Talk:Reading/Web/PDF Functionality/2017/06

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.

js snippets

see https://zh.wikipedia.org/wiki/User:Shizhao/asPDF.js zh:User:Shizhao/asPDF.js Shizhao (talk) 08:57, 2 June 2017 (UTC)

Why limited to PDF functionality?

In the day and age of the electronic document, I wonder why we limit the scope of the document generation to PDF alone. Wouldn't and shouldn't we be thinking about PDF / EPUB / MOBI / ... collectively in our outputs rather than looking to generate a document that is set to a standard page size and is not greatly extensible? — billinghurst sDrewth 05:00, 6 June 2017 (UTC)

Easy.. resources! Who's doing it ? :)

Currently all of this is hardly maintained to begin with. And the previous extension is failing because no one is resourced for it. I don't see this easily improving in the short term. —TheDJ (Not WMF) (talk • contribs) 14:43, 9 June 2017 (UTC)

Wikisource and PDF

Currently, the PDF functionality is not suitable for Wikisource users (because of multi-page texts). Instead, we are relying on the WSExport tool (which works fine, but with all the pros and cons of external tools). Are there any plans about adapting the PDF functionality to Wikisource's needs? Candalua (talk) 07:24, 6 June 2017 (UTC)

Definitely agree, though part of the difference is that the WSes are generally creating works that are self-contained, so a user can publish a whole work, or a specific subpage. There seems less call/need to build one's own compilation, as a wikivoyager or wikipedian may wish to do. — billinghurst sDrewth 02:11, 7 June 2017 (UTC)

The Wikisource projects normally create texts that are several pages long. With the PDF functionality that is now available for testing , page numbers are shown but are covered by the text. This is apparently a problem with the functionality itself, not the PDF reader. Thuresson (talk) 22:43, 9 June 2017 (UTC)

{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/es.wikipedia.org/v1/page/pdf/Playa_Gir%C3%B3n"} 190.247.153.175 (talk) 23:12, 10 June 2017 (UTC)

@Candalua - could you give us some detail on why the PDF functionality is not suitable for wikisource? You mention multi-page texts, but is the focus on providing page numbers and table of contents (which we will be including in the final version of the new renderer), or is it that you would prefer more other file type options, such as mobi? OVasileva (WMF) (talk) 13:38, 15 June 2017 (UTC)

@OVasileva (WMF) Generally, Wikisource texts (excluding very short ones) are composed of a "base page" (let's say "Alice in Wonderland") and many subpages ("Alice in Wonderland/Chapter 1", "Alice in Wonderland/Chapter 2" etc.). The WSExport tool takes the base page as input, collects the subpages and packs them together into a single file, all automatically. On the other hand, the "canonical" PDF functionality requires the user to manually select each chapter page. Providing all the most used file formats is also another thing that WSExport does and PDF functionality does not. I suggest taking a look at WSExport code and see if it's possible to embed all or some parts in Mediawiki (you can ask the author, @Tpt, for more details). Candalua (talk) 13:57, 15 June 2017 (UTC)

@OVasileva (WMF) Here is an example of PDF generated by Wsexport from a Wikisource page. Wsexport have been created in order to have the following features not covered by the previous PDF export systems provided on Wikimedia projects:

1. Make the exportation tool work with Wikisource special tags (especially <pages>). It was mostly relevant when the PDF generation backend what the one developed by PediaPress with a custom MediaWiki parser.
2. Being able to download a book in one click from the Wikisource main page of the book, without having to create "collection" pages for each book (that would be a tedious process and would duplicates contents
3. Having more formats than PDF (ePub, mobi...)
4. Being able to extract and display the relevant metadata: the author of the content is the author of the original work (Shakespeare...) and not the wiki contributors that are only credited in a "about" section at the end of the PDF. It is very relevant for ePub which embeds metadata and to create front pages.
5. Customize a bit the rendering for Wikisource needs.

If you are interested here are some statistics on Wsexport usage. Feel free to reach me if you want to help. I would be very pleased to see the hacky Wsexport replaced by something on production cluster. Tpt (talk) 14:45, 15 June 2017 (UTC)

@OVasileva (WMF): do you have any comment? Candalua (talk) 10:24, 11 July 2017 (UTC)

We're working on updating our styles to fit some of the wikisource requirements. However, it might take us a while to get to a specialized solution. For the initial solution, we will attempt to allow the renderer to function with wikisource text (which should take care of 1 and 4). 2 (customizing output) is currently scheduled for the second portion of this effort, which will most likely be during next year. During this stage, we will also consider allowing multiple formats, although I have to warn that due to limited resources, we will probably not be able to support this in the near future. OVasileva (WMF) (talk) 13:12, 14 July 2017 (UTC)

@OVasileva (WMF) we have no hurry :), the good news is that you're working on it. If you need any information or want feedback from users, just ask on the Wikisources' Village Pump and we'll try to help you :) Thank you. Candalua (talk) 15:36, 14 July 2017 (UTC)

Hi,

created my wikipedia/media-account 2 days ago.

Used to work as data&database-analytic in hightech/highend for some years.

Meanwhile working on my private database, I´m coming near to downloading

some thousand wiki-features. For several reasons I want to obtain an offline-version of my database.

My database contains the chapter e-book/epub as well ( and full linux/base fumctionality will

be required later). NOW I´m looking for some folks, involved in the pdf-export/generation-story,

because the pdf-rendering-(bug)/problems take my time.

If you know some wikipedians in charge of the compilation issue, please tell me.

Meanwhile I´ll be comparing (wiki) source-codes on my hunt for the export-bugs.

Greets,

Stefan Stefan Sakowitz (talk) 01:13, 12 July 2017 (UTC)

The new function of forming PDF in one column is much more convenient 92.126.206.215 (talk) 10:00, 12 July 2017 (UTC)

I always wanted downloaded PDFs with only one column instead of two. Thanks for considering the changes. The last page is completely blank, the size of the PDF can be reduced by removing it. Thanks.

Regards,

Pooja 49.34.2.143 (talk) 12:50, 15 July 2017 (UTC)

Pdf and lilypond

With wikisource we are able to transcribe musical scores using lilypond, but when using the download/export as pdf tools the musical scores aren't rendered at all at wikisource, and also testing it on wikipedia the scores are rendered but are all over the place. From what I know lilypond has built in functionality for creating pdfs. For wikisource it would be great to be able to download and print a ready made pdf made by lilypond with a click of a button. Jpez (talk) 04:10, 7 June 2017 (UTC)

Proposed New PDF Format Choices

I like the two (2) column page format of PDF for the downloaded presentation of formation that has been used by Wikipedia for years. I like it just things just the way there are right now without any changes. Except for the use of even more illustrations in the articles. If it is helpful to others, I would like to see both a one (1) and two (2) column per page format choice to go forward. Having a choice is nice. However, given only one (1) choice, I would prefer the two (2) column presentation much better. Whatever decision Wikipedia makes, I would like to make it clear that I would prefer a PDF that does not lose the ability to transfer charts, graphs, or tables in either format. There is far too much information that is presented in the many subjects available Wikipedia that would be lost if this kind of information were not transferred from Wikipedia's articles to a PDF. I can not stress this point strongly enough. Consider how many useful lists are included in Wikipedia articles or provided as stand-alone articles, how many things or groups of information that can not be put into words that charts and graphs explain, or how is easy it is to compare and contrast an endless number of things when it is portrayed in a visual manner. Therefore, please keep the visual information in and add as much as possible to both the Wikipedia articles themselves as well as the PDF version of Wikipedia's article. Thank you for your consideration on this matter. 2601:586:201:8B4:A509:2B6C:AC0E:36F4 (talk) 20:34, 8 June 2017 (UTC)

Before the change, the pdf downloaded to a browser window so it could be viewed and checked to see if the page was rendered correctly (to determine whether a browser file/print/save to pdf was required).

Now it just downloads the pdf with a save as dialog. Please set the default to download to a browser window as before, perhaps with a checkbox for direct download. 68.98.170.156 (talk) 14:02, 9 June 2017 (UTC)

ROLL BACK!

The old version is fine! 115.193.4.29 (talk) 04:16, 10 June 2017 (UTC)

We'd be very happy to leave the old version if we had the option. However, quoting the page:

"Our current PDF rendering service, the offline content generator, is no-longer maintainable. Simply put, it's breaking down. Originally created by a third party, it currently runs on outdated code which may introduce security vulnerabilities and other major issues in the future. If we're to have the PDF functionality, we unfortunately have to replace it, or we might suddenly find ourselves in a situation where we'd have to take it down without having planned to do so."

We're not doing this because we want to, we're doing it because we need to if we're to make sure we still have a PDF service.

Is there anything in particular you think the old version has that you're afraid we'll be missing that we should work hard to include? Johan (WMF) (talk) 16:59, 10 June 2017 (UTC)

Funktioniert leider nicht

Funktioniert leider überhaupt nicht :-( 2A02:8070:A1A1:5400:C19D:7B9D:C8D9:F67F (talk) 13:50, 10 June 2017 (UTC)

Was genau funktioniert nicht? / What exactly does not work? Malyacko (talk) 16:33, 10 June 2017 (UTC)

das erstellen der pdf Datei funktioniert nicht 94.216.146.75 (talk) 16:58, 10 June 2017 (UTC)

Und wo? Deutschsprachige Wikipedia oder woanders? (And where? German Wikipedia or somewhere else?) Johan (WMF) (talk) 17:04, 10 June 2017 (UTC)

Invalid page at Google Chrome

At wikipedia, I tested Single column at chrome Version 58.0.3029.110 (64-bit) and Version 59.0.3071.86 (Official Build) (64-bit) at both of them it downloads *.htm file which has this error

Invalid page The specified page is not valid. Yamaha5 (talk) 14:19, 10 June 2017 (UTC)

On which of the Wikipedias? And on which exact page? Malyacko (talk) 02:37, 11 June 2017 (UTC)

on fa.wikipedia and en.wikipedia i tested it on random article Yamaha5 (talk) 05:47, 11 June 2017 (UTC)

Non-indicaton of hyperlinks

Is it possible to indicate in PDFs, by coloration, where there are hyperlinks in the original. Not essential, but it would be nice. Thanks PaulGEllis (talk) 19:12, 10 June 2017 (UTC)

Thanks for your feedback. We will be including hyperlinks in the final version of the new renderer. OVasileva (WMF) (talk) 13:39, 15 June 2017 (UTC)

Prefer single column

I much prefer the single column format. I've found in the past that having two columns complicates the reading. I also agree with PaulGEllis below that it would be nice to indicate where hyperlinks are in the original. El Guero (talk) 19:57, 10 June 2017 (UTC)

العاب 129.45.112.114 (talk) 13:29, 12 June 2017 (UTC)

Thank you for your feedback. We're currently investigating adding hyperlinks to the final version of the single column format. OVasileva (WMF) (talk) 19:18, 12 June 2017 (UTC)

I am not as skilled as many who are replying to changes being made to downloaded PDF's, I only moments ago had my first exposure to the new single column format, and "I Loved It!" Desiderata, and keep up the good work. Not a Geek, more an Infoeek, 72.169.96.111 (talk) 19:37, 14 June 2017 (UTC)

Nearly as good as it was three years ago!

The loss of tables in PDF files was brought to your attention when you changed to the current (now old) method, yet nothing was done about it. The single column method at least gets us tables back in PDF's, but would it not have been easier to revert to the original (2013 or thereabouts) method? 194.75.13.186 (talk) 21:41, 10 June 2017 (UTC)

Table support got investigated (see e.g. https://phabricator.wikimedia.org/T135643 or https://phabricator.wikimedia.org/T73808 ). Some issues with the "original" method are listed in https://en.wikipedia.org/wiki/Wikipedia_talk:Offline_Content_Generator Malyacko (talk) 02:44, 11 June 2017 (UTC)

True, this brings back the simple and working functionality there was some time ago. I do vividly recall, when I downloaded wiki docs as PDFs and I didn't have any missing fields or data! I don't know why they ever changed that service, at least now they realise that the current service is whack and needs replacing!! 41.57.217.71 (talk) 13:28, 11 June 2017 (UTC)

If you want to "know why they ever changed that service", you could click the last link in my previous reply to read C. Scott Ananian's explanation why the "original" method had to get replaced. Malyacko (talk) 18:15, 11 June 2017 (UTC)

How long does it take to fix it?

I would like to use book creator as fast as possible. If it's take too long, can I save the book creator? 60.50.249.11 (talk) 02:20, 11 June 2017 (UTC)

I'm not sure I completely understand your question. Is your problem that the book creator is too slow when it creates the PDF and you would want to try later, to see if works better then? Or are you talking about the step where you collect the articles that go into the book? Johan (WMF) (talk) 14:48, 14 June 2017 (UTC)

Or do you mean that you won't be able to use the book creator? Johan (WMF) (talk) 15:36, 14 June 2017 (UTC)

Double column

Will we able to create double column pdfs with tables and infoboxes as well ? Please do so, double column is so much better than single column. I am sure it can be done by merging the two columns together so that a table takes the entire page. 125.253.101.199 (talk) 05:04, 11 June 2017 (UTC)

Plantage du PDF sur impression

Bojour,

Choix du PDF > Télégarger > et on obtient une page blanche avec cette ligne :

<HTML><CODE>{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/fr.wikipedia.org/v1/page/pdf/Staffelberg"}</CODE></HTML>

Mon navigateur est le YANDEX russe basé sur Chromium et autres composants.

Cordialement. 128.79.114.16 (talk) 09:44, 11 June 2017 (UTC)

Functionend excellently - very quick!!! 178.202.52.91 (talk) 10:21, 11 June 2017 (UTC)

Single Column

Single column download is not working Agarwalpravin7 (talk) 17:58, 11 June 2017 (UTC)

Could you please provide an exact link to the page where this problem happens, so someone else could try to reproduce the problem that you see? Malyacko (talk) 18:07, 11 June 2017 (UTC)

Unable to make a document pdf or text

Please provide steps to reproduce (for example a link) so others can try to find out if they get the same problem. Malyacko (talk) 22:01, 11 June 2017 (UTC)

Who needs a 2 column download  ??

What makes a 2 Column download useful or even desireable ? > and most importantly worth the time (and money) to program it ??  I download a lot of information in PDF format; simple text 'tables' are edited to fit an 8.5 x 11 page with margins. When I search for data which I want to analyze, Most sites offer a spread-sheet in EXCEL format ( *.xls or *.csv) If you really want what is displayed on the website do a capture, paste it into an Excel worksheet; where you remove all of the Internet links before editing for content. If you want (need) to present a document that will not have anyones eyes "glaze over" , save each worksheet in PDF formate. You be happy to know that many sites have PDF versions of their data.. Allan Mackey-51 (talk) 23:55, 11 June 2017 (UTC)

It's more readable. —Hugh (talk) 22:18, 28 August 2017 (UTC)

I concur. I prefer a one-column format - I never want 2 columns. 72.196.221.41 (talk) 15:47, 29 August 2017 (UTC)

2 columns are important to read the pdf well on a e-book! 87.157.152.136 (talk) 09:37, 27 October 2017 (UTC)

The major problem with the 1 column format is that it is NOT readable without manually scrolling back and forth on every line of text on smaller smart phones (e.g., Samsung Galaxy S7 USA). Reducing the font size down so that the 1 column format fits across, even in wide screen rotation, makes it very difficult to read. The two column format makes a block of text beautifully readable, please bring it back! 67.61.192.182 (talk) 09:12, 29 October 2017 (UTC)

Anyone who is used to reading academically, will be more than familiar with the two-column layout that is endemic in academic papers.

Its' actually a space saving tool and promotes readers to skim the information much quicker. 82.16.167.155 (talk) 11:46, 13 January 2018 (UTC)

I write papers for school in 12 pt font, double spaced. And theses and dissertation still use this standard, so as to be able to read a 12 pt sentence across the whole line, and then, on to the next sentence with ease. Also, it allows space for revisions in marking up the text. But this revision point, should not apply to finish texts.

It is bad typography to reduce 12 pt to single space, as this becomes unreadable on 8 1/2 by 11 paper (though people get used to it, and do read it).

So then there is the use of larger font with single spacing, or smaller font with some extra spacing beyond single. The latter is currently done.

I think the rule of thumb in typography I read somewhere long ago, is like 7 words per line. Newspaper columns used less.

I print out all my final papers in book format, double column, 9 pt font (like the printed Encyclopedia Britannica). I still make corrections on these, with symbols, and along the margins. And, I can see my whole paper more clearly (good paragraphing helps here). Double column feature was fantastic. I've printed a lot of single column PDF articles on a lot of paper, and it looks OK, and appears seamless with photos. Double column would make the sentences more readable, and save paper.

People aren't use to the double column look. I forget how photos were handled with it.

--A.A.

PS. Line spacing feature? 2600:1700:E120:C800:ED8D:5365:8808:EEBD (talk) 20:54, 21 January 2018 (UTC)

Beth Orton - Central Reservation

For the first time in ages I was able to capture the entirety of a wikipedia page in .pdf format. (Single column print). Thank you for replacing the previous ineffective method, which treated tables with contempt.

Windows 7 PC using Palemoon non-bloatware fork of Firefox, with Adobe pdf add-in. 103.17.198.204 (talk) 02:56, 12 June 2017 (UTC)

Thank you. Johan (WMF) (talk) 11:52, 12 June 2017 (UTC)

Single Column Failure

Every time Wikipedia tries to make single column PDF from this link [1], it encounters unknown error. 103.25.251.118 (talk) 06:27, 12 June 2017 (UTC)

I have found the present single column PDF is really excellent. I am sure I get more information 2001:8003:54AB:F700:BCE6:B2CB:A1DE:D608 (talk) 08:43, 12 June 2017 (UTC)

{"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Amphetamine"}

This is not specific to the topic but more frequent with heavy Internet traffic and security filters. When bad 50% failure rate common. Pressing button again on TouchPad mobile crashes through at the expense of double copies. One can be deleted but none is gone forever in the frantic rewriting now typical. The alterations look a lot like vested interests or government fact control incompatible with information sharing. Usually the error is accompanied by delays in appearance of Android 'download' of 5-15 seconds The original issue could be linked to the failure of linkages from pictures which worked a year ago. It has solved the problem of cut and paste into Word destroying chemical formulas and blank cross referenced pictures..

Apologies for anon but an appropriate atavar for different postings troubles me. I have a professional identity as well as personal and not happy with unique identity tag for abuse by Government. 49.178.25.159 (talk) 11:26, 12 June 2017 (UTC)

Thank you for reporting the problem. Johan (WMF) (talk) 11:52, 12 June 2017 (UTC)

lines cut off edge of page

This Wikibook (wikibooks:Professional and Technical Writing/Print version) has lines of text cut off the edge of the page when exported as single-column PDF in Google Chrome for Linux Mint. Nicole Sharp (talk) 13:48, 12 June 2017 (UTC)

Even with Google Chrome print-to-PDF as horizontal landscape pages, the text is still cut off. Might be formatted badly. Nicole Sharp (talk) 13:55, 12 June 2017 (UTC)

Attempting to export a single wikipage to PDF results in a error message (wikibooks:Professional and Technical Writing/Rhetoric/Assumptions). However, print-to-PDF works for individual pages, unlike the single-file version. Nicole Sharp (talk) 14:03, 12 June 2017 (UTC)

I was finally able to successfully export to PDF the Wikibook without errors by compiling the pages into a Collection: wikibooks:Wikibooks:Collections/Professional and Technical Writing. Nicole Sharp (talk) 14:12, 12 June 2017 (UTC)

It is still missing images and the chapter numbers are wrong, but at least the text is readable as PDF. Nicole Sharp (talk) 14:20, 12 June 2017 (UTC)

Thank you for reporting this. Johan (WMF) (talk) 14:49, 12 June 2017 (UTC)

Top info box not in all current .PDF files

The top info box has excellent info which is not always covered in the article. It may be that old .pdf files which were generated before the new system went into effect are the only ones that have this problem. e.g., Seriema doesn't have the info box and important information is missed (e.g., distribution map). On the other hand Montenegro does include that box in it's pdf. Hal Molitor (talk) 21:21, 12 June 2017 (UTC)

Thanks for your feedback and you are correct - only the new (single-column) version of the PDF renderer will be able to consistently include infoboxes. OVasileva (WMF) (talk) 13:28, 15 June 2017 (UTC)

Single-column files downloading fine. I prefer two columns though.

Don't know if this is any help, but whatever bugs are affecting your PDF function appear to be more pronounced when downloading files in two-column format. Single-column files are downloading fine if one chooses to ignore the error messages, as I have done.

Technical issues aside, let me just say: Long live Wikipedia. I couldn't imagine the modern world without this laudably idealistic resource!

Any problems attendant to sharing knowledge in a democratic, egalitarian fashion are as nothing compared to the potential hazards of leaving the generatiion of information to the power mongers and their ilk, who have a long history of generating fake news to further their megalomania. 2003:55:6C1C:3901:8D89:E18F:36EB:91FF (talk) 22:52, 12 June 2017 (UTC)

my computer `s pdf is very well done therefore in my opinion

my computer `s pdf is very well done therefore in my opinion the problem i have about using pdf  is wiki'online could you explain how to deal with pdf'malfuction?  125.139.182.203 (talk) 23:51, 12 June 2017 (UTC)

I'm afraid we're not quite capable of helping you based on the information you've shared with us. If you have problems with the PDF service, you have to explain where (on which wiki, e.g. "English Wikipedia" or "Faroese Wiktionary"), which article/page and exactly what happens that you're not happy with, i.e. what do you see, any messages, if not, what it is that tells you it's not working. Johan (WMF) (talk) 02:22, 13 June 2017 (UTC)

Copied from elsewhere

In this thread I'll paste edits replying to the question that ended up in the wrong place. Johan (WMF) (talk) 10:54, 13 June 2017 (UTC)

"Die Funktion hat keine Rückmeldung, wie weit sie ist. Wenn sie nicht fertig ist, sagt sie einfach, das Dokument existiert nicht." – anonymous edit of the board description Johan (WMF) (talk) 10:57, 13 June 2017 (UTC)

Translation from de to en: The function(ality) provides no feedback how far it [the process] is. If it is not done/ready [yet?], it just says that the document does not exist. AKlapper (WMF) (talk) 11:08, 13 June 2017 (UTC)

"I think it is brilliant!" – Special:Diff/2486666 Johan (WMF) (talk) 10:59, 13 June 2017 (UTC)

Rendering Graphics ( references )

https://de.wikipedia.org/wiki/Toulouse

the downloaded PDF doesn't show the graphics from page 3 onwards, page 1-2 are ok.

strange, since the downloa for Carcassonne works without problems

Heribert 178.191.119.190 (talk) 11:56, 13 June 2017 (UTC)

Thanks. Johan (WMF) (talk) 12:20, 13 June 2017 (UTC)

Thanks for your report. I downloaded the PDF for this page and images are showing up without problem on my end. Could you tell us a bit more about the browser and operating system you were using when you encountered this? OVasileva (WMF) (talk) 13:23, 15 June 2017 (UTC)

Proposed New PDF Format Choices #2

Proposed New PDF Format Choices #2

Previously, I wrote in to Wikipedia to add my opinion, for whatever it is worth, hoping that it would be helpful to everyone while serving my own self-interests for the maximum transfer of information from an article in Wikipedia into the presentation of information in the PDF format and for a desire for even more maps, photographs and illustrations in both Wikipedia’s articles and PDF’s, as well as for all of the other great things Wikipedia provides. Visual information has been repeatedly proven to be a key component to learning and understanding, because of the way it can help convey information and in a manner quite differently in the human brain than the printed word.

Speaking of the written word, I am writing again because I am embarrassed by my casual agreement with the spell-checking editing device in Microsoft Word that I used to edit my comments before sending them to the many good people at Wikipedia. The very same good people that have provided me with all of the information that was either new to me to learn or as a review of information that I already knew. As much as I would love to blame all of the errors on the monster that is Microsoft’s Office Suite, which Bill Gates created and is continually perpetuated by Microsoft on a regular and routine timeframe filled with tens of thousands of errors, the ultimate responsibility is mine. As such, I take the blame for sending the many fine people at Wikipedia a message with three (3) errors and I am resending my earlier comments to Wikipedia properly edited. I am sorry for any inconvenience on their part for having wasted any time trying to decode and properly understand my previous message send on 2017 06 08, written below:

I like the two (2) column page format of PDF for the downloaded presentation of formation that has been used by Wikipedia for years. I like it just things just the way there are right now without any changes. Except for the use of even more illustrations in the articles. If it is helpful to others, I would like to see both a one (1) and two (2) column per page format choice to go forward. Having a choice is nice. However, given only one (1) choice, I would prefer the two (2) column presentation much better. Whatever decision Wikipedia makes, I would like to make it clear that I would prefer a PDF that does not lose the ability to transfer charts, graphs, or tables in either format. There is far too much information that is presented in the many subjects available Wikipedia that would be lost if this kind of information were not transferred from Wikipedia's articles to a PDF. I can not stress this point strongly enough. Consider how many useful lists are included in Wikipedia articles or provided as stand-alone articles, how many things or groups of information that can not be put into words that charts and graphs explain, or how is easy it is to compare and contrast an endless number of things when it is portrayed in a visual manner. Therefore, please keep the visual information in and add as much as possible to both the Wikipedia articles themselves as well as the PDF version of Wikipedia's article. Thank you for your consideration on this matter.

To be rewritten, for submission, as the following:

I like the two (2) column page format of PDF for the downloaded presentation of information that has been used by Wikipedia for years. I like it just things just the way they are right now without any changes. Except for the use of even more illustrations in the articles. If it is helpful to others, I would like to see both a one (1) and two (2) column per page format choice to go forward. Having a choice is nice. However, given only one (1) choice, I would prefer the two (2) column presentation much better. Whatever decision Wikipedia makes, I would like to make it clear that I would prefer a PDF that does not lose the ability to transfer charts, graphs, or tables in either format. There is far too much information that is presented in the many subjects available in Wikipedia that would be lost if this kind of information were not transferred from Wikipedia's articles to a PDF. I can not stress this point strongly enough. Consider how many useful lists are included in Wikipedia articles or provided as stand-alone articles, how many things or groups of information that can not be put into words that charts and graphs explain, or how is easy it is to compare and contrast an endless number of things when it is portrayed in a visual manner. Therefore, please keep the visual information in and add as much as possible to both the Wikipedia articles themselves as well as the PDF version of Wikipedia's article. Thank you for your consideration on this matter.

As stated before, thank you for your website and your consideration on the matter of format on PDF’s. 2601:586:201:8B4:A509:2B6C:AC0E:36F4 (talk) 16:21, 13 June 2017 (UTC)

Thank you so much for your feedback. We are currently investigating the ability to provide an option for 2-columns which include tables and particular templates, however, at this point it seems we might not be able to provide this option for our current iteration. That said, we are experimenting with different styles that will preserve the more formal look and feel of the 2-column format. We will be updating this page with this options once the design is complete. OVasileva (WMF) (talk) 13:18, 15 June 2017 (UTC)

Thank you. Johan (WMF) (talk) 14:48, 14 June 2017 (UTC)

PDF creation errors?

Hello, I encountered several times the same error (?) when trying to create a PDF from some articles, resulting in>> {"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Brexit"} <<

I am not sure if the error is on your side or located i my laptop . . .

Usually going back to the PDF creation section and clicking 'single'/ download again, fixed this, but on 'Brexit' it took about five or six attempts.

As for the resulting PDFs: ALL are MISSING the BOOKMARKS, but page internal links function properly, as far as I have tested.

I hope this helps.

Anyway - thanks a lot for everything that DOES work - One never can say thanks enough for all your good work! THANK YOU!!! 81.154.4.82 (talk) 18:34, 13 June 2017 (UTC)

Thank you for your report. Johan (WMF) (talk) 20:46, 13 June 2017 (UTC)

Headlines on PDF

Some Headlines appears on the last Line of a Page, and the following Text was shown on the next Page, This ist vexing. - (Sorry my bad English) Rosehnal (talk) 03:52, 14 June 2017 (UTC)

Thanks for reporting this! Could you provide a link to an example article where this problem happens? Malyacko (talk) 06:55, 14 June 2017 (UTC)

Perfect result, showing that the topic is not a general one. 2A01:CB19:552:A400:D1B0:80E3:D2BF:3AA1 (talk) 12:45, 14 June 2017 (UTC)

much better

well this new upgrade is much better than the two column .

i used to copy the whole page and then past it to Microsoft office word in order to make a single column page . now it will save me much time , loved it . thank you wiki`s people . Ali A. Alahdal (talk) 09:13, 15 June 2017 (UTC)

Better (could be even better still

I should point out at the start that I am one of the 10% of the British population who does not have internet access at home and, now being retired, no longer have it at work.  So I download stuff from public library cybercafe to memory stick for viewing at leisure home. (In short I am not your young professional who has easy access to internet.  There are others like me and especially in less-developed parts of the world. Just saying in case wikipedia is into diversity of access.)

Rendering tables is a huge improvement.  Well done. In the old days I'd save wiki pages as .mht single pages and these had the look of the original page, and then wiki 'improved' the system losing the ability to save .mht with non-current versions of Explorer maintainng the page style.  Huge shame.  So I moved to saving as PDFs but this made a mess of tables, so I then moved to copying and pasting pages as MS Word and then (once sorting out images etc) saved as PDF.  However this new wiki PDF system is almost as good.  I'd like links to be shown so that when you hover over them or click on them they work.

Also what would be nice is at either the top or bottom to have a line saying something like 'this page downloaded [date] from [URL]' (As wikipedia pages are temporally dynamic.) 89.241.116.59 (talk) 10:50, 15 June 2017 (UTC)

Thank you. Johan (WMF) (talk) 11:12, 15 June 2017 (UTC)

Page size while rendering PDFs.

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.

---

Currently the new service seems to run faster and seems all around nicer, however it'd be nice if it allows to render PDF's in different page sizes. Europe uses A series papers, and US uses Letter / Legal. The conversion always seems a bit ugly on either side of the pond and if an A series printout option is provided it would be nice.

I've downloaded a PDF and it seems US sizing for papers and it'd be hard for me to print and read it with proper formatting. Hbayindir (talk) 18:30, 15 June 2017 (UTC)

Sorry, my bad. The PDF pages are in fact A4. Hbayindir (talk) 18:34, 15 June 2017 (UTC)

The discussion above is closed. Please do not modify it. No further edits should be made to this discussion.

...

J'ai toujours eu une préférences pour les pages à colonne unique et je n'ai pas encore constaté de défauts sur les données converties donc mon jugement aurait pu être hâtif.

Je penserai à examiner le résultat et vous réécrire...

P.s: Conservez le choix des nombres de colonnes, vous avez rendu certaines personnes accros aux colonnes double !

philet26 Philet26 (talk) 00:52, 16 June 2017 (UTC)

Possible to open PDFs directly in browser?

I've seen other websites handle like this, but if you didn't want to download a PDF, would it be possible to just open it in the browser? I believe this handles with the MIME type of the page where text/html pages can open in the browser, but application/pdf (or something else) forces a download (because the browser doesn't have an app designed to open PDFs, relatively speaking). Simsnet (talk) 07:14, 16 June 2017 (UTC)

Première conversion : parfait 91.180.1.193 (talk) 08:14, 16 June 2017 (UTC)

Missing many images within long list

While exporting list of Frank Lloyd Wright works, many of the images do not make it into the PDF

"https://en.wikipedia.org/wiki/List_of_Frank_Lloyd_Wright_works"

I would like to eventually put this PDF'd list with pictures on my cell phone, computer or tablet for viewing and showing others when planning possible road trips to these places, when we don't have internet access. It would be great if all of the pictures were included. 216.178.75.56 (talk) 13:10, 16 June 2017 (UTC)

Thank you. Johan (WMF) (talk) 13:24, 16 June 2017 (UTC)

wikibooks:Linear Algebra/Print version/Part 2

This Wikibook previously exported to PDF without errors, but now has display errors (page 7+ of exported PDF). Nicole Sharp (talk) 19:51, 17 June 2017 (UTC)

Also, if there was a way to not get mathematics to show in bold text, that would be great. Bold text has a specific meaning in mathematics and physics (for vector notation), and should not be used otherwise. Nicole Sharp (talk) 19:51, 17 June 2017 (UTC)

wikibooks:Linear Algebra/Print version/Part 1 will not export to PDF at all (error message displayed). Nicole Sharp (talk) 19:56, 17 June 2017 (UTC)

I was able to successfully save Part 1 as PDF without errors using Google Chrome print-to-PDF. Though technically having mathematics always displayed in bold text is still an error, and also makes it more difficult to read. Nicole Sharp (talk) 20:04, 17 June 2017 (UTC)

I pressed the left hand side as suggested, nothing happens

I pressed the left hand side of the two and nothing happened, reckon still glitches Widget506 (talk) 16:22, 18 June 2017 (UTC)

equations is broken into the text

equations is broken into the text 176.58.96.140 (talk) 12:55, 19 June 2017 (UTC)

Could you give an example of the article where you encountered this? Also, which option you chose (single column, or 2-column) to render? Thank you! OVasileva (WMF) (talk) 14:09, 20 June 2017 (UTC)

Thanks. Johan (WMF) (talk) 23:23, 19 June 2017 (UTC)

Download as PDF

The single column options works OK, but the new two column option hangs up at 15% and doesn't proceed, even with refreshing the page. 185.16.161.49 (talk) 14:26, 19 June 2017 (UTC)

The Way You Have It In The Left Hand Column To Download And Read 1 Page At A Time Is The Way It Should Have Been From The Start. Going Back Up Finding Where You Where Last Was A Pain In The Ass., Forget The Links, People Can Do That Later. One Page Not Split In Half Is The Way I Would Have Designed It From The Start, Like I Said The Way It Is In The Right-Side Should Be Done Away With Now Today. Simplicity Is What People Want, Give It To Them. 50.74.108.19 (talk) 17:46, 19 June 2017 (UTC)

Thanks. Johan (WMF) (talk) 23:23, 19 June 2017 (UTC)

I am prefer a single column form!!! 2-columns is just a nuisance. 72.196.221.41 (talk) 02:41, 20 June 2017 (UTC)

Leaves a blank page at the end, which is annoying for those who wish to save paper 75.76.46.134 (talk) 17:44, 20 June 2017 (UTC)

Pictures in the pdf files are low quality and pixelated Smellpolkadots (talk) 01:10, 21 June 2017 (UTC)

The layout should be modified, especially its margins. 147.46.102.54 (talk) 05:51, 21 June 2017 (UTC)

I prefer the single column format with infoboxes. I would normally have to go to a lot of trouble to include them. I found no substantial problem with this when downloaded. Spellczecher (talk) 16:09, 21 June 2017 (UTC)

147.46.102.54: Could you specify more exactly how you'd want the layout to be modified? Johan (WMF) (talk) 22:28, 25 June 2017 (UTC)

thank you very very much 2405:204:E50F:8327:E877:92D3:D637:3519 (talk) 09:20, 26 June 2017 (UTC)

Single column works very well and is my preference.

Thank you very much 60.242.146.35 (talk) 07:11, 27 June 2017 (UTC)

Main problem with double column is LONG EQUATIONS, usually they don't fit in the margin. 192.76.8.64 (talk) 23:01, 27 June 2017 (UTC)

The new version of "Download as PDF", in the present stauts, seems almost the same with the PDF document which can be made via "printable version" (maybe the difference would be the activated hyperlink). Even, through the "printable version", the margins of the document can be modified, which is absent in the new "Download as PDF". 147.46.211.86 (talk) 07:45, 28 June 2017 (UTC)

Internal links within the article dont work Windows 10 (IE v11)

Internal links within the article dont work after pdf download Windows 10 (IE v11) 96.239.88.89 (talk) 23:11, 19 June 2017 (UTC)

Thanks. Johan (WMF) (talk) 23:22, 19 June 2017 (UTC)

Margins on single column

The page marign columns for single column print is too narrow. Punching holes in the paper (for putting it in a binder) will come in conflict with the text. Both right and left margin needs to be wider to facilitate two-paged printing. 51.174.130.183 (talk) 09:13, 20 June 2017 (UTC)

I have to agree: both margins (left & right) are too narrow. 2003:D1:63DC:3F00:E48B:8297:5286:4BB (talk) 13:48, 20 June 2017 (UTC)

The program doesn't designate .PDF for downloads

PLEASE include .PDF in downloads; I'm getting 'files' that need to be 'opened by' (select a program) - which is silly It's easier to add the .PDF - which changes the 'open-with file' to a .PDF with my reader icon on it. 72.95.203.55 (talk) 15:03, 20 June 2017 (UTC)

062917 - Still not downloading with '.PDF' file designator; this produces an undifferentiated 'file' that needs '.PDF' manually attached. I am using Windows 7. Jetblanchard3 (talk) 19:20, 29 June 2017 (UTC)

Concur. Had to manually add the .pdf extension for it to be recognized.

Formatting is a big improvement over the previous creator. 64.174.11.41 (talk) 18:30, 30 June 2017 (UTC)

Double Column

Please do not omit implementation of double column PDFs!  If I just want a simple pdf print of a page then I can simply right click press print and save as pdf with chrome or firefox, why would I need media wiki to do that for me, the useful part is double-column pdf ! 125.253.101.199 (talk) 16:27, 20 June 2017 (UTC)

Thank god for google translate. I didn't say implement double column instead of single, I said implement both 125.253.101.199 (talk) 15:55, 22 June 2017 (UTC)

The double column doesn't have the graphics or the charts which make a large portion of the Wikipedia information vital for users. Rhapsodiefm (talk) 00:00, 25 June 2017 (UTC)

@Rhapsodiefm then implement it ! 125.253.101.199 (talk) 07:26, 25 June 2017 (UTC)

How difficult is to have tables and charts in the bottom of the page as single column and text as double column ? 125.253.101.199 (talk) 07:27, 25 June 2017 (UTC)

Main problem with double column is LONG EQUATIONS, usually they don't fit in the margin. Thanks! 192.76.8.64 (talk) 23:02, 27 June 2017 (UTC)

Paper Size and PDF Download

I poked around briefly in the frequently asked questions and did not find any references regarding page size (they might exist, I just didn't find any).

When I request a PDF download, the PDF download occurs as expected. The PDF is, however, in A4 paper size. It is nearly impossible to find A4 paper in the United States, where "U.S. Letter" (8.5" x 11") is the standard size. Scaling A4 PDF to Letter size PDF does print, etc, however, it also miniaturizes the text and figures.

I'm certain I'm not the only one who experiences this page size issue. Is there a plan to allow generating or downloading PDF in U.S. Letter (8.5" x 11") paper size? TimothyTFrank (talk) 17:49, 20 June 2017 (UTC)

Error with the font on exported PDF

Hello,

I tried to export some pages in Vietnamese. Although the grammar still works-well, but the font is mix of san-serif and serif. 2405:4800:2076:C439:4459:261A:9DE0:4FD2 (talk) 10:52, 21 June 2017 (UTC)

Thank you. Would you be able to give a link to the page you tried to create a PDF of, as well as tell us which of the two methods to create a PDF you used? Johan (WMF) (talk) 12:55, 21 June 2017 (UTC)

I also found errors with the font when creating a PDF Single Column. The errors were in the Footnotes section of https://en.wikipedia.org/wiki/History_of_the_Shakespeare_authorship_question Roy.clarke (talk) 10:56, 22 June 2017 (UTC)

External links

In the External Links section it would be useful to have the actual links shown as text. 78.69.210.201 (talk) 08:30, 22 June 2017 (UTC)

Thanks. Johan (WMF) (talk) 10:58, 22 June 2017 (UTC)

Or maybe as blue hyperlinks ? 125.253.101.199 (talk) 03:18, 24 June 2017 (UTC)

Fonts seemed less sharp than before.

the new PDF generator is part of chromium web browser which is not dedicated to professional PDF generation. the text in PDF files generated in the new way seem to be less accurate than the previous PDF renderer did, one can see on many occasions where font are bold, esp. maths formula and equations, making it both ugly and hard to distinguish ! Lelandyang (talk) 04:42, 23 June 2017 (UTC)

Is table printing working or not?

I read this page: Reading/Web/PDF Functionality but could not understand it as its technical knowledge. I am confused, is table printing working or not in Book Creator?

I was making a book with this page https://en.wikipedia.org/wiki/List_of_mathematical_symbols and it only printed the sections headings for all symbol tables.

Is it possible that if it is not working then display a message on https://en.wikipedia.org/wiki/Special:Book so that everybody knows about this. I guess that table is important thing. 219.91.148.202 (talk) 13:32, 24 June 2017 (UTC)

Yes. Worked well. Very nice, clean PDF copy. Thanks. 72.189.0.118 (talk) 14:02, 24 June 2017 (UTC)

table rendering works in 1 col, thx.

double col may be useful for long text and little illustration.

if i'd to finally choose between 1 or 2 col, i'd take the first, since i'm loading scientific stuff mostly (many tables and illustrations).

if both is possible with option to the user (like right now :-) ) -- that would be great! 193.159.32.93 (talk) 13:55, 25 June 2017 (UTC)

We're going to have to get rid of at least one of the current options (OCG), as the code is too difficult to maintain, unfortunately. Johan (WMF) (talk) 22:31, 25 June 2017 (UTC)

This page doesn´t work! 92.73.149.60 (talk) 23:01, 25 June 2017 (UTC)

Random requests

First of all, I'm immensely grateful of your huge work here.

I have some requests:

• is it possible to enable an option which you can use to manipulate the resulting PDF by deleting the references, or some paragraphs the user may consider useless?
• also, can you let the user decide the size of the resulting text? I'm asking this because sometimes I get a perfectly perfect big text ([2]), and more often the text gets really tiny ([3]).

Thanks for your help! Ruislip Gardens (talk) 14:11, 24 June 2017 (UTC)

Thank you for your suggestions. (: Johan (WMF) (talk) 22:28, 25 June 2017 (UTC)

Great Johan! Also, could you please add a feature you can use to remove the index? Especially with huge articles it may become quite troublesome

Thanks! Ruislip Gardens (talk) 14:02, 26 June 2017 (UTC)

also, I found a bug! [4] if you download that PDF and head to p. 9 you can see how one of the lateral images gets overlaid with the text.

Can you fix that! THANKS! Ruislip Gardens (talk) 14:56, 26 June 2017 (UTC)

Download to single column worked fine using Edge. Thank you. I believe single column is a far better format for electronic material since it requires less scrolling. Royal G. 70.188.169.9 (talk) 15:14, 26 June 2017 (UTC)

Now it's getting better

Now it's getting better 46.133.52.16 (talk) 19:07, 24 June 2017 (UTC)

PDF's not downloading

The HTML, XHTML, Binary Programming Language documents that were to be downloaded today came up as a piece of computer language, with no documents. Rhapsodiefm (talk) 00:03, 25 June 2017 (UTC)

PDF's downloading

The documents are loading so fast compared to the older programs, that's great! THANKS Rhapsodiefm (talk) 00:04, 25 June 2017 (UTC)

some time the download is not on pdf form its not working like that 213.55.73.205 (talk) 07:23, 25 June 2017 (UTC)

Unbulleted lists and templates in infoboxes.

I tried downloading this article: https://en.wikipedia.org/wiki/Halo_Wars_2 as a .pdf. I noticed that the infobox on the .pdf doesn't show the same way as on Wikipedia. Fields containing https://en.wikipedia.org/wiki/Template:Unbulleted_list and https://en.wikipedia.org/wiki/Template:Video_game_release are indented on the .pdf. The1337gamer (talk) 12:47, 25 June 2017 (UTC)

Formatting and Equation Print-out

The single column PDF for topic "Fourier Transform" fails to print out equations on pp. 8-10; the double column print out has problems with keeping graph and text within the column, extending from the left column over the right (pp. 4,9,11) or from the right column off the page (pp. 16,18). Skh141954 (talk) 13:17, 26 June 2017 (UTC)

Thank you for your feedback! Could you tell us what browser and OS you are using? I tried to replicate the issue, and the equations are rendering okay of me on Firefox 54.0, osx 10.11.6. OVasileva (WMF) (talk) 14:21, 29 June 2017 (UTC)

Split paragraphs

Can you implement a feature that avoids splitting paragraphs in this way? [5][6] thanks! Ruislip Gardens (talk) 06:44, 27 June 2017 (UTC)

I experienced a similar issue while exporting the shunting yard algorithm using the single-column format.

The paragraph heading, Graphical Illustration, appeared at the bottom of page one, while its contents continued on page two as seen in this image. Alex Essilfie (talk) 08:17, 27 June 2017 (UTC)

Doesn't work if question marks are found in the title

What happens is that if the title contains question marks (but I guess this happens with all punctuation marks) the PDF maker is not able to recognize it and so fails to download the PDF. Look at this article > [7], for instance, and try to download the resulting PDF >https://it.wikipedia.org/w/index.php?title=Speciale:ElectronPdf&page=Aha+oe+feii%3F&action=show-selection-screen&coll-download-url=%2Fw%2Findex.php%3Ftitle%3DSpeciale%3ALibro%26bookcmd%3Drender_article%26arttitle%3DAha%2Boe%2Bfeii%253F%26returnto%3DAha%2Boe%2Bfeii%253F%26oldid%3D88607897%26writer%3Drdf2latex it won't download it.

Thanks for your amazing work! Ruislip Gardens (talk) 20:08, 27 June 2017 (UTC)

Thank you for your report! I've filed this bug here: https://phabricator.wikimedia.org/T169223 OVasileva (WMF) (talk) 14:15, 29 June 2017 (UTC)

wonderful! thanks for the help :) Ruislip Gardens (talk) 14:56, 29 June 2017 (UTC)

Fotos, imágenes.

No se ven algunas fotos. El resto bien en general. 94.198.88.37 (talk) 10:17, 28 June 2017 (UTC)

I get this error message when downloading as PDF : {"type":"https://mediawiki.org/wiki/HyperSwitch/errors/unknown_error","method":"get","uri":"/en.wikipedia.org/v1/page/pdf/Inflation_in_India"} 2405:204:90AD:174F:1DF4:F1C:754B:C625 (talk) 07:16, 2 July 2017 (UTC)

Web/ PDF Functionality - Different Approach

I think you are making too complicated. In Linux they have this thing called containers. I don't know about Linux and I am no expert. But I learned that day made me think that Linux based systems can work with any platforms or OS's. They are currently working on our smartphones. So I don't think there would have a problem of incompatibility issues. All you need to make is make a virtual box available for each of your customers such as myself. I noticed I had several books saved on this account but they are gone. I guess that was part of the problem.

So what I would propose would be create a virtual box. I am customer xyz with id code and whatever identifying parameters to determine my bonafides so this box does not become virus infecting machine which it could theoretically. So after determining I am legitimate, I am given a temporary virtual box to snatch or snapshot copies of whatever information from Wikipedia I desire or variations of the same or to add hyperlinks to the same information in the Wikipedia page I am modifying for own use. Then I add this into the v-box and virtual information is then dump into format creator of my choice which could be pdf or word. After I make my choice and save the information and the format I choose, then I save it to my account under the book I desire. After that I discard the container and virtual box creating form goes dormant again till the next client.

I wished Google Chrome or Firefox would do that and I which it was a built feature with the Linux system I am using. It would also make saving browsing history so much easier than it is now. Anyway those are some of thoughts on the matter. I think it would work and work well and be a great feature to add to Wikipedia or any browser for that matter. Look forward to see what comes next. Rhodes Scholar Rhodes.Scholar77 (talk) 20:01, 28 June 2017 (UTC)

Two Columns are more professional

Single Column just looks sloppy and childish, designed for dumb little kids. Claverhouse (talk) 13:51, 29 June 2017 (UTC)

Your preference is noted, and thank you for your feedback – but please, let's avoid statements that can easily be read as implying that those who prefer one column are on the mental level where they pick designs for dumb little kids. Even if that's not what we meant. (: Johan (WMF) (talk) 14:24, 29 June 2017 (UTC)

Article title parsing bug?

Noether's Theorem got the title 's_Theorem. Article title parsing bug? 68.98.170.156 (talk) 21:01, 29 June 2017 (UTC)

Thanks for reporting this. Johan (WMF) (talk) 11:08, 30 June 2017 (UTC)

Could you give us the browser and OS you were using when you encountered this? I attempted to replicate and it rendered okay for me. Thanks! OVasileva (WMF) (talk) 13:55, 30 June 2017 (UTC)

Talk: Reading/Web/PDF functionality

Downloaded just fine, and with the PDF format.  Was able to open it with Adobe Acrobat CuriousINTJ (talk) 01:42, 30 June 2017 (UTC)