Jump to content

Talk:Reading/Web/PDF Functionality/2017/07

Add topic
From mediawiki.org
Latest comment: 7 years ago by Steelpillow in topic Exculsion of certain links

About giving feedback

Please read Reading/Web/PDF Functionality and comment on the plans we lay out there, to tell us what you need from the PDF service. We're especially interested in what you need in the future that doesn't exist in the plans laid out there – if there's a bug with something that should work right now (e.g. you get an error message when you try to create a PDF), we need to fix it, of course, but that would have been on the agenda anyway.

Charts and Page Breaks

[edit]

The single column version formats well with the exception of handling charts that need more than a single page to display. It seems that if the PDF were zoomed out further, the HTML breaks would not create the partial pages I'm seeing around charts. Maybe the PDF service could implement a line break in the chart if it's over 50 rows... Jkellner6504 (talk) 13:06, 1 July 2017 (UTC)Reply

[edit]

It's impossible to highlight text without ending up on a web browser. Thanks!! 89.251.186.211 (talk) 15:12, 1 July 2017 (UTC)Reply

Original Choice

[edit]

Could you please keep the original choice of "printable version" that actually displayed in my browser? It worked perfectly for my needs. Now it doesn't display the printable version in browser, but immediately wants to print a PDF which I cannot use directly for my work.

Thank you for all the tremendous amount of work that you do. Rkmcfry (talk) 15:33, 2 July 2017 (UTC)Reply

Thanks for your feedback! Let me make sure I understand your request. We have not altered any of the behavior of the printable version link, only the behavior for PDF rendering - accessed from the download as PDF link. We you having issues with the printable version link? OVasileva (WMF) (talk) 15:09, 4 July 2017 (UTC)Reply
Ms. Vasileva, The commenter to whom you replied [Rkmcfry] is ABSOLUTELY RIGHT -- the original choice of the "printable version" doesn't display in the browser anymore!!!! NOW instead pops-up the window, which offers PRINTING to MICROSOFT ONE NOTE, which is COMPLETE JUNK (мусор, барахло, халтура, ит.д.) with extremely limited features.
Looks like, you were not informed about that "novelty." (;'-(((((
Could you please return the PREVIOUSLY FUNCTIONAL "PRINTABLE VERSION", and let it stay, regardless your modifications to the PDF printing, which doesn't work for me as a USER by MANY reasons?!
I realize that this is some "team work", so please submit my request to whom it may concern.
Thanks in advance for your consideration AND HELP with this matter!!! 2606:6000:6129:7A00:C182:7027:FB9E:B012 (talk) 20:57, 31 August 2017 (UTC)Reply

download problems

[edit]

Please devise a way to download to a pdf without hyperlinks. Once downloaded and imported to an application such as notabililty, I am unable to highlight or notate because the hyperlink starts to operate. 100.47.38.62 (talk) 11:17, 3 July 2017 (UTC)Reply

Thank you. Johan (WMF) (talk) 11:59, 3 July 2017 (UTC)Reply

This problem occurred after you installed the press charge. Strange isn't it?

[edit]

This problem occurred after you installed the press charge. Strange isn't it? 150.217.179.198 (talk) 08:11, 4 July 2017 (UTC)Reply

I must admit I'm not entirely sure I understand what you refer to, but we've had the problem with the OCG code crumbling and needing a lot of maintenance for quite some time. Johan (WMF) (talk) 09:10, 4 July 2017 (UTC)Reply

Single column download

[edit]

The single column download ( into .pdf file) doesn't work for me but the double column download was perfectly executed. I use Google Chrome Version 59.0.3071.115 175.176.222.17 (talk) 09:27, 4 July 2017 (UTC)Reply

Thank you. Johan (WMF) (talk) 09:43, 4 July 2017 (UTC)Reply

Occasional failure with single column

[edit]

Single column PDF creation has been working well about 85-90% of the time, but occasionally fails with an error message. I just received this one:

{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Executable_and_Linkable_Format"} 68.118.182.189 (talk) 12:46, 4 July 2017 (UTC)Reply

New PDF

[edit]
I like the newer method because it shows the info square to the right just like your page is set up. I research a lot of films on Wikipedia along with IMDB. I like having the file handy and the info box has the movie poster and other info summarized for easy viewing. You guys are the best. My dad was a door to door salesman for encyclopedias back in the1950s. Wikipedia is 100% better than those encyclopedias I grew up with. Keep up the great work. DanielDLessard (talk) 22:22, 4 July 2017 (UTC)Reply

Finally a single column, but a small problem

[edit]

I think this new option is nice, because it is closer to the original Wikipedia experience and, for me at least, improves legibility.

I small problem I encountered on the first ever page I converted using this method. The picture on the top is probably too wide, and so the text flows in a very small and awkward column to the left of it. Either the picture should become smaller, or it should take up the whole width of the page with the text starting right under it.

Anyway, good work on what is there so far, and I hope it only gets better with constant improvement.

Cheers. Rushtake (talk) 22:45, 4 July 2017 (UTC)Reply

[edit]

The links on words are not evident. We cannot see them on the Pdf rendered. 88.11.63.241 (talk) 12:22, 5 July 2017 (UTC)Reply

Thank you. Johan (WMF) (talk) 12:38, 5 July 2017 (UTC)Reply
Links should not be shown, surely? They make a printed book look horrible. Steelpillow (talk) 19:17, 23 December 2018 (UTC)Reply
[edit]

It'd be helpul if some topics like 'Reference' could be remved from all the pages 117.204.124.110 (talk) 14:42, 5 July 2017 (UTC)Reply

References are necessary for the reader to check the facts for themself. It would be unacceptable to remove them. Steelpillow (talk) 19:19, 23 December 2018 (UTC)Reply

Headers should be "connected" with their text

[edit]

I've just downloaded the PDF of the following article:

https://it.wikipedia.org/wiki/Bullone

I've figured out that some sub-headers are in the bottom part of the page, and the text they refer to are in the next page because there is no more space in the header page (the header is the last available line of the page).

The header shall be in the next page, together with the text it refers to.

For better understanding, please download the PDF of the article and check the bottom of page 8. 93.145.198.130 (talk) 15:25, 5 July 2017 (UTC)Reply

letters in PDF overlapping

[edit]

for https://en.wikipedia.org/wiki/Regular_language downloaded in single cocolumn version.

There are letters overlapping, not formated well . 2601:241:8201:163E:8543:2A7D:EA85:BF42 (talk) 16:02, 5 July 2017 (UTC)Reply

If you see this – what browser and OS do you use? Johan (WMF) (talk) 16:21, 5 July 2017 (UTC)Reply
printed perfect internet explorer, windows 7,Dell laptop, canon printer 174.19.167.58 (talk) 16:32, 5 July 2017 (UTC)Reply

footnotes

[edit]

one cannot go back from footnotes to the text by clicking the number 2A02:1203:ECB0:2230:2DD9:B49F:C61B:1100 (talk) 08:32, 6 July 2017 (UTC)Reply

Thanks. Johan (WMF) (talk) 10:42, 6 July 2017 (UTC)Reply

mediawiki2latex Web Server

[edit]

We already got at server that exports mediawiki to odt, epub, LaTeX and pdf. Its here

http://mediawiki2latex.wmflabs.org/ Dirk Hünniger (talk) 13:22, 6 July 2017 (UTC)Reply

Text Size

[edit]

Single column text size too big - document unattractive and ungainly. 109.151.254.239 (talk) 14:04, 6 July 2017 (UTC)Reply

Thanks for your feedback. Johan (WMF) (talk) 14:47, 6 July 2017 (UTC)Reply

Comentario sobre el pdf generado.

[edit]

Está muchísimo mejor que antes en el modo sencillo (el otro no lo probé). 200.37.4.213 (talk) 15:58, 6 July 2017 (UTC)Reply

Gracias. Johan (WMF) (talk) 17:43, 6 July 2017 (UTC)Reply

I prefer the new pdf format, but ? font size

[edit]

I find the new one-column pdf format more readable than the two-column version. A smaller font may be better for a document. The present font is OK for viewing on a computer, as you can zoom out, and then you have a reasonable column width for reading on screen. If you wanted to print it, however, the font would be huge. If the font for the pdf were smaller, you could still zoom in to make it bigger on a computer screen, but it would print out better. However, you would have more words per line, which would be less readable. 71.8.170.118 (talk) 18:19, 7 July 2017 (UTC)Reply

Thank you for your feedback. Johan (WMF) (talk) 20:30, 7 July 2017 (UTC)Reply

Funktioniert nur rudimentär

[edit]

Statt des Buches bekomme ich beim runterladen eine PDF mit Inhaltsverzeichnis und 1 (in Worten einem!) Link zur Webseite. Ich wollte das Buch als PDF, um es auch lesen zu können, wenn ich mal nicht online (unterwegs) bin. Wenn ich mir das Inhaltsverzeichnis mit copy & paste in ein Word-Dokument kopiere, habe ich immer noch mehr davon, nämlich den Link zu jedem einzelnen Kapitel. Ich muss sagen, ich bin enttäuscht! 

Uwe 2.247.247.62 (talk) 06:57, 8 July 2017 (UTC)Reply

 
Hi,Hey,Hallo 
Vielleicht hilft euch ja diese Fehler/Operationsbeschreibung weiter
Probably these errors/operations will be bringing you forward
>xxxxxxx< anonymisiert
>xxxxxxxx< anonymised
 
1.    Beispiel: (1ex)
 
Layout 1spaltig (mit Rahmen und Tabellen)
Layout1 (with frames and tables)
 
Fehler: Datei nicht gefunden
Error: Page not found
Die Dateien unter https://de.wikipedia.org/api/rest_v1/page/pdf/xxxxxxxxxxx konnten nicht gefunden werden. 
Bin zu Zweispalten-layout gewechselt
Turned to two-column-layout 
Worked !
Ging! 
2.    Beispiel: (2nd Ex)
 
Layout 1spaltig (mit Rahmen und Tabellen)
Layout1 (with frames and tables)
 
Fehler: Datei nicht gefunden
Error: Page not found
Die Dateien unter https://de.wikipedia.org/api/rest_v1/page/pdf/xxxxxxxxxxx konnten nicht gefunden werden. 
Bin zu Zweispalten-layout gewechselt
Turned to two-column-layout 
Worked !
Ging! 
Zurueck zu Layout 1spaltig
Back again to one-column-layout 
a)      Einmal ging es nun
Worked once
b)      Einmal ging es nicht
One time didn´t work again  
3.    Beispiel: (3rd Ex)
Habe den Neuladebutton im browser benutzt
Used refresh-Button in browser
 
https://de.wikipedia.org/w/index.php?title=Spezial:ElectronPdf&page=xxxxxxxxx&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpezial%3ABuch%26bookcmd%3Drender_article%26arttitle%3DKlaus%2Bxxxxxxxxxxx%26returnto%3DKlaus%2Bxxxxxxxxx%26oldid%3D166473913%26writer%3Drdf2latex
  
Meine  unerheblichen  Schlussfolgerungen:
My  unimportant conclusions: 
a)     Someone got hacked by stately crackers from >we don`t like wikipedia<
a)     Cracker; gibt Staaten die wikipedia nicht moegen 
b)     Nach etwaigen browserprozessen bedarf es mitunter >Neuanstossbefehlen< im script; wenn die nicht erfolgen, blockiert´s. Deshalb koennte ich mit meinen operativen Neuanstoessen erfolgreich gewesen sein 
b)     Following diverse browser-processes there might be a necessity for >restart/continue-orders< in the script ; if none, then >error<. Probably my successes are due to my >program-re-kicks/starts<
 
Grts, 
Stefan 78.42.121.245 (talk) 01:56, 9 July 2017 (UTC)Reply
Einspaltige PDF-Download und auch 'Druckversion' scheinen nicht zu erfolgen und werden nirgendwo als Datei sichtbar. Doch beim 2.Mal Anklikken des 'Herunterladen'-Buttons erscheint die Abfrage, ob die Seite gedruckt oder gespeichert werden solle. Und dann erfolgt die Speicherung im 'Download'Ordner.
Gruß, Dieser 92.74.177.95 (talk) 05:54, 18 July 2017 (UTC)Reply

getting overlapping or scrunched up characters

[edit]

Did a 'Single Column' PDF of this wiki article and it rendered the overlapping characters on page 4, on a couple of lines. :

https://en.wikipedia.org/wiki/Digital_camera_modes

Bug or something? that has to do with the long citation entries?

Otherwise, quite satisfied. 50.90.138.130 (talk) 07:19, 8 July 2017 (UTC)Reply

Were you using the single page renderer? If so, could you tell us the browser and operating system you were on? I tried to recreate the error in firefox 54, osx 10.11.6, but did not notice any overlapping characters. Thank you for your report! OVasileva (WMF) (talk) 13:21, 14 July 2017 (UTC)Reply

Overlapped text

[edit]

Hi! I've experienced a quite annoying issue: the text of a quote, if aligned next to an image, gets all overlapped with it (https://ibb.co/nz8xAF).

Can you fix that? Thanks! 95.247.30.245 (talk) 11:05, 8 July 2017 (UTC)Reply

Thank you for your report! Could you tell us what article you experienced this on? OVasileva (WMF) (talk) 13:15, 14 July 2017 (UTC)Reply

Error creating PDF

[edit]

Get an error trying to create a PDF of an article using the single column format:

{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Anatomy_of_a_Murder"} 32.210.178.11 (talk) 17:50, 8 July 2017 (UTC)Reply

 
Hi,Hey,Hallo 
Vielleicht hilft euch ja diese Fehler/Operationsbeschreibung weiter
Probably these errors/operations will be bringing you forward
>xxxxxxx< anonymisiert
>xxxxxxxx< anonymised
 
1.    Beispiel: (1ex)
 
Layout 1spaltig (mit Rahmen und Tabellen)
Layout1 (with frames and tables)
 
Fehler: Datei nicht gefunden
Error: Page not found
Die Dateien unter https://de.wikipedia.org/api/rest_v1/page/pdf/xxxxxxxxxxx konnten nicht gefunden werden. 
Bin zu Zweispalten-layout gewechselt
Turned to two-column-layout 
Worked !
Ging! 
2.    Beispiel: (2nd Ex)
 
Layout 1spaltig (mit Rahmen und Tabellen)
Layout1 (with frames and tables)
 
Fehler: Datei nicht gefunden
Error: Page not found
Die Dateien unter https://de.wikipedia.org/api/rest_v1/page/pdf/xxxxxxxxxxx konnten nicht gefunden werden. 
Bin zu Zweispalten-layout gewechselt
Turned to two-column-layout 
Worked !
Ging! 
Zurueck zu Layout 1spaltig
Back again to one-column-layout 
a)      Einmal ging es nun
Worked once
b)      Einmal ging es nicht
One time didn´t work again  
3.    Beispiel: (3rd Ex)
Habe den Neuladebutton im browser benutzt
Used refresh-Button in browser
 
https://de.wikipedia.org/w/index.php?title=Spezial:ElectronPdf&page=xxxxxxxxx&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpezial%3ABuch%26bookcmd%3Drender_article%26arttitle%3DKlaus%2Bxxxxxxxxxxx%26returnto%3DKlaus%2Bxxxxxxxxx%26oldid%3D166473913%26writer%3Drdf2latex
  
Meine  unerheblichen  Schlussfolgerungen:
My  unimportant conclusions: 
a)     Someone got hacked by stately crackers from >we don`t like wikipedia<
a)     Cracker; gibt Staaten die wikipedia nicht moegen 
b)     Nach etwaigen browserprozessen bedarf es mitunter >Neuanstossbefehlen< im script; wenn die nicht erfolgen, blockiert´s. Deshalb koennte ich mit meinen operativen Neuanstoessen erfolgreich gewesen sein 
b)     Following diverse browser-processes there might be a necessity for >restart/continue-orders< in the script ; if none, then >error<. Probably my successes are due to my >program-re-kicks/starts<
 
Grts, 
Stefan 78.42.121.245 (talk) 01:55, 9 July 2017 (UTC)Reply

Intermittent functionality

[edit]

Have been downloading numerous pages on the same topics, all have similar layout, tables and content. 80% of them download immediately and perfectly. Others remain stuck in download mode forever (for no apparent reason). Length or content of the page doesn't seem to affect the fault- just occurs randomly on some pages. 2001:8003:8609:6E00:2CDA:B708:4B1D:62BD (talk) 06:02, 9 July 2017 (UTC)Reply

Related to the same fault, occasionally get the following text after activating 'save as pdf'- {"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Victory_Medal_(United_Kingdom)"}
This sometimes persists over several attempts, sometimes the next attempt is successful, or sometimes the next attempt reverts to the 'forever downloading' fault described. 2001:8003:8609:6E00:2CDA:B708:4B1D:62BD (talk) 08:35, 9 July 2017 (UTC)Reply
I get the same. Some pages work, some spin forever, some say "Not found." I seem to have problems all the time with heavy math pages such as "Wave Equation" or "Del Operator." 71.163.239.245 (talk) 16:39, 9 July 2017 (UTC)Reply

Bonjour,

[edit]

Nous avons téléchargé l'article sur Joyce Carol Oates, en français, sous forme de fichier .pdf.

Nous l'avons imprimé avec le pilote Samsung Universal Printer, sur une imprimante SAMSUNG 1630, laser noir et blanc.

Nous avons choisi la forme livret, 2 pages par feuille. Soit 12 pages, imprimées sur 3 feuilles recto verso.

Aucune erreur à signaler, au chargement ou à l'impression. Impression très satisfaisante.

Jean-Philippe Guerder 80.12.42.64 (talk) 09:13, 9 July 2017 (UTC)Reply

Bonjour, aussi,
d`accord.... pas des problemes avec les fiches francaises. Pres d`aujourd´hui. :-)
Stefan 5.56.197.64 (talk) 00:18, 10 July 2017 (UTC)Reply
|.de    >eine Spalte< +pic
|Geht direkt/works directly/ca va
|https://de.wikipedia.org/w/index.php?title=Spezial:ElectronPdf&page=Joe+Dassin&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpezial%3ABuch%26bookcmd%3Drender_article%26arttitle%3DJoe%2BDassin%26returnto%3DJoe%2BDassin%26oldid%3D165819386%26writer%3Drdf2latex 
|-
|.fr     >colonne simple< +pic
|Geht direkt/works directly/ca va
|https://fr.wikipedia.org/w/index.php?title=Sp%C3%A9cial:ElectronPdf&page=Joe+Dassin&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSp%25C3%25A9cial%3ALivre%26bookcmd%3Drender_article%26arttitle%3DJoe%2BDassin%26returnto%3DJoe%2BDassin%26oldid%3D138823134%26writer%3Drdf2latex 
|-
|.en   >single column< +pic
|Geht nicht/didn´work/ne marche pas
|https://en.wikipedia.org/w/index.php?title=Special:ElectronPdf&page=Joe+Dassin&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpecial%3ABook%26bookcmd%3Drender_article%26arttitle%3DJoe%2BDassin%26returnto%3DJoe%2BDassin%26oldid%3D776802876%26writer%3Drdf2latex 
|} 
Comment:
ABER: nach dem parsing/rendering der deutschen und franzoesischen Version ging die Englische auf einmal auch
BUT: after parsing/rendering of the german and french pdf, the english version worked too
MAIS: Apres finisser les versions allemands et francaises, le pdf anglais marche aussi  
2nd attempt / zweiter Versuch / essaie No 2 
|.de    >eine Spalte< +pic
|Geht direkt/works directly/ca va
|https://de.wikipedia.org/w/index.php?title=Spezial:ElectronPdf&page=Voltaire&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpezial%3ABuch%26bookcmd%3Drender_article%26arttitle%3DVoltaire%26returnto%3DVoltaire%26oldid%3D166938709%26writer%3Drdf2latex 
|-
|.fr     >colonne simple< +pic
|Geht direkt/works directly/ca va
|https://fr.wikipedia.org/w/index.php?title=Sp%C3%A9cial:ElectronPdf&page=Voltaire&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSp%25C3%25A9cial%3ALivre%26bookcmd%3Drender_article%26arttitle%3DVoltaire%26returnto%3DVoltaire%26oldid%3D138335712%26writer%3Drdf2latex 
|-
|.en   >single column< +pic
|Geht nicht/didn´work/ne marche pas
|https://en.wikipedia.org/w/index.php?title=Special:ElectronPdf&page=Voltaire&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpecial%3ABook%26bookcmd%3Drender_article%26arttitle%3DVoltaire%26returnto%3DVoltaire%26oldid%3D789829099%26writer%3Drdf2latex 
|} 
Comment:
Deutsch:    a) zuerst einmal gleiches Ergebnis (deutsch/franzoesisch geht; english haengt)
                    b) der franzoesische download funktionierte sogar ohne Wartezeit
                    c) die englische Version funktionierte als Einspaltenversion erst, nachdem ich eine   
                        Zweispaltenversion gerendert hatte 
English:      a) at first: same result (german/french worked; english failed)
                    b) the french download worked without even waiting (directly)
                    c) the english version had to be rendered as two-column-version, before I was able to  
                        download the one-column-version  
Francais:     a) premier le meme resultat (allemand/francais marche; anglais ne marche)
                     b) le telecharger francaise marche sans esperer / directement
                     c) la version anglaise doit etre telecharche que >deux colonnes< avant la version  
                         >colonne simple< marche (excuse mon francais!)  
Schlussfolgerungen: KEINE. Ich vergleiche jetzt mal die Seitenquelltexte.
Conclusions: NONE. I´ll be trying to compare the source texts
Deductions: NE RIEN. J`essaie  de lire le texte source 5.56.197.64 (talk) 02:25, 10 July 2017 (UTC)Reply

PDF Generation And Wikisource..

[edit]

Wikisource does some things a bit differently... Would it be possible to look into ways of consulting with Wikisource about various specifics? Like for example

  • many mainspace articles on Wikisource are multi page transclusions and thus considerbaly longer that Wikipedia articles.
  • (English Wikisource ) Has some specfic formatting templates, that English Wikipedia doesn't, as well as some workarounds (given wikitext markup) has still not been revised to cope with a multi-page environment.
  • There is a specfic Table-style template not known by English Wikipedia

and so on ShakespeareFan00 (talk) 18:36, 9 July 2017 (UTC)Reply

Bug in Implementation.

[edit]

While the idea for a new implementation of OCG is needed. There are some flaws that need to be addressed before a complete rollout.

A) No delay when the generation is taking place. This leaves a hole where the file should be, but because of timing, the rendering engine has not completed the file creation. Error message is presented to the user when the 'file not found'.

In the old system , there was a two step process , where the article was rendered, then the article was downloaded to the user's PC as a PDF file.

This two step function has been reduced to one, however, if there is a significant amount of article to render, (ie >7 pages), then the current tool gives an error message, and offers to re-render the file again. It does not (like the former tool) offer the ability to wait until the rendering engine is complete.

An interim solution (for me anyway ) is to go back one page, then hit 'Download PDF' again, and this time the file is 'found'.

B) No way to know if the PDF generation is complete.

See A above,

However this is a little bit of functionality, that was removed in the new implementation,

Since there is no way to know when the rendering engine is complete (a rotating Icon or spinner?) The user gets fooled into thinking the file is ready, when it is not.

C) PDF storage of a current version.

I do not know how the back-end process of the former rendering tool worked, but I do think, that once a version of the document has been rendered, it is far less resourceful (or maybe not-depending on the price of storage) to just save most current version with a date-time marker embedded in the pdf, Whenever an initial request comes in that version is saved, and when a subsequent request arrives, that file is served to the user.

IF it never has been rendered before, then add a "please wait .... rendering now" to the file generating phase. Richard416282 (talk) 19:24, 9 July 2017 (UTC)Reply

Hi, 
In fact: You´re perfectly right. Most pdf-problems (page not
available) relate to the rendering-time. And as long, as I´m
spending  time,  clicking-around, I´m the >lucky moron<,
giving the necessary time to the machine. 
But: Even with the former functionality (display of rendering-
process on english wiki-pages), I had to face, that sometimes
rendering was not started. There might be a second issue. 
Greets,
Stefan   5.56.197.64 (talk) 22:06, 9 July 2017 (UTC)Reply
|.de    >eine Spalte< +pic
|Geht direkt/works directly/ca va
|https://de.wikipedia.org/w/index.php?title=Spezial:ElectronPdf&page=Joe+Dassin&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpezial%3ABuch%26bookcmd%3Drender_article%26arttitle%3DJoe%2BDassin%26returnto%3DJoe%2BDassin%26oldid%3D165819386%26writer%3Drdf2latex 
|-
|.fr     >colonne simple< +pic
|Geht direkt/works directly/ca va
|https://fr.wikipedia.org/w/index.php?title=Sp%C3%A9cial:ElectronPdf&page=Joe+Dassin&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSp%25C3%25A9cial%3ALivre%26bookcmd%3Drender_article%26arttitle%3DJoe%2BDassin%26returnto%3DJoe%2BDassin%26oldid%3D138823134%26writer%3Drdf2latex 
|-
|.en   >single column< +pic
|Geht nicht/didn´work/ne marche pas
|https://en.wikipedia.org/w/index.php?title=Special:ElectronPdf&page=Joe+Dassin&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpecial%3ABook%26bookcmd%3Drender_article%26arttitle%3DJoe%2BDassin%26returnto%3DJoe%2BDassin%26oldid%3D776802876%26writer%3Drdf2latex 
|} 
Comment:
ABER: nach dem parsing/rendering der deutschen und franzoesischen Version ging die Englische auf einmal auch
BUT: after parsing/rendering of the german and french pdf, the english version worked too
MAIS: Apres finisser les versions allemands et francaises, le pdf anglais marche aussi  
2nd attempt / zweiter Versuch / essaie No 2 
|.de    >eine Spalte< +pic
|Geht direkt/works directly/ca va
|https://de.wikipedia.org/w/index.php?title=Spezial:ElectronPdf&page=Voltaire&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpezial%3ABuch%26bookcmd%3Drender_article%26arttitle%3DVoltaire%26returnto%3DVoltaire%26oldid%3D166938709%26writer%3Drdf2latex 
|-
|.fr     >colonne simple< +pic
|Geht direkt/works directly/ca va
|https://fr.wikipedia.org/w/index.php?title=Sp%C3%A9cial:ElectronPdf&page=Voltaire&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSp%25C3%25A9cial%3ALivre%26bookcmd%3Drender_article%26arttitle%3DVoltaire%26returnto%3DVoltaire%26oldid%3D138335712%26writer%3Drdf2latex 
|-
|.en   >single column< +pic
|Geht nicht/didn´work/ne marche pas
|https://en.wikipedia.org/w/index.php?title=Special:ElectronPdf&page=Voltaire&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpecial%3ABook%26bookcmd%3Drender_article%26arttitle%3DVoltaire%26returnto%3DVoltaire%26oldid%3D789829099%26writer%3Drdf2latex 
|} 
Comment:
Deutsch:    a) zuerst einmal gleiches Ergebnis (deutsch/franzoesisch geht; english haengt)
                    b) der franzoesische download funktionierte sogar ohne Wartezeit
                    c) die englische Version funktionierte als Einspaltenversion erst, nachdem ich eine   
                        Zweispaltenversion gerendert hatte 
English:      a) at first: same result (german/french worked; english failed)
                    b) the french download worked without even waiting (directly)
                    c) the english version had to be rendered as two-column-version, before I was able to  
                        download the one-column-version  
Francais:     a) premier le meme resultat (allemand/francais marche; anglais ne marche)
                     b) le telecharger francaise marche sans esperer / directement
                     c) la version anglaise doit etre telecharche que >deux colonnes< avant la version  
                         >colonne simple< marche (excuse mon francais!)  
Schlussfolgerungen: KEINE. Ich vergleiche jetzt mal die Seitenquelltexte.
Conclusions: NONE. I´ll be trying to compare the source texts
Deductions: NE RIEN. J`essaie  de lire le texte source 5.56.197.64 (talk) 02:26, 10 July 2017 (UTC)Reply

Download fails with Chrome and Edge

[edit]

I just tried downloading a PDF in Chrome and got a line of HTML code instead. Tried using Edge and got a "could not be downloaded" message. 174.44.82.40 (talk) 22:03, 9 July 2017 (UTC)Reply

Book Creator Loses Pages & Content

[edit]

Book Creator loses a lot of pages & occasionally page content. I added over 100 Wikipedia pages to the Book Creator. I went to rearrange the pages & there were only 77. I also noticed that a little page content was lost very infrequently... e.g., lists, pictures, shortened tables. 75.91.240.244 (talk) 20:51, 15 July 2017 (UTC)Reply

Thank you for your comment. Any additional details (web browser and version, how long time it took you, your OS) are welcome, if you see this. Johan (WMF) (talk) 22:43, 16 July 2017 (UTC)Reply

Export with iMac

[edit]

I can't make helpful recommendations regarding the current problem at Wikipedia. BUT ... there is a very simple alternative if you are using an Apple computer, an iMac. {I don't know whether this method is available with a Windows platform}. Use the pull down menu from the "File" shown in the basic menu across the top of your screen. In the pull down menu you will see "Export as PDF." This has worked quite well for me. ........... I would attach a screen clip/view to show that, but I can't find a way to attach an image here. Good luck, Carl Thomas, Knoxville, TN 2601:840:8201:54F0:D82B:9D8E:F619:7258 (talk) 20:57, 15 July 2017 (UTC)Reply

Thank you, Carl. We'll continue working on making sure there's a good solution for everyone, regardless of OS, though. (: Johan (WMF) (talk) 22:42, 16 July 2017 (UTC)Reply

Black cover when downloading to iPad

[edit]

Downloaded a Norwegian Article, Norges_tekniske_h%C3%B8gskole , from the Wikipedia Desktop. The first attempt a had a black overlay on pages 2 and 3 (of 3) Second attempt using the "Open in IBook" button opened the PDF in iBook, but with page 1 covered by a black overlay. Tried again and the PDF in iBook was perfect. IPad 64 G, 1.2 GB available IOS 9.3.5 Used single column

Thank you for the new PDF functionality. It is very helpful. 92.220.145.86 (talk) 09:05, 16 July 2017 (UTC)Reply

Thank you. Johan (WMF) (talk) 22:41, 16 July 2017 (UTC)Reply

PDF download durch einen Trick, dann geht`s!!!

[edit]

Immer wenn ich ein Thema als PDF-Format herunterladen möchte mit einer Spalte und es funktioniert nicht, mache ich folgendes: Erst drucke ich es mit 2 Spalten und wenn ich das Dokument dann habe, hole ich es mir danach mit einer Spalte herunter. So klappt es meistens, also zu 99,99%!!! Es dauert halt etwas länger. Einfach warten. Es klappt!!!

Wenn das auch nicht funktioniert, tja dann über den PDF-Drucker von Windows drucken! Das bringt das Programm Adobe Acrobat Professionell mit. Luciferius Infernalis (talk) 17:36, 16 July 2017 (UTC)Reply

Danke. Johan (WMF) (talk) 22:41, 16 July 2017 (UTC)Reply

Keine Funktionalität in Google Chrome

[edit]

Nach Aufruf "C++-Programmierung: Template-Klassen" als .pdf einspaltig herunterladen (https://de.wikibooks.org/w/index.php?title=Spezial:ElectronPdf&page=C%2B%2B-Programmierung%3A+Template-Klassen&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpezial%3ABuch%26bookcmd%3Drender_article%26arttitle%3DC%252B%252B-Programmierung%253A%2BTemplate-Klassen%26returnto%3DC%252B%252B-Programmierung%253A%2BTemplate-Klassen%26oldid%3D564049%26writer%3Drdf2latex) erhalte ich nur folgende einzeilige Fehlermeldung:

{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/de.wikibooks.org/v1/page/pdf/C%2B%2B-Programmierung%3A_Template-Klassen"} Dstockinger (talk) 07:30, 17 July 2017 (UTC)Reply

pdf download error

[edit]

I receive the following error message when attempting to download a pdf:{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Gas_in_a_box"} 128.173.38.192 (talk) 18:20, 17 July 2017 (UTC)Reply

Thanks.
(If you see this, we always welcome more information, such as which web browser you have, which version of it, and which OS you use.) Johan (WMF) (talk) 21:08, 17 July 2017 (UTC)Reply
It would be really efficient to offer mini books that we could download as pdfs. For example, one mini book could be called electronic components. 66.227.213.107 (talk) 21:11, 17 July 2017 (UTC)Reply
Same problem:
using Chrome Version 59.0.3071.115 (Official Build) (64-bit)
on Windows 10.
Started on: https : // en.wikipedia.org/w/index.php?title=Special:ElectronPdf&page=Strategy&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpecial%3ABook%26bookcmd%3Drender_article%26arttitle%3DStrategy%26returnto ...
Selected Single Column
Clicked Download.
Got a page saying: {"type":"https : // mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Strategy"} Hmeijeri2 (talk) 01:40, 18 July 2017 (UTC)Reply

Fails to generate PDF

[edit]

page contains the following text instead of the expected single column PDF of the Lodash entry.

{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Lodash"} 69.92.118.32 (talk) 07:08, 18 July 2017 (UTC)Reply

Same results here ... Mpeetoom (talk) 07:36, 18 July 2017 (UTC)Reply
article about ISO 10993 did Show up with table in one column view. However, parts of the table were missing.
Seems that the right few columns of the table did not make it onto the page. 108.171.129.188 (talk) 07:43, 18 July 2017 (UTC)Reply
Thank you. Johan (WMF) (talk) 12:11, 18 July 2017 (UTC)Reply
Did not generate PDF for Vista Editions, always worked fine with the previous PDF generator. DownThemAll stuck starting the connection but never starts downloading.
Browser Firefox 54.01 x64 using DownThemAll plugin 3.08. 2601:C8:4004:8A47:CC23:B8D6:355F:705B (talk) 11:17, 20 July 2017 (UTC)Reply

pdf funktioniert unregelmäßig

[edit]

Statt pdf erhalte ich

{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/de.wikipedia.org/v1/page/pdf/Pulsoxymetrie"} 193.170.2.70 (talk) 10:46, 18 July 2017 (UTC)Reply

Jetzt funktioniert die file-Erzeugung, aber nur die erste Abbildung wird eingebettet 193.170.2.70 (talk) 11:18, 18 July 2017 (UTC)Reply
Danke!
(Wenn du meine Antwort siehst: Es wäre uns hilfreich wenn du uns erzählen könntest, welcher Browser, welche Browserversion und welches OS du verwendest.) Johan (WMF) (talk) 12:11, 18 July 2017 (UTC)Reply
leider etwas spät: jetzt funktioniert es mit Safari und Firefox (jeweils aktuelle Versionen) 193.170.2.70 (talk) 13:47, 18 July 2017 (UTC)Reply
Danke. (: Johan (WMF) (talk) 02:50, 19 July 2017 (UTC)Reply

Fails to generate PD

[edit]

FНе работает :( 178.134.99.71 (talk) 18:06, 18 July 2017 (UTC)Reply

book

[edit]

fails to create book if more than 100 pages 103.21.127.78 (talk) 09:45, 19 July 2017 (UTC)Reply

Thank you. If you see this – could you tell us what browser, browser version and OS you use? Johan (WMF) (talk) 20:37, 19 July 2017 (UTC)Reply

Readable - much better in single column.

[edit]
This new format, single column, is LOTS better and much easier to read than the other dual-column format. 121.44.110.73 (talk) 11:24, 19 July 2017 (UTC)Reply
Agreed! 91.7.13.2 (talk) 09:02, 22 July 2017 (UTC)Reply
Thank you for your feedback. Johan (WMF) (talk) 20:37, 19 July 2017 (UTC)Reply
I agree. 24.158.90.205 (talk) 13:47, 21 July 2017 (UTC)Reply
It's pretty much better now... 49.35.20.165 (talk) 15:40, 21 July 2017 (UTC)Reply
I find it much more readable and attractive than the old double column PDFs. 67.1.177.196 (talk) 19:09, 21 July 2017 (UTC)Reply

Two columns with graphics, table, picture ....is better.

[edit]

Two columns with graphics, table, picture ....is better. 176.186.159.222 (talk) 02:26, 22 July 2017 (UTC)Reply

Thank you for your feedback. Johan (WMF) (talk) 16:35, 24 July 2017 (UTC)Reply

De momento funciona bien. Gracias

[edit]

De momento funciona bien. Gracias por la novedad 81.32.143.3 (talk) 10:51, 22 July 2017 (UTC)Reply

Gracias. Johan (WMF) (talk) 16:34, 24 July 2017 (UTC)Reply

Single and double columns: both are equally useful

[edit]

Hi!

Firstly, I would like to thank everyone working on this issue for their time and commitment to maintaining and improving Wikipedia. Your efforts are significantly appreciated.

After reading the project plan overview, I would like to provide my input that I think having the option of double columns should be a long-term priority, even if it isn't viable at the moment. While acknowledging the urgency of the issue, and also the limits currently at hand, Wikipedia benefits from having the two options. I find that the effectiveness of the format really depends on the article (and personal preference) when it comes to what works best--different articles having different criteria to consider. For example, long articles with large amounts of text can sometimes benefit from having the text broken into two columns, but shorter articles or articles with info boxes and tables may work best with the single column format.

In any case, I recognize that the priority is to quickly find a functional alternative, so once again, I send my words of appreciation!

Best,

Sturgeontransformer Sturgeontransformer (talk) 05:14, 23 July 2017 (UTC)Reply

Readability

[edit]

Had no problem downloading and saving the new single column PDF (using Windows 10 OS). All went well. However, please, continue offering the 2 column version as well. Years ago in publishing we were taught, and it has continued to prove true in personal experience, that the human eye and brain can best see and comprehend without fatigue a maximum of 45 characters (including spaces) per line. Those of us with less than perfect sight and cognition, but who are nonetheless avid reader/researchers, will be extremely grateful. The double column version of the same article was also successfully downloaded (took a bit of time) and saved to replace the single column version. In agreement with "Sturgeontransformer," thank you to all the providers at Wikipedia! Reader Eve (talk) 13:37, 24 July 2017 (UTC)Reply

Download failed

[edit]

Download failed with following error response :

{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/de.wikipedia.org/v1/page/pdf/Kr%C3%BCmmung"} 95.208.35.190 (talk) 14:21, 24 July 2017 (UTC)Reply

Thank you for reporting this. Johan (WMF) (talk) 16:34, 24 July 2017 (UTC)Reply

"Printable Version" works; "Download as PDF" inadquate

[edit]

Thanks for working on this. I've long considered the "feature" that "Create a book" and "Download as PDF" omitted tables without warning as "user hostile". Fixing that is really important. I'm very glad it's finally getting attention.

In my tests just now, I found that "Download as PDF" was still largely unacceptable, because of the absence of page numbers and other information in headers and footers; at least I now get a warning that infoboxes and tables will not be printed if I select two columns.

Fortunately, the "Printable version" provides reasonable defaults for headers and footers:

  • Date in upper left.
  • Article title and Wikimedia project (e.g., Wikiversity) centered in the header.
  • URL in lower left.
  • Page number and count in lower right.

Why can we get headers with "Printable version" but not with "Download as PDF"? I would seem like an easy fix to direct them both to the same code.

Longer term, may I suggest that the dates (at least optionally) should be in the format of ISO 8601 -- or at least should have that as an option. (By the way, your sexy WYSIWYG editor did NOT want to let me link to the Wikipedia article on "ISO 8601": I could get a link to a non-existent Wikiversity article on "ISO 8601", which displayed in red as a broken link. I failed to find a way to link to Wikipedia -- until I figured out how to switch to "Source editing". Before that, I wrote [[w:ISO8601|ISO 8601]] outside of Wikiversity, then pasted that string in place. That worked but was clumsy.)

Without ISO 8601, dates are a problem from multiple perspectives.

  • Most obviously, is "6/7/2017" the seventh of June or the sixth of July of 2017? It depends on the context: It could be either -- and could contribute to stupid mistakes. Worse, is "01/02/03" the second of January of 2003 or the first of February of 2003 or the second of March of 2001 or the third of February 2001? It's probably NOT the first of March of 2002 nor the third of February of 2002; that helps some ;-)
  • More subtly, "allows dates to be naturally sorted" using lexicographical order. Otherwise, "6/7/2013" sorts between "5/6/2015" and "7/6/2016". That seems pretty dumb to me. [It's not as dumb as the fact that the "[[w:October Revolution|October Revolution]" occurred in November (1917-11-07, New Style). It's also not as dumb as the October, the "eighth month," according to its name, is actually the tenth ;-) But I digress.]

Again, thanks so much for working on this. DavidMCEddy (talk) 15:25, 24 July 2017 (UTC)Reply

Margin control is inadequate

[edit]

The "Printable version", which is the only "Print / export" function I currently find useful, could use improved margin control, especially since the default margins are something like 0.38 inches on all sides. These margins are fairly narrow for many purposes. They are unacceptable for book binding, because binding the book typically consumes part of a margin; this is often followed by trimming the other three edges to make the product look more smooth and professional.

Allowing the user to modify those margins may not be easy, because the defaults seem to vary depending on which browser one is using.

In Google Chrome, I found that I could theoretically adjust the margins with "More settings" > Margins > Custom. However, changing those settings did NOT change the placement of headers and footers, part or all of which might be lost in book binding, as noted above.

I noticed this was different in Safari and Firefox. DavidMCEddy (talk) 15:52, 24 July 2017 (UTC)Reply

Unfortunately, the positioning of headers and footers are completely browser specific as well (you can however disable them in most print dialogs). —TheDJ (Not WMF) (talkcontribs) 17:52, 24 July 2017 (UTC)Reply

great idea going from two to one column layout.

[edit]

great idea going from two to one column layout. 2602:306:BDA7:DFA0:5DA9:17AE:ECCC:F6DF (talk) 12:55, 25 July 2017 (UTC)Reply

I agree. Just downloaded a pdf of an article. It's neat, well-spaced and easy to read. Thanks for the improvement! 71.121.196.196 (talk) 15:18, 25 July 2017 (UTC)Reply
I agree. 59.99.111.228 (talk) 18:12, 28 July 2017 (UTC)Reply
Yes! Good to have so much improvements with the platform: Options of 1 or 2 colonnes not to mention the book creator function so useful in compiling related topics. Big up to Sir Jimmy Wales.
We can only give our full support to his project of "news wiki" for a reliable democratic web content. Thanks Njiebeye (talk) 20:33, 28 July 2017 (UTC)Reply
The single column is much more readable, and the typeface easy on the eye.
However, the links are not coming through. This was a very useful part of the earlier downloads. 86.113.59.24 (talk) 14:33, 30 July 2017 (UTC)Reply
As long as there's the option for two column layouts still :P 136.24.86.171 (talk) 21:22, 31 July 2017 (UTC)Reply

2nd time mis-parse of title with " 's "

[edit]

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.


Hubble's_law pdf showed up in save dialog file name as " 's_Law " making this the second time I have seen this error of leaving off characters before " 's". 68.98.170.156 (talk) 19:48, 25 July 2017 (UTC)Reply

Thanks for reporting. Johan (WMF) (talk) 21:46, 25 July 2017 (UTC)Reply
Thanks for reporting this. We were also able to reproduce this and filled the following bug: https://phabricator.wikimedia.org/T171747 OVasileva (WMF) (talk) 14:50, 26 July 2017 (UTC)Reply
This has been resolved in production. Thanks for the report! Gabriel Wicke (GWicke) (talk) 20:28, 27 July 2017 (UTC)Reply
The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.

Successful Download

[edit]

The new methodology worked easily and the One Column format is preferred for the application I needed. Two column will have its place so please keep these options. The only issue noted is incorrect pagination even when space was available for correct alignment of article text & titles. 165.156.39.52 (talk) 21:30, 25 July 2017 (UTC)Reply

successful download

[edit]

worked fine single column here: Retrieved from "https://commons.wikimedia.org/w/index.php?

title=Chlorogalum_purpureum&oldid=178831711" 2602:304:B35A:C5D0:F136:F5C6:51A0:D505 (talk) 18:22, 26 July 2017 (UTC)Reply

Regarding PDF-creation functionality on the web: bug

[edit]

You probably already know this, but there is no header and no (visible) document footer in created PDF files, although the paper format seems to be appropriate (Letter/A4 slight differences); also, page break (in respect to tables) is aesthetically bad, although no table text seems to be missing/lost. Best regards.

[1] 213.198.236.239 (talk) 20:04, 26 July 2017 (UTC)Reply

Much more reliable and consistant downloads

[edit]

This new "Download as PDF" has been more consistantly reliable when formating the downloading in PDF file format. So far no error messages about not being able to format the web page content for downloading of the page as a PDF file with active links. Keep up the quality. Thanks. 2601:181:4101:66EA:1431:11A5:866C:304 (talk) 12:30, 28 July 2017 (UTC)Reply

Thank you. Johan (WMF) (talk) 14:09, 28 July 2017 (UTC)Reply

#1 add a warning notice / #2 github open source

[edit]

#1 Consider a warning notice that says "If error occurs generating pdf of page, suggest using your OS 'Print to PDF' options"

Chrome has "print to PDF" that works pretty well (which is what I use when the process fails).

#2 Make your PDF routines an open source effort on GitHub w/ some oversight. 170.170.59.138 (talk) 21:18, 30 July 2017 (UTC)Reply

#1 FYI, this new "Download as PDF" is basically serverside Chrome print to pdf :)
#2 Everything we do is open source and mirrored on GitHub —TheDJ (Not WMF) (talkcontribs) 09:09, 4 August 2017 (UTC)Reply
DJ - can you reply here with a link to the GetHub folders for the pdf effort? thx 170.170.59.138 (talk) 02:04, 13 August 2017 (UTC)Reply
Extension:ElectronPdfService
Diffusion browse
GitHubTheDJ (Not WMF) (talkcontribs) 04:26, 13 August 2017 (UTC)Reply

References are not set properly

[edit]

I have noticed that when for example asking for PDF on "Emily Bronte" entry. The numbered references are in order. However the alphabetised references are converted to numbers as well and in the actual article the reference numbers are shown as "lower-alpha 1", "lower-alpha 2", etc. The error is specifically occurring in the two-column layout and does not occur in the one column layout. Tirab (talk) 09:01, 31 July 2017 (UTC)Reply