Help:Books/Bug reports

From WikiEducator
Jump to: navigation, search
Vista Community Help.png

Collections - Bug Reports

--Cstumpf (talk) 17:25, 27 January 2015 (UTC): When trying to download a PDF of a book I've created I get the following message: "There was an error while attempting to render your book." This happens even when attempting to render a PDF of a single page, no matter which page a try it with.

--Gladys Gahona C. 14:53, 1 August 2011 (UTC): Got the same error message "missing or malformed collection" for The script? Templates? HTML code?

I think you had exactly the same problem as Kirstys immediately below. Your saved book did not have a title, and the packaging script could not supply a default. I've changed the packager to make up a title when the author does not provide one. -- JimTittsler 05:51, 2 August 2011 (UTC)
Okay Jim, I will try it once more. --Gladys Gahona C. 05:55, 2 August 2011 (UTC)
Wowowow... it worked! The resource looks great in the Moodle, however I found two issues: broken links and WikiQuiz not working. Take a look please. --Gladys Gahona C. 06:46, 2 August 2011 (UTC)

Not sure if this is the right place - When trying to use the service to generate an IMS Content Package, I get an error message "missing or malformed collection page". When tested with the example urls it works fine. Trying to use but also having difficulty removing pages from that as per Missan's bug below with pages persisting after clicking Clear Collection. I have manually edited the collection page with no positive result. PDF generation does work.

--kirstys 00:07, 12 October 2009 (UTC)

Update: User error! When saving a collection that you wish to export using the IMS packager, you must give the collection a title otherwise it won't work. Thanks to Wayne Mackintosh for the patient troubleshooting. --kirstys 00:58, 13 October 2009 (UTC)

I can’t clear "My collection". I click “Clear Collection”, select “OK” and it shows as cleared but as soon as I navigate away the collection has persisted. Same with saving a collection, I can delete some of the articles in my collection but when I save it saves them all anyway.

I am getting the following message:

Could not generate PDF
From WikiEducator
An error occured during PDF generation.

Return to Main Page.

Retrieved from ""

--Missan 19:46, 25 December 2007 (CET)

when you click the radio button for Personal or Community Collection, the Action button does not go live.

--Randy Fisher 15:35, 12 December 2007 (CET)

Did you enter a page title for the collection? --Johannes Beigel 17:03, 12 December 2007 (CET)

Also, when this function is ready (without bugs), you might consider naming: "Create PDF", and put it in the navigation bar (left) under "my collection".

--Randy Fisher 15:37, 12 December 2007 (CET)

Which function exactly do you mean? The "Download as PDF"-Link which is currently in the toolbox? --Johannes Beigel 17:07, 12 December 2007 (CET)
Also perhaps a "my collections" could go on the list of "my" things along the top for logged in users that would take you to the page that has all your collections on it. Right now its hard to find where that page is as it seems to be created automatically but no permananent link is created. brent 22:24, 12 December 2007 (CET)
I think that is a very good idea. --Randy Fisher 22:28, 12 December 2007 (CET)

Bottom of printed out PDF says "All Articles originate from - for usage see the GFDL". brent 22:11, 12 December 2007 (CET)

Yikes! Don't know what you mean...

- Randy

It's the same mechanism as puts the CC-BY-SA license text into the PDFs from this site, except that brent must be using the code from PediaPress's repositories on another wiki. This code has the license description set to the GNU FDL. This is currently hard-coded into, so can only be altered by editing the source code to mwlib.rl and re-installing it.

--Vtaylor 13:22, 29 June 2008 (UTC) : Sometimes there is an "Add article" link in my collection box, sometimes not. I'm trying to make a WikiEducator "cheat sheet" so I want to include some of my own pages and some of the Special pages - which don't get the "add article" link.

Also - the process for adding to and saving an existing Collection is error prone - requires retyping an existing name and saving in the right panel. It would be nice to have a save button in the left panel to save the current list with the modifications.

I was surprised to find that the "saved" collection didn't reflect new additions. I had also created several Collections because I had used a different name in the right panel to save while trying to figure out how this worked.

Nice feature and I really want to use it a lot! I think this could be a huge "value add" - allowing "collectors" to discover great content and publish collections with reviews, comments, ratings, ... The problem with OER isn't the availability of content - it is already there. It's a problem to find the good stuff and use it. This is where Collections will provide an invaluable service. Collectors will build reputations in their areas of expertise. Users know they can trust the collectors recommendations, limiting their search to a manageable set of OER content. Then OER adoption will accelerate.

Bug list

Info about submitting bugs:
Whenever submitting a bug, try to be specific about what goes wrong.
Supply a sample article and specify the text/markup that causes the
pdf to "break".
If you plan on testing the pdf toolkit intensivly please register on and submit bugs to the bugtracker there. 
That facilitates issue tracking for us a lot.
btw: thanks to everybody who has contributed so far ;-) 
-- Volker Haas 16:27, 21 December 2007 (CET)
  1. The names of all authors are missing on the generated PDF. I think it is absolutely nessecary to give attribution to the authors of any article, which gets converted to PDF. The License CC-By-SA and the GFDL makes this clear. --SvH 14:48, 28 December 2007 (CET)
Listing all authors of an article can - in principle - be done. Unfortunately this involves 
quering the database and could therefore be a time consuming process. Also it might not be 
desirable to list all authors but rather the "main" authors. Needless to say that this is
even more complicated.  -- Volker Haas 14:55, 4 January 2008 (CET)
  1. WikiEducator uses a default CC-BY-SA license unless otherwise indicated. In cases of pages using CC-BY we should find a mechanism to identify whether any pages are using CC-BY. Perhaps we could insert a template on all CC-BY pages and use a workaround using the "print-specific versions of templates" feature to identify these pages in a pdf output. In this scenario the copyright page would need to include a statement that some pages in the collection are licensed under CC-BY.
The above mentioned should be doable. -- Volker Haas 14:55, 4 January 2008 (CET)
  1. Images do not resize proportionally with their relative size in the web display. Images currently use a default size when output to pdf. This is not ideal as the user should be able to manipulate the size of the output image.
  2. The following wiki format (often used for display code) does not render properly. The text is lost entirely and the box is not rendered in the pdf version.
My example text
Fixed, but boxes are not rendered around preformatted text Volker Haas
  1. Text and images added to a page using a template (example: Template:BES) is not rendered on the pdf. I use the template to add general disclaimers, a link back to to the central orgainzing page, and a logo. It would be nice to have those in the pdfsDmccabe 02:40, 15 December 2007 (CET) (also = Declan above) -- Hi Declan - I've arranged a temporary fix with this print specific template Template:BES_print - when printing it will automatically substitute this template.
known bug the reason is, that instead of 
{{BES}} the templatename is prefixed with "Template:" on invokation: {{Template:BES}} Volker Haas
  1. Text wrapping around an image within a table cell does not render according to specified image attributes. See specifically the images in the activity tables here: Wikieducator tutorial/Creating an Account/Instructions - note that the print version uses the substitute print template: template:Activity_print
known bug. this specific case is: Volker Haas
  1. An example of a template incorporating an image with placement specified by the parameters in the template not rendering in the pdf. See: Wikieducator tutorial/Communication and Interaction/Discussion
  2. Implementing typographical conventions
    • In this application its better to use left justified text for readability - see also below. (We don't have control over kerning, hyphenation, leading etc - so its better to use a left justification default).
    • Because we're printing books -- the numbering of odd-numbered pages should be right justified and even numbered pages left-justified - (thats the actual number -- not the text on the page :-).
    • The ability to start a new section or chapter on a new page. In most books - the typographical convention is to start a new chapter or unit on the odd numbered page -- therefore the option to start on a new page is important. The parser should check, if the new page is even -- then insert a blank page and start the new page on the first odd numbered page.
    • The placement of the copyright page is done immediately after the title page (not the cover.) We could treat the first page as the title page (or is this the cover of the book) -- If this is intended to be the cover -- then we must add the ability to create a title page and then insert the copyright page immediately after the title page. Title pages are usually the first odd-numbered page in the book with the copyright page printed on the reverse. The placement of the full GPL at the end is correct.
text justification can easily be customized: the file contains the 
definitions of the styles used for different text-nodes: 
TA_JUSTIFY has to be changed to TA_LEFT Volker Haas 
Thanks Volker -- on WikiEducator -- could you implement the TA_LEFT as the default. Cheers - --Wayne Mackintosh 18:55, 20 December 2007 (CET)
  1. All pages in a collection start one after the other - is there any way of starting a new page from the wiki on a new page in the pdf (apart from chapters)?
currently there is no way of doing that Volker Haas 10:06, 2 January 2008 (CET)
  1. Creating chapters seemed to be hit and miss - I tried adding a chapter with different numbers of pages and in different places but it sometimes worked with one setup and not with another where the only differences were the individual pages I had selected - keeping structure constant. (When it "didn't work", I got an error message after trying to download the pdf file).
please supply a link to example collection (one that works and one that breaks) Volker Haas 10:06, 2 January 2008 (CET)
  1. The attribution message at the end of the pdf file - and on the bottom of each page - is generic, and doesn't specify where the page is (at least to someone not familiar with page naming on a wiki). This might mean simply adding another line in the generic message about how to find the page (eg. "You can find the original page by adding each page's title after the url" - hmm, I'm sure this could be phrased better). Cormaggio 10:33, 26 December 2007 (CET)
Adding article urls in the page footer doesn't really help, since multiple articles can be on one page. 
Instead, the article url could be added after the article title (in the "content" are) or at the end of an article. Would that be an option?
Volker Haas 10:06, 2 January 2008 (CET)

Fixed Bugs

Please note that "fixed" only refers to the code available from 
--> The mwlib software needs to be updated on wikieducator in order for the changes to take effect.
Volker Haas
  1. A few image placement issues -- centered image doesn't center and inline images don'r render well in the pdf - see Wikieducator tutorial/Basic Text Formatting/Bold Italics and Headings
-there is one known bug (probably a reportlab issue) with the horizontal placement of 
inline images - thats one thing you see. 
-centered images are now displayed correctly. Volker Haas
fixed now - Volkerhaas 12:33, 21 January 2008 (CET)
  1. Sub-bullets within a numbered list don't render properly in the pdf. See: Wikieducator tutorial/What is a wiki/About wikieducator#Types of content on WikiEducator - under the heading "Types of content on WikiEducator
known bug, see and Volker Haas
fixed now: Volkerhaas 12:30, 21 January 2008 (CET)
  1. The gallery feature is not yet implemented in the pdf generator.
  1. This PDF has still tags in it(startfeed endfeed/), which should be completly removed in the pdf. I could not upload the pdf, so you have to believe me. ;-) --SvH 14:48, 28 December 2007 (CET)
fixed ( bug seen in article FLNW2/Feeds ) Volker Haas 9:58, 2 January 2008 (CET)
  1. There are some image issues. The text wrapping is sometimes a little off such that text sometimes overlaps the bottom of framed images (example An_Inside_Look_at_Apples! - Posted by Declan
  2. Text does not wrap correctly around this Thumbnail image (.png) see:
  3. When bold and italics formatting are applied - the standard serif font renders in sans serif in the pdf output -- which is not consistent. Serif font should be used throughout body text.
  4. The <nowiki> tags, if embedded in a template do not render in the pdf output - but render OK to the browser. Take a look at the pdf output of the 2nd tip on this page (which contains a "nowiki" tag):

Must have features

  • Currently the pdf output uses full justification. Based on readability studies in the distance education field and taking typeface size into account, the pdf output should use left justification for the text. See page 3 of the pdf out output for this page (See below the activity in the pdf version).
  • We must find a way to implement an option which ignores printing the article header - see for example this study guide collection:

WikiEducator:Collections/Cost_and_Finance_Unit_3 (Cool hey -- I can just point to the saved collection!)

Workaround/implementation suggestion: Given that a collection uses a wikitext format -- what about the possibility of adding piped text. In the case where a pipe exists, then the pipe text is used as the string for parsing to pdf. In the event of no pipe -- the article page is used as the string for parsing. So for example:
[[Cost and Financing in Open Schooling/Capital Costs/Depreciating Capital Costs|Depreciation]] should parse as "Depreciation"
We should also cater for a situation where no text is parsed. --Wayne Mackintosh 19:03, 20 December 2007 (CET)
  • A generic div eg.
    would be very useful in managing things which we know will break the pdf generator, are unlikely to be implemented in the near future but not mission critical in the print version. For example, on this page our media Idevice embeds the MP3 player extension. In such cases we could use the NoPrint div tag and wrap this around the player. So when printing the nested html is ignored -- but in the browser will display normally. I would prefer this behaviour rather than the pdf breaking or rendering garbage. Another example would be the green navigation arrows at the bottom of this page: - I see this as a generic instance of the print specific template.

  • Is there any way you can insert an save a single space on a wiki page? This is useful for suppressing the printing of a navigation template in Template:Navigation_print -- at the moment I'm using a hard carriage return -- which is not ideal because it impacts on the spacing of the text between sub-sections.
  • The ability to manipulate the header text -- For example:
    • To Display the specified chapter name on the odd numbered pages
    • On even number pages to only display the actual page title in a structured text -- Therefore when printing the page "Cost and Financing in Open Schooling/Capital Costs/Depreciating Capital Costs" the even page header should only print "Depreciating Capital Costs"

Feature Wishlist

  1. The ability for the user to specify the logo/image used on the title page. For example The University of Foo might want to use the Foo logo instead of the WikiEducator logo -- we're supportive of this idea.
  2. Basic table formatting eg the ability to turn on and turn off a single border would be a great help in manipulating layout for print-specific templates.
  3. Allow user to specify some white space around "page" - currently fully left and right justified. I'm using this to include pages as Resources in Moodle that doesn't add any borders so text goes right to the edges of the window. --Vtaylor 13:16, 29 June 2008 (UTC)
Retrieved from ""