Re: LyX behaves buggy when spellings-checker not installed
"R. Lahaye" <[EMAIL PROTECTED]> writes: | R. Lahaye wrote: | > | > I run LyX on a machine that has no spellings checker installed! | > The following happens when I start LyX from an xterminal: | | There are more spelling checher nuissances: | | When I [Close] the spelling checker, I always get a message | dialog that says: | | "Spell checking completed!" | words checked | [OK] | | This message dialog is a pain! Let me guess that nobody | ever really reads the number of words checked, but everybody | hits that [OK] button as quick as possible. Wrong. A lot of people use this _just_ to get the number of words checked. -- Lgb
Re: frequent polling in LyX (plus LyX under DRT screenshots)
Amir Michail wrote: > I also found another problem with getting rid of cursor blinking. > If you type a sentence and press return, the cursor will not appear > for a while until it is told to blink on. By then, the after picture > has already been taken, so cursor tracking is broken. So it's not sufficient > to disable blinking. We also need to make cursors appear immediately when > they are moved in various contexts. The cursor handling is a bit broken. We _always_ should hide *and* show the cursor, but sometimes we just "forget" to show it as it really should show as then we get cursor trails :(. IMO we need some sort of counter or some sort of flag (as the undo does) in which we see if the cursor is already hidden and don't do it again. So the first one hiding it is responsible for showing it again. When we changed this behaviour it should be quite easy tracking down where we "forgot" to put a showCursor call. Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Mitterstrich 151/A Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39050 SteineggWeb: http://www.lyx.org/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._
Re: Updated da.po for lyx 1.2.2cvs
Claus Hindsgaul wrote: > Hello, > > I have attached an updated da.po against the lyx-devel lyx-1_2_X branch. Where? > Please apply Really compact this danish language, it's so short that we don't even need a da.po file ;) Jug -- -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._ Dr. Jürgen VignaE-Mail: [EMAIL PROTECTED] Mitterstrich 151/A Tel/Fax: +39-0471-450260 / +39-0471-450253 I-39050 SteineggWeb: http://www.lyx.org/~jug -._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._-._
Re: LyX behaves buggy when spellings-checker not installed
R. Lahaye wrote: > > I run LyX on a machine that has no spellings checker installed! > > The following happens when I start LyX from an xterminal: There are more spelling checher nuissances: When I [Close] the spelling checker, I always get a message dialog that says: "Spell checking completed!" words checked [OK] This message dialog is a pain! Let me guess that nobody ever really reads the number of words checked, but everybody hits that [OK] button as quick as possible. That the spell checking is completed, it totally non-informative. Let's get rid of this useless message dialog!! Instead, we could show the number of words checked, while spelling-checking, inside the spellingchecker dialog. There it's much more useful too! For example, replace the "0%" and "100%", underneath the slider, by: [_slider__] 0 total where "" is the current number of words checked, and "total" the total number of words (to be checked) in the document. But, is that information available in the spellingchecker dialog code? The dialog should interogate the controller about the total number of words that are possibly going to be checked. With partialUpdate(), the dialog gets the current number of words checked. With that info, the dialog can update [slider] in percentage and [xxx] in number of words. But all that needs some recoding of the controller, I suppose. Regards, Rob.
[Patch] Better xforms dialogs.
Hi, I have improved and reshaped three Xforms dialogs: form_citation.fd form_ref.fd form_spellchecker.fd I made these few dialogs less bulky with a better layout. Patch attached. I used some of John's improvements in Qt, added tooltips, and set the proper gravity/resize policy (NOTE: present fdesign 1.0.0rc4 needs to be patched for the gravity/resize to work!). The dialogs are smaller because of: - re-arranging a few widgets. - making default height of input/buttons 25 (was 30), and use "12-normal" font. (Yes, I checked that the one-size larger font still fits in!). - initial design can be smaller, because resize works properly. If these improvements are appreciated, I will continue with the other dialogs. Please let me know. Regards, Rob. BetterDialogs.diff.gz Description: application/gzip
Re: frequent polling in LyX (plus LyX under DRT screenshots)
Hi, Perhaps a configure flag could be used to build LyX in a way that works well under DRT? Although the polling change I mentioned (and others) may not be bug free, they do seem to get LyX working under DRT. You can see some preliminary results here in the following screenshots: http://www.cse.unsw.edu.au/~amichail/lyx8.png >From the function sequence in the picture above, we see that the user probably moved the cursor up. However, that's not the whole story as made obvious by the before/after screenshots. In fact, the cursor moved out from math mode as well. This shows the importance of before/after pictures in explaining past actions (possibly performed by others). BTW, could someone tell me if the functions shown in the pictures are particularly relevant to the action? The darker the action label, the more relevant it is perceived to be to the action shown in the before/after pictures. To see how we arrived at the previous screenshot, see the following: http://www.cse.unsw.edu.au/~amichail/lyx1.png http://www.cse.unsw.edu.au/~amichail/lyx2.png http://www.cse.unsw.edu.au/~amichail/lyx3.png http://www.cse.unsw.edu.au/~amichail/lyx4.png http://www.cse.unsw.edu.au/~amichail/lyx5.png http://www.cse.unsw.edu.au/~amichail/lyx6.png http://www.cse.unsw.edu.au/~amichail/lyx7.png http://www.cse.unsw.edu.au/~amichail/lyx8.png I also found another problem with getting rid of cursor blinking. If you type a sentence and press return, the cursor will not appear for a while until it is told to blink on. By then, the after picture has already been taken, so cursor tracking is broken. So it's not sufficient to disable blinking. We also need to make cursors appear immediately when they are moved in various contexts. Amir
LyX behaves buggy when spellings-checker not installed
Hi, I run LyX on a machine that has no spellings checker installed! The following happens when I start LyX from an xterminal: *Immediately* after selecting the spellings checker, the xterminal gets the message "LyX: Failed to start ispell!". Then LyX seems to hang for a while. After that while, the xterminal gets "Ispell read timed out, what now?" and TWO identical error dialogs, one after the other, appear on my display; dialog screenshot attached. Why does LyX hang/wait, when it detects so quickly that "LyX: Failed to start ispell!" ?? Having two identical error messages (the second appears as soon as I close the first) is one too many. BTW is the beginning of the sentence missing in the error message? Another peculiar detail: selecting "ispell" or "aspell" in preferences both pop up in the xterminal "LyX: Failed to start ispell!". Something is not working properly here! Regards, Rob.
Re: Bug in lyxconvert_217.py
On Sat, Aug 31, 2002 at 12:49:23AM +0300, Dekel Tsur wrote: > On Fri, Aug 30, 2002 at 03:26:47PM +0100, José Abílio Oliveira Matos wrote: > > On Thursday 29 August 2002 19:21, Dekel Tsur wrote: > > [...] > > > Another solution may be write a function find_end_of() which find the > > > correct end of the tabular. find_end_of() should be a generalization of > > > find_end_of_inset() which have two additional arguments: start token & end > > > token (for find_end_of_inset, start token = \begin_inset, and end token = > > > \end_inset). > > > > And be able to find only the top level close. I agree. I dind' need until > > now to take in account the structure present in lyx files. > > You don't need that, at least for case. > An example for a table conversion code would be: >for each line in file: >if line begins withdo the appropriate conversion. > > The only problem with this is speed (doing the check for each line). > A faster solution is as follows: > while 1: > find find the matching for each line between the two lines above: > if line begins with do the appropriate conversion. Ok, you are right. > (*) This step is simple > > > I also need that for the paragraph inside cells. Nothing prevents me from writing or Why? -- José Abílio Matos LyX and docbook a perfect match. :-)
Re: Feature freeze 2002-09-13
On Thu, Aug 29, 2002 at 12:23:54PM +0200, Jean-Marc Lasgouttes wrote: > > "Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes: > > Lars> Do anyone have a problem with that? > > I will not have time to finish the 'read floats from layout' things > before this date. Now that everything is moved into LyXTextClass, it > should not be difficult, in fact. So if somebody is willing to finish > the work, I'd be grateful. > > JMarc Great! I was planning to look into that, as I have a personal stake in it. _Please_ try to get this into the next stable. Martin msg43900/pgp0.pgp Description: PGP signature
Re: MathsSymbols status
On Thu, Aug 29, 2002 at 09:05:11AM +0100, John Levon wrote: > Hopefully, all symbols now exist, with the exception of AMS nrel. The > xbm does not have remotely static heights for the symbols, os it goes > horribly wrong. Martin, can you please look at this ? Do you mean that 320 is not divisible by 9? Would it help to go to 324? And then reposition the icons? > wrt xforms, I think it would be a lot better for xforms to use these > xpm's as well. This means we can *delete* MathsSymbols and all can > rejoice. But nrel needs fixing first and I'm worn out. > > Is there a kind volunteer willing to split images/ams_nrel.xbm into > individuals xpms ?? Perhaps, but not right now. > regards > john Martin msg43899/pgp0.pgp Description: PGP signature
Re: MathsSymbols.C - Martin !
On Thu, Aug 29, 2002 at 06:27:49AM +0100, John Levon wrote: > > case 7: > case 8: > case 9: > case 10: > // to be added -- MV > break; > } AARGH!! I remember this. I tried to faithfully add code as I added the amsmath panels stuff, without actually understanding what is going on here. This looks *extremely* cludgy. I assume you understand dw and dh -- the panel dimensions in icons. What happens to id is apparently, that it is converted per-panel into an inside-subpanel serial number. What happened IIRC was, that at some point I got the feeling that this code didn't actually do anything (i.e., was not called), so I stopped adding my newer panel work to it, with the firm resolve to return to it if any bugs got reported -- first thing being a clean-up. And that's where we stand today... > Please do. I have hacked code to split up each panel into xpm's with the > "arg" filename. What are the correct settings for these options ? I suggest you (a) find out if this code gets actually used (I suspect not) and (b) nuke it if so. It's *UGLY*. So ugly that even if it does do something, I still suggest you roll your own :-) I suspect the actual work is done elsewhere. > thanks > john > > -- > "Take the ideas you find useful. Try not to get hung up on the labels." > - Jonathan S. Shapiro Good luck Martin msg43898/pgp0.pgp Description: PGP signature
Re: Polishing away...
On Tue, Aug 27, 2002 at 12:01:35PM +0200, Jean-Marc Lasgouttes wrote: > > "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: ... > As far as I know, the lfun is already disabled when you are not at the > relevant place. The code for that is in lyxfunc.C:368. Ah! I'm just slow... great. > Concerning cursor-jumping: I do not want the cursor to jump in this > case. I think this could be really annoying for people. > > So my question is again: WHY do we want this inset to be at start of > the paragraph? Is there one good reason to enforce this?? No... but some people want it to be in that position and me being the nice guy I am... but it's a kludge, of course. I agree: this is not an inset tied to the start of the paragraph, so let's just honestly have it all over the place :-) > JMarc > Martin msg43897/pgp0.pgp Description: PGP signature
Updated da.po for lyx 1.2.2cvs
Hello, I have attached an updated da.po against the lyx-devel lyx-1_2_X branch. Please apply Claus Hindsgaul
