Forum Replies Created
-
Posts
-
#26682
We have included the pd4ml lib in an application and externalized the template of the document itself. Now we have created below document, but it appears that the transform tag does not work; all other tags do work. The main goal we are trying to achieve is to get the page HTML body to fit the PDF exactly because of some images (cut out in example code).
Can anyone help us and get this fixed?
<br /> <pd4ml:transform<br /> pageFormat="A4"<br /> screenWidth="710"<br /> pageOrientation="portrait"<br /> pageInsets="0,0,0,0,points"><br /> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"<br /> "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd"><br /> <html xmlns="http://www.w3.org/1999/xhtml"><br /> <head><br /> </head><br /> <body style="width: 710x"><br /> <pd4ml:page.header height="140"><br /> <div id="header" /><br /> </pd4ml:page.header><br /> <pd4ml:page.footer height="50"><br /> <div id="footer" /><br /> </pd4ml:page.footer><br /> Content goes here.<br /> </body><br /> </html><br /> </pd4ml:transform><br />#28878The tag is a wrapper for PD4ML API, so all its attributes have equivalents in PD4ML class methods.
#28794It is going to be released this week.
#28873We’ll address the issue tomorrow. Hopefully it will be fixed with the beta release planned this week.
#28811Thanks a lot! Is this now available in the latest release?
#28880There is no dedicated tag for that. It can be configured with PD4ML API: pd4ml.setHtmlWidth(), pd4ml.setPageSize() and pd4ml.setPageInsets()
there is also a possibility to change the parameters on-a-fly with
#28877We’ve just added the feature. It will be available with the forthcoming beta release (hopefully, this week)
#28882No, there is no tag to define initial conversion parameters.
As long as you do not use the JSP taglib, I guess you instantiate PD4ML object and call render() method. Why don’t you define needed parameters via the API before the render() call?
#28874We’ve just refined pd4ml.clearCache() method to remove ALL cached objects, including harmless parsed CSS properties.
But it seems the issues with the cache reset happened only if you reuse PD4ML instance for multiple conversions (which is not recommended, in general). So I am not sure if we addressed exactly the issue you faced with
Also -Dpd4ml.cache.enable=false prevents downloaded objects from being cached. Bearing in mind the image size has been changed in the debug output, I’ve got no explanation from where the old image comes from.
Should we send you an actual development build to test the caching issues on your side?
#28745PD4ML does not test if a particular resource (image or stylesheet) has been changed: if it has a cached version – it uses it.
You may explicitly disable caching with pd4ml.cache.enable=false environment variable, or with a dynamic parameter.
[language=java:2wg7x71d]Map m = new HashMap();
m.put(PD4Constants.PD4ML_CACHE_ENABLE, false);
pd4ml.setDynamicParams(m);[/language:2wg7x71d]We’ve just refined clearCache() method in the development build, but as I got it worked for you before.
Another workaround would be to add a harmless dynamically generated parameter to the stylesheet reference:
./styles/main.css?rnd=
It should distinguish a new stylesheet from ones loaded before. As a drawback of the approach it will accumulate stylesheet versions in the cache. It is not that dramatic, as they are cached as soft references, but anyway it is a good idea to perform clearCache() from time to time.
#26683Hi,
We are generating large PDFs from a selection of web pages on our site. The resulting PDF files are quite large, around 60 MB. Using the Acrobat PDF Optimizer (http://tv.adobe.com/watch/acrobat-x-tips-tricks/how-to-optimize-your-pdf-file/) shrinks the files down to around 2 MB.
It seems that PD4ML is not very economical when creating the PDF, but perhaps there are some settings that we can use to make it better?
Please advice on how to use PD4ML to create PDFs that are more reasonably sized.
Cheers, Jon
