configure.in
===
RCS file: /cvs/lyx/lyx-devel/Attic/configure.in,v
retrieving revision 1.120.2.16
diff -u -r1.120.2.16 configure.in
--- configure.in 2002/12/16 15:07:44 1.120.2.16
+++ configure.in 2003/01/08 14:19:24
@@ -363,6 +363,7 @@
boost/Makefile \
config/Makefile \
development/lyx.spec \
+ development/lyx.desktop \
lib/Makefile \
intl/Makefile \
po/Makefile.in \
Index: development/lyx.spec.in
===
RCS file: /cvs/lyx/lyx-devel/development/lyx.spec.in,v
retrieving revision 1.14.2.1
diff -u -r1.14.2.1 lyx.spec.in
--- development/lyx.spec.in 2002/07/26 12:17:20 1.14.2.1
+++ development/lyx.spec.in 2003/01/08 14:19:25
@@ -2,16 +2,20 @@
Name: @PACKAGE@
Version: @VERSION@
Release: 1
-Copyright: see COPYING file
-Group: X11/Editors
+License: see COPYING file
+Group: Applications/Publishing
Url: http://www.lyx.org/
Packager: Kayvan A. Sylvan <[EMAIL PROTECTED]>
Source: ftp://ftp.lyx.org/pub/lyx/stable/lyx-%{PACKAGE_VERSION}.tar.gz
BuildRoot: %{_tmppath}/%{name}-root
Icon: lyx.xpm
Prefix: /usr
-Requires: xforms >= 0.88, tetex-xdvi, tetex, tetex-latex
+BuildRequires: xforms-devel >= 1.0
+%{!?_without_pspell:BuildRequires: pspell-devel}
+Requires: tetex-fonts, tetex-xdvi, tetex, tetex-latex
+%{!?_without_pspell:Requires: aspell}
Obsoletes: tetex-lyx
+PreReq: /usr/bin/perl
%description
LyX is a modern approach to writing documents which breaks with the
@@ -40,15 +44,17 @@
CXXFLAGS="$RPM_OPT_FLAGS" CFLAGS="$RPM_OPT_FLAGS" \
./configure --prefix=%{_prefix} --mandir=%{_mandir} \
--bindir=%{_bindir} --datadir=%{_datadir} \
- --without-warnings $ISTRING
+ --without-warnings \
+ %{!?_without_pspell:--with-pspell} \
+ $ISTRING
make
%install
unset LINGUAS || true
rm -rf ${RPM_BUILD_ROOT}
install -d -m 755 ${RPM_BUILD_ROOT}
-make DESTDIR=${RPM_BUILD_ROOT} install
-gzip -f9 ${RPM_BUILD_ROOT}%{_mandir}/man?/*
+# transform shouldn't be needed, but something is doing strange things
+make DESTDIR=${RPM_BUILD_ROOT} transform='' install
#
# Set up the lyx-specific class files where TeX can see them
@@ -64,6 +70,12 @@
cp -a lib/images/lyx.xpm ${RPM_BUILD_ROOT}%{_datadir}/lyx/images/
cp lib/reLyX/README README.reLyX
+#
+# Desktop file
+#
+mkdir -p ${RPM_BUILD_ROOT}/etc/X11/applnk/Applications
+cp development/lyx.desktop ${RPM_BUILD_ROOT}/etc/X11/applnk/Applications
+
%clean
rm -rf ${RPM_BUILD_ROOT}
@@ -71,6 +83,9 @@
# Make TeX understand where LyX-specific packages are
texhash
+# Register fonts
+[[ -n "$(which chkfontpath 2>/dev/null)" ]] && chkfontpath -a %{_datadir}/%{name}/xfonts
+
# Before configuring lyx for the local system
# PATH needs to be imported
if [ -f /etc/profile ]; then
@@ -95,11 +110,15 @@
#
texhash
+# Unregister fonts
+[[ -n "$(which chkfontpath 2>/dev/null)" ]] && chkfontpath -r %{_datadir}/%{name}/xfonts
+
%files
%defattr(-,root,root)
%doc ABOUT-NLS ANNOUNCE COPYING
%doc README UPGRADING ChangeLog NEWS
%doc lib/CREDITS README.reLyX
+/etc/X11/applnk/Applications/lyx.desktop
%{_bindir}/*
%{_mandir}/man?/*
%{_datadi
Re: [PATCH] Re: still can't import any latex tabulars
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: | > "Juergen" == Juergen Spitzmueller <[EMAIL PROTECTED]> writes: | | Juergen> Dr. Richard E. Hawkins wrote: | >> Yes; I can't even reproduce the order or pattern of blank/nonblank. | >> All I have so far is that it seems to be affected by a failed | >> import, abut also by using anything on the menu. | | Juergen> Here's what I figured out: | | Juergen> Start LyX: all is ok (5 import options) if I: a.) choose a | Juergen> menu command without opening a doc (e.g. TeX information), | Juergen> the whole "Import" item is greyed out. b.) open a doc, all | Juergen> but ASCII (as lines, as pars) disappears | | Juergen> Having done b.), I get all 5 import options back if I move on | Juergen> the menue bar to the right (from file to help) and back. If I | Juergen> chose a menue command now (open a dialog), they disappear | Juergen> again. This is with qt. | | This happens with xforms too. Actually it is not needed to open some | dialog, just moving around in menus is needed. | | After some investigation, it turns out that the following patch cures | the problem. I am not sure why this code caused these particular | symptoms, but it was broken anyway. | | Lars, since the "(3)" is your doing, feel free to replace that with any | C++ally correct construct that may make you feel good. What was in place before I changed it? anyway it is fine as you changed it. -- Lgb
Re: [patch] remove unconditional error message
Andre Poenitz <[EMAIL PROTECTED]> writes: | Ok? sure. -- Lgb
Re: Compiler warnings
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> On Wed, Jan 08, 2003 at 05:26:48PM +0100, Jean-Marc Lasgouttes John> wrote: >> This one is weird: the code reads >> >> // is any optimization possible? if ((y - old_first) < >> owner_.workHeight() && (old_first - y) < owner_.workHeight()) >> >> is there a reason why the test is not >> >> // is any optimization possible? if (std::abs(y - old_first) < >> owner_.workHeight()) >> >> Or maybe the test is not what it intended to be. John> Please be very careful. My attempt to fix this in qscreen.C led John> to several problems. That's why I posted here. However, I fail to see how the two forms I posted may have any difference. Am I blind? JMarc
Re: [PATCH] Re: still can't import any latex tabulars
> "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes: Lars> What was in place before I changed it? anyway it is fine as you Lars> changed it. You just added the (3), probably because you thought that the size was known in advance. Unfortunately the effect was quite different. I'll apply the patch. JMarc
Re: [PATCH] Re: cut table cells doesn't work in float
> "Jean-Marc" == Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes: > "Edwin" == Edwin Leuven <[EMAIL PROTECTED]> writes: Edwin> in the qt frontend that is. The minibuffer says "Command Edwin> disabled" when I try to cut a range of cells of a tabular in a Edwin> table float. Cutting cells when the tabular is not in a float Edwin> works just fine. Edwin> Is this a qt specific problem? Jean-Marc> No, it is general. Here is a patch. Jean-Marc> Lars, is that OK? Lars? JMarc
