[geeklog-devel] PDF support, was [Fwd: further span in comments & problem with security listserv signup]

[Tony Bibbs]

Dirk Haun wrote:

> To be a bit more precise:
>
>- generated PDFs not being deleted
>  
>
Should be easy enough to fix.  Main problem, though, is the currently 
implementation gives a timestamp as the name of the PDF.  Sounds fine 
but if the same one is clicked over and over we regenerate copies of the 
PDF.  This was donated code so I'll have to monkey with it.  My 
inclination is to use the SID and append the timestamp to the end of the 
file.  That would give something like 
<some_story_id>_<some_timestamp>.pdf.  The codecan then easily modify 
the timestamp portion on subsequent views so that the purge process works.

>- uses XHTML, even though htmldoc doesn't understand it
>- as a consequence of the above: Do we need the Tidy support then?
>  
>
No.  Not sure WTF I was thinking.

>And, most importantly, it doesn't work - at least not for me:
>  
>
Lol, works for me ;-).  More testing will be needed after the above 
changes are needed.

Vinny, what's your plate looking like?  I'm tied up tonight, might have 
time tomorrow and won't have time after that until Saturday morning.

--Tony