font havoc

font havoc

[Hans van der Meer]

Suddenly it seems, I am getting bad messages for quite normal math  
fonts:

Warning: pdfetex (file cmsy8): kpathsea: Running mktexpk --mfmode / -- 
bdpi 72 --mag 1+7/72 --dpi 79 cmmi10
mktexpk: Mismatched mode ljfzzz and resolution 72; ignoring mode.
mktexpk: Can't guess mode for 72 dpi devices.
mktexpk: Use a config file, or update me.  (<<<< Why is this?)
Font cmsy8 at 72 not found

Warning: pdfetex (file cmmi10): Font cmmi10 at 79 not found
{/usr/local/teTeX/share/texmf-dist/fonts/enc/dvips/psnfss/8r.enc}</ 
usr/local/te
TeX/share/texmf-dist/fonts/type1/urw/helvetic/uhvr8a.pfb></usr/local/ 
teTeX/shar
e/texmf-dist/fonts/type1/urw/helvetic/uhvb8a.pfb></usr/local/teTeX/ 
share/texmf-
dist/fonts/type1/urw/courier/ucrr8a.pfb>

I am not aware having changed to resolution 72 for ljfzzz or doing  
something different from previous installations of new context  
versions. The mode for ljfzzz was set with texconfig.
I was running the 2005-10-27 context version, but reverting tot  
2005-08-31 did not solve the problem.
What has happened to my fonts?
Has it something to do with the fact that tetex and context seem to  
use directories .texmf-config and .texmf-var differently?
Why does something that ran without a hitch suddenly is broken?

Does someone know how to restore things to normal?

Hans van der Meer

Re: font havoc

[Taco Hoekwater]

Hans van der Meer wrote:
> Suddenly it seems, I am getting bad messages for quite normal math  fonts:
> 
> Warning: pdfetex (file cmsy8): kpathsea: Running mktexpk --mfmode / -- 
> bdpi 72 --mag 1+7/72 --dpi 79 cmmi10

Most likely cause: pdftex cannot find the map file that contains
the cmmi10 entry, and therefore assumes it is a bitmap font. In my
system, there are cmmi10 lines in the following three map files:

   fonts/map/pdftex/context/context-base.map
   fonts/map/pdftex/context/original-ams-cmr.map
   fonts/map/pdftex/context/original-public-lm.map

Either pdftex cannot find those (the third one is normally the
used one), or it loads an incorrect one.

In any case, you should check the 'map file' part of context and
pdftex's messages. Or you can try posting the log of the run

   \starttext Hello World!\stoptext

if you can't figure out the messages yourself.

Cheers,
Taco

Re: font havoc

[Hans Hagen]

Taco Hoekwater wrote:

> Hans van der Meer wrote:
>
>> Suddenly it seems, I am getting bad messages for quite normal math  
>> fonts:
>>
>> Warning: pdfetex (file cmsy8): kpathsea: Running mktexpk --mfmode / 
>> -- bdpi 72 --mag 1+7/72 --dpi 79 cmmi10
>
>
> Most likely cause: pdftex cannot find the map file that contains
> the cmmi10 entry, and therefore assumes it is a bitmap font. In my
> system, there are cmmi10 lines in the following three map files:
>
>   fonts/map/pdftex/context/context-base.map
>   fonts/map/pdftex/context/original-ams-cmr.map
>   fonts/map/pdftex/context/original-public-lm.map
>
> Either pdftex cannot find those (the third one is normally the
> used one), or it loads an incorrect one.
>
> In any case, you should check the 'map file' part of context and
> pdftex's messages. Or you can try posting the log of the run
>
>   \starttext Hello World!\stoptext
>
> if you can't figure out the messages yourself.


btw, in the tui file you can see what map files are expected (loaded)

Hans

Re: font havoc

[Hans Hagen]

Hans van der Meer wrote:

> Suddenly it seems, I am getting bad messages for quite normal math  
> fonts:
>
> Warning: pdfetex (file cmsy8): kpathsea: Running mktexpk --mfmode / -- 
> bdpi 72 --mag 1+7/72 --dpi 79 cmmi10
> mktexpk: Mismatched mode ljfzzz and resolution 72; ignoring mode.
> mktexpk: Can't guess mode for 72 dpi devices.
> mktexpk: Use a config file, or update me.  (<<<< Why is this?)
> Font cmsy8 at 72 not found
>
> Warning: pdfetex (file cmmi10): Font cmmi10 at 79 not found
> {/usr/local/teTeX/share/texmf-dist/fonts/enc/dvips/psnfss/8r.enc}</ 
> usr/local/te
> TeX/share/texmf-dist/fonts/type1/urw/helvetic/uhvr8a.pfb></usr/local/ 
> teTeX/shar
> e/texmf-dist/fonts/type1/urw/helvetic/uhvb8a.pfb></usr/local/teTeX/ 
> share/texmf-
> dist/fonts/type1/urw/courier/ucrr8a.pfb>
>
> I am not aware having changed to resolution 72 for ljfzzz or doing  
> something different from previous installations of new context  
> versions. The mode for ljfzzz was set with texconfig.
> I was running the 2005-10-27 context version, but reverting tot  
> 2005-08-31 did not solve the problem.
> What has happened to my fonts?
> Has it something to do with the fact that tetex and context seem to  
> use directories .texmf-config and .texmf-var differently?
> Why does something that ran without a hitch suddenly is broken?
>
> Does someone know how to restore things to normal?

probably due to the changes in latin modern map/enc naming; older 
versions don't mix well currently

what puzzles me is:

pdfetex (file cmsy8): kpathsea: Running mktexpk --mfmode / -- bdpi 72 
--mag 1+7/72 --dpi 79 cmmi10
                     
^^^^^                                                                                                                                
^^^^^^^

which are diferent fonts

is automapfile enabled on your system?

Hans

Re: font havoc

[Taco Hoekwater]

Hans Hagen wrote:
> what puzzles me is:
> 
> pdfetex (file cmsy8): kpathsea: Running mktexpk --mfmode / -- bdpi 72 
> --mag 1+7/72 --dpi 79 cmmi10
>                     
> ^^^^^                                                                                                                                
> ^^^^^^^
> 
> which are diferent fonts

That is just a timing error. What you see is the stdout and stderr
messages from two separate programs intermixed.

Taco

Re: font havoc

[Hans Hagen]

Taco Hoekwater wrote:

> Hans Hagen wrote:
>
>> what puzzles me is:
>>
>> pdfetex (file cmsy8): kpathsea: Running mktexpk --mfmode / -- bdpi 72 
>> --mag 1+7/72 --dpi 79 cmmi10
>>                     
>> ^^^^^                                                                                                                                
>> ^^^^^^^
>>
>> which are diferent fonts
>
>
> That is just a timing error. What you see is the stdout and stderr
> messages from two separate programs intermixed.

ok,

my guess here is that cm is used in the definitions while lm should be 
used (the cm -> lm mapping file is not in the zip, but i added it again 
for good old times sake)

i'll upload a new distribution later today which has more redundant lm 
map definitions; better play safe and not let context users suffer from 
all those intermediate changes in lm related files -)

Hans