mailing-list for TeXmacs Users

Text archives Help


Re: [TeXmacs] Some other quirks


Chronological Thread 
  • From: Joris van der Hoeven <address@hidden>
  • To: address@hidden
  • Subject: Re: [TeXmacs] Some other quirks
  • Date: Thu, 24 Aug 2006 10:40:04 +0200

On Wed, Aug 23, 2006 at 11:36:42PM +0200, address@hidden wrote:
> - On my system, the CygTexmacs ignores the "Look and feel" setting, and
> always uses Emacs mode (it uses the German localized menus), even after
> a restart.

Strange; I have to look at that. The default should be "windows" anyway.

> - While it is nice that Texmacs keeps track if a document has been
> modified after the last save, I think it is a bad idea not to save the
> document if the user wants it ("No changes need to be saved").
> For example, the old method for editing composite documents had a bug,
> where the composite document was scattered with source code fragments if
> one of the included documents was set to "source view" (haven't checked
> it on the current version, though). Thus: 1. open the included file, 2.
> switch off source view, 3. save the file. BUT that did not work, since
> TeXmacs still thought the document was unchanged, and ignored the save
> command.
> So this feature opens up the possibility for very nasty bugs leading to
> data loss. The other point is, you never know WHY the user wants to
> save, so let him save. For example, he might have deleted the file and
> wants it to be recreated, or he wants a current date/time stamp on the file.

Yes, you are probably right; I copied the behaviour from Emacs.

> - The linux version scans quite aggressively for font files. In my case,
> it found Truetype versions of the "CM" fonts in a MATLAB for Windows
> directory mounted in the Linux system. It then preferred these files to
> the Type1 CM-fonts. Unfortunately, the TTF fonts used a different
> character encoding (they were created to be used my MATLAB only), which
> then lead to spurious character replacements (for example, the "rubber"
> left and right brackets were rendered incorrectly for just a specific
> size range, causing me headaches for several days).

Hmm, that is annoying indeed. I probably have to look at some standard
locations first. Anyway, we have to get out of the TeX font business
as soon as possible and move to a more standard font selection program.

> - Another font related question: With the current version, my thesis
> converts to PDF beautifully, with the exception of a few composite
> characters, like ":=", which still seem to get included as raster fonts.
> Is there a way to work around that?

I am afraid not, since these characters are virtual.
One might hack the way they print for the superposed characters,
but not as easily for other transformations (rotations, clippings, etc.).

Best wishes, Joris



Archive powered by MHonArc 2.6.19.

Top of page