Manipulation of jpg data

[jbv]

Xavier,

Thanks for the suggestion.
It actually crossed my mind, but I have the feeling that a
simpler solution exists, especially because the original
files I want to re-arrange into a new, bigger file are rather
small (10 x 10 pixels), and they already exist.
And also because new files will have to be created on the
fly, upon cgi requests from a browser...
So I can store the content of the original samll files on disk,
each file containing a serie of pixel values. Then I open the
needed files, organize the content in a variable and save the
result as a new jpg file.

In fact, what I need is a good jpeg compression algo that
I can implement in Transcript.
I'm presently doing a google search, but found only jpeg
compression theory stuff so far...
JB

> JB
>
> Why not have a separate MC/RR instance running on your
> server and periodically do a check for data to process?
>
> When the files are detected or present you process
> them with RR. I have a few background application
> services that do this at work. Works like a charm!