Bug 40120 - font-base wrongly interpreted when embedding font and input in folder
Summary: font-base wrongly interpreted when embedding font and input in folder
Status: CLOSED FIXED
Alias: None
Product: Fop - Now in Jira
Classification: Unclassified
Component: general (show other bugs)
Version: 0.92
Hardware: PC other
: P2 normal
Target Milestone: ---
Assignee: fop-dev
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-07-26 18:29 UTC by Jeroen
Modified: 2012-04-01 06:25 UTC (History)
0 users



Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Jeroen 2006-07-26 18:29:12 UTC
Environment: jre1.5.0_06, any fo (or xml-xsl) file rendered to pdf using a non
standard and embedded font.
Under the following conditions:

- NOT specifying a font-base in the userconfig
- Specifying base as "." (probably irrelevant)
- Embedding a font through userconfig.xml, assuming the current directory for
both the .ttf as well as the metrics xml, ie embed="file:gothic.ttf" and of
course those font files sitting there
- Actually using that font in the fo
- the fo (or xml in a xml/xsl setup) being in a separate folder (ie fop "-fo
store/my.fo"). The file being specified relatively to the current directory is
probably irrelevant.

the resulting pdf is generated without complained, but actually does not have
the .ttf embedded, resulting in a font not found when opening the pdf. Further
investigating showed if the .ttf is ALSO copied to the relative path of the
source .fo (or xml), it DOES embed in the pdf.

Main problem is the renderer does not complain, combined with the fop
documentation stating "(font-base) specifies the base URL based on which
relative font URLs will be resolved. If not specified defaults to the base URL
above", which is obvisouly not true when the source file is in a folder.

Workaround is to specify a font-base in the userconfig, even if it's a simple
".". Refering to the .ttf in an absolute way worked too (eeww).
Comment 1 Jeroen 2006-07-26 18:31:41 UTC
excuse me, where it says

embed="file:gothic.ttf"

please read

embed-url="file:gothic.ttf"
Comment 2 Jeroen 2006-11-30 05:47:20 UTC
Erm erm erm, any news on this one? Tx /Jeroen
Comment 3 Simon Pepping 2006-11-30 12:32:41 UTC
(In reply to comment #1)
> embed-url="file:gothic.ttf"
Is this a valid URI? I think it should read either embed-url="gothic.ttf" or 
embed-url="file:/path/to/gothic.ttf".
Comment 4 Jeroen 2006-11-30 13:55:18 UTC
Oh gosh, that solved the problem too, just as my workaround did. I keep on being
confused about url's and uri's, esp when it is file based. Thank you.

But..... I still think it's a bug as using the "file:xxx.ttf" form it will
generate a pdf without any complaint and the resulting file doesn't work. There
is probably a semantic explanation why that behavior is completely valid, but
still. Anyway, no big deal as the issue can be fixed/worked around. I humbly
suggest at least an error message if the format given in this specific scenario
leads to a crippled PDF file.

Thanks Simon for pointing out! Won't change the status as I don't know if the
developers will incorporate that.
Comment 5 Simon Pepping 2006-12-03 12:50:49 UTC
(In reply to comment #4)
> Oh gosh, that solved the problem too, just as my workaround did. I keep on being
> confused about url's and uri's, esp when it is file based. Thank you.
> 
> But..... I still think it's a bug as using the "file:xxx.ttf" form it will
> generate a pdf without any complaint and the resulting file doesn't work. There
> is probably a semantic explanation why that behavior is completely valid, but
> still. Anyway, no big deal as the issue can be fixed/worked around. I humbly
> suggest at least an error message if the format given in this specific scenario
> leads to a crippled PDF file.

It is a bug if "file:xxx.ttf" is silently accepted and leads to a useless result
file. It is an erroneous URI and should be signalled as such.
Comment 6 Adrian Cumiskey 2007-02-01 07:36:57 UTC
This bug should be fixed/addressed by [PATCH]
http://issues.apache.org/bugzilla/show_bug.cgi?id=41514
Comment 7 Vincent Hennebert 2007-02-14 06:30:10 UTC
Should be fixed in rev. 507539.
Comment 8 Glenn Adams 2012-04-01 06:25:39 UTC
batch transition pre-FOP1.0 resolved+fixed bugs to closed+fixed