[tex-live] extractbb can not read the pdf file generated by mutool

ABE Noriyuki abenori at math.sci.hokudai.ac.jp
Wed Jan 6 10:55:06 CET 2016


Dear all,

I see. Now my question is

* should extractbb strictly reject an unexpected input?

I think Hirata-san's answer is "yes". If the answer is "yes", then I 
agree with that we can't use mfgets. But the current extractbb (without 
my patch) accepts an unexpected input.
This is a problem of the policy of dvipdfmx.


2016/01/06(Wed) 06:50:15, Shunsaku Hirata <shunsaku.hirata74 at gmail.com>:
> Hi,
> 
> > Do you mean that, for example, "x ref" or "xre\0f" should be also
> > accepted? Otherwise, I think mfgets works.
> 
> No, both "x ref" and "xre\0f" are invalid.
> 
> mfgets() is supposed to return a pointer to null-terminated byte strings so
> it should not be used for reading data which can contain null characters
> within that.
> 
> Consider the following example,
> 
> xref
> \0 0 2
> 0000000000 00001 f
> 0000000032 00000 n
> 2 1 foo
> 0000000182 00000 n
> 3 1\0bar
> 0000000322 00000 n
> 4\01
> 0000001236 00000 n
> 
> The first subsection is valid, the second subsection is not, and the
> third is not valid too. The last one is subtle. (I think it is invalid but
> if you want to treat null characters just like spaces it might be valid)
> 
> I think the proposed patch won't treat all cases properly.
> 
> 
> Thanks,
> 
> Shunsaku Hirata

Best,
Noriyuki

-- 
abenori at math.sci.hokudai.ac.jp
Department of Mathematics, Hokkaido University


More information about the tex-live mailing list