[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#130285: tetex-bin: xdvi cannot display some DVI file



tag 130285 + woody
thanks,

Julian Gilbey <J.D.Gilbey@qmul.ac.uk> writes:

> On Fri, Jun 21, 2002 at 09:58:58AM +0200, Laurent Bonnaud wrote:
>> 
Julian> So looks like he needs to find the font, or that someone
Julian> created the DVI file will a local font which is not generally
Julian> available.
>> 
>> It is even non-free, purchaseable from several web sites.
>> 
Julian> If this is all the bug report is about (haven't checked), then
Julian> we can close it.
>> 
>> How about font substitution ?

> xdvi used to do this.  Hmm.

Yes, it's supposed to replace the font with cmr10. This was a bug
in (the t1lib module of) 22.40f, which was fixed in (some version
before) 22.40v.

The following minimal .dvi file shows the bug:

begin 644 test2.dvi
M]P(!@Y+`'#L``````^@;(%1E6"!O=71P=70@,C`P,RXP-RXP.3HR,S(WBP``
M``$```````````````````````````````````````````````#_____C9_R
M``".H`*#,]J-H/V&S":-D10``/,R2_%@>0`,````"@````)G;MUT97-TCHZ?
M&```C9(`Z&"C\P!+\6!Y``H````*`````F=NJS&.C/@````J`8.2P!P[````
M``/H`ILSV@'5P4<``@`!\S)+\6!Y``P````*`````F=N\P!+\6!Y``H````*
3`````F=N^0```*8"W]_?W]_?WP``
`
end


For this file, 22.40v will report the following:

xdvi: gn at 720 dpi
kpathsea: Running mktexpk --mfmode ljfour --bdpi 600 --mag 1+120/600 --dpi 720 gn
mktexpk: don't know how to create bitmap font for gn.
kpathsea: Appending font creation commands to missfont.log.
xdvi: Can't find font gn; using Type1 version of cmr10 instead.
xdvi: gn at 600 dpi
kpathsea: Running mktexpk --mfmode ljfour --bdpi 600 --mag 1+0/600 --dpi 600 gn
mktexpk: don't know how to create bitmap font for gn.

The bug was that the second use of the same font in a different
size (720 vs. 600 dpi) doesn't trigger a new error condition with
t1lib (note the missing error message for the second loading
attempt), which 22.40f incorrectly interpreted as success,
leading to a later failure to display the font. A workaround for
the error in 22.40f is to use the `-not1lib' option.

Since the bug doesn't occur any more with 22.40v (in sid), I'd
suggest to tag the bug as `woody'.

Best,
Stefan



Reply to: