On Wed, Aug 29, 2001 at 02:58:39PM +0300, Dekel Tsur wrote:
> Wouldn't it be simpler to just add the following to in_word_set:
>
> if (wordlist.empty())
> ReadSymbols();
apart from the endless loop if reading fails, certainly. Use at least a
static 'bool initialized' variable
> > The problem is that the constructor will be called before executing main(),
> > so the location of the system dir has not yet determined.
>
> So put it into some static var in some function. That will be initialized
> the first time the function is called.
Wouldn't it be simpler to just add t
> this only affects mathed right?
It requires writes outside src/mathed, but the added functionality is for
mathed only.
Andre'
--
André Pönitz . [EMAIL PROTECTED]
On Wed, Aug 29, 2001 at 11:33:39AM +0200, Lars Gullik Bjønnes wrote:
> then no objections from me.
Does this mean Dekel's patch could be committed if he thinks his patch is ok?
Andre'
--
André Pönitz . [EMAIL PROTECTED]
> yes. but in this case... do we fall gracefully?
Yes. X symbols are used if available, otherwise ERT is used. I.e. exactly
the same solution as we had all the time.
Andre'
--
André Pönitz . [EMAIL PROTECTED]
On 29-Aug-2001 Lars Gullik Bjønnes wrote:
>| Yes. That's what Dekel's current implementation is doing.
>| So the problem is bogus.
>
> It is not bogus if we do not have a decent fallback.
We have a decent fallback and if you followed the mailinglist you know
which one ;)
Jürgen
--
-
On Wed, Aug 29, 2001 at 10:47:31AM +0200, Juergen Vigna wrote:
> IMO you're right we are asking the XServer to provide us the font (this
> is normal fontmechanism in X). We don't need to know which fonts are installed
> on the machine LyX are running and the "symbols" table is NOT generated on
> r
On 29-Aug-2001 Andre Poenitz wrote:
> Wait... is this problem real?
>
> We do not really use fonts from the machine we are running on, do we?
> Aren't we asking the _server_ to give us a font? And if it does not
> have it, we fall back anyway.
IMO you're right we are asking the XServer to pro
> | If not, we could probably settle for a lyxrc switch 'useamsfonts' or
> | something like that...
>
> Why can't we dynamically discover and fallback?
I don't know how
Wait... is this problem real?
We do not really use fonts from the machine we are running on, do we?
Aren't we asking t
> Andre Poenitz <[EMAIL PROTECTED]> writes:
> | Uuh... excellent question... but a real problem will only arise if you are
> | on a machine with the fonts installed and use an X server running on a
> | machine without the fonts.
>
> this wil not be uncommon...
I guess so. Is there a possibility
On Wed, Aug 29, 2001 at 11:43:14AM +0900, Shigeru Miyata wrote:
> Anyway I have a question. LyX reads the symbols file from the machine
> on which it is running. On the other hand, the fonts must be available
> at the machine on which X server is running. Are we going to have a
> facility to sy
> The problem is that the constructor will be called before executing main(),
> so the location of the system dir has not yet determined.
So put it into some static var in some function. That will be initialized
the first time the function is called.
> > Why the use of LyXLex to read that simp
In-Reply-To: <[EMAIL PROTECTED]>
Andre Poenitz <[EMAIL PROTECTED]> wrote:
> Fallback to X symbols does not really seem to work (i.e. I get \sum in red
> if I take away the latex fonts from the fontpath?)
Have you update "~/.lyx/symbols" accordingly?
Anyway I have a question. LyX reads the symb
On Mon, Aug 27, 2001 at 11:56:42AM +0200, Andre Poenitz wrote:
> I do not like the call to ReadSymbol form lyxmain. This breaks
> encapsulation... What is wrong with the "static variable constructor
> trick"?
On Sat, Aug 25, 2001 at 12:26:57AM +0300, Dekel Tsur wrote:
> An here is a slightly improved version.
I do not like the call to ReadSymbol form lyxmain. This breaks
encapsulation... What is wrong with the "static variable constructor
trick"?
Why the use of LyXLex to read that simple file forma
15 matches
Mail list logo
