Ouch, that could explain it. No, no changes in that area. Actually, images could be written to the file immediately and then released instead of having to wait until the next page-sequence is finished. Should be easy to fix. I see no reason to defer writing the image to the PDF.
On 21.06.2006 22:24:44 J.Pietschmann wrote: > Jeremias Maerki wrote: > > Given the source code, this should actually work with PNGs, too, because > > the image cache is able to release memory through the use of a > > WeakHashMap. The PDFRenderer signals the image cache immediately that it > > doesn't need the reference to the image anymore when it has embedded the > > image in the PDF. But it seems that somewhere along the road something's > > still going wrong. Nice task for filling a boring, rainy Sunday.... :-) > > Last time I checked, there was a pointer from the PDF XObject > to the raw image data in the cache, and the XObject was held > until the object dictionary was written, after all the content > from the FO. I.e. no chance to release the image data until > basically the whole PDF was complete (font embedding might > have come afterwards, I didn't check). I didn't track changes > to this code, have there been changes in this area? Jeremias Maerki --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
