Wikisource:Scriptorium/Help

I can see the displayed ornaments at Page:Great essays of all nations by Francis Henry Pritchard (1929).pdf/7 to be slightly cut on both sides. What can be the reason? -- Jan Kameníček (talk) 14:14, 30 October 2025 (UTC)Reply

I do not see anything cut off. It could be your browser or its settings. I do see that the margin images are set to 450px width, which is wider than I would use. --EncycloPetey (talk) 15:41, 30 October 2025 (UTC)Reply

Here is how I can see it in the mainspace (and the same in page NS too):
--Jan Kameníček (talk) 16:05, 30 October 2025 (UTC)Reply

That issue is not happening for me in Firefox. Again, it may be your browser or personal settings, because I am not seeing any such issue. --EncycloPetey (talk) 17:44, 30 October 2025 (UTC)Reply

Probably... Just strange that I am experiencing it in both Chrome and Firefox, and I am not seeing the issue with other pictures. --Jan Kameníček (talk) 18:33, 30 October 2025 (UTC)Reply

• This is also happening to me (I’ve checked IE and Edge). It shows the sides of the image if I remove the width parameter. I couldn’t figure out a way to fix it. TE(æ)A,ea. (talk) 00:18, 31 October 2025 (UTC)Reply

Has someone created a phab task for this? (Intend to if not, asking to avoid creating a duplicate.) — Alien  3
3 3 14:57, 8 November 2025 (UTC)Reply

This was an issue here (switch to Layout 2 or Layout 4, and go about halfway down). Sidenotes within {{fine block}} are not properly offset, and instead cut in to the main text. TE(æ)A,ea. (talk) 03:19, 2 November 2025 (UTC)Reply

@TE(æ)A,ea. Not 100% sure if this solution will work, and it would be a bit tedious, but you could switch all the fine block/s and /e templates in the headers and footers to Line-height2/s and /e, and then apply the font size reduction using the fine template (not fine block) on every paragraph. This would then let you break the fine template usage mid-paragraph, whenever a sidenote appears there, so that none of the sidenotes are ever inside a template that reduces the font-size, which seems to be (part of) the issue in this case. Sorry that I don't know how to fix this through the sidenote template, as my proposed workaround is less than ideal. Regards, TeysaKarlov (talk) 19:47, 6 November 2025 (UTC)Reply

 This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. — rae^5e <talk> 21:49, 6 November 2025 (UTC)Reply

I'm proofreading pages for the monthly challenge and I've run into a problem where using {{multicol}} causes continuations of paragraphs on one column to be split into separate rows when they should be combined into one paragraph (or row). See here for the issue. On the leftmost column, "ĉiuj" and "multegaj" should be in the same paragraph, and on the rightmost column the same should apply for "artificial" and "one". Is there a way I can fix this or will it have to remain this way, or should I use something other than multicol? — rae^5e <talk> 09:30, 5 November 2025 (UTC)Reply

@Theki I expect your options are either a) to convert everything to a table, which may allow paragraphs to combine across pages, with the appropriate use of nopt in headers/footers (I can probably hunt down an example if you are considering tables), or b) you can keep using multicol, and add noinclude/include only to ensure the text looks correct in both page and main spaces. If you go with the multicolum approach, see Page:Blessedbegodcomp00call.pdf/97 and Page:Blessedbegodcomp00call.pdf/98 for an example (the text on /97 is placed in the footer, so does not appear in mainspace, and is also copied to /98 using include only so that it appears as continuous paragraphs). Regards, TeysaKarlov (talk) 20:16, 5 November 2025 (UTC)Reply

The latter suggestion worked, thank you! I assume it will cause the page indicators on the leftmost side of transcluded works (the ones that show page boundaries when you hover over them) will be made inaccurate due to this, but I assume that is inconsequential. — rae^5e <talk> 21:48, 6 November 2025 (UTC)Reply

@Theki Yes, the page indicators/boundaries may end up slightly out of place, and yes, this is generally inconsequential. Glad it is otherwise working. Regards, TeysaKarlov (talk) 20:12, 7 November 2025 (UTC)Reply

Hi all! I'm working on transcribing this right now, and have run across a couple of issues (particularly with footnotes).

Firstly, there have been a couple of times already (here—1 and here—2) that the book contains a correction from the errata that appears in the footnotes. Both of these happen to be in a multi-page footnote, which I think will just make it more difficult to solve, but please correct me if I'm wrong. The green box appears as it should, and the corrected version appears in the transcluded version, but due to the fact that they are both in the footnote to begin with, and the errata template is set up to appear as a footnote, I'm getting an error and the note about change is not appearing in either the page or the main namespace.

Second, here—3, there is an occurrence of a footnote within a footnote, which kind of ties into the first issue, but I am just looking for the best way to transcribe that so it appears as it was intended.

Finally, I have had a couple of times now where I have words that include the {{ls}} character when they happen to be broken across pages. These transclude properly (I'm using {{hws}} and {{hwe}}, but they don't appear correctly in the Page namespace (For example, as in here—4 and here—5. Any workarounds for that would also be appreciated.

Thanks so much in advance for your thoughts and comments! — Erblanchard57 (talk) 06:48, 6 November 2025 (UTC)Reply

Hello @Erblanchard57,

For the errata, I am not sure of the exact fix. I think you have to add group="errata" to the smallrefs call, which at least gets rid of one of the errors. But I am not sure if the errata is then supposed to display as a separate footnote, but isn't, or if it is then correct (I haven't used errata much). It also seems like the use of group="errata" has messed with the font size of the reference, which you could manually adjust. I have held off doing so for the moment, in case there are other issues with the errata. Consider this not entirely fixed at present.

For the footnotes in footnotes issue, see the example in the nested footnotes section of this page: Help:Footnotes and endnotes. Note that you have to use the refn template, and have to call it twice (empty the first time) so that the references appear in the correct order. When the next page is proofread, continue using refn with the follow parameter, so that the reference on the next page joints to the previous when transcluded. Note that you could also use refn for the errata issue above. I did try this, and then a separate footnote with the errata appeared. But it appeared out of order, and the dummy call to refn (i.e. with refn being empty on first use) didn't seem to fix the ordering issue.

For the {{ls}} character in hws/hwe, the workaround is to not use hws/hwe. Really, hws/hwe should only be used in special circumstances these days, as Wikisource has come far enough to be able to join words across pages if you just type the hyphen (with the exception of if the hyphenated word is in a footnote, then the workaround gets more complicated). P.S. If you ever want the hyphen to stay once transcluded, use either the {{peh}} or {{upe}} templates, but you don't need either in this case.

As an aside, for paragraph breaks in references, the template to use is {{pbr}}, rather than {{nop}}.

Regards, TeysaKarlov (talk) 20:11, 6 November 2025 (UTC)Reply

Hi TeysaKarlov! Thank you so much for the helpful pointers! Based on the information that you provided, I was able to get those other pages fixed with the long s, and I also ended up figuring out how to fix the errata problems using {{refn}}. I ended up just having to format the note on the first page as {{refn|name="example"|text of note}}, and on the second page {{refn|follows="example"|text of note}}. I was then able to add two different {{smallrefs}} tags in the footer of each Page namespace, and also in the Main namespace (with the first being {{smallrefs}} and the second being {{smallrefs|group="errata"}}.)

P.S. I realized that I should be using {{pbr}} instead of {{nop}} in footnotes actually just a few minutes before you had made that comment; I started doing that. Thanks!

 This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. Erblanchard57 (talk) 07:05, 7 November 2025 (UTC)Reply

A while ago (well, many years ago...) some content (translation of a Polish PD text) was deleted due to what appears to be misunderstanding about licensing; the author (translator) is still active in Wikimedia projects but gave up trying to get it restored as they find Wikisource too hard to naviate. From what I see this was mostly a result of some old choice of using GFDL rather than CC-BY-SA; the author has no problems relicensing this. See User_talk:Xover#Regarding_User_talk:Nihil_novi#Fables_and_Parables. I discussed this with an editor who was involved in trying to restore it a while ago, but he seems to be inactive these days. Maybe someone here could look into this and restore this? Piotrus (talk) 14:54, 7 November 2025 (UTC)Reply

The correct place to raise the issue of undeletion is Wikisource:Proposed deletions. --EncycloPetey (talk) 16:08, 7 November 2025 (UTC)Reply

@EncycloPetey Proposed deletions? But I am proposing undeletion, not deletion. Piotrus (talk) 01:21, 8 November 2025 (UTC)Reply

Yes. Undeletion requests are held there. Just create a section called "Undelete [name of pages in question]". — Alien  3
3 3 14:54, 8 November 2025 (UTC)Reply

This is partially a request as well; when looking through this document I noticed that all of the references are broken. Should this be fixed by manually re-working all of the references, or just transcluding all of the sub-chapters as individual chapters? TE(æ)A,ea. (talk) 02:41, 8 November 2025 (UTC)Reply

@TE(æ)A,ea. The endnotes don't use note, required as part of template ref, to enable the functionality of the references. Instead, the endnotes are just numbered lists. So transcluding all of the sub-chapters as individual chapters on its own wouldn't help. And I expect any fix (e.g. authority references) would thereby be a pain to implement. Regards, TeysaKarlov (talk) 19:32, 9 November 2025 (UTC)Reply

@TE(æ)A,ea.: a variant on {{pent}} could work, but all the footnotes would need to be reworked to fit it. ltbdl (talk) 05:55, 12 November 2025 (UTC)Reply

Confusingly some of the end notes do use note but not all of them. Regardless its still a pain to fix.

Also is there any reason why this work uses layout 2? I can't see any reason why this would need or benefit from it.

For whoever decides to fix the footnotes on this work I also wanted to point out that the footnotes on Introductory Material to the Final Report of the Select Committee to Investigate the January 6th Attack on the United States Capitol are broken too. ToxicPea (talk) 01:17, 13 November 2025 (UTC)Reply

Hello all!

I’m working on a book that has a section that goes across several pages (starting at page 269—here and ending at page 273—here) which prints two parallel columns of English text. The columns stay aligned at paragraph starts (though sometimes have multiple paragraphs for one paragraph in the other column). The closest thing I can think of to compare it to is a Polyglot Bible layout (that allows for blank space between paragraphs when matching text doesn't exist for a certain column). I'm looking for a way to reproduce that on Wikisource so the alignment matches the page layout.

What I’m looking for is a template (or an existing tool I haven’t discovered) that would:

• Handle two independently flowing columns. (But if something has to be built possibly with the possibility of more columns, so it could be used for Polyglot books, etc.)

• Allow each column’s text to continue properly across pages. (Multiple starts and stops on the same page.)

• Let users place paragraph breaks where needed so the columns align properly while reflecting the occasional blank space.

• Show catchwords at the bottoms of the columns (multiple per page) (Possibly using a form of the current {{continues}} template?)

• Work at “book scale”—allowing multiple pages of text that will need to transclude properly. (Possibly even the ability to transclude individual columns, again, if something had to be built for this.)

I don’t have the technical knowledge to build a new template myself, so I’m asking for guidance:

1. Is there an existing template or method that already supports this kind of independent two-column parallel layout that I can utilize so something doesn't have to be built for this?

2. If not, would someone be willing to design a template or module that can?

Thank you for any advice or direction! — Erblanchard57 (talk) 03:51, 8 November 2025 (UTC)Reply

• I tried something out here, but it’s very convoluted and isn’t really easy or scalable. I would normally use a table, but that runs into problems when the first column continues over. There is no way, that I know, to use section-based transclusion to break up tables like that—the software always messes up somewhere. It will probably have to be a complicated module—or one template (with /s and /e) per paragraph, although I’m not sure a template will work across page breaks. TE(æ)A,ea. (talk) 14:51, 8 November 2025 (UTC)Reply

  I tried the template you shared, and it did a really good job at making it look on the Page namespace what it looks like originally in the book. Thanks! The one thing though that it seems like is still struggling is getting the independent columns to "follow" each other. Is there a way to get that to happen, maybe if I were to put the sections that break pages in some sort of wrapper that allows me to have one follow the other, similar to <ref name=> and <ref follow=>? Erblanchard57 (talk) 20:51, 9 November 2025 (UTC)Reply

  • I tried to fix it; check your sandbox. There’s an issue with the cross-page parser that randomly adds new lines when they’re not desired, and it took a lot of work to try to get around it. This is the best I could get: it mostly matches, but there’s a gap after the paragraph at the top of each page. I might be able to remove it with some more messy workarounds, though, but no promises. Another potential solution, although it doesn’t look as nice, is to move the content at the top of the paragraphs to the end of the preceding page. TE(æ)A,ea. (talk) 23:22, 9 November 2025 (UTC)Reply

    Thank you so much for your attempts to help fix this! I ended up just taking your final suggestion to move the content from the succeeding page to the preceding one to make it look as natural as possible on the Main namespace. Hopefully we can figure something out better and do that in the future, but until then I guess that's probably the easiest way.

     This section is considered resolved, for the purposes of archiving. If you disagree, replace this template with your comment. Erblanchard57 (talk) 04:23, 10 November 2025 (UTC)Reply

  As it happens I've also been looking for something similar for one of my transcription projects. Will have a look to see whether it's worth replacing the table approach I have right now. Arcorann (talk) 05:52, 10 November 2025 (UTC)Reply

how do you proofread a 6-column toc, like in Page:Pentagon-Papers-Part IV. C. 2. a.djvu/10? ltbdl (talk) 05:31, 9 November 2025 (UTC)Reply

I recommend doing this as a multi-level list rather than trying to do as a table. Beeswaxcandle (talk) 06:09, 9 November 2025 (UTC)Reply

• I would just add manual indents (with {{gap}} or {{em}}) and then adjust the hi parameter to account for this. TE(æ)A,ea. (talk) 14:35, 9 November 2025 (UTC)Reply

I'd say the best bet would be using lists, and the type attribute of OLs because it can automate using I/II/III or A/B/C as list marker (maybe a template for this)? Hanging-indent should probably be set, but to avoid it being too cumbersome it should probably be done with index CSS. — Alien  3
3 3 16:12, 9 November 2025 (UTC)Reply

@Beeswaxcandle, @TE(æ)A,ea., @Alien333: thanks for the answers, but they're too technical for me. do you have a demonstration? ltbdl (talk) 07:02, 10 November 2025 (UTC)Reply

• ltbdl: I’ve created that page with my preferred approach, using {{dtpl}} and {{tpl}}. TE(æ)A,ea. (talk) 15:03, 10 November 2025 (UTC)Reply

Hello! I would like to start a transcription project for Le Viandier. Because of the age of the work (14th-15th century), I am fairly certain that the text content of the book is in the public domain. However, the hosting database (BNF) states that commercial use (I assume of the scan itself?) is prohibited, so I don't think the scan can be uploaded to Commons? What are my options here? It's somewhat challenging to transcribe without the WS transcription tools that require Commons as a host, and it couldn't be validated by other editors. Thank you! —Kittycataclysm (talk) 01:33, 15 November 2025 (UTC)Reply

That doesn't belong en.wikisource; it would have to go to fr.wikisource, which may have different rules than us. As for Commons, see C:Commons:When to use the PD-Art tag. Anything like that is fine to upload to Commons.--Prosfilaes (talk) 03:15, 15 November 2025 (UTC)Reply

Got it—thank you! —Kittycataclysm (talk) 03:29, 15 November 2025 (UTC)Reply

So I'm looking do this File:Geneva Convention on Road Traffic.pdf, however, the United Nations formatted the document with two columns, English on the left, French on the right.

Am I supposed to try and recreate the two column format, or just discard the French portion and only retain the English and eliminate the two column formatting?

There are also some places, such as the title pages (PDF Page 6, for example) where the two languages are presented in non-column format. Should these be handled differently than above?--The Navigators (talk) 20:42, 30 November 2025 (UTC)Reply

Fixed an issue with the linked file name.--The Navigators (talk) 22:03, 30 November 2025 (UTC)Reply

• Generally speaking, we do not transcribe text which is not in English. When the text is on multiple pages, {{iwpage}} is used on the non-English pages; I’m not sure that there’s a settled policy for mixed-language pages, but I personally just ignore the foreign-language text in parallel columns. TE(æ)A,ea. (talk) 04:41, 1 December 2025 (UTC)Reply

  Partly because there are a such a variety of different cases. E.g. are we talking about language-learning aides like an exercise in a textbook for English speakers? An English commentary on a foreign language text like the Bible? A book discussing translation of poetry? Keep in mind that it is straightforward to set up links between different language Wikisources, so full texts like this you can transcribe the French text on FR wikisource and just link to it if someone does want to compare the two versions. MarkLSteadman (talk) 05:02, 1 December 2025 (UTC)Reply

  It's a United Nations treaty on road transport. It was apparently published in English and French; and when it was printed they put both languages into the same document, instead of printing an English version and French version. The majority of the document, each page has two columns: The left column is English, and the exact same text in French on the right column. I basically want to make sure what I'm expected to do with the French text before I start trying to transcribe it to avoid adding extra work having to redo it.

  There is a fair chance that I could take the French Text and send it over to FR wikisource, since it's all there, and I likely know enough French from high school to work my way through the Wikisource interface. That might be the best solution.--The Navigators (talk) 05:55, 1 December 2025 (UTC)Reply

Hi,

In checking some use of ppoem, I am in need of a tool that checks the start and end tags of a series of ppoem calls in a given page range, and gives an indication where the start and end tags of successive pages do not agree. (This is not detected as a LintError until the Page:'s are transcluded.

The current workflow is to check every single page, This is error prone and time consuming.

A better workflow is needed.

Special:ExpandTemplates can't expand a <Pages>...</Pages> tag, which makes one possible workflow, more tedious.

It's also shame there isn't an Special:ExpandTransclusion portion which would allow display ofthe generated output of given PAGES tag, with options for just the wikitext, or the whole HTML expansion as with Special:ExpandTemplates

ShakespeareFan00 (talk) 12:54, 1 December 2025 (UTC)Reply