[Revisited] Memory Limitation Question

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   This is a reprise of a question I asked last year, but to my embarrasement I can't remember the answer given then, or find it on search.  So, as it looks like there've been a few developments since then anyway, I thought I'd cross my fingers and repeat it...

We have a program that uses QuickPDF to merge and bookmark a number (possibly a large number) of smallish PDF documents.  The merging's done with MergeFiles and works fine for any number of documents, but the bookmarking requires the merged document to be loaded into memory and this can fail if things turn out to be too large.

The size limit above which the loading fails appears to be independent of actual memory installed on a PC, so presumably is determined by a hardcoded buffer size in the QuickPDF code.

As we have a non-source version (site licence) for QuickPDF I can't investigate this myself, so is anyone able to confirm whether this is the case?  If so, does increasing the limit look like it'll be an easy thing to do?

If anyone wants to/can send me _just_ the relevant files or code fragments needed to look into this issue (don't want to turn into a pirate!) I'd be happy to do the investigation, and report back with the findings.

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
IanK Members Profile Find Members Posts Beginner Joined: 30 Nov 05 Status: Offline Points: 8	Post Options Post Reply Quote IanK Report Post Thanks(0) Quote Reply Topic: [Revisited] Memory Limitation Question Posted: 17 May 06 at 9:40AM
	This is a reprise of a question I asked last year, but to my embarrasement I can't remember the answer given then, or find it on search. So, as it looks like there've been a few developments since then anyway, I thought I'd cross my fingers and repeat it... We have a program that uses QuickPDF to merge and bookmark a number (possibly a large number) of smallish PDF documents. The merging's done with MergeFiles and works fine for any number of documents, but the bookmarking requires the merged document to be loaded into memory and this can fail if things turn out to be too large. The size limit above which the loading fails appears to be independent of actual memory installed on a PC, so presumably is determined by a hardcoded buffer size in the QuickPDF code. As we have a non-source version (site licence) for QuickPDF I can't investigate this myself, so is anyone able to confirm whether this is the case? If so, does increasing the limit look like it'll be an easy thing to do? If anyone wants to/can send me _just_ the relevant files or code fragments needed to look into this issue (don't want to turn into a pirate!) I'd be happy to do the investigation, and report back with the findings.

Wheeley Members Profile Find Members Posts Senior Member Joined: 30 Oct 05 Location: United States Status: Offline Points: 146	Post Options Post Reply Quote Wheeley Report Post Thanks(0) Quote Reply Posted: 17 May 06 at 4:41PM
	If you break the bookmarking into stages of around 500 at a time it will work. I had the same issue you are having. I believe it is some memory leak with the bookmarking process. So here is what we did. 1. Load pdf 2. do 500 bookmarks 3. save and remove file from memory 4. start at 1 again This will work. Using this process I was able to create a 32 MB pdf file created with 3000+ bookmarks. Yes this is an extreame pdf, but sometimes things like this happen in the real word. Give it a try Wheeley

IanK Members Profile Find Members Posts Beginner Joined: 30 Nov 05 Status: Offline Points: 8	Post Options Post Reply Quote IanK Report Post Thanks(0) Quote Reply Posted: 18 May 06 at 5:11AM
	Thanks for the suggestion Wheeley, but I'm afraid this isn't applicable in my particular case. For these the files are mid-100's of MB large, and are too big for KiSEDQuickPDF.LoadFromFile to load in the first place. Ideally QuickPDF would start using virtual memory, so arbitrarily large files could be opened and bookmarked, but as things actually seem to be (a fixed upper limit on file size) I'd settle for simply increasing that upper limit to something more appropriate to present day 1GB+ systems.