[Libreoffice] newish to git: howto recover from bad merge?
Hullo List, It must be something simple, but I'm having a devil of a time recovering from a bad merge. Given my transcript below, how do I recover? It usually happens after I've committed a change to my local repository, that I then sent in as a patch. That patch got applied with a slight modification, and then the conflict. - $ cd .../rawbuild/binfilter $ git pull -r First, rewinding head to replay your work on top of it... Applying: EasyHack: RTL macro from createFromAscii Using index info to reconstruct a base tree... Falling back to patching base and 3-way merge... Auto-merging binfilter/bf_svx/source/items/svx_xmlcnitm.cxx CONFLICT (content): Merge conflict in binfilter/bf_svx/source/items/svx_xmlcnitm.cxx Failed to merge in the changes. Patch failed at 0001 EasyHack: RTL macro from createFromAscii When you have resolved this problem run git rebase --continue. If you would prefer to skip this patch, instead run git rebase --skip. To restore the original branch and stop rebasing run git rebase --abort. $ vim bf_svx/source/items/svx_xmlcnitm.cxx # AS NORMAL, HERE I MANUALLY FIX THE BAD MERGE $ git status # Not currently on any branch. # Unmerged paths: # (use git reset HEAD file... to unstage) # (use git add/rm file... as appropriate to mark resolution) # # both modified: bf_svx/source/items/svx_xmlcnitm.cxx # no changes added to commit (use git add and/or git commit -a) $ git add bf_svx/source/items/svx_xmlcnitm.cxx $ git commit # Not currently on any branch. nothing to commit (working directory clean) $ git rebase --continue Applying: EasyHack: RTL macro from createFromAscii No changes - did you forget to use 'git add'? When you have resolved this problem run git rebase --continue. If you would prefer to skip this patch, instead run git rebase --skip. To restore the original branch and stop rebasing run git rebase --abort. - Any pointers for the uninitiated would be awesome. Thanks! Kevin ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PUSHED] Re: tools/ RTL easy hack
On Thu, 2010-11-18 at 17:43 -0500, Kevin Hunter wrote: Hiya List, Here's an RTL easy hack for tools/ . Looks good, all pushed. Thanks for this. C. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PUSHED] Re: EasyHack: svtools/ RTL macro conversion
On Thu, 2010-11-18 at 18:50 -0500, Kevin Hunter wrote: Hullo List, An easy hack for RTL conversion strings, against svtools/ . Excellent, looks good. Pushed this. I simplified the syntax of a few of the constructor here and there on the way. C. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PUSHED] Re: Easy hack: removed commented-out code
On Thu, 2010-11-18 at 18:06 -0500, Kevin Hunter wrote: Hullo List, Another easy hack of cluttering comments removal. Looks good to me, pushed. Thanks for this. C. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PUSHED] Re: RTL easy hack, look over please
On Thu, 2010-11-18 at 17:49 -0500, Kevin Hunter wrote: Hi List, Here's an RTL easy hack, that affects a macro. I believe it okay because it affects strings created like this: string = prefix Something else; but through a macro. Please look over. Yeah, the original would only work if m was a literal, so the new one is equivalent. Pushed. C. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] use dbglevel=1 for --enable-debug builds?
On Thu, 2010-11-18 at 13:01 +0100, David Tardon wrote: Hi all, currently, configuring with --enable-debug means that dbglevel is set to 2, which in turns means that OSL_DEBUG_LEVEL is set to 2. This enables all the OSL_ debugging stuff there is (well, maybe there is some really rare stuff that is only enabled for OSL_DEBUG_LEVEL 2, but I don't know about any), which spews unbelievable amount of stuff on stdout. If we used dbglevel=1 as default, only assertions would be enabled and the output would be much more manageable (practically nonexistent in most cases, from my experience). I never use --enable-debug ( some my opinion aint worth much ) but the above seems more sensible that adding even more options Noel ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] feature/rip-build-repo - branch for removal of the need of the 'build' repo
Hi Sebastian, On 2010-11-19 at 08:41 +0100, Sebastian Spaeth wrote: In case you are interested to try to build without the 'build' repo (the rawbuild/ way), you can try to get the feature/rip-build-repo branch which I hope is going to become the 'official' way of building soon ;-) git clone git://anongit.freedesktop.org/libreoffice/bootstrap libo cd libo/ git checkout -b rip-build-repo origin/rip-build-repo git checkout -b rip-build-repo origin/feature/rip-build-repo ./autogen.sh [flags that you normally use in rawbuild/] ./g clone BTW, this doesn't work right yet ? Or is it some git failure ./g clone = artwork = Initialized empty Git repository in /media/lo/artwork/.git/ fatal: The remote end hung up unexpectedly ./g: line 202: cd: /media/lo/clone/artwork: No such file or directory Sorry for that, fixed now. Regards, Kendy ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] newish to git: howto recover from bad merge?
Hi, On Fri, 19 Nov 2010 03:11:51 -0500, Kevin Hunter hunt...@earlham.edu wrote: $ git commit # Not currently on any branch. nothing to commit (working directory clean) $ git rebase --continue Applying: EasyHack: RTL macro from createFromAscii No changes - did you forget to use 'git add'? When you have resolved this problem run git rebase --continue. If you would prefer to skip this patch, instead run git rebase --skip. To restore the original branch and stop rebasing run git rebase --abort. - It would be helpful to retrospect what happened as follows. Assume that you are on the master branch in the git directory like: A--*B* [master] where both A and B are changesets in the history and *B* means B is HEAD. Then you have committed a changeset as Y to master in order to make a patch for that: *Y* [master] / A--B which is all green so far. But the upstream introduced C which resembles and thus ironically conflicts Y. Then running `git pull -r` can *not* result in *Y'* [master] / A--B--C [remotes/origin/master] as you want, it fails merging after fetching and making a temporary branch: Y [master] / A--*B* [(no branch)] \ C [remotes/origin/master] So `git add` has no luck. You can check where you are by `git branch`, if it looks like $ git branch * (no branch) master then the above guess matters. A workaround for the situation might be: $ git reset --hard # nuke dirty status $ git checkout master $ git merge origin/master $ vi bf_svx/source/items/svx_xmlcnitm.cxx # resolve the conflicts $ git add bf_svx/source/items/svx_xmlcnitm.cxx # add resolved $ git commit HYH. Cheers, -- Takeshi Abe ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] installation and package for Linux SYS distro sucessful
I reclaimed about problems with the installer. Meanwhile, after many work by hand, I get it installed (the most complete installation I got, was with 'make install', not with 'bin/ooinstall'). The build program needs to be improved very. libreoffice runs nice on my computer, appearently without problems, even the quickstart program too, and the program opens fast. I dont get it compiled with the --with-lang option, because of the error what I reclaimed earlier today. So I currently have it only in english. The installer should have the possibility to compile each module (each .tar.bz2 source file) separately, specially language packages. I made a .tgz package, for the SYS distro, what very likely runs also on Slackware etc. Currently its rsync-ing to the mirror ftp://ftp5.gwdg.de/pub/linux/install/sys With this, the distro changed from Oo to libO. I hope now, that you bring forward good libO . w.landgraf --- Professional hosting for everyone - http://www.host.ru ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] installation and package for Linux SYS distro sucessful
On Fri, 19 Nov 2010 07:04:56 -0400, werner wrote: I reclaimed about problems with the installer. Meanwhile, after many work by hand, I get it installed (the most complete installation I got, was with 'make install', not with 'bin/ooinstall'). The build program needs to be improved very. Glad to see you managed to install it finally. We will be the first to agree that the current build/make install system is far from optimal. But it has grown for 10 years as part of OpenOffice.org and although we would like to change it to something more standard rather today than tomorrow, it is an incredibly huge and intertangled mess of code which is not easy to change. We welcome any help to get the build system sane, but as I said, it is not easy and cannot be fixed within a day. libreoffice runs nice on my computer, appearently without problems, even the quickstart program too, and the program opens fast. Good that it seems to work for you. I dont get it compiled with the --with-lang option, because of the error what I reclaimed earlier today. So I currently have it only in english. The installer should have the possibility to compile each module (each .tar.bz2 source file) separately, specially language packages. Yes, it should. AFAIK, and I am no expert in packaging, debian at least manages to get independent language .deb packages, so it must be possible already today. I made a .tgz package, for the SYS distro, what very likely runs also on Slackware etc. Currently its rsync-ing to the mirror ftp://ftp5.gwdg.de/pub/linux/install/sys Great to see libO spreading. With this, the distro changed from Oo to libO. \o/ I hope now, that you bring forward good libO . We'l do our best. But it is 10m lines of code and changes are not easy. Help is always welcome Sebastian pgpK9yKYr4mjk.pgp Description: PGP signature ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] installation and package for Linux SYS distro sucessful
On Fri, 2010-11-19 at 07:04 -0400, werner wrote: I dont get it compiled with the --with-lang option, because of the error what I reclaimed earlier today. If you use --with-lang after configuring once without it you need to do the ./download step again (because download pulls down the i10n module only after with-lang has been used) C. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] [PUSHED] More clean code at writer [source/core/{docnode, draw, edit}]
On Thu, Nov 18, 2010 at 12:12:41PM -0800, Kayo Hamid wrote: Sorry ;-) Hello, sending this for review. I think that cleaniness is better than the previous. revol_ Pushed. D. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] [PATCH] Install to the correct directory based on PREFIX
Hi Robert, thanks for the patch. It is committed as http://cgit.freedesktop.org/libreoffice/build/commit/?id=0cf0c0db05295aa1cc1b7728183b9cbc57e37408 Also I have cherry picked it for the libreoffice-3-3 branch. Note that I added back the $OODESTDIR. It is needed when you build RPMs. In this case, we need to install the files into extra prefix, e.g. /var/tmp/XXX to do not affect the system stuff. See also http://rpm5.org/docs/api/buildroot.html The variable is empty during normal installation, so it should not cause any harm. Best Regards, Petr ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] 5 times LGPL installed
On Wed, 2010-11-17 at 16:18 +0100, Robert Nagy wrote: Well remove it from the menu then. It needs to be there; go read the LGPLv3 :-) Personally, I love the LGPL - so, having it five times could be a feature ;-) Having said that it does seem somewhat excessive. Having it simply twice - once as plain-text, once as odt seems much more reasonable. I'll add an easy hack to remove the other versions, unless someone here wants to do it. Thanks, Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] newish to git: howto recover from bad merge?
Hi Kevin, On Fri, 2010-11-19 at 03:11 -0500, Kevin Hunter wrote: It usually happens after I've committed a change to my local repository, that I then sent in as a patch. That patch got applied with a slight modification, and then the conflict. Grief - what a pain; sorry about that - best if you push yourself clearly. Yes - I too fall over this dumb thing every time as well; you cannot 'git add' and 'git commit' the file if it is identical to the original :-) Sometimes I cheat by doing some trivial whitespace change (which is no doubt evil ;-). I -believe- if there are no changes after the edit you want to do: git rebase --skip But - of course, I could be confused [ this is a perennially irritating git issue for me too - I would -love- someone to file it with the git authors with a trivial repo test - IMHO it should auto-skip the patch in that case ;-]. Any pointers for the uninitiated would be awesome. It rather depends if you have your own changes, if not you could do git reset --hard previous-version to throw away your changes, and then git pull; otherwise it is harder. HTH, Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] newish to git: howto recover from bad merge?
On Nov 19, 2010, at 0:11, Kevin Hunter hunt...@earlham.edu wrote: Hullo List, It must be something simple, but I'm having a devil of a time recovering from a bad merge. Given my transcript below, how do I recover? It usually happens after I've committed a change to my local repository, that I then sent in as a patch. That patch got applied with a slight modification, and then the conflict. - $ cd .../rawbuild/binfilter $ git pull -r First, rewinding head to replay your work on top of it... Applying: EasyHack: RTL macro from createFromAscii Using index info to reconstruct a base tree... Falling back to patching base and 3-way merge... Auto-merging binfilter/bf_svx/source/items/svx_xmlcnitm.cxx CONFLICT (content): Merge conflict in binfilter/bf_svx/source/items/svx_xmlcnitm.cxx Failed to merge in the changes. Patch failed at 0001 EasyHack: RTL macro from createFromAscii When you have resolved this problem run git rebase --continue. If you would prefer to skip this patch, instead run git rebase --skip. To restore the original branch and stop rebasing run git rebase --abort. $ vim bf_svx/source/items/svx_xmlcnitm.cxx # AS NORMAL, HERE I MANUALLY FIX THE BAD MERGE $ git status # Not currently on any branch. # Unmerged paths: # (use git reset HEAD file... to unstage) # (use git add/rm file... as appropriate to mark resolution) # #both modified: bf_svx/source/items/svx_xmlcnitm.cxx # no changes added to commit (use git add and/or git commit -a) $ git add bf_svx/source/items/svx_xmlcnitm.cxx $ git commit # Not currently on any branch. nothing to commit (working directory clean Do not commit, let git rebase --continue do that Norbert $ git rebase --continue Applying: EasyHack: RTL macro from createFromAscii No changes - did you forget to use 'git add'? When you have resolved this problem run git rebase --continue. If you would prefer to skip this patch, instead run git rebase --skip. To restore the original branch and stop rebasing run git rebase --abort. - Any pointers for the uninitiated would be awesome. Thanks! Kevin ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Fwd: Re: cppcheck : snprintf size is out of bounds
Hi Julien, On Thu, 2010-11-18 at 22:44 +0100, Julien Nabet wrote: The bug on cppcheck has been fixed : https://github.com/danmar/cppcheck/commit/66c2825b2309d21474b22eb4e73e75a4b4ee150f Nice work :-) great to see the cleanup of LibO's twistedness having a positive effect on shared tooling as well. Thanks ! Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] setting ulimit before running soffice
On Wed, 2010-11-17 at 10:57 -0500, Kevin Hunter wrote: What's the story with the ulimit set in 'ooenv' right before running a freshly built LO: As Caolan says - without this, your app will simply not dump core. Of course, mostly people can run under gdb just fine, and core files can be a PITA to deal with. So - it is not a huge issue if it fails, we could simply pipe 2 /dev/null in the script if you want :-) It is also only used in debug (linkoo'd) mode. HTH, Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Git server-side hooks
On Fri, Nov 19, 2010 at 12:31:14PM +, Michael Meeks michael.me...@novell.com wrote: We should be using 'git merge' unless it is for a purely local branch that we are rebasing - surely ? at least - encouraging wider use of re-base is likely to burn time and motivation - that thing simply doesn't work for complex cases where there have been multiple merges to a feature branch [ I got -really- burned with Evolution and a long running branch here ]. Or did I totally miss the point ? :-) No, you are right. As I mentioned earlier, initially I did not think about multiple developer may work in a feature branch. pgpryCgYUSef3.pgp Description: PGP signature ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Git server-side hooks
On Fri, 2010-11-19 at 12:31 +, Michael Meeks wrote: On Thu, 2010-11-18 at 08:28 -0500, Kohei Yoshida wrote: Now that we promote the concept of feature branches, I hope we can revert this. We should be using 'git merge' unless it is for a purely local branch that we are rebasing - surely ? at least - encouraging wider use of re-base is likely to burn time and motivation - that thing simply doesn't work for complex cases where there have been multiple merges to a feature branch [ I got -really- burned with Evolution and a long running branch here ]. Yes, that's exactly the point I was trying to make (to make rebase optional, not default), and I believe this is what we are going to do. The problem was that rebasing was forced even when you didn't specify it during git pull. Kohei -- Kohei Yoshida, LibreOffice hacker, Calc kyosh...@novell.com ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] [PUSHED] Re: EasyHack: svtools/ RTL macro conversion
At 4:59am -0500 Fri, 19 Nov 2010, Caolán Mcnamara wrote: On Thu, 2010-11-18 at 18:50 -0500, Kevin Hunter wrote: An easy hack for RTL conversion strings, against svtools/ . Excellent, looks good. Pushed this. I simplified the syntax of a few of the constructor here and there on the way. Och, Thanks. More missed. I appreciate that. Kevin ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] setting ulimit before running soffice
At 7:59am -0500 Fri, 19 Nov 2010, Michael Meeks wrote: On Wed, 2010-11-17 at 10:57 -0500, Kevin Hunter wrote: What's the story with the ulimit set in 'ooenv' right before running a freshly built LO: As Caolan says - without this, your app will simply not dump core. Of course, mostly people can run under gdb just fine, and core files can be a PITA to deal with. Heh, well, actually from long ago I said, These are dumb! I'll turn this on when I need it! and put in ulimit -c 0 in my .bash_profile. Of course, that was before I started religiously commenting (you should see my .bash* files, e.g.), so I forgot about it. In the meantime, I think I've sussed out that warning message is unique to me. So - it is not a huge issue if it fails, we could simply pipe 2 /dev/null in the script if you want :-) Yeah, I picked that up after the responses, and reading a bit more about it. I actually knew peripherally what it did, but I was not aware that I could set up allowances on a per-user basis. Thus, I asked the (in retrospect) dumb question if others were running LO as root. I don't think getting rid of that output is a good idea (in this case). For something like '$ find /', it might be more appropriate, where it's clear there will be some permission denied listings, etc. As I said above, I think it's only me, so problem solved. Thanks though! Kevin ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PATCH] update URLs
Hi, Found a couple of links outdated in README. Cheers, -- Takeshi Abe From e2fbb2ca9e15d0e32db79e46e6cbc49f12be86a8 Mon Sep 17 00:00:00 2001 From: Takeshi Abe t...@fixedpoint.jp Date: Fri, 19 Nov 2010 23:24:30 +0900 Subject: [PATCH] update URLs --- README |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/README b/README index b263eef..d77f4aa 100644 --- a/README +++ b/README @@ -16,10 +16,10 @@ can be obtained by anonymous Git with this command: $ git pull -r ; bin/g pull -r Packaged releases can be found from: - http://libreoffice.org/download/ + http://www.documentfoundation.org/download/ More documentation is in the wiki: - http://www.freedesktop.org/wiki/Software/LibreOffice/HowToBuild + http://wiki.documentfoundation.org/Development/How_to_build Finally, you can subscribe to the development mailing list at http://lists.freedesktop.org/mailman/listinfo/libreoffice and please -- 1.7.2.3 ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] RFE: abolish memory settings
Hi there, On Fri, 2010-11-19 at 09:30 +0100, Sebastian Spaeth wrote: I hate the options dialog in general :), but one thing that has always confused the heck out of me is the memory settings and I propose that could be a nice project for someone: :-) Sounds good to me. HOW am I supposed to know if 20MB for the LO graphics cache is plenty? Quite - and then there is the meme about upping the limit improving performance ;-) And why should care about the space we want to allocate to a single graphics object? And even worse, what are the implications of throwing out these cached objects after 10 minutes? That is too much black magic. Right - someone gave up on auto-detecting the right options. LO should adapt itself to the memory available on the system, and it should find a sensible cache expiry policy without me having to hardcode some arbitrary minute number. Right; so there are some complexities here, but probably nothing that we cannot address. My first wonder is: is it -really- slow to re-load the images as/when we need them ? [ or when the come 'near' the visible region ] ? Can we do away with those settings and even better replace the under-the-hood with some sensible adaptive guesses? Right; Mozilla has exactly the same problem - though it is not that clear to me that they do a wonderful job there either - but we should definately look at that code too. Could you add a (not very easy - ie. at the bottom) hack to investigate this, and reply CCing Taras Gleck with the link - he is the Mozilla I/O, memory, performance genius :-) and can no doubt help us find whatever piece(s) of code we could re-use for this. Incidentally - the process of finding bogus, obsolete and/or uninteresting settings is a really good one. I would love to see them triaged into: * commonly useful [ ie. stay exposed ] * useful but not-so-useful [ ie. for an advanced config UI ] * non-useful compromises that should be removed (cf. above) Anyone interested in picking that up ? Thanks, Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] RTL easy hack against bridges/
Hullo List, A simple one against but test cases, but, as I said in the commit message, they were in my (grep) road! Kevin 0001-EasyHack-RTL_CONST-macro-from-createFromAscii.patch.xz Description: application/xz ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] [PATCH] [PUSHED] update URLs
On Fri, 2010-11-19 at 23:35 +0900, Takeshi Abe wrote: Hi, Found a couple of links outdated in README. Thanks a lot! Pushed. Kohei -- Kohei Yoshida, LibreOffice hacker, Calc kyosh...@novell.com ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] quick starter applet/systray defaults and various confusion
There is something of a mess going on at the moment I think. a) If you look at desktop/source/app/app.cxx see DoRestartActionsIfNecessary If there is no ~/.libreoffice dir, then on first start we register the extensions, this triggers a restart of the app. Simply because there is no explict -noquickstart then the quickstarter is launched. Once the quickstarter is launched it will reappear on every login because there we're passing true, true to the quickstarter, i.e. start now, always start next time Now, disable the quickstarter, shutdown LO and rm -rf ~/.libreoffice/3/user/extensions/bundled to emulate adding a new bundled extension. and launch LO again, the quickstarter should reappear, i.e. even though it was disabled I believe adding a new extension to LO and being forced to restart LO will reenable it. b) If you launch LO, enable the quickstarter and it appears in the systray. Now exit the quickstarter and exit LO, restart LO, and no entry in the systray. Logout, login, and ta-da its back. So, I reckon that: 1) DeRestartActionsIfNecessary at a minimum should only *show* the quickstarter, not set it to restart next time. i.e. single argument true case, not true, true. Which is what InitializeQuickStartMode does. 2) DeRestartActionsIfNecessary is actually bogus, and its an attempt to fix (b) for the specific restart-after-updated-extensions case, but doesn't address the normal case. And even with 1 will display the quickstarter applet regardless of your settings if you happen to add a new extension. 3) We should probably instead modify InitializeQuickStartMode to effectively || Sfx::ShutdownIcon::GetAutostart() with the IsQuickStart command line argument so that b gets fixed. If we were to do all through, the first thing people would notice is that the quickstart applet/systray would appear to change from defaulted on to defaulted off. So on a fresh install, do we want the systray quickstarter to be enabled or disabled by default ? C. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] 5 times LGPL installed
Hi there, On Fri, 2010-11-19 at 13:53 +0100, Sebastian Spaeth wrote: - Module readlicense_oo needs to be modified to not produce them. Patch is in git (and reverted again). oh - shame - why reverted ? - Module scp2 needs to be modified to not copy them. Patch is in git (and reverted again) Ditto ;-) you do realise scp2's dependencies are rather sub-par, it is well worth removing unxlngi6.pro in scp2/ at a whim. - instsetoo_native then complained about a missing LICENSE-EN_es.html.orig or something, even after I had done a make clean. Not sure where that came from. Perhaps a reboot would have fixed it (our environment variable fudging is amazing). Reboots ? just start a new clean login shell (in an xterm eg.) would be just as good; it is ~impossible for one shell to corrupt another's environment (unless it is its child). So I had to revert this. If someone tackles it, can we put the README and LICENSE in the existing directories please? Currently 8 out of 13 topdir entries in our installation refer to some readme or license. Right - this is really silly. And for the bundled extensions there is no need to have separate licenses either, nor a separate license per language IMHO. HTH, Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Git server-side hooks
On Fri, 2010-11-19 at 12:31 +, Michael Meeks wrote: We should be using 'git merge' So, part of the problem was that I was always using git pull, even where git merge would've been appropriate. To be honest, I didn't even know of 'git merge' until now. I always thought pull was merge in git universe... :-/ Kohei -- Kohei Yoshida, LibreOffice hacker, Calc kyosh...@novell.com ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Contributing test cases
On Fri, 2010-11-19 at 01:32 +0100, Miklos Vajna wrote: Is 3) about cppunit as well? If yes, what about just having 3) in the tree as well, and then it could be invoked with a 'make check' or similar that would run all those tests. I'm asking because I guess the more complex to run it, the fewer people will actually try it out at all. True. I was sort of wondering about potential colossal size. But sure I guess a separate repo for megatest, optional like l10n in the normal course of things, and a make check would do the trick for me. C. ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] quick starter applet/systray defaults and various confusion
On 19/11/10 15:37, Caolán McNamara wrote: There is something of a mess going on at the moment I think. a) If you look at desktop/source/app/app.cxx see DoRestartActionsIfNecessary If there is no ~/.libreoffice dir, then on first start we register the extensions, this triggers a restart of the app. Simply because there is no explict -noquickstart then the quickstarter is launched. Once the quickstarter is launched it will reappear on every login because there we're passing true, true to the quickstarter, i.e. start now, always start next time Now, disable the quickstarter, shutdown LO and rm -rf ~/.libreoffice/3/user/extensions/bundled to emulate adding a new bundled extension. and launch LO again, the quickstarter should reappear, i.e. even though it was disabled I believe adding a new extension to LO and being forced to restart LO will reenable it. b) If you launch LO, enable the quickstarter and it appears in the systray. Now exit the quickstarter and exit LO, restart LO, and no entry in the systray. Logout, login, and ta-da its back. So, I reckon that: 1) DeRestartActionsIfNecessary at a minimum should only *show* the quickstarter, not set it to restart next time. i.e. single argument true case, not true, true. Which is what InitializeQuickStartMode does. 2) DeRestartActionsIfNecessary is actually bogus, and its an attempt to fix (b) for the specific restart-after-updated-extensions case, but doesn't address the normal case. And even with 1 will display the quickstarter applet regardless of your settings if you happen to add a new extension. 3) We should probably instead modify InitializeQuickStartMode to effectively || Sfx::ShutdownIcon::GetAutostart() with the IsQuickStart command line argument so that b gets fixed. If we were to do all through, the first thing people would notice is that the quickstart applet/systray would appear to change from defaulted on to defaulted off. So on a fresh install, do we want the systray quickstarter to be enabled or disabled by default ? Disabled by default, don't be intrusive by default :) -- Andrew ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] Git detached HEAD (no branch) was: newish to git: howto recover from bad merge?
On Fri, 2010-11-19 at 19:53 +0900, Takeshi Abe wrote: $ git branch * (no branch) master So, I'd like to expand on this mysterious detached HEAD thingie in git, since I just learned about this yesterday and it might help other people understand what the heck is going on when you are detached. Basically, in git, the term branches roughly equals tips, or the last commits of the branches. They are sometimes referred to as heads (in lowercase) though not as frequently as branches. The term HEAD refers to the last commit of the current branch. Now, you are in a detached HEAD when you are not in the tip of a branch. For example, when you check out a particular commit via its hash, or tag, your checkout is no longer the last commit of any branch, hence you are detached. And, when you start committing while you are in that state, you are committing to this detached special branch called (no branch). To re-apply these changes from this detached branch to the tip of a branch, you can use git rebase e.g. git rebase master to re-apply your changes over the top of the master branch. In this particular case, since your detached branch is entirely local, it's safe to use git rebase. Anyway, this knowledge helped me understand git a bit more. So, I hope this will help others too. Kohei -- Kohei Yoshida, LibreOffice hacker, Calc kyosh...@novell.com ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] crash report and Oracle url
On 19/11/10 08:21, Thorsten Behrens wrote: Jean-Baptiste Faure wrote: Yesterday I got the crash report in LibO after a test of docking the navigator on the right edge of the LibO window. It contained a link to the privacy policy of Oracle. I did not sent the crash report because it was no clear for me who is the recipient of this crash report. Hi Jean-Baptiste, ugh - that one should be disabled. What build/platform did you use? Besides that, the text you mention is here: crashrep/source/all/crashrep.ulf Any takers for rewording that? I might have a go if you can give me some way to invoke the crash report detector - so I can see what it looks like. -- Andrew ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] git via http ...
Hi Marc, On Wed, 2010-11-03 at 10:16 +0100, Thorsten Behrens wrote: sorry, freedesktop git is only available via native git protocol - So - we now have a public http version (thanks to Tollef) - and it is here: git clone http://anongit.freedesktop.org/git/libreoffice/build.git/ Unfortunately, it still will not work with download ... we should add some code to download.in such that it will fallback to http: if git: is not available. I added this as an easy hack: http://wiki.documentfoundation.org/Development/Easy_Hacks#Make_download_fall-back_to_http It would be wonderful if you, or someone else, could help out with that. Thanks, Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] crash report and Oracle url
On Fri, 2010-11-19 at 16:33 +, Andrew wrote: I might have a go if you can give me some way to invoke the crash report detector - so I can see what it looks like. We need more crashes ! [ there's an app for that ;-]. You could (I guess) try killing soffice.bin thusly: pkill -11 -f soffice.bin (simulating a SEGV) - really we should have some regression tests that do this sort of thing as well - to test the crash recovery code. HTH, Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] crash report and Oracle url
On 19/11/10 17:39, Michael Meeks wrote: On Fri, 2010-11-19 at 16:33 +, Andrew wrote: I might have a go if you can give me some way to invoke the crash report detector - so I can see what it looks like. We need more crashes ! [ there's an app for that ;-]. You could (I guess) try killing soffice.bin thusly: pkill -11 -f soffice.bin (simulating a SEGV) - really we should have some regression tests that do this sort of thing as well - to test the crash recovery code. HTH, Michael. Hmm, that crashed LibreOffice, and then when I started it again, the Document Recovery Tool opened. At the last step, where it says 'Click next to open the Error Reporting Tool', I clicked next, but the error reporting tool didn't open and it just went to Writer. I am using the latest build from the LibO website - how can I access the error reporting tool? -- Andrew ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Presenter Console
Hi Everyone. As part of a bit of polishing of the UI of the PresenterConsole extension I wish to undertake, I want to improve the icons used at the bottom. Could you give me your opinion, as to which is best? 1st - As it currently is 2nd - Smaller font, no use of bold, 'filled' icons 3rd - Smaller font, no use of bold, 'line drawing' icons -- Andrew attachment: presenter-console.png___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Presenter Console
I like number 2 but i think the text needs to be a bit lighter. i feel its a bit on the dark side imho. On Fri, Nov 19, 2010 at 7:19 PM, Andrew rugby...@gmail.com wrote: Hi Everyone. As part of a bit of polishing of the UI of the PresenterConsole extension I wish to undertake, I want to improve the icons used at the bottom. Could you give me your opinion, as to which is best? 1st - As it currently is 2nd - Smaller font, no use of bold, 'filled' icons 3rd - Smaller font, no use of bold, 'line drawing' icons -- Andrew ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice -- Jonathan Aquilina ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PATCH] remove unused SvXMLNumFmtDefaults methods
Hi Everybody, My first mail on this liste :-) I was grepping for some dead code using 'grep #if 0' and I have found a bunch of unused method in xmloff package. I have attached 2 patches for review. Regards, Joachim. From a545c47c769e1b380d8670093bfdc6bc19243d1d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Joachim=20Tr=C3=A9mouroux?= joachim.tremour...@gmail.com Date: Fri, 19 Nov 2010 19:11:14 +0100 Subject: [PATCH] removed unused SvXMLNumFmtDefaults methods --- .../bf_xmloff/source/style/xmloff_xmlnumfi.cxx |5 - 1 files changed, 0 insertions(+), 5 deletions(-) diff --git a/binfilter/bf_xmloff/source/style/xmloff_xmlnumfi.cxx b/binfilter/bf_xmloff/source/style/xmloff_xmlnumfi.cxx index aa92804..1aa873a 100644 --- a/binfilter/bf_xmloff/source/style/xmloff_xmlnumfi.cxx +++ b/binfilter/bf_xmloff/source/style/xmloff_xmlnumfi.cxx @@ -1100,11 +1100,6 @@ void SvXMLNumFmtElementContext::EndElement() break; case XML_TOK_STYLE_ERA: rParent.UpdateCalendar( sCalendar ); -#if 0 -//! I18N doesn't provide SYSTEM or extended date information yet -if ( rParent.IsFromSystem() ) -bEffLong = SvXMLNumFmtDefaults::IsSystemLongEra( rParent.GetInternational(), bLong ); -#endif rParent.AddNfKeyword( bEffLong ? NF_KEY_GGG : NF_KEY_G ); // HasEra flag is set break; -- 1.7.1 From 1387cc4ef6e2ee3ebb5f74c0a68141d1b75b1923 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Joachim=20Tr=C3=A9mouroux?= joachim.tremour...@gmail.com Date: Fri, 19 Nov 2010 19:11:14 +0100 Subject: [PATCH] removed unused SvXMLNumFmtDefaults methods --- xmloff/inc/xmloff/xmlnumfi.hxx |7 xmloff/source/style/xmlnumfi.cxx | 68 -- 2 files changed, 0 insertions(+), 75 deletions(-) diff --git a/xmloff/inc/xmloff/xmlnumfi.hxx b/xmloff/inc/xmloff/xmlnumfi.hxx index fc8a663..eae060a 100644 --- a/xmloff/inc/xmloff/xmlnumfi.hxx +++ b/xmloff/inc/xmloff/xmlnumfi.hxx @@ -115,13 +115,6 @@ public: class SvXMLNumFmtDefaults { public: -static sal_Bool IsSystemLongDay( const SvtSysLocale rSysLoc, BOOL bLong ); -static sal_Bool IsSystemLongMonth( const SvtSysLocale rSysLoc, BOOL bLong ); -static sal_Bool IsSystemTextualMonth( const SvtSysLocale rSysLoc, BOOL bLong ); -static sal_Bool IsSystemLongYear( const SvtSysLocale rSysLoc, BOOL bLong ); -static sal_Bool IsSystemLongEra( const SvtSysLocale rSysLoc, BOOL bLong ); -static sal_Bool IsSystemLongDayOfWeek( const SvtSysLocale rSysLoc, BOOL bLong ); - // return value is NfIndexTableOffset static sal_uInt16 GetDefaultDateFormat( SvXMLDateElementAttributes eDOW, SvXMLDateElementAttributes eDay, SvXMLDateElementAttributes eMonth, diff --git a/xmloff/source/style/xmlnumfi.cxx b/xmloff/source/style/xmlnumfi.cxx index 7f38b03..527769d 100644 --- a/xmloff/source/style/xmlnumfi.cxx +++ b/xmloff/source/style/xmlnumfi.cxx @@ -1246,74 +1246,6 @@ void SvXMLNumFmtElementContext::EndElement() //- -sal_Bool SvXMLNumFmtDefaults::IsSystemLongDay( const SvtSysLocale, BOOL bLong ) -{ -// TODO: merge system information and defaults into i18n locale data -#if 0 -return bLong ? rIntn.IsLongDateDayLeadingZero() : rIntn.IsDateDayLeadingZero(); -#else -return !bLong; -#endif -} - -sal_Bool SvXMLNumFmtDefaults::IsSystemLongMonth( const SvtSysLocale, BOOL bLong ) -{ -// TODO: merge system information and defaults into i18n locale data -#if 0 -if (bLong) -{ -MonthFormat eMonth = rIntn.GetLongDateMonthFormat(); -return ( eMonth == MONTH_ZERO || eMonth == MONTH_LONG ); -} -else -return rIntn.IsDateMonthLeadingZero(); -#else -return !bLong; -#endif -} - -sal_Bool SvXMLNumFmtDefaults::IsSystemTextualMonth( const SvtSysLocale, BOOL bLong ) -{ -// TODO: merge system information and defaults into i18n locale data -#if 0 -if (bLong) -{ -MonthFormat eMonth = rIntn.GetLongDateMonthFormat(); -return ( eMonth == MONTH_SHORT || eMonth == MONTH_LONG ); -} -else -return sal_False; -#else -return bLong; -#endif -} - -sal_Bool SvXMLNumFmtDefaults::IsSystemLongYear( const SvtSysLocale, BOOL bLong ) -{ -// TODO: merge system information and defaults into i18n locale data -#if 0 -return bLong ? rIntn.IsLongDateCentury() : rIntn.IsDateCentury(); -#else -return bLong; -#endif -} - -sal_Bool SvXMLNumFmtDefaults::IsSystemLongEra( const SvtSysLocale rSysLoc, BOOL bLong ) -{ -// TODO: merge system information and defaults into i18n locale data -return IsSystemLongYear( rSysLoc, bLong ); // no separate setting -} - -sal_Bool SvXMLNumFmtDefaults::IsSystemLongDayOfWeek( const SvtSysLocale, BOOL bLong ) -{ -// TODO: merge system information and defaults into i18n locale data -#if 0 -return ( bLong rIntn.GetLongDateDayOfWeekFormat() == DAYOFWEEK_LONG ); -#else -
Re: [Libreoffice] Presenter Console
At 1:19pm -0500 Fri, 19 Nov 2010, Andrew wrote: Could you give me your opinion, as to which is best? 1st - As it currently is 2nd - Smaller font, no use of bold, 'filled' icons 3rd - Smaller font, no use of bold, 'line drawing' icons I like the icons from 1, and the lighter colors. However, take away the bold. The text of 2 is needs to be brighter (like number 1), but has it right in terms of thickness of the glyph lines. Kevin ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] High Contrast Cleanup
Hi there, On Sun, 2010-11-14 at 23:46 +0100, Sebastian Spaeth wrote: Copying them over would mean, our download tarballs qould bloat by another 10 MB or so only transiently, until we remove the duplicates. ( I guess git would be able to handle that efficiently though). Sure, -but- git since it stores -all- history for all time, is a real issue. If you checkin and delete a 10Mb compressed tar-ball, and then delete it, then you add 10Mb to the git pull size forever - so, I would prefer to keep the repo size small, even if it means we have to (for one release) have a larger source archive. the very end, complaining about a missing crystal-icon.tar or Odd; did you run 'download' in the top-level to fetch them ? ATB, Michael. -- michael.me...@novell.com , Pseudo Engineer, itinerant idiot ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PATCH] RTL_CONSTASCII_USTRINGPARAM for libs-core/connectivity/drivers
Hello, Here's patch for the last changes to do that I've seen in libs-core/connectivity/drivers Julien. (LGPLv3+ / MPL) connectivity_source_drivers.txt.bz2 Description: application/bzip ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] git via http ...
On Fri, Nov 19, 2010 at 05:34:34PM +, Michael Meeks michael.me...@novell.com wrote: not available. I added this as an easy hack: http://wiki.documentfoundation.org/Development/Easy_Hacks#Make_download_fall-back_to_http It would be wonderful if you, or someone else, could help out with that. Hmm, I don't want to question you, but maybe that's not really necessary. If we are going to get rid of the build repo for development, then we'll get rid of the download script as well, I guess. And then bin/g clone will be used for download, which will just use the same repo base url which was used to clone bootstrap.git, ie. no fall-back will be needed. Please correct me if I missed something. :) pgpNog4ComfXM.pgp Description: PGP signature ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Git server-side hooks
On Fri, Nov 19, 2010 at 11:03:21AM -0500, Kohei Yoshida kyosh...@novell.com wrote: So, part of the problem was that I was always using git pull, even where git merge would've been appropriate. To be honest, I didn't even know of 'git merge' until now. I always thought pull was merge in git universe... :-/ Don't worry, in the past there were no git merge, and just pull. :) Then later it got split to fetch + merge to make git more distributed (fetch is always a trivial operation, then you can merge whenever you want, even when you're offline). :) pgpueikOdapMf.pgp Description: PGP signature ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PATCH] binfilter/ : remove dead code
Hullo List, Removing some dead/non-compiled code in binfilter. Did some drive by comment translation, and minor tab removals. Cheers, Kevin 0001-EasyHacks-kill-dead-code-translate-formatting.patch.bz Description: application/bzip ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PATCH] try to correct some timers in sc + question
Hello, Here's a try to correct some timers in scmod.cxx. Here's what I did and why. 1) In IMPL_LINK( ScModule, SpellTimerHdl, Timer*, EMPTYARG ) I just added aSpellTimer.Stop() when there's neither input from keyboard nor ContinueOnlineSpelling. + I changed the function to have 1 return only. 2) In IMPL_LINK( ScModule, IdleHandler, Timer*, EMPTYARG ) I removed all the things with bSpell since it seems that ContinueOnlineSpelling is managed by SpellTimerHdl I'd like to add a stop but i'd like first to understand what's the use of all the SC_IDLE_STEP, SC_IDLE_MAX, nIdleCount... I repeat, this patch is just a try. If I'm all wrong, sorry for the noise and please, explain my mistakes. Julien. (just in case what I did is ok, license is as usual LGPLv3+ / MPL) diff --git a/sc/source/ui/app/scmod.cxx b/sc/source/ui/app/scmod.cxx index 1bc98b5..1f164e9 100644 --- a/sc/source/ui/app/scmod.cxx +++ b/sc/source/ui/app/scmod.cxx @@ -1912,11 +1912,8 @@ IMPL_LINK( ScModule, IdleHandler, Timer*, EMPTYARG ) BOOL bLinks = pDoc-IdleCheckLinks(); BOOL bWidth = pDoc-IdleCalcTextWidth(); -BOOL bSpell = pDoc-ContinueOnlineSpelling(); -if ( bSpell ) -aSpellTimer.Start(); // da ist noch was -bMore = bLinks || bWidth || bSpell;// ueberhaupt noch was? +bMore = bLinks || bWidth; // ueberhaupt noch was? // While calculating a Basic formula, a paint event may have occured, // so check the bNeedsRepaint flags for this document's views @@ -1956,15 +1953,27 @@ IMPL_LINK( ScModule, SpellTimerHdl, Timer*, EMPTYARG ) if ( Application::AnyInput( INPUT_KEYBOARD ) ) { aSpellTimer.Start(); -return 0; // dann spaeter wieder... } - -ScDocShell* pDocSh = PTR_CAST( ScDocShell, SfxObjectShell::Current() ); -if ( pDocSh ) +else { -ScDocument* pDoc = pDocSh-GetDocument(); -if ( pDoc-ContinueOnlineSpelling() ) -aSpellTimer.Start(); + +ScDocShell* pDocSh = PTR_CAST( ScDocShell, SfxObjectShell::Current() ); +if ( pDocSh ) +{ +ScDocument* pDoc = pDocSh-GetDocument(); +if ( pDoc-ContinueOnlineSpelling() ) +{ +aSpellTimer.Start(); +} +else +{ +aSpellTimer.Stop(); +} +} +else +{ +aSpellTimer.Stop(); +} } return 0; } ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] LibreOffice Beta 3
I am running LibreOffice Beta 3 under Ubuntu 10.10 (64-bit) and it seems to gradually bog down with use, so that I have to exit and restart it several times a day. It gets so that cursor movement is very slow. This was also a problem with Beta 2, and I was hoping that Beta 3 would fix it, but not so... ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Git detached HEAD (no branch) was: newish to git: howto recover from bad merge?
Hi *, On Fri, Nov 19, 2010 at 5:31 PM, Kohei Yoshida kyosh...@novell.com wrote: On Fri, 2010-11-19 at 19:53 +0900, Takeshi Abe wrote: $ git branch * (no branch) master So, I'd like to expand on this mysterious detached HEAD thingie in git, [...] I'll add to that: Just never ever thing you're smart and git clone that directory first to try things. When you do, you're screwed once you delete the original copy, since the clone doesn't know anything about the branches after that thing anymore, I had to spend too much time to fixup with that. detached heads in git are dangerous!!! Don't use them (always name it, even if it is stupidnametonotredoeverythingbecauseofdetachedheadmadness). I learned the hard way… Especially when you're used to mercurial and thing what's so bad about multiple heads, that's a no-brainer: GIT is very different in this regard. You have been warned. ciao Christian ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] [PATCH] Removal of obsolete option --with-unstable-wp
MPL 1.1 / GPLv3+ / LGPLv3+ see: http://cgit.freedesktop.org/libreoffice/build/commit/?id=9e0f4d66e7b219baa142fe86f2167e37d8f8e6f7 From dc6697dec2cd2ad3ba9e1b21bd4f73699bb80b33 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niko=20R=C3=B6nkk=C3=B6?= ron...@iki.fi Date: Tue, 16 Nov 2010 16:50:30 +0200 Subject: [PATCH] Removal of obsolete option --with-unstable-wp Removed obsolete option --with-unstable-wp. See: 9e0f4d66e7b219baa142fe86f2167e37d8f8e6f7 --- configure.in |5 - 1 files changed, 0 insertions(+), 5 deletions(-) diff --git a/configure.in b/configure.in index c2a2a01..a20045c 100755 --- a/configure.in +++ b/configure.in @@ -494,11 +494,6 @@ AC_ARG_WITH(system-libwps, --with-system-libwpsUse system libwps* library.], ,) -AC_ARG_WITH(unstable-wp, -[ - --with-unstable-wp Use new unstable libwp* libraries.], -,) - AC_ARG_WITH(win32, [ --with-win32Define the build to be on Win32; this sets a number of -- 1.7.0.4 ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] Problem building sw (writer) in debug mode under Linux
David Tardon wrote: On Tue, Nov 16, 2010 at 10:34:41AM +0100, Giuseppe Castagno wrote: Hi all, per subject, with a pull -r from this morning (CET), compiling sw with: build -P2 debug=true Try pull again; commit f19fb8630f6dbf4b525bdd4ba775f05bb9d7e1a0 should fix this. ok that fixed, but then I hit more problems on the same area. Now what it's the right flag to use to surround a debug specific code: OSM_DEBUG_LEVEL 1 or DBG_UTIL ? I confess I'm a little confused :-). Though is seems, but I'm not entirely sure, that debug=true in the compile command line gives a -DOSL_DEBUG_LEVEL=2 as a compiler flag. beppec56. -- Kind Regards, Giuseppe Castagno Acca Esse http://www.acca-esse.eu giuseppe.castagno at acca-esse.eu beppec56 at openoffice.org ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] [Patch][EasyHacks] Replace ScfRef with boost::shared_ptr in calc filter code
Hi Nigel, On Fri, 2010-11-19 at 18:29 +, Nigel Hawkins wrote: Hi, Attacking another item on the easy hacks page. The attached patches should remove ScfRef (which was used more than I thought it would be) and replace it with boost::shared_ptr. Great work! I've applied all your patches and committed as a single commit. I hope that's okay. I've applied your patches mostly as-is, except for one part which I explain below. Most of them are very simple. The last one is to replace all uses of the ScfRef.is() function. Here, you replaced all instances of xFoo.is() with xFoo.get(), which is itself not incorrect. But boost::shared_ptr overloads the operator bool() which returns the wrapped pointer value when a boolean value is expected, so in this example simply replacing xFoo.is() with xFoo is sufficient (and cleaner). I've made this change while applying your last patch. ScfRef had a note about being used instead of shared_ptr because it was faster. Some simple profiling suggested it was about two and a half times faster, but that unless we're throwing these around in their millions, there wouldn't be any significant (or even noticeable) performance hit. Agreed. That was my idea as well. In case using boost::shared_ptr causes any significant performance hit, we could easily replace that with boost::intrusive_ptr for better performance. Thanks a lot for your work. Really appreciated. :-) Kohei -- Kohei Yoshida, LibreOffice hacker, Calc kyosh...@novell.com ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
[Libreoffice] Discuss Easy_Hacks: Remove Sheet F rom File?
Is this the place to discuss this proposal? I don't see any discussion on the wiki page itself. http://wiki.documentfoundation.org/Development/Easy_Hacks#Remove_.27Sheet_From_File27_menu_item_from_Calc I, for one, find Insert Sheet From File a useful shortcut for importing data from a text file, and would hate to see it removed. Yes, the same function is available through Insert Sheet, but importing data is a common operation and the two extra clicks are a noticeable bit of hassle. If Calc had a clean way to import data from a text file, then I'd love to see this go, but right now, it would be missed. Joe ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
Re: [Libreoffice] More links inherited from OOo
Hi Thorsten, all, On 16/11/2010 08:14, Thorsten Behrens wrote: Hi folks, after some more grepping, also adapted the readmes to only reference documentfoundation/libreoffice links. File is attached, would be cool if you could review it for correctness, language - and whether the links as such make sense. Thanks for the work on this. Some remarks: - the paragraph about changes on BerkeleyDB engine and extensions database incompatiliby seems strange when speaking about [for ${PRODUCTNAME} versions prior to 3.2]. May be we should namely speaks about OpenOffice.org. - could we replace the link www.libreoffice.org/access by www.libreoffice.org/accessibility? Is there anybody already working on this page? - I think we should completely remove the paragraph about Registration and User Survey - it is mention that IssuZilla is on our website, this should be corrected by giving the right BugZilla address on Freedesktop. Hope this helps :) Kins regards Sophie -- Founding member of The Document Foundation ___ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice