![](https://secure.gravatar.com/avatar/d0e94fb08d4620cfbd2f8f59a7868637.jpg?s=120&d=mm&r=g)
Maybe one further element that would suggest Context is using a cached copy of the tfm-file instead of the newly created file: when I move the file Myfont.tfm out of the $home/texmf/fonts/tfm directory, pdftex gives me the expected error "! I can't find file `Myfont'." However, Context will still happily produce output--with an outdated copy of the tfm! So context must be caching it somewhere, and I need to flush that cache. Anyone an idea how this can be done? Thanks Thomas
![](https://secure.gravatar.com/avatar/49e63acb01f2ca80efce7eed08310ce8.jpg?s=120&d=mm&r=g)
At 12:46 05/10/2003, you wrote:
Maybe one further element that would suggest Context is using a cached copy of the tfm-file instead of the newly created file: when I move the file Myfont.tfm out of the $home/texmf/fonts/tfm directory, pdftex gives me the expected error "! I can't find file `Myfont'." However, Context will still happily produce output--with an outdated copy of the tfm! So context must be caching it somewhere, and I need to flush that cache. Anyone an idea how this can be done?
maybe issuing the command "mktexlsr" helps Hans
![](https://secure.gravatar.com/avatar/d0e94fb08d4620cfbd2f8f59a7868637.jpg?s=120&d=mm&r=g)
On Tuesday, October 7, 2003, at 09:12 AM, Hans Hagen wrote:
maybe issuing the command "mktexlsr" helps
Well, I solved my immediate problem by renaming the tfm file to MyFontNew.tfm and editing the appropriate line in the map file. But I'm still curious what caused the problem. No, mktexlsr wouldn't help since all the files are in my local tree which doesn't use ls-R indexes. I had a dim impression that Context was kinda stubborn before: I had tried something new, given up and switched off the computer. The next day--bamm, the thing would work without a hitch. So, is Context using some internal caches for stuff like this?
![](https://secure.gravatar.com/avatar/ce8f97ad3112ba0b6bf9fdc3d3fbb5af.jpg?s=120&d=mm&r=g)
On Sun, Oct 05, 2003 at 12:46:50PM +0200, Thomas A.Schmitz wrote:
Maybe one further element that would suggest Context is using a cached copy of the tfm-file instead of the newly created file: when I move the file Myfont.tfm out of the $home/texmf/fonts/tfm directory, pdftex gives me the expected error "! I can't find file `Myfont'." However, Context will still happily produce output--with an outdated copy of the tfm!
Are you sure it is the TFM file that is out of date? Or might it be a PK font? If it is the latter, you can simply find the PK font cache (often something like /var/lib/texmf/fonts/... on Linux systems) and delete any suspicious files. -- Matt Gushee When a nation follows the Way, Englewood, Colorado, USA Horses bear manure through mgushee@havenrock.com its fields; http://www.havenrock.com/ When a nation ignores the Way, Horses bear soldiers through its streets. --Lao Tzu (Peter Merel, trans.)
![](https://secure.gravatar.com/avatar/d0e94fb08d4620cfbd2f8f59a7868637.jpg?s=120&d=mm&r=g)
Are you sure it is the TFM file that is out of date? Or might it be a PK font?
No, it is a (postscript) font I'm creating myself, so there is no pk-file. I'm having the same trouble again this morning: edited pfb-file, saved, new file is used in pdftex testfont, Context is still using the old version. Thomas
![](https://secure.gravatar.com/avatar/6af1f724a401c9f6f746a973f33ff595.jpg?s=120&d=mm&r=g)
Hi,
No, it is a (postscript) font I'm creating myself, so there is no pk-file. I'm having the same trouble again this morning: edited pfb-file, saved, new file is used in pdftex testfont, Context is still using the old version.
... of what? the tfm file or the pfb file? ConTeXt does not cache anything. So the probability that your system is misconfigured is > 0. I've forgotten: what system are you using? Patrick -- You are your own rainbow!
![](https://secure.gravatar.com/avatar/49e63acb01f2ca80efce7eed08310ce8.jpg?s=120&d=mm&r=g)
At 11:07 08/10/2003, you wrote:
Are you sure it is the TFM file that is out of date? Or might it be a PK font?
No, it is a (postscript) font I'm creating myself, so there is no pk-file. I'm having the same trouble again this morning: edited pfb-file, saved, new file is used in pdftex testfont, Context is still using the old version.
you can use kpsewhich to see what font pdftex finds first, it's probably a file search problem (did you run mktexlsr?) Hans
![](https://secure.gravatar.com/avatar/d0e94fb08d4620cfbd2f8f59a7868637.jpg?s=120&d=mm&r=g)
\startembarrassment I found out what caused the strange behavior. I was testing the font in a directory where there was also an old instance of the .tfm-file. So Context would search the working directory, find everything it needed to produce the font, and thus neglect all the new stuff I had modified and installed into $home/texmf. Performing the tests in a different directory removed all the unexplicable behavior. \stopembarrassment Thanks to all for looking into this, and sorry for the confusion I created. On an unrelated note: please forgive if I seem sluggish in responding to posts. For some reason I can't figure out (and Erik looked into this and tried to help), some of my posts take three days or even longer before they get posted.
![](https://secure.gravatar.com/avatar/49e63acb01f2ca80efce7eed08310ce8.jpg?s=120&d=mm&r=g)
At 13:23 11/10/2003, you wrote:
\startembarrassment I found out what caused the strange behavior. I was testing the font in a directory where there was also an old instance of the .tfm-file. So Context would search the working directory, find everything it needed to produce the font, and thus neglect all the new stuff I had modified and installed into $home/texmf. Performing the tests in a different directory removed all the unexplicable behavior. \stopembarrassment
in texmf.cnf you can see how paths are configured, some paths indeed have . as first path to consult; in general, context itself will look for source files in ./ ../. ../../. tex input paths but fonts and alike are out of context's control Hans
participants (4)
-
Hans Hagen
-
Matt Gushee
-
Patrick Gundlach
-
Thomas A.Schmitz