From Wikipedia, the free encyclopedia
Jump to: navigation, search
How to report issues How to escalate issues
  • please check whether this issue is already known
  • provide a link to an article that is affected by the issue
  • report only one issue per section
  • select a descriptive title
  • consider to prefix the title with Bug, Proposal, Question, Comment, Task ...
  • remember to sign your post with ~~~~
  • consider to warn readers of the problem by placing |bug=... in the {{saved book}} template of relevant books

There is a central page at gathering all major issues with this extension. Issues that can't be solved and are not yet covered on the page at meta should be added there.

For obvious bugs the issue tracker is the preferred place to directly issue and check tickets.

At IRC #pediapress some immediate support might be available.

Links in PDF Table of Contents[edit]

The table of contents in generated PDF files looks nice but it doesn't link to the articles in the book. I'd create a user page and basically copy the book page, remove the semicolons and colons, then use that in my book as a table of contents. That way I'd have a table of contents that I could click on and jump to the articles in the book. I can't disable the generated table of contents as far as I know, so that would suck having two TOCs in a book. It would be really cool if the generated TOC preserved the relative links in the same way that links inside articles do.Matthew Kastor (talk) 03:03, 18 October 2013 (UTC)

Sounds like an awesome idea...any progress after 4 months? Please see my new entry below regarding an Index.--Graham Proud (talk) 02:32, 4 March 2014 (UTC)

Render server error[edit]

Whenever I export this book with 300+ articles to PDF, around 70-80% it always says "Render server error". Why is that like that? The limitation is 500 articles but it's only 300+. Please fix as soon as possible, thanks! Here's the link of the book: Thatpopularguy123 (talk) 15:49, 22 May 2013 (UTC)

You have another limitation 2000 max pages and 400 mb. You can put only 20-30 countries /book. -- (talk) 16:11, 22 May 2013 (UTC)

Muchas gracias!Thatpopularguy123 (talk) 04:12, 23 May 2013 (UTC)

I keep getting this error message: An error occurred on the render server: error executing command 'zip_post': <urlopen error [Errno 113] No route to host>

Return to Main Page.; whenever I try to preview any of my books on Pedia Press. Mind you it doesn't matter which of the 20 or so books I try to preview, how many pages or articles are in the books and many of them have been successfully accessed and printed previously. Very perplexing problem. Help!! I have no idea how to fix this and have not been able to find any info. on Wikipedia that is helpful.

Kkhemet (talk) 07:28, 17 November 2013 (UTC)

Hyperlinks not included in PDF[edit]

I agree to including all references in the article being "PDF'ed," and I'd like hyperlinks to also be included. If it's OK to do so and it's a shortage of volunteers, I'll help. — Preceding unsigned comment added by LaRaine Mae (talkcontribs) 05:51, 31 August 2013 (UTC)

epub format books contain [edit source | edit] beside each section title[edit]

Creating books is wonderful but the epub format (haven't tried PDF) of late will display a link (unclickable in e-reader) to [edit source | edit] beside each section title. This is needed when viewing the article/page in wikipedia of course, but when creating a book I think this should be hidden shouldn't it? Makes books look a bit odd... Additionally, bold header text is a bit squashed (not enough line spacing) on Kobo ebook reader. — Preceding unsigned comment added by (talk) 01:31, 14 August 2013 (UTC)

Category:Exclude in print[edit]

Is it true that using Category:Exclude in print no longer works? If so, shouldn't the documentation be updated to reflect that? Powers T 13:27, 11 September 2013 (UTC)

Unfortunately, yes. See bugzilla:48052, bugzilla:50750 and bugzilla:45861. Helder 21:33, 11 September 2013 (UTC)
Feel free to help in updating the documentation. I did a little - you can improve my changes further. Wiki, wiki, wiki. There's a hack (workaround) discussed and linked to here. Boud (talk) 00:20, 31 October 2013 (UTC)

Book Splitting[edit]

Wikipedia clearly warns new editors that the Book Creator does not support large books with more than 500 pages. However, an alternative option is to fork a book just prior to the 500 page limit, by saving it under a unique title (or revision), prior to proceeding to adding more pages and subsequent topics, although later saves of the book may fail. Again, as already indicated, this method is highly like to error out for many users and is not recommended. This is not a problem with technology, it is a problem with editorship.

For instance, most users cannot plan their book out in advance, such that each saved volume contains 500 or fewer pages (give or take), because most books grow in an utterly random fashion similar to the Bell Curve of a pile of dung dripping from a cave ceiling, but with a tail that skews to the right. In other words, book size (as number of pages) grows non-linearly as a function of numerous random variables, including the grow of semantic topics included in the book. Perhaps the correlation closest to a linear relationship is the growth RATE in pages, against the growth RATE of topical scope, although this would be difficult to operationalize. Thus, central topics fill more pages added, in early-stage book growth, with topical scope widening at a fast rate, then narrowing again at a slower rate (of pages added per change in scope). Additionally, more fringe topics tend to fill in gaps between central topics, at a nearly steady rate per click throughout the process of book creation, but represent very nearly the ONLY added pages, near the final stages of book creation.

It is significant to note here that most users [whether planned or not] alphabetically organize their books, as a last step before saving them, although almost half of all books do not get saved permanently, and another smaller percentage of books never even get saved. This is theorized to represent compensation for lack of organization of the book. However, a much better method for compensating for lack of organization, is to actually organize the book, which might require segmentation into more manageable chapters and volumes first. Thus, for a typical non-linear, poorly planned, and unpredictable 'non-central growth' model and given the likelihood that few pages will be deleted from most created books, either as drafts or in a final pruning or quality control stage, editors can save lower quality final works as multiple volumes instead of higher quality single volumes, and still retain the option of future refinement, without any immediate compromise in total pages included.

The best approach to content splitting (for the average editor) is to save a work-in-progress multiple times (under 2 titles), and then delete pages from each volume accordingly, prior to adding pages to each volume. By such a method therefore, a multi-volume book might grow indefinitely through iterative splits. For example, at 500 pages, one could save one's book with the title "Big:Volume 1", and then immediately save the exact same book again as "Big:Volume 2" (still, with the exact same 500 pages). Next, the user would delete pages 250-500 from Volume 1, and delete pages 1-250 of Volume 2. Then the user could proceed (once again) with the task of randomly surfing and "filling in" their book with accidentally discovered candidate pages for each of the two volumes (technically, now two separate books), via the navigation patterns of click-through behavior documented by web analytic research. Of course, an even superior method (albeit unlikely) would simply be to plan one's editorial work out in advance, in terms of topical coverage, order, audience, goals, etc, and use an iterative PAGE-DELETION methodology with at least two drafts, thus excluding less critical pages and creating a final piece of higher value.

Article with complex markup gives pdf mess[edit]

I realize that this article ([1] -- specific revision linked here for definiteness, though this has been a problem for years) is highly marked up and may be something of a stress test for pdf conversion. But other article must be I hope that sooner or later the very serious problems in the conversion machinery can be addressed. Among these are (very partial list):

  • Markup debris in image captions
  • The string < /ref > scattered throughout text
  • Superscript ref callouts missing or incomplete
  • Ref groups (Arabic vs upper-alpha etc.) apparently mixed together into one long list of mis-numbered notes
  • Some notes run into article text
  • Honestly, so many bizarre bugs it boggles the mind

Will this stuff ever be fixed? I notice that today's FA [2] renders into pdf with all its refs missing. Really, the feature should just be disabled until it is somewhere close to functional. EEng (talk) 19:01, 25 September 2013 (UTC)

<bump> EEng (talk) 03:29, 31 October 2013 (UTC)
@EEng: that page is a trainwreck for formatting and such. It is beyond a "stress test" for it breaks even my browser. There are so many issues with that page that it needs to be completely rewritten and addressed to make it accessible in the first place - much less try and use it in a book. I'll give it a shot and see what happens. ChrisGualtieri (talk) 05:07, 9 November 2013 (UTC)
As mentioned in my original post above, the pdf converter has made a mess of this page for many years, and it's my observation that it does the same to many other pages, particularly with regard to the citations. Go to User:EEng/sandbox/Gage2012July30_0144 which has a version from more than a year ago (containing none of the unusual markup you complain about) and look at the Notes and References sections; then do a pdf export and compare those sections.

As for breaking your browser, I have tested in IE, Chrome, Firefox, and Safari and see no problems. What browser are you using? I don't think I've ever done a pdf export of any but a very simply page that there wasn't something obviously wrong, so blaming the page instead of the converter is a mistake.
You now seem to be in the process of removing all markup you don't understand. You've removed the figure designations, changed he section head sizes and image sizes, changed the lettered Note designations e.g. [A] to e.g. [note 1], and seem to be converting the internal citation syntax from the one in use to one with which, I assume you're more familiar. And it looks like you plan to put the sources list in random order instead of the current sectioned, alphabetized order. I appreciate any and all help, but please take the time to understand the markup before assuming it's all garbage. It's not, it's all there for a purpose, and as far as I know it's all legal.
You've also begun to make content changes that are flatly incorrect, such as changing place-of-death from "in or near San Francisco" to just plain "San Francisco", and describing one of the portraits as a "daguerreotype of a portrait", which makes no sense.
Do you think it's appropriate to make such sweeping changes without at least inquiring as to why the page is the way it is? Certainly the page's markup is far more complex than that of most pages, but it's that way in order to present the material most effectively. Instead of just ripping out what you don't understand, wouldn't it be more appropriate to find cleaner ways to get the same effect (by creating templates and so on), or to add clarifying notes to explain how things work? EEng (talk) 07:15, 9 November 2013 (UTC)
Huh? I fixed that already. Though the markup is indeed extremely over complicated, I'd like you to point me to a more complex markup case. Look. I can flip it back like a switch. I got to take a break for awhile. I'll ask Mag to take a look at it. It's amazingly inefficient and needlessly complex... but he knows this better than I. ChrisGualtieri (talk) 08:24, 9 November 2013 (UTC)
Huh? As is clear from this disucssion you don't know what you're talking about. The markup is legal and the fault is with the pdf renderer. EEng (talk) 10:50, 11 November 2013 (UTC)

<bump> again. EEng (talk) 07:58, 17 November 2013 (UTC)

References not included in PDF[edit]

I enjoy having the references on hand and each time I go to save an article as a PDF it only includes a couple references out of many. Oddly, they're not always the same references each time either. Is this by design & if so is there a way to have the full ref list included? Coinmanj (talk) 07:26, 20 June 2013 (UTC)

In my case, I've noticed that the PDF generator CREATES references from inline external HTTP links, but ignores references to citations. This is pretty bad, since a technical article has many citations. An example is the Mabel Hokin biography: the PDF only includes an external HTTP URL and a WikiMedia image URL. The many citations vanish, both from the reference list and the article itself. They are present in the "printable version" of course. Sammyjava (talk) 22:33, 19 November 2013 (UTC)
Multiple citations in the ref name style (<"ref name="X"/">) simply do no longer appear in pdfs recently, while they used to work perfectly before. Can something be done about this, as most articles include such multiple references, especially the more advanced ones, like FAs and GAs. Buchraeumer (talk) 13:18, 22 November 2013 (UTC)
This is just more of the problems I mention in the section just above. The pdf renderer as it stands is a joke. EEng (talk) 17:31, 22 November 2013 (UTC)

some articles are excluded from rendered book[edit]

In my book even after several tries I was unable to get last few articles in the ZIM file. Also unable to download complete PDF file, so no question of openning it. The book in question is:

Articles "Lawrence Lessig" and below are excluded everytime. The same occured in my previous book:

Can anybody help me to find out the actual problem, please. --E99plant (talk) 15:35, 6 February 2014 (UTC)

Is this book creation tool (or MediaWiki Collection extension) is only for PediaPress _printed_books_? --E99plant (talk) 14:36, 9 February 2014 (UTC)

Import a List of Articles from Browser Favourites[edit]

I collect my favorite wikipedia articles as favorites with my browser. Is there a possibility to import a list of URLs to the Book Generator? — Preceding unsigned comment added by (talk) 14:30, 25 February 2014 (UTC)

Comments Section of Signpost (and other Signpost issues)[edit]

The Signpost books include the comments section of each article, which ought to be excluded. Is there a way to mark a section of an article for exclusion from the books? Additionally each article should ideally begin on its own page. Is there any way to set the book renderer to do that? Zell Faze (talk) 15:01, 28 February 2014 (UTC)

@Zellfaze: I tried Book:Wikipedia Signpost/2014-02-26, and, yes, the comments sections are included in the PDF. That's odd, because they are included via Wikipedia:Signpost/Template:Signpost-article-comments-end, which is in Category:Exclude in print. If no one else comments here, I suggest you raise this at WP:VPT. I don't know the answer to your second question. -- John of Reading (talk) 08:15, 1 March 2014 (UTC)

Index builder? Suggestion[edit]

Clearly, as noted above, the Table of Contents MUST have clickable links in the PDF. Similarly, I would like to raise a suggestion for an Index Builder. A book should have a list of index words, much the same as it works with Microsoft Word and other word processors. With Wikipedia we have an opportunity to do something really out of the box here. In much the same way as the Book Creator already scans the book's pages to create a Suggested Pages list, could I suggest that an Index Builder could scan the pages for important words and Statistically Improbable Phrases to create the first-draft index word list?--Graham Proud (talk) 02:37, 4 March 2014 (UTC)

An example is available from PDF Colony--Graham Proud (talk) 02:45, 4 March 2014 (UTC)


Images used in the Book will be referenced with the username of the uploader but not attributed to the creator if they differ. Agathoclea (talk) 11:13, 10 April 2014 (UTC)