Some more information... We fell back and faxed the document, which uses a soon to be obsolete part of our system. This generates the document in Powerbuilder, creates the file as a TIFF and then faxes the TIFF. The resulting document is 5 pages long, other 5 page documents we create are about 55K. We captured the TIFF and converted it to PDF (with PDFWriter) and created a file 138K large.
-----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Sternbergh, Cornell Sent: Wednesday, July 21, 2004 14:24 To: [EMAIL PROTECTED] Subject: [iText-questions] PDF file on steroids Good Afternoon This afternoon we had a problem. We created a 38 Meg PDF which shouldn't be more than 136K. We have a Java process which reads data from a database and formats into a PDF. In the past 9 days we've created 9,700 documents, the largest is 136K, the smallest is 24K. 38M is way out of line. We use iText to generate the PDF. As we've created more than 120,000 documents with this process without a hitch, we don't expect so much a coding problem as some data related problem. And I need to get a lead on where to start. This came to our attention because a second process, which prints or emails the PDF, got hung up on this file, and distribution of other files got held up. The file size was approximately 37.9M. We tried to open the PDF with Adobe Acrobat, which claims the file needs repair and then says it can't be repaired. We deleted the document from the queue and deleted the file, and the process continuted handling other files. Someone requested that this document be re-printed, which caused the process to recreate it. This time, it was approximately 38M. First we note that we've created two PDF's from the same data, but the size of the resulting files are different, not to mention way too large. Second, we observe, from looking at the data in the database, that this document's data is almost identical to another, successfully generated document, except that one field has more characters in it. This string would be retrieved from the database (DB2) and then put into a cell in a table. Could it be possible that we've blown a length limitation on table cells? Is there a convenient way for a human to understand the contents of a PDF file? We opened it in UltraEdit (a text/hex editor) and note that there are no strings which correspond to the text of the document. I do observe ASCII strings such as (0Ah refers to hex 0A): /Subtype /Image0Ah /Type /XObject0Ah /Filter /FlateDecode0Ah /Width 2830Ah /Height 2980Ah /BitsPerComponent 80Ah /Length 55140Ah /ColorSpace which I assume to be PDF commands or attributes. How are the strings of characters which make up the actual text stored? TIA Cornell Sternbergh ------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_idG21&alloc_id040&op=ick _______________________________________________ iText-questions mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/itext-questions ------------------------------------------------------- This SF.Net email is sponsored by BEA Weblogic Workshop FREE Java Enterprise J2EE developer tools! Get your free copy of BEA WebLogic Workshop 8.1 today. http://ads.osdn.com/?ad_idG21&alloc_id040&op=click _______________________________________________ iText-questions mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/itext-questions
