Re: can fop handle this (memory question)
Jeremias Maerki schrieb: On 16.03.2006 00:17:28 Paul Tremblay wrote: On Wed, Mar 15, 2006 at 09:22:36AM +0100, Jeremias Maerki wrote: It's always difficult to estimate if there will be a problem. Trying it out will help a lot. :-) Anyway, given that the latest code hasn't received much attention on memory consumption, yet, there's a good chance that your document might not work. Having a lot of RAM (and an Hey Jeremias, you're way too pessimistic here, in fact I just tried a document with 120 pages of text, a TOC at the beginning and a table with 99 512x512 images (~200k each) at the end and it worked without a problem. I've used JDK1.4 (without jvm tuning) and the result I got after 1:15 min was a 20MB PDF. So if you are using jpeg as image format and PDF as output format FOP should handle your document (In any case please report back your result ;-) -- Christian - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: can fop handle this (memory question)
On Wed, Mar 15, 2006 at 09:22:36AM +0100, Jeremias Maerki wrote: It's always difficult to estimate if there will be a problem. Trying it out will help a lot. :-) Anyway, given that the latest code hasn't received much attention on memory consumption, yet, there's a good chance that your document might not work. Having a lot of RAM (and an increased VM size) helps a lot which you obviously don't really have. The images are probably less of a problem than the forward references. FOP has a special area tree model which can serialize pages to disk but I've only fixed it, not done any extensive tests with it. It also needs Java knowledge to activate it. In the end, if your TeX solution works, use it. There's a good chance that you run into the problems you noted below with bigger documents and the latest FOP release. Sorry to sound ignorant, but what do you mean by VM size and Java knowledge? I may convert the document anyway if I have time and then report back here and let you know. In the future, do you envision FOP being able to handle longer documents? Paul -- *Paul Tremblay * [EMAIL PROTECTED]* - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: can fop handle this (memory question)
On 16.03.2006 00:17:28 Paul Tremblay wrote: On Wed, Mar 15, 2006 at 09:22:36AM +0100, Jeremias Maerki wrote: It's always difficult to estimate if there will be a problem. Trying it out will help a lot. :-) Anyway, given that the latest code hasn't received much attention on memory consumption, yet, there's a good chance that your document might not work. Having a lot of RAM (and an increased VM size) helps a lot which you obviously don't really have. The images are probably less of a problem than the forward references. FOP has a special area tree model which can serialize pages to disk but I've only fixed it, not done any extensive tests with it. It also needs Java knowledge to activate it. In the end, if your TeX solution works, use it. There's a good chance that you run into the problems you noted below with bigger documents and the latest FOP release. Sorry to sound ignorant, but what do you mean by VM size and Java knowledge? VM size: The Java VM provides only 64MB of RAM to the application by default even if you have more RAM available. The -Xmx option for the Java VM can be used to increase the maximum available memory: http://xmlgraphics.apache.org/fop/latest/running.html#memory Java knowledge means you need Java programming skills to activate the special component that can temporarily off-load pages to disk. It is not yet properly intergrated in FOP and tested so everyone can use it. I may convert the document anyway if I have time and then report back here and let you know. In the future, do you envision FOP being able to handle longer documents? Sure, as soon as people start actively improving FOP in this direction. My own focus is on business documents where memory consumption is less of a problem. Jeremias Maerki - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
can fop handle this (memory question)
I want know if FOP will run out of memory if I try to convert a thesis to PDF. I will be using .91 beta and am running java 1.4.1. I am using linux on an older box with a speed of 400 MHZ and ram of 500 Mb. The document will be about 80 pages long. It will have a table of contents at the beginning. The appendix will contain a table of 100 rows, with each row containing a small jpg graphic, probably around 100 Kb in size. Right now I am using TeX to convert the document. I would like to use FOP instead, but before I do all the work of changing the XSLT stylesheets, I want to know if FOP can handle the memory. I know that a table of contents at the beginning uses a lot of memory, right? Doesn't a table of contents require FOP to hold the whole document in memory? Likewise, I am not sure if the number of images (though small) will require too much memory. Thanks Paul -- *Paul Tremblay * [EMAIL PROTECTED]* - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
