Re: Index dialog in qt frontend

2003-01-15 Thread Juergen Spitzmueller
Angus Leeming wrote: > Having said that, I do not think that this should be done before the 1.4 > cycle. There's a lot of tweaking gone into this stuff which we'd almost > certainly break. So we should take my patch for now which cures the obvious problems in qt and do your change in the next cyc

Re: Index dialog in qt frontend

2003-01-15 Thread Angus Leeming
John Levon wrote: > On Wed, Jan 15, 2003 at 05:46:57PM +0100, Juergen Spitzmueller wrote: >> if (isValid()) >> bc().valid(); >> else >> bc().invalid(); ButtonController::valid takes an optional arg (bool = true), so simplify your code with bc().valid(isValid()); >> there cures the probl

Re: Index dialog in qt frontend

2003-01-15 Thread John Levon
On Wed, Jan 15, 2003 at 06:48:35PM +0100, Juergen Spitzmueller wrote: > Any other cases I missed? Is it intentional that an empty label can be > inserted? Yes. We should not impede workflow for the sake of "correctness" regards john -- "Anarchy in development, tyranny in integration."

Re: Index dialog in qt frontend

2003-01-15 Thread Juergen Spitzmueller
John Levon wrote: > Why doesn't xforms have the same problem because it calls bc().valid() directly from within update(). The attached patch cures the index and url problems. Any other cases I missed? Is it intentional that an empty label can be inserted? Jürgen Index: src/frontends/qt2/ChangeLo

Re: Index dialog in qt frontend

2003-01-15 Thread John Levon
On Wed, Jan 15, 2003 at 05:46:57PM +0100, Juergen Spitzmueller wrote: > if (isValid()) > bc().valid(); > else > bc().invalid(); > > there cures the problem with the index and URL dialog, but now the buttons are > enabled even when nothing has been changed.

Re: Index dialog in qt frontend

2003-01-15 Thread Juergen Spitzmueller
John Levon wrote: > can you try adding, at the end of show() in Qt2Base.C, Adding if (isValid()) bc().valid(); else bc().invalid(); there cures the problem with the index and URL dialog, but now the buttons are enabled even when nothing has been c

Re: Index dialog in qt frontend

2003-01-15 Thread John Levon
On Wed, Jan 15, 2003 at 05:08:14PM +0100, Juergen Spitzmueller wrote: > This seems to be a general bug. E.g. in the URL dialog, OK is enabled on popup > even if there is no entry. Only if you type something and delete again, it > will be disabled. isValid seems to check only after the first keyp

Re: Index dialog in qt frontend

2003-01-15 Thread Juergen Spitzmueller
John Levon wrote: > Oh I see. Yes this is bad and it's a bug. I'm not sure of its source, > however. This seems to be a general bug. E.g. in the URL dialog, OK is enabled on popup even if there is no entry. Only if you type something and delete again, it will be disabled. isValid seems to check

Re: Index dialog in qt frontend

2003-01-15 Thread John Levon
On Wed, Jan 15, 2003 at 05:04:59PM +0100, Jean-Marc Lasgouttes wrote: > You mean QIndex::isValid? But this checks whether the entry is empty, > doesn't it? What I see here is that OK is disabled even though there > is some entry. OK is enabled only if I change something. But obviously > most of th

Re: Index dialog in qt frontend

2003-01-15 Thread Jean-Marc Lasgouttes
> "John" == John Levon <[EMAIL PROTECTED]> writes: John> On Wed, Jan 15, 2003 at 04:50:17PM +0100, Jean-Marc Lasgouttes John> wrote: >> If I use Insert>Index entry in lyx/qt, the OK button of the dialog >> is grayed out, which means that I should use escape to close the >> dialog instead of re

Re: Index dialog in qt frontend

2003-01-15 Thread John Levon
On Wed, Jan 15, 2003 at 04:50:17PM +0100, Jean-Marc Lasgouttes wrote: > If I use Insert>Index entry in lyx/qt, the OK button of the dialog is > grayed out, which means that I should use escape to close the dialog > instead of return. This seems very unintuitive to me... remove isValid for the dia

Index dialog in qt frontend

2003-01-15 Thread Jean-Marc Lasgouttes
If I use Insert>Index entry in lyx/qt, the OK button of the dialog is grayed out, which means that I should use escape to close the dialog instead of return. This seems very unintuitive to me... JMarc