Re: using multiple page sequences
On Thu, 2004-03-11 at 22:02, Fleetwood, Brett (SSABSA) wrote: Hi all, Quick question: Have you increased the memory available to your JVM ? If you are using the JVM default, your easiest solution might be to increase the memory available. This depends on your system having enough real memory to support it and varies by JVM vendor. So what is your environment ? OS, real memory available, Java Version ? What is your command line used to invoke Fop ? I am having out of memory issues when transforming my FO - PDF using fop-0.20.5rc2. I read that using multiple page sequences in the XSL and There is a production release of 0.20.5 therefore in the FO means that fop will release some memory. I don't see how I can do this. My XML file is generated dynamically from a database so I don't know how big it will be. Is there any solutions I can use that uses multiple page sequences or possible change the xml structure, if not is there another FOP transformer that isn't as memory intensive. Thanks in advanced for any help. If I havn't provided enough info please ask and I can get back to you. See above -- John Austin [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: using multiple page sequences
Thanks for you response John, hopefully I can answer your questions. From what I understand, we have a web-based application that uses Linux Apache-Tomcat as the web-server to do the rendering. This is a decent machine and we have allocated extra memory (approx 1Gb) but being a web-app possibly many users can request data to be rendered into PDF at the same time therefore I need to keep things as small as possible. I'm not sure throwing more resources at it will ultimately fix it, I think that the actual XSL needs to change or possibly use a different rendorer. Has anyone had experience in setting up a machine to solely perform transformations? -Original Message- From: John Austin [mailto:[EMAIL PROTECTED] Sent: Friday, 12 March 2004 1:01 PM To: [EMAIL PROTECTED] Subject: Re: using multiple page sequences On Thu, 2004-03-11 at 22:02, Fleetwood, Brett (SSABSA) wrote: Hi all, Quick question: Have you increased the memory available to your JVM ? If you are using the JVM default, your easiest solution might be to increase the memory available. This depends on your system having enough real memory to support it and varies by JVM vendor. So what is your environment ? OS, real memory available, Java Version ? What is your command line used to invoke Fop ? I am having out of memory issues when transforming my FO - PDF using fop-0.20.5rc2. I read that using multiple page sequences in the XSL and There is a production release of 0.20.5 therefore in the FO means that fop will release some memory. I don't see how I can do this. My XML file is generated dynamically from a database so I don't know how big it will be. Is there any solutions I can use that uses multiple page sequences or possible change the xml structure, if not is there another FOP transformer that isn't as memory intensive. Thanks in advanced for any help. If I havn't provided enough info please ask and I can get back to you. See above -- John Austin [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: using multiple page sequences
Fleetwood, Brett (SSABSA) schrieb: I read that using multiple page sequences in the XSL and therefore in the FO means that fop will release some memory. The ML archives seem to indicate that, but I'm unsure how significant the savings would be. It might not be worth the effort. I'm currently looking in this area for our newer version (1.0) and hopefully band-aids such as these will not be necessary in the future. Is there any solutions I can use that uses multiple page sequences or possible change the xml structure, if not is there another FOP transformer that isn't as memory intensive. RenderX and AntennaHouse are the main commercial implementations. fo:layout-master-set fo:simple-page-master master-name=A4 /fo:simple-page-master /fo:layout-master-set fo:page-sequence master-reference=contents Hmmm...I guess I'm missing something here--doesn't the master reference (contents) need to refer to the master-name of the simple-page-master (A4)? Glen - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: using multiple page sequences
Cheers Glen, Are you on the FOP dev team? Is 1.0 far away? I'll look into these other implementations. You are right the XSL won't work, I was originally playing around with page-sequence-master master-name = contents but removed it as it wasn't useful, and obviously forgot to change the rest. Thanks again. bF. -Original Message- From: Glen Mazza [mailto:[EMAIL PROTECTED] Sent: Friday, 12 March 2004 3:07 PM To: [EMAIL PROTECTED] Subject: Re: using multiple page sequences Fleetwood, Brett (SSABSA) schrieb: I read that using multiple page sequences in the XSL and therefore in the FO means that fop will release some memory. The ML archives seem to indicate that, but I'm unsure how significant the savings would be. It might not be worth the effort. I'm currently looking in this area for our newer version (1.0) and hopefully band-aids such as these will not be necessary in the future. Is there any solutions I can use that uses multiple page sequences or possible change the xml structure, if not is there another FOP transformer that isn't as memory intensive. RenderX and AntennaHouse are the main commercial implementations. fo:layout-master-set fo:simple-page-master master-name=A4 /fo:simple-page-master /fo:layout-master-set fo:page-sequence master-reference=contents Hmmm...I guess I'm missing something here--doesn't the master reference (contents) need to refer to the master-name of the simple-page-master (A4)? Glen - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: using multiple page sequences
Yes, 1.0 is far away, but we keep working on it. Every day it's getting better formed. Glen Cheers Glen, Are you on the FOP dev team? Is 1.0 far away? I'll look into these other implementations. You are right the XSL won't work, I was originally playing around with page-sequence-master master-name = contents but removed it as it wasn't useful, and obviously forgot to change the rest. Thanks again. bF. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: using multiple page sequences
Hi Brett, It's not about faith, spanning datas across multiple page sequences brings an huge memory saving. We solve issue of constant OutOfMemory errors (according to heap allocation) going to about 2 MB memory consumption on fully written 100 pages documents. You have to identify where in your document could be inserted a page break or decide to break intentionally after N rows of datas (1 record on your XML file can be M rows on document so M*N is the total rows on document) In the second case put logic for break when ciclying your datas templates using something like this ($STOP is max number of data rows for page): xsl:for-each select=DATA_ROW[position() mod $STOP = 1] fo:page-sequence master-reference=YOUR_MASTER_REFERENCE initial-page-number=auto fo:flow flow-name=xsl-region-body fo:table table-layout=fixed width=480pt fo:table-column column-width=85pt/ fo:table-column column-width=15pt/ fo:table-body fo:table-row line-height=30pt fo:table-cell fo:block text-align=center font-size=10pt font-family=Couriertitle 1/fo:block /fo:table-cell fo:table-cell fo:block text-align=center font-size=10pt font-family=Couriertitle 2/fo:block /fo:table-cell /fo:table-row /fo:table-body /fo:table xsl:for-each select=.|following-sibling::DATA_ROW[position() lt; $STOP] xsl:apply-templates select=./ /xsl:for-each /fo:flow /fo:page-sequence /xsl:for-each xsl:template match=DATA_ROW fo:table table-layout=fixed width=480pt fo:table-column column-width=85pt/ fo:table-column column-width=10pt/ fo:table-body fo:table-row line-height=13pt fo:table-cell fo:block text-align=left font-size=10pt font-family=Courier xsl:value-of select=YOUR_ROW_KEY_1/ /fo:block /fo:table-cell fo:table-cell fo:block text-align=center font-size=10pt font-family=Courier xsl:value-of select=YOUR_ROW_KEY_2/ /fo:block /fo:table-cell /fo:table-row /fo:table-body /fo:table /xsl:template Hope this help fabrizio At 02.32 12/03/2004, you wrote: Hi all, I am having out of memory issues when transforming my FO - PDF using fop-0.20.5rc2. I read that using multiple page sequences in the XSL and therefore in the FO means that fop will release some memory. I don't see how I can do this. My XML file is generated dynamically from a database so I don't know how big it will be. Is there any solutions I can use that uses multiple page sequences or possible change the xml structure, if not is there another FOP transformer that isn't as memory intensive. Thanks in advanced for any help. If I havn't provided enough info please ask and I can get back to you. Defend your privacy! Encrypt your email today! http://www.bytefusion.com/secexmail - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: using multiple page sequences
On Thu, 2004-03-11 at 23:32, Fleetwood, Brett (SSABSA) wrote: Thanks for you response John, hopefully I can answer your questions. From what I understand, we have a web-based application that uses Linux Apache-Tomcat as the web-server to do the rendering. This is a decent machine and we have allocated extra memory (approx 1Gb) but being a web-app possibly many users can request data to be rendered into PDF at the same time therefore I need to keep things as small as possible. I'm not sure throwing more resources at it will ultimately fix it, I think that the actual XSL needs to change or possibly use a different rendorer. Has anyone had experience in setting up a machine to solely perform transformations? I asked about memory on the machine and about the command line arguments used to invoke your environment as a sanity-check to make sure that your Java VM (in this case Tomcat ?) is configured with a large enough heap to support your program. It could explain why your application runs out of heap. There are reasons WHY it requires this much memory, that we don't want to go in to here. Simply plugging in a few memory cards isn't enough. Java has to know the size of the heap (free store) that it is allowed to create. For the Sun Java, use the commands: java -j and java -X to see the options. In Tomcat 5, you could use the environment variable JAVA_OPTS to pass in a value for -Xmx, say JAVA_OPTS=-Xmx800M to allow an 800 Mb heap. This isn't a perfect solution, but it could help you get to the next level and keep you alive while you consider your commercial options. -- John Austin [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
