Difference between revisions of "Issue talk:Export to pdf"

Latest revision as of 19:07, 17 November 2009

I am also trying to create pdf files and am also having problems. I am using MW 1.14. I upgraded DPL to v1.8.9 (r58681) without benefit. I also converted from using a dpl tag expression to a parser function format without solving the problem of the dpl-generated table not appearing in the pdf output.

My users want to be able to create not only pdf pages but collections of pages. Having to print page by page would discourage them, I think.

I looked up the program you mentioned, PDF Creator, on Wikipedia. That particular program, being windows-only, and having some issues, as described in the article, does not seem like a general solution.

In any event, following your suggestion on the issue page, I was able to print a page and create postscript from my browser without problems. I think the problem may be in the PDF Writer extension that we would like to use. We are currently using their default, remote server. This appears to fetch the page and convert it. It may be a problem with how the page is fetched or rendered. It might be a problem in the api that, I suppose, it uses to do that.

I think the Collection extension installed the Create a book and Download as PDF links in the print/export box on my wiki page that I am trying to use. I see from your Version special page that you do not have that extension installed. So I think you might not have actually seen this particular problem.

I think this PDF extension has attracted a lot of interest because many people still feel that they have to create stand-alone documents, and they would like to do it from their wiki pages, and this pdf extension seems to make that easy.

It would help if you could confirm that the solution to this problem is to be found downstream of dpl, perhaps in the api or the pdf writer. Any insight you could contribute to the solution will be welcome.

--AJim 04:05, 14 November 2009 (UTC)

OK. I found this discussion through the mailing list link on the Collection extension wiki (Pedia Press). As of 26 Oct 09 they have no plans to support DPL expansion. It appears that they have a python library, mwlib, that functions as an alternate parser, so dpl capabilities would have to be patched into that library.

--AJim 04:46, 14 November 2009 (UTC)

From a quick glance at collection extension, pdfWriter and mwlib I got the impression that mwlib seems to be the crucial point. Personally I made good experience with PediaExpress (having ordered two books with more than 100 articles each). I think to SOME EXTENT they must be able to process mediawiki extensions in their mwlib. Otherwise they could not produce books from official Wikipedia sites. If someone could tell me the deeper reason why DPL is not supported (or positively explain which other extensions are supported by mwlib) I might be able to track the problem down. Generally speaking I think that DPL reports printed in pdf are kind of a "natural alliance" - so it would be worth investing some time if there was a chance to get that running. If it required some skill in Python, however, I would be the wrong candidate ... -- Gero 08:54, 14 November 2009 (UTC)

it looks like the mwlib folks are interested in helping support this. http://groups.google.com/group/mwlib/browse_thread/thread/81acda38cd01e6e5/9a1eb771c90feab6

Jbreiding 18:07, 17 November 2009 (UTC)