[tex-live] Bug report for updmap

[Philip TAYLOR]

Jean-Marc Bourguet wrote:

> I think I cut and pasted from the PDF showed by `texdoc updmap` (at least I can reproduce it that way)

This is interesting; I had a case of semi-colons that weren't semi-colons
recently. Having created TeX source from the PDF output of Surveymonkey,
I made semi-colon active, and it had no effect on the semi-colons in
the document.  I did not realise until your message that what I was
getting back from PDF might not be a "real" semi-colon at all, and now
that you have alerted me to the fact, I have just extracted one and
asked Richard Ishida's rather nice Unicode utility [1] to tell me what
it is.  Richard's utility says :

	; U+037E GREEK QUESTION MARK

Now why would Surveymonkey use that rather than the real semi-colon
that I entered when creating the survey in the first place ?

Incidentally, a rather worrying event occured when trying to copy
the putative semi-colon from the PDF using Acrobat; each time I tried,
I was security-prompted (UAC) as to whether I wanted to allow Oracle's
"ssvagent" to run; clearly I did not, nor did I have any idea why it
was trying to, so I carried out some research and found :

> ssvagent is run when java detects registry is not correct and then ssvagent tries to recover it.

So, why is Acrobat trying to launch (presumably) Java ?  Knowing
of the recent security breach discovered in Java I had disabled
it in all browsers but did not even think it might also be a plug-in
to Acrobat.  I have now removed it completely from this system (VM)
and will remove it from all the other VMs, the host and the server
immediately.  But are Surveymonkey PDFs infected with something nasty ?

Philip Taylor
--------
[1] http://rishida.net/tools/conversion/