[reportlab-users] PDF without file

[John J. Lee]

On Tue, 30 May 2006, Robin Becker wrote:
> François Pinard wrote:
>> [Michael Hipp]
[...]
>> Hi, people.  I did not follow the thread and may be fully out of context, 
>> so take me with a few grains of salt :-).
>> 
>> Contents of a CStringIO may not be Unicode (unless Python changed 
>> recently).  Now that Reportlab is Unicode-aware, one may sometimes have to 
>> resort to the slower StringIO.
>> 
> Being in utterly duh Homer mode I completely forgot that there is indeed an 
> existing string output for the code in the canvas.
> Just use
>
> pdf = c.getpdfdata()
[...]

Just by the way, the PDF data itself is actually binary, so the lack of 
unicode support in cStringIO is not relevant.  But Robin's .getpdfdata() 
is better.


John