Template talk:Rp/Archive 1

Archive 1

Page numbers next to note-letter in refs/footnotes section?

Resolved

 – Query answered.

Is there any way to have the page numbers appear next to the letter of the note in the references/footnotes section rather than in the body beside the number? LaraLove 02:23, 21 September 2007 (UTC)

No; the display of stuff in the <references /> (or {{Reflist}}) section is entirely determined by the core MediaWiki software. There are alternative means, however. You may be looking for {{Ref harv}} (though I warn you that the {{Ref}}-based citation system is very geeky and a bit complicated.) — SMcCandlish [talk] [cont] ‹(-¿-)› 14:34, 21 September 2007 (UTC)

Can we remove the space generated after the citation?

Resolved

 – No objections.

Note these examples (there are two examples total) of citations. When the {{rp}} is used within a sentence or after a sentence, everything looks fine. However, if the citiation follows a fact within a parenthetical, there is additional space added by the template. Therefore, perhaps the space generated after the citation by the template can be removed, leaving the spacing to each individual editor? kilbad (talk) 02:02, 8 February 2009 (UTC)

• That seems reasonable to me. (And it may even have been my recent changes that added the space, if so I apologize.) --❨Ṩtruthious ℬandersnatch❩ 04:48, 9 February 2009 (UTC)

• No problem. I love this template when compared with the alternative never-ending list of page references, so keep up the good work. With regard to removing the space, I will defer that to someone else who edits templates frequently, as I am not sure how to do that. Thanks again! kilbad (talk) 19:43, 9 February 2009 (UTC)

• So will someone fix this spacing issue for us? kilbad (talk) 19:20, 13 February 2009 (UTC)

• Ok, I went ahead and removed the blank space. If you have any questions or concerns, please post them here. kilbad (talk) 22:47, 19 February 2009 (UTC)

Making superscripts too long

The superscript links to the footnotes are already quite bulky, especially in controversial articles where some statements are followed by four or more three digit numbers. Rather than adding to the size of the footnote links, it would be better, IMO, to find a way to add the page numbers into the footnote itself. It also makes more sense from a data layout point of view. The other information about the reference (author, date, etc.) is in the footnote itself, so why should page numbers be in the footnote number instead of the footnote itself? Gronky 07:07, 23 July 2007 (UTC)

Harvard referencing, which is used by journals all over the world, and is certainly a well-respect and -accepted referencing style in Wikipedia itself, puts the page numbers in the (sometimes superscripted, sometimes not) inline reference citations. I think you have picked a much, much larger fight that you think. :-) The scientific community, who use Harvard referencing massively, do not seem to have any "data layout" problems with it. The underlying issue is not this template at all, which simply enables such citation styles with the presently rather technologically deficient <ref> and <references /> MediaWiki citation code.

This template isn't very frequently used, so the problem you envision isn't likely to arise much if at all. And frankly, how problematic could it really be? Is there any serious difference between

• Blah blah blah,^{[7][8][9][14][17][37][38][42][43][124]} yack yack yack

and

• Blah blah blah,^{[7][8][9][14]:23-25 [17][37][38][42][43][124]} yack yack yack

? I don't see one.

If {{rp}} ever became so broadly used that it was seen all over the place and was genuinely annoying, this would simply spur The Developers to fix <ref> and <references /> to be more useful, and a bot would do the conversion within a few days of the new features going live. The {{rp}} template solves actual problems well-explained in its documentation. A few extra inline characters are far, far preferable to a 400-character reference citation no one can read, or 50 redundant reference citations doubling the length of the page for no good reason; there isn't any problem adding page numbers to references directly inside <ref>, on a limited basis; as the documentation explains, the code simply isn't presently in place to allow one to cite a single reference at multiple specific pages or page ranges, without making either one kind of mess or the other, without using this template. Again, as the template's own dox explain, page numbers normally should be in the body of the reference citation; {{rp}} is for a special case (and my quick review of "What links here" shows that it is not being abused and applied to other than that case). No enmity intended at all, but I find it hard to take your objection as anything but a "what if it were abused" point, which leads me to cite WP:BEANS - we don't need to worry about things that are not demonstrably problematic. — SMcCandlish [talk] [cont] ‹(-¿-)› 08:04, 23 July 2007 (UTC)

The problem between the two examples you give is that the first can be collapsed if multiple numbers link to the same book (when someone fixes the reference system to allow this), but the latter can't. As for Harvard does X, Harvard journals probably don't try to write articles about Hugo Chavez with 100 authors :-) I agree that this template isn't the problem, but it seems to be developing tools for heading down a wrong path, so I'd like to discuss this with the tool makers in these early stages. Gronky 09:57, 23 July 2007 (UTC)

I'm not sure what you mean by "the problem between" the two examples; do you believe the examples conflict with each other? Getting the developers to fix this problem would be great; I don't know any of them by name, myself, and Wikipedia:Village pump (technical) doesn't seem to be particularly effective at getting anything fixed (maybe I didn't try long/hard enough). As this template's dox say explicitly, it is intended that this template be replaced by better <ref></ref>, <ref /> and <references /> functionality. PS: I wasn't talking about what Harvard journals do, but what Harvard referencing does, and that Harvard referencing is one of several accepted citation styles on Wikipedia. I'm not even a huge fan of it, but its inline page numbers in references is the only way presently around the <ref> limitations problem, in the particular circumstance of an article citing the same source many times at different source pages/page ranges (as at Glossary of cue sports terms which cites Shamos 1999 again and again and again. See the talk page at that article for the "origin story" of this template, in fact. :-) — SMcCandlish [talk] [cont] ‹(-¿-)› 18:08, 23 July 2007 (UTC)

A comment from the viewpoint of reading articles: the recommendations for WP:rp at present suggest that it should only be used if there are a great many references to different pages in the same work. I find it useful even if there are only 2 references, as I can see in the text that the two references are from the same work. And there is one comprehensive entry in the list of references, no need to put "Smith (2004)", then have to find the comprehensive information elsewhere. I can see the problems with a great long list of references interrupting the text and won't argue my case there (though personally I don't think it a serious problem). I find that this template solves a problem that I had been thinking about for a while; until I found it I was going to suggest that multiple page numbers from one work were a problem (for which I didn't have a solution). This template may not be the best thing since sliced bread, but it's certainly the best thing since ref. Pol098 (talk) 00:09, 2 March 2009 (UTC)

Is this the best solution?

Resolved

 – No consensus against this template; wrong venue.

The better solution to the footnote problem that you correctly identify is to put short form citations in footnotes (at least when a single reference is cited multiple times with different pages), and putting full references in a bibliography. I see this being done in some articles that have a Notes or Footnotes section followed by a References or Bibliography section. This template has nothing to do with Harvard referencing, which uses inline, parenthetical short form references to a source plus page numbers, and leads the reader to a bibliography for full citation of the source. Harvard referencing is another, MoS-approved solution to the problem you identify, which does away with the footnotes.

While innovation and creativity are to be cherished, innovation that is contrary to accepted conventions sows confusion. I am not aware of any accepted citation system that uses a footnote call followed by page numbers. Are you? I have never seen this citation system used in any form of printed publication. Have you? How is the typical Wikipedia reader supposed to know what these numbers trailing the inline footnote calls are (at least p. or pp. would be a clue)? And, it is typographically ugly. And, it is typographically horrid between consecutive footnote calls. In the absence of support in a referencing system that Wikipedia endorses, use of this referencing system (with or without the template) is against the MoS, and should be. Please consider withdrawing this template and fixing the footnotes in the relatively small number pages in article space that use it. Thanks. Finell (Talk) 05:09, 29 November 2007 (UTC)

Yes, it is, de facto: This is the only solution for now. When I say "multiple times" I don't mean 5 times. I mean 105 times. Putting "short form" ibid. type citations does not solve the problem; you'll still end up with a refs or notes section screens and screens long. As the template's documentation clearly says, it is not for general, but rather exceptional, usage. It is principally for use in glossaries with one major reference used to source a hundred or more individiual entries (Glossary of cue sports terms will certainly be one of these when Fuhghettaboutit and I get done sourcing it down to every last entry; still a work in progress). It has also been sanely used at article that cite one source a dozen times or whatever but also cite many, many other sources, resulting in a strong desire to not have any redundant refs section entries.

This template has been in use for many months now with virtually no objections, and was quite heavily "advertised", in case anyone would object to it, ergo it is now one of the "accepted conventions" "that Wikipedia endorses", per WP:CONSENSUS. Some day (don't hold your breath) the developers will make the <ref> system more functional, and this can all simply be done away with. Until then, we do what we have to. How is the typical reader to know? The same way they know what it means in "(Johson 1994:14-17)". If you wish to use this template and are concerned that the readers would be confused, you can always just add "p." or "pp."; I wouldn't do that, but I can only speak for myself, and this template was intentionally designed so that the field was flexible (it has to be for other reasons, because some pages do not have numbers and have to be identified by other means, e.g. "frontispiece").

It is not "against" the MoS. I edit the MoS more than probably anything else here (and more than anyone else other than User:Tony1, probably, other than the last few weeks; been busy offline), so I know its intent pretty well. This is a workaround for a technical problem, in addition to, that is, supplementing, what the MoS and WP:CITE have to say about reference citations. I appreciate that it somehow just seems to bug you, but there isn't really anything to be done about that in the short term; the root problem is the MediaWiki software's limitations, which I really, really hope get worked on in this regard soon, but I'm not betting on it. It is better for the encyclopedia for one editor to be annoyed by a workaround, than for a great many readers to be annoyed by articles with reams and reams of redundant citations ("The good of the many..."). It solves the problem it needs to solve for now, and I have to think that you have more important concerns than this. :-) I agree it is not the most ideal thing in the world, but many things on WP aren't, right?

— SMcCandlish [talk] [cont] ‹(-¿-)› 08:34, 29 November 2007 (UTC)

PS: I also have to note that this cannot logically be argued both ways. Your anti-Rp position is (in part) "it's hardly used on any templates", ergo somehow not worthy; this is a Wikipedia-internal combination of the underlying problems with the WP:AADD#This number is big or not big enough, WP:AADD#I've never heard of it, and WP:AADD#Nobody's working on it deprecated arguments, (with a strong dose of evident WP:AADD#I don't like it as well, from what I can tell). Meanwhile the only other Rp detractor I can recall, Gronky up top, essentially makes the opposite argument, that it is "bad" because it might spread to articles all over the place. The template dox themselves thwart the latter, and with regard to the former argument, this template is intentionally not widely deployed; it is a special tool for special circumstances. — SMcCandlish [talk] [cont] ‹(-¿-)› 08:39, 29 November 2007 (UTC)

Dear S: I know you are very intelligent, so please do not misrepresent what I say or raise inapplicable objections, while failing to address the specific criticisms that I did make.

I did not say that your system is bad because it is only used on a handful of pages. I gave several reasons why your system is bad, and said that fortunately it would be easy to clean up the mess because it is only on a handful of pages. I did not say that your system is bad because WP:AADD#I've never heard of it. I said it is bad because it is not used anywhere in the real world, so no one has heard of it or will know what to make of it. I invited you to show me where it is used elsewhere; you didn't. Many people already know Harvard referencing (for the reasons you so eloquently state), and it is documented on Wikipedia (and elsewhere), and it is fairly intuitive. No one will have a clue what your numbers following the footnote calls mean because no one has ever seen anything like them. It is true, obviously, that I don't like it, but I gave specific reasons. Your position is, essentially, "I [you] like it, that is why I invented it," so pity the reader who cannot decipher it.

The only solution you say? Books and journals of all types do just fine without resorting to anything like what you propose. How is your system preferable to the example below my signature, which is a standard format, which the current footnote system can do, and which most everyone would understand? If there is more than one reference by one author, there are many ways to distinguish them (such as by year or abbreviated title). Scat singing uses Harvard references in footnotes followed by a bibliography; that is a perfect solution and requires no invention. By the way, ibid. should never be used on Wikipedia because any editor can insert another footnote between the first citation and the ibid., so the latter will point to the wrong ref. Op. cit. works fine, however.

Have you discussed your idea at Wikipedia talk:Citing sources, where citation methods are discussed, and where some very knowledgeable people hang out? Care to try it out there, and see what reactions you get?

By the way, one big problem with the MoS, which should be relatively stable, is that hobbyists are continually hacking away at it. Two or three people agree on some little change and declare themselves a consensus when no one else is paying attention.

Won't you please reconsider your position on this? Thanks. Finell (Talk) 10:42, 29 November 2007 (UTC)

Example of another solution:

Footnotes

¹Ball p. 47.

²Devlin pp. 108–12.

³Devlin p. 217 n. 8.

References

Devlin, Keith (1994). Mathematics: The Science of Patterns. New York: Scientific American Library. ISBN 0-7167-5047-3.

Ball, W.W. Rouse (1960). A Short Account of the History of Mathematics (4th ed. [Reprint. Original publication: London: Macmillan & Co., 1908] ed.). New York: Dover Publications. ISBN 0-486-20630-0.

Whether someting is used in the real work isn't very germane. See reams of decisions at WT:MOS in that regard; WP:NOT#PAPER and we are not bound by what The Chicago Manual of Style says or any other style guide, though of course MOS editors pay attention to such works and we attempt not to deviate from them unless it is necessary to do so in Wikipedia's unique context. That is the case here. What I meant by WP:AADD#I've never heard of it is that the underlying meaning of that AADD passage, when extrapolated out of the AfD context, can be distilled as "delete because the material in question is unfamiliar or obscure". This usage certainly is obscure, as its own documentation indicates. It is an adaptation of Harvard referencing's ":123–28" inline page-number annotation to the funky referencing system that WP has. By your reasoning WP's entire <ref> system should simply be scrapped, because it is not like any real-world referencing system to begin with. Objecting to this template on the grounds that it is neither Harvard nor MLA nor X published style is a moot point; nothing about Wikipedia's <ref> system is "standard"; it's just something new that Wikipedia is doing (for whatever reason; I was not party to the discussions that led up to its creation.)

Re: "but I gave specific reasons...": And a I responded to them as best I was able at the time. The one you re-raise, that no one will understand that they are page numbers has seemed to prove untrue, since to my knowledge not a single person has asked, here or any talk page of an article using this template, what they mean (I haven't looked in a while though; I did monitor that for a while because of the newness of the template). Anyone who has seen HR will instantly recognize it, and anyone who has not seen HR won't understand it in HR either, so that too seem like a moot point to me. WP does have some rather difficult-to-use HR templates like {{ref harv}} and its footer counterpart, but that template will not address the problem; you'd simply end up with 5 pages of redundant references in that format instead of the other.

Re: "Books and journals of all types do just fine without resorting to anything like what you propose." I honestly think you need to re-read the documentation from scratch and look at the template's source code and look at how it is used on a page like WP:CUEGLOSS, and make a copy of that in your sand box and convert all of the Rp instances in it to some other notes system, either <ref> as usually used or a {{Ref}} variant, and look at the results. You do not appear to me to understand Rp's purpose, or you wouldn't have said something like that. "Books and journals of all types" do not use Wikipedia's half-baked [ref> system. The problem lies in the <ref> system, and it has to be fixed by the developers. Until they do, this is the only (extant, anyway) solution for the problem that this template was created to solve. Using pre-<ref> WP citations systems does not solve the problem. Having a forest of redundant footnotes in a Notes section does not solve the problem. None of that is meant in any way as any sort of slight; I mean it literally, that you seem to be misintuiting why this template exists, and not fully understanding that it is only used for working around an unsual problem that affects very few articles.

WP:CITE was one of the first places this was posted to (and it was in the "See also" section there for a while, but eventually all templates got removed from there in favor of one line to Wikipedia:Citation templates apparenty), along with Template talk:Ref harv (which I think is just a redir to Template talk:Ref, and the Village pump, and... It was quite broadly announced.

Agree with you about MoS. Tony and I and few others try to keep the churn to a minimum, as far as substantive changes go. It's difficult.

Hope this round was a little clearer!

PS: If you see this show up on 1000 pages then we really do have a problem!

— SMcCandlish [talk] [cont] ‹(-¿-)› 12:38, 29 November 2007 (UTC)

Dear S: The issue that I am raising is not fundamentally about the template itself, but the non-standard, invented new citation system that it supports. Please point me to the discussion on WT:CITE and the other discussions; I would like to see what the response was to your idea and if any consensus was reached. Exactly what is the problem in the <ref> system to which you refer? Have you raised it with the developers (if so, please point me to the discussion)? What is wrong with the systems used in William Shakespeare and Scat singing (please do look at them) and other articles, which follow conventions and are easily understood? In my opinion, Wikipedia really is not the place for experimentation in something so fundamental as WP:Citing sources. The prescribed solution for the problem you raise is here: WP:Citing_sources#Short_footnote_citations_with_full_references. Finell (Talk) 06:52, 30 November 2007 (UTC)

I understand what you are saying, but it is not a "non-standard, invented new citation system", but an adaptation of real-world citation systematics to WP's own actually non-standard, invented new citation system, which is quite inadequate and causing real problems. Basically, I have moved it one tiny step closer to Harvard referencing, and perhaps the MediaWiki software developers will take the hint. We can but hope. I do not have time to spend an hour digging around in edit histories to find stuff for you. The response, from all "advertising" combined, including at CITE and at the {{Ref harv}} talk page and so on, was dead silence, and in WP we generally take silence as assent, per WP:BOLD and WP:BRD. I did not expect a roar of "holy crap, you are a genius!" for resolving a simple and rare issue, and (predictably) did not get one. I was certainly open to the possibility of negative outcry, but there was none, and the template started being actually used (by others besides myself) immediately (on, of course, a very limited basis, because this template's applicability is, by the nature of the problem it solves, very narrowly applicable to begin with). No "experimentation" is taking place; this is a normalizing and problem-solving exercise, not a Gedankenexperiment much less a WP:POINT. I'm aware of the solution proposed at WP:Citing sources#Short footnote citations with full references but it is an even clumsier kluge, grossly time-consuming, not particularly useful in all situations even if one has oodles of time to waste, and still produces a redundant mess. This template was created for a reason, with very limited applicablity, with very limited deployment (i.e., only where actually needed) even months after the fact, ergo no evidence of abuse. Therefore, this is (pretty much by definition in the WP context) a non-issue. I think we both have better things to invest our time and energy in. — SMcCandlish [talk] [cont] ‹(-¿-)› 07:34, 3 December 2007 (UTC)

(outdent:) This is indeed a "non-standard, invented new citation system," thoroughly unintuitive, and extraordinarily ugly to boot. I completely agree that there are other, better ways of addressed the supposed problem. --jbmurray (talk • contribs) 18:13, 1 September 2008 (UTC)

I have to suggest that if there were a consensus for your viewpoint, the template would not be widely used, yet it is. If you continue to feel strongly on the matter, take it up at WP:TFD and it will be settled permanently one way or the other. — SMcCandlish [talk] [cont] ‹(-¿-)› 20:25, 8 April 2009 (UTC)

Why not just use short footnotes?

I do not think that this is a better solution than using shortened footnotes, could someone explain why this template is superior given that it lengthens the text in superscripts, and it breaks the link between reference tag and the page number which has potential maintenance issues. --PBS (talk) 12:09, 2 April 2009 (UTC)

It isn't necessarily superior, any more than Harvard style is "superior" to AMA. It's simply different. If I needed to cite something 5 times at 5 different pages, I would probably use the CITESHORT style, but if I had to cite it 87 times, I would use Rp style, because CITESHORT style would still produce a ridiculously redundant "References" section. — SMcCandlish [talk] [cont] ‹(-¿-)› 20:05, 8 April 2009 (UTC)

Imagine what the references section would looks like if this was cited using short citations instead of Rp.--Fuhghettaboutit (talk) 22:01, 8 April 2009 (UTC)

Missing page numbers

Resolved

 – Need met, and documented.

Hitherto missing page numbers in references have been flagged by following the reference with the Page number template and the month. What is the recommendation with the Rp template: to use a reference without page number and add the Page number template as is usually done (i.e., not use the Rp template at all), or what? I have been simply entering the page number as "?" in normal or bold text in the Rp template and not bothering with a date, which seems reasonable to me. It's easy enough to work out my own solution, but it might be useful for a standard procedure to be recommended in the template documentation and avoid everyone working out their favourite technique. If a question mark is too short, maybe or or even the full Monty, , though that does seem so long as to interrupt the text. Pol098 (talk) 00:30, 2 March 2009 (UTC)

Since {{Page number}} adds the article to a category it seems best to me to simply use that. So I'd say go ahead and add a note endorsing such to this template's documentation if you want to. --❨Ṩtruthious ℬandersnatch❩ 01:39, 2 March 2009 (UTC)

Another possibility (which I won't add to the documentation, unless anyone here recommends it) is to nest templates <ref name=xyz/>{{rp|{{Page number}}}} which generates, may be useful? It might prompt people to enter the page number in the existing Rp template, where they might not know to add it if not already there. Pol098 (talk) 00:25, 4 March 2009 (UTC)

In spite of what I said in the previous paragraph, as there's been no comment against using nested templates (nor for), I'll modify the documentation to suggest this technique. Anybody seriously against can revert. Advantage of nesting the Page number template with Rp: the article will be added to the PageNumber category. Disadvantage: rather long second-level superscript note. Pol098 (talk) 13:34, 7 March 2009 (UTC)

I don't know enough about templates to say whether this is feasible, or whether the consensus will think it a good idea, but "page number needed" is very long to insert as a superscript. I take the point that articles with references needing page numbers should be added to a category; is it feasible to modify the Rp template so that a standardised form (I suggest that "p?" should be used in place of a page number) could cause the article to be added to the appropriate category? In other words a reference such as this would be recognised by the template code and add the article to the page number missing category.Pol098 (talk) 13:56, 7 March 2009 (UTC)

I like your idea, but would change it up just a little. I would suggest that when there is no page number included after the pipe separator in the template (i.e. {{rp|}}) then the category should be added and "page number needed" displayed. kilbad (talk) 14:24, 7 March 2009 (UTC)

I see what you mean, but I'm suggesting that the short and reasonably clear should be displayed for readers instead of the long and intrusive or any other superscript with the long form ^{"page number needed"}. But let others comment and a consensus be reached. Assuming someone is willing to modify the template according to this eventual consensus! Keeping the displayed message short is particularly important for references in narrow infoboxes. Pol098 (talk) 22:37, 7 March 2009 (UTC)

I understand now. Well, I would still favor {{rp|}} to generate , and would also want the "p?" to be its own footnote stating that a page number is needed. kilbad (talk) 00:28, 8 March 2009 (UTC)

A good idea, but one footnote may apply to a number of references not all of which lack page numbers, so the footnote should say ^{"some references lack page numbers"}? Or maybe more ideas are needed? Pol098 (talk) 11:49, 8 March 2009 (UTC)

Nesting an inline template inside of Rp like that is a no-go. Because the small-font CSS is doubled by doing this, tt makes the nested template too small to read on many if not most systems at default font sizes. Even at 1024x768 monitor resolution! I'm removing that recommendation from the /doc file. — SMcCandlish [talk] [cont] ‹(-¿-)› 21:39, 8 April 2009 (UTC)

{{Editprotected}} However, I have figured out, tested and documented a simple solution, and implemented it at Template:Rp/sandbox. Test cases are posted to the bottom of Template:Rp/testcases. Summary: Use {{rp|needed=y|{{subst:DATE}}}}, or if you are really lazy, just {{rp|needed=y}}. The docs have also been updated at Template:Rp/sandbox/doc, which should replace the content in Template:Rp/doc after the code change, and then speedily delete /sandbox/doc.

The benefits of doing it this way:

1. We are not trying to dictate the content of the "page needed" message; that is a matter for consensus at Template talk:Page needed.
2. As a result; the text here would not have to be kept in synch with another template.
3. It keeps the results exactly as legible as they would be without nesting.
4. It preserves the use of {{Rp}} in articles using it, which should help minimize inconsistent citations styles within the article.
5. It does not insert a pointless colon.
6. It uses {{Page needed}} correctly, including passing a date parameter so that the articles get properly categorized by month and year.

— SMcCandlish [talk] [cont] ‹(-¿-)› 21:39, 8 April 2009 (UTC)

• Looks good. Thanks for your work on this. kilbad (talk) 21:41, 8 April 2009 (UTC)

Y Done; changes implemented to both template and documentation subpage and doc sandbox deleted.--Fuhghettaboutit (talk) 22:10, 8 April 2009 (UTC)

Donkey shins! — SMcCandlish [talk] [cont] ‹(-¿-)› 00:12, 9 April 2009 (UTC)

Double lang-links

Resolved

 – Removed lang-link from Template:Rp.

Just info. This template seems same lang-links twice, one is in Template:Rp, and another one is in Template:Rp/doc. First one should be removed, but Template:Rp is protected.　--NISYAN (talk) 09:20, 26 July 2009 (UTC)

Thanks. --NISYAN (talk) 14:30, 26 July 2009 (UTC)

GA and FAC

Is using this template acceptable for GA and FAC articles?--Work permit (talk) 05:58, 29 August 2009 (UTC)

Absent any brand new guideline prohibiting it, of course it is. :-) "That which is not forbidden is permissible." — SMcCandlish [talk] [cont] ‹(-¿-)› 00:28, 27 November 2009 (UTC)

How about informing the reader about the "strange numbers" in the superscript?

I just stumbled (for the first time) upon a page where the Rp template is used. I wondered what these numbers mean and it took me a while to find out. Even though they stand just next to the superscript linking to the source it's not at all obvious what they mean. At first I thought it was just a mistake, my second guess was that it could be a count of the number of times a source is used, but a look at the abcde... in the references section could not confirm that. So I had a look at the source and saw this stuff. Next, I had to look for a template on wikipedia.org which includes this Rp thing and I finally found the meaning - quite an odyssey if you ask me. Are there any suggestions yet about how the readers could be informed about these numbers in a less cumbersome way? --Marsupilami04 (talk) 22:21, 9 May 2010 (UTC)

If you convert these to AMA style (Template:Rp#With_parentheses, it will start with p or pp to indicate that it means pages. So that's a solution. II | (t - c) 22:34, 9 May 2010 (UTC)

Thanks, I didn't look at it so closely. But I'm still wondering why the possibility with colons exists in the first place? In other words, what's the case against using "page" all the time so that "everyone" would understand it (because there are certainly a lot who don't even understand p/pp)? --Marsupilami04 (talk) 19:32, 11 May 2010 (UTC)

In response to Marsupilami04 above: I'm the one who added the AMA style and I agree with you, I think this template should simply default to AMA as it's an established standard externally. Another possibility is to add the HTML for a tooltip that says "page X" when you hover over the page numbers, I can try to figure out how to do that if it sounds like a good idea to anyone. --❨Ṩtruthious ℬandersnatch❩ 14:51, 22 June 2010 (UTC)

I just ran into this ref template in William Hanna and I was completely baffled by it. I'm strongly opposed to this type of referencing, which I find needlessly complicated and contrived. It's bad enough that Wikipedia has invented several new and novel formats (like "stacked" notes, repetitions of notes and even multiple sets of notes) which are used entirely on the whims of individual primary contributors, but they are at least tolerable since they're not confusing. This, however, is utterly cryptic even to those who are used to academic referencing. Judging by some of the reations above it's perfectly obvious that it requires that even readers who are used to notes to look up what it actually means.

This strikes me as a blatantly contrived and overly complicated solution to a non-problem; if you have a huge list of unique page references to the same work, you should deal with it with reasonably recognizable normal methods rather than invent a system that is oblique and completely unheard of in real life. I see no particular problem with long lists of notes if the number of unique notes is obviously large. This is an obvious case of making an quite uncomplicated situation immeasurably more complicated entirely on account of editor preferences. This template should in my opinion be aboided at all cost.

Peter ^Isotalo 11:14, 22 June 2010 (UTC)

It would be really great if the existing {{citation}} template and {{reflist}} implemented your "reasonably recognizable normal methods" for doing this; to my knowledge, it does not, nor does it even accommodate parameters to associate each note with particular page numbers. Once it does it'll be pretty easy to get a bot to churn through and convert {{rp}} to {{citation}}. So the sooner someone gets to work on that, the sooner the instances of this template can be merged with everything else.

I think you're being kind of hyperbolic; it really isn't that hard to figure out what it means, especially once you see ones that say something like "[4]:69-72". And even if you didn't figure that out... there's still a link in the superscript for each source and you click on that and it tells you where the information comes from and you have all of the exact same information you'd have otherwise. It seems much more to me that you're the one getting all in a froth about a non-problem. As far as it being used in "real life", as noted right above you're comment it's pretty much the same thing as the AMA style.

I think that the thing you and others are possibly up in arms about is that you think it doesn't look pretty or something; well, whatever. As long as the information (including the page number for each instance) is stored in a structured way in the wiki code we can make it look pretty at any point in the future. This isn't paper. --❨Ṩtruthious ℬandersnatch❩ 14:51, 22 June 2010 (UTC)

Other FAs use this RP format and it's a valid format for wiki. That Peter's personal preference is not this one doesn't invalidate the RP format nor William Hanna being on the main page. This is nothing but a case of IDONTLIKEIT by Peter. — Rlevse • Talk • 20:27, 22 June 2010 (UTC)

Well, don't get me wrong, it could probably be made to look more pretty or more user-friendly, I don't think that as a citation style it's completely beyond criticism or anything like that. It's just that people assert that it's easily interchangeable with some other approach to storing the same information but I don't think it's that straightforward. --❨Ṩtruthious ℬandersnatch❩ 21:32, 22 June 2010 (UTC)

I doubt that there is a workable one-size-fits-all bot-able conversion strategy. AFAICS, converting all {{Rp}} instances into individual {{Citation}} instances would create a big mess. In many cases, WP:CITESHORT would probably be closer but still not 100%. (note: {{Citation}} and CITESHORT do accommodate page numbered references)

Re the tooltip, replacing

|:{{{1}}}

with

|<span title="page(s) {{{1}}}">:{{{1}}}</span>

seems to work, at least with Firefox. Wtmitchell (talk) (earlier Boracay Bill) 21:54, 22 June 2010 (UTC)

The bottom line is that this is a valid ref format and until that changes and a bot can change it to another format without making a mess objecting to it being used at all and/or using its usage to stop an article from being on the main page is baseless. — Rlevse • Talk • 10:07, 23 June 2010 (UTC)

A suggestion

Having only stumbled upon this feature a few days ago and then having used it extensively in my just completed rework of the 4-6-2 article, a comment and a suggestion:
Comment: I like it. No matter that it's not been used before in "real life" (read: printed media), it's an excellent tool in electronic media.
Suggestion: It does tend to clutter up the text, though. So, why not hide it and only make the page numbers (and the complete reference, for that matter) show up in a box that only appears when the pointer is hovered over the reference? Same as the boxes that appear when hovering over any links in the text.
André Kritzinger 01:26, 21 October 2012 (UTC)

Hoovering text is not consistent between browsers and hoovering on a tablet maybe difficult. Regards, Sun Creator^(talk) 01:37, 21 October 2012 (UTC)

Doing it by adding a title= attribute to the <sup> tag is working for me in IE, Firefox, and Chrome:

Another asserted fact.

Tablets and other browsers that aren't able to do hovering text presumably have their own way to display the "title" attribute text of HTML tags. This seems like a good idea but might be better to add to the standard citation system or through another template meant to be wrapped around the ref tag, rather than modifying or adding options to {{Rp}}. --▸∮truthious ᛔandersnatch◂ 19:23, 21 October 2012 (UTC)

Sanity check

While reading Two-streams hypothesis I found "<ref name=Schenk2009/>{{rp|62}}" at the foot of the article. But the cited reference does not include a page 62. I am inclined to think this 62 is misinformation. Am I missing something about the rp template? If this 62 is spurious, I will flag it. Thank you. --Ancheta Wis   (talk | contribs) 08:33, 24 December 2013 (UTC)

When Acrobat Reader page number mismatches displayed page number

PDF URLs permit the addition of a "#page=N" suffix, to direct Acrobat Reader to scroll to the given page.

For instance, the following URL nicely gets you to "Page 89" of a 230-page transcript:

https://dspace.njstatelib.org/xmlui/bitstream/handle/10929/30031/b8512013a.pdf#page=89

Or at least, that's what Acrobat Reader considers as Page 89. However, the page number visible at the bottom of the page is "86".

This occurred because Acrobat Reader is unaware that the cover page, a preface page, and the table of contents page appear before the page labeled "1".

On the Fort Lee lane closure scandal page, there are multiple references into the one transcript, but with distinct page numbers, so the "#page=N' suffix is not appropriate in the (shared) URL, so we need to use RP notation at each cite.

The question is how best to note the page in RP notation at the spot for Acrobat Reader's page 89 (visible as page 86). Among the possibilities:

just 86

just 89

mixed to indicate mismatch, eg

What is the best approach? Thanks. JackGavin (talk) 18:53, 29 January 2014 (UTC)

Actually, this issue is also relevant for any time a citation is used where a page number needs to be cited within the citation template (i.e. | page = 89 |) for an online source, which also has physical page numbers shown on the sourced document, in addition to the electronic page numbers for the cited electronic file document. Most of the time, I've seen the electronic page number cited rather than the physical page number. This would seem to make the most sense since a lot of times the cited electronic document has several pages for a table of contents, a title page, and/or an introductory acknowledgment that uses Roman numerals prior to starting the regular page numbering of 1,2, etc. It would be easier to insert the electronic page number to go directly to the sourced page, rather than manually trying to figure out what electronic page to go to that would correspond to the physically shown page number in the document itself. Obviously, in many cases the electronic page numbers match up with the physically shown numbers when the document page numbering (1, 2, etc.) starts with the first page of the document so that there would be no questions about which page to show.

If anyone else, with sourcing expertise, can verify if this approach is the preferred correct way, or if there are acceptable alternate methods, that would be helpful. Wondering55 (talk) 20:28, 29 January 2014 (UTC)

Actually, my experience with upper-level articles is the reverse: the page number cited is the one printed on the page. For scanned copies of sources that lack page numbers, no page number is used. Imzadi 1979 → 20:43, 29 January 2014 (UTC)

Yep. Pagination of PDFs is a user-agent matter and varies by device. In non-geekspeak: There are lots of PDF viewers, not just Adobe Acrobat Viewer, and how they paginate by number (if at all) varies. I agree that this should be noted elsewhere, probably at WP:CITE, since it's important and not about this template.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  04:19, 28 April 2014 (UTC)

Seriously

Why are we still using this individualistic curisoity of a reference standard? It has to be the most deliberately annoying citation format ever invented. Never mind the relative obscurity of its use in print and likeliness to astonish readers familiar with reference notes. Is it some sort of token indulgence just to make a point about editorial individualism?

Peter ^Isotalo 15:42, 9 March 2014 (UTC)

@Peter Isotalo: Have you stopped beating your wife? Ignoring the characterizations in the question in answering it, because it serves a purpose that no other referencing system has solved—when you need to use one reference a huge number of times and provide verification to the page numbers. Using shortened footnotes in such situations would result in a ridiculous display. For example (and the article that was the actual impetus for this template's creation), see Glossary of cue sports terms#cite note-TIEOB-1, where there is 70 references to one book, and Glossary of cue sports terms#cite note-BCA-5 where there is 113 references to another. If you see this being used somewhere where the problem it is intended to address does not exist, but as an alternative to shortened footnotes, then you're just seeing a misuse.--Fuhghettaboutit (talk) 16:51, 9 March 2014 (UTC)

The use of this format can hardly be more complicated or "ridiculous" than simply having plenty of notes. For the digital version of an article, it makes absolutely no sense since you can either get hovering ref or links to notes, which makes the superscript clutter pointless. And for a printed or pdf version, it's an outright nuisance.

If you offer this format as a possibility with the vague caveat that it should be used only when referencing "many" times to a single source, users are guaranteed to interpret "many" as anything from five to a five hundred. Having lots of notes really isn't a problem to begin with. It really just boils down to the completely arbitrary idea "this looks annoying when I reference a lot". It's the equivalent of lopping of text from an article with the motivation "there were just too many letters for me to handle".

Peter ^Isotalo 23:39, 9 March 2014 (UTC)

You're assuming all use will be by sighted users, with complex visual browsers that support Javascript and have pointing devices that can "hover" a cursor over something. None of these are safe assumptions. It's not comparable in any meaningful way to lopping text from an article, since that would be deletion of content. {{Rp}} doesn't remove any content from anywhere. In fact it actually saves a lot of sourcing content, in the form of specific page numbers needed for verifiability, that many people would not bother putting in otherwise because it would be too cluttering. See the first discussion on this page. It's the intent that Rp will be replaced by better built-in reference citation functionality. Until this happens, it's the only practical solution we have for this particular sort of case. If it's being used where it shouldn't be, then just fix it there. How could a template –, i.e. code intended to be used again and again in a uniform manner by thousands of editors – possibly be an "individualist" exercise, which is a strange mischaracterization you felt compelled to make twice in the space of only 58 words? How can:

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum. ^{[1][2][3][4][5]:3–11}

be "an outright nuisance" in print or otherwise, if:

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum. ^{[1][2][3][4][5]}

is perfectly fine? Since you like metaphors: That's like saying a whole bag of candy is okay, but a bag of candy plus one gummmy bear to finish it off is going to put you in diabetic coma. Or maybe drop the hyperbole??  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  03:59, 28 April 2014 (UTC)

Bit late reply here, but I forgot about this before I ran across yet more usage at right- and left hand traffic and Delaware Basin that baffled me by the addition of paragraphs as well as pages. We have different views about this and concede that I'm letting sheer annoyance affect my posts here. Facilitating future machine-readability and helping non-seeing users or those on simpler systems I understand. But I don't see how any of this justifies this completely made-up notational systems. It's complicated and unintuitive and isn't found anywhere else. It violates basic tenets of simplicity like KISS and the principle of least astonishment. It's pretty much the technical anathema of what is specifically recommended against in guidelines like WP:JARGON, WP:ASTONISH and WP:COMMONSENSE.

Regarding the examples given, instances of ^{[1][2][3][4][5]} are not ideal, but they can't possibly be worse than perfectly plausible examples like this: ^{[1][2]:45–66[3][4]:563–578[5]:3–11}. Saying the former is just as bad as the latter just doesn't strike me as fair. This problem here is obviously solved by using one note for multiple references rather than any sort of stacking, as you can see here. Adding another layer of superscripted information is just adding to the complexity. The problem as I see it is that the notation system is erroneously conflated with the instances of references themselves. A note is a typographical indication that there is more to read, be it comments or references. That's it. It has never been intended to indicate a "count" of cited sources or to provide information in itself, because it's highly impractical both on paper and in digital text. Information should not be indicated at more levels than is actually necessary. Attempting to make notes or refs less repetitive or "annoying" is also seriously misguided. Redundancy in referencing is perfectly excusable and should never, ever be a reason to starting messing with the prose. Notes and refs don't need to be pretty since they have just one primary purpose: to be as clear, consistent, informative and unambiguous as possible.

Peter ^Isotalo 12:49, 3 August 2014 (UTC)

Reduced intstructions

I edited[1] the documentation of the template since it had gotten sidetracked into some rather complicated argument about why this template is necessary. The biggest problem was the portrayal of existing, extremely common standards of notation as somehow convoluted and complicated with words like "basically unreadable and certainly useless" or "reader-annoying". This focuses on extreme examples (100+ cites, but linking to a mere 13 cites as horrid in itself) and completely ignores that superscripted page notation is unheard of outside of Wikipedia. Quite simply, using template documentation as platform for making arguments about typography is not helpful to anyone.

Peter ^Isotalo 13:05, 3 August 2014 (UTC)

But that's precisely what you're doing in stripping the template documentation of instructions about when and when not to use it, and why. You're clearly opposed to this usage, on the basis that it's "unheard of outside of Wikipedia" (of course it is - there aren't any serious competiting online encyclopedias that need to deal with this issue!), and removing the material in question serves your interest in seeing it further opposed. Your change of "Overuse of this template will make prose harder to read," to the grossly overgeneralized "Use of this template may be perceived as making prose harder to read," followed by the suggestion that it anyone should revert it any use of it on that basis, is a telling, WP:SOAPBOXy change. I'm thus reverting your excessive changes, though also moderating the tone of the material in question, since it was actually a bit polemical in places. Note also that what you deleted contained other important information, such as that the solution provided by this template is explicitly intended to be temporary, and replaced as Cite.php is improved (at some point) to deal with the problems that this template works around. Finally, there actually are sources that are cited 100+ times in the same article. This is most common in WP:GLOSSARIES, and that's precisely the context in which this template was created in the first place. Anyway, I did retain some of your changes as well as addressed the tone issue you raised; it's not a blanket revert.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  01:30, 4 August 2014 (UTC)

You're kinda missing the point here. I tried to trim down text that clearly focused on arguments. The paragraph "The problems addressed..." is very obviously not a neutral description nor any kind of useful instruction. Calling 20+ different citations from a single soruce a "severe problem" is really just a subjective opinion. Many, many users, quite possibly the great majority, don't seem to have much of a problem with it. And the example of 20+ pages in one note is pretty much a fictitious solution to a subjective problem. None of it is actually instructive.

If you're not prepared to accept the rather mild rewording of "use" over "overuse" (I thought a general caution was justified), you can't argue that your preferred views are unproblematic. I'm suggesting a toned down version, so please consider the arguments before reverting.

Peter ^Isotalo 02:52, 4 August 2014 (UTC)

@Peter Isotalo: I did consider them, as I've already clearly indicated. I'm not missing any point, but you missed several, even after they were spelled out pretty clearly, such as that {{Rp}} addresses not one but two different problems associated with citing the same source numerous times. I've restored the second. Changing "overuse" (excessive, wrongful, misapplied use) to "use" (any use at all, ever) is not a "mild rewording", it's an assault on the purpose and legitimacy of the template. It's also in keeping with other edits you made that seem intended to downplay the very broad support and wide implementation this template has, and imply that it's some weird nonsense rather than directly borrowed in function if not precise stylistic form from Harvard, AMA, and APA referencing. Not to mention your efforts to promote Harvard style as a Wikipedia "standard" (it's no such thing here, just one of various styles). Several of your objections seem petty and pointless; you objected to 100 so I changed it to 20 (despite 100 actually being based on real usage), but now you object to 20. If you don't like the word "severe", take it out; there's no need to delete the entire passage. WP:BRD need not be an exercise in tooth-gnashing, nor nit-picking. I've tried in good faith to produce another compromise version, but please note that BRD is bold-revert-discuss, not bold-revert-try-again-and-again-but-avoid-discussion. This template and its documentation have been stable for a very long time; it's a consensus that you're trying to change. The least you can do is try to clearly articulate what you think the problems are.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  11:33, 4 August 2014 (UTC)

As far as I can see, we've done the BRD dance quite nicely. I still don't quite understand example 2 (who actually does that?), but at least it's easier to parse now. My complaint about 100 repeats was about exemplifying it with an instance of just 13. The former is somewhat awkward, the latter perfectly normal. But everything is much more tightly focused on explaining the merits of this template now. What should be avoided is portraying other standards as excessively cumbersome, because the exact same can be said about this. Focus should always be on actual traits, not "this is better than that"-arguments. If you want to establish original reference standards, you don't do it by trying to make other standards look crappy.

Btw, please note that I have not claimed any one style to be the Wikipedia standard. Harvard and the likes are common are established standards, though. Take a bit more care with your conclusions.

Peter ^Isotalo 12:28, 4 August 2014 (UTC)

Problem type 2 was part of the genesis of the template. It was happening quite frequently in longer articles on obscure topics (i.e. those with a few editors devoted to writing and sourcing a lot, but not enough to ensure that everything was done to WP best practices), and it's a serious matter because it basically means that material has to be re-researched and the citations fixed, which can be prohibitively time-consumptive, especially in something like a long glossary or list article with hundreds of entries. {{Rp}} helps ensure that it's done right the first time; the incidence of problem type 2 has gone down markedly since the introduction of {{Rp}}, and it seems so who-actually-does-that weird to you because {{Rp}} is doing its job. It's a bit like saying we don't need vaccinations because we hardly ever get sick. ;-) I get what you mean about "a" vs. "the" standard, but no citation styles are "standards" at WP, they're just "styles". I don't "want to establish" anything; {{Rp}} is already quite well established, without being promoted in any way, simply because it works and is the easiest solution that works. I don't think there's any expectation on anyone's part that it'll become some formal policy here, nor that it'll supplant off-WP styles in external publications like law reviews and science journals. (I have actually seen it used on other websites, but promoting it off-wiki isn't a "goal"). Other styles are excessively cumbersome (i.e. crappy), in some contexts, and (as you say) this one can be, too. I think this was already clear, but if we both think it's even clearer now, then thumbs up.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  12:56, 4 August 2014 (UTC)

Protected edit request on 7 August 2014

Please replace the template's code thus:

New code

Current code

{{#if:{{{needed|}}} | {{Page needed |date={{{date|}}}}} | <sup class="reference" style="white-space:nowrap;"><!-- -->{{#if:{{{page|}}} | {{#if:{{{nopp|}}} |( |(p}}{{{page}}}) | {{#if:{{{pages|}}} | {{#if:{{{nopp|}}} |( |(pp}}{{{pages}}}) | {{#if:{{{at|}}} |({{{at}}}) | :{{{1}}} }} }} }}<!-- --></sup> }}<noinclude><!-- Note: Please do not change the parameter names "page", "nopp", "pages" and "at" as they are also used by [[Template:Citation]]. --> {{Documentation}} </noinclude>

{{#if: {{{needed|}}} |{{page needed|date={{{date|}}} }} |<sup class="reference" style="white-space:nowrap;">{{#if: {{{page|}}} |{{#if: {{{nopp|}}}|(|(p}}{{{page}}}) |{{ #if: {{{pages|}}} |{{#if: {{{nopp|}}}|(|(pp}}{{{pages}}}) |{{ #if: {{{at|}}} |({{{at}}}) |:{{{1}}} }} }} }}</sup> }}<noinclude> <!-- Parameter names "at", "page", "pages", and "nopp" are for interoperability with [[Template:Citation]], please do not change them. --> <noinclude> {{Documentation}} <!-- PLEASE ADD CATEGORIES AND INTERWIKIS TO THE /doc SUBPAGE, THANKS --> </noinclude>

The "New code" is the version currently in the sandbox. Its alignment has been amended, the comment rephrased and the duplicate <noinclude> and surplus blanklines before the Documentation removed. No functionality etc has been altered. Sardanaphalus (talk) 11:23, 7 August 2014 (UTC)

Is it your intention that the new version of the template should match the version currently in the sandbox? --Redrose64 (talk) 14:09, 7 August 2014 (UTC)

Apologies; I removed a newline as an experiment but forgot to revert. The sandbox should now match the "New code" above, which is what's intended. Sardanaphalus (talk) 17:23, 7 August 2014 (UTC)

There's still a difference. The sandbox has

| <nowiki>:</nowiki>{{{1}}}

whilst the code blob above has

| :{{{1}}}

To eliminate this confusion, please ensure that the sandbox has the desired code, and remove both of the code blobs above (whether they be correct or not). --Redrose64 (talk) 17:41, 7 August 2014 (UTC)

• I'm confused now – which version do you recommend? (Perhaps I shouldn't use code "blobs"...) Sardanaphalus (talk) 20:21, 7 August 2014 (UTC)

That's why we have a sandbox page and WP:TESTCASES. You put your proposed change in the sandbox version, and set up demonstrations at Template:Rp/testcases that show that not only does your proposal do the extra things that you want it to do, it is also not deleterious to current usage. --Redrose64 (talk) 20:53, 7 August 2014 (UTC)

The <nowiki>...</nowiki> does exist in this talk page "blob", but isn't escaped. To be clear, it should stay. Incidentally, I think that's way too much whitespace, and I've edited the sandbox accordingly. —WOFall (talk) 23:44, 7 August 2014 (UTC)

I partially implemented the changes. Let me know if there is still something that needs to be fixed. Thanks! Plastikspork _―Œ^(talk) 18:14, 9 August 2014 (UTC)

The recent edit to this template seems to have a bug, forcing page numbers into the main article body, rather than appearing in the footnotes as they should. For an example, see Albert Einstein#Travels abroad, 1921–1922--Pharos (talk) 17:24, 12 September 2014 (UTC)

This is the purpose of the template; one footnote per reference, with relevant page numbers marked at point of reference. I don't think the behaviour has changed. —WOFall (talk) 17:52, 12 September 2014 (UTC)

The only edit in the last three years is this one. It did nothing to the template output. Footnotes are generated from whatever is between <ref>...</ref>, but the documented usage of {{rp}} is to put the template after the closing </ref> tag; that is to say, in the article text, not in the footnote. --Redrose64 (talk) 18:15, 12 September 2014 (UTC)

Deprecate?

Shouldn't this be replaced with a "page=" parameter in most citations nowaday? Thanks. Mike Peel (talk) 21:17, 19 August 2015 (UTC)

No. Did you read this template's documentation?  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  01:13, 20 August 2015 (UTC)

Potential replacement for {{Rp}}

FYI

 – Pointer to relevant discussion elsewhere.

The new improvement to <ref>...</ref> proposed at meta:WMDE Technical Wishes/Book referencing/Call for feedback (May 2018) would obviate the need for the {{Rp}} template, as well as provide various other enhancements. The discussion is presently swamped by people who just don't like fully-inline citations and only want to use {{sfn}} and page-bottom referencing, but this is a false dichotomy. The discussion isn't about which citation style is better (the answer to that is "it depends on the article"); the question is whether this feature would be good to have for referencing that is fully inline, and the answer is clearly "yes". I would be delighted if my old {{Rp}} template was finally superseded by an actual (and more tidy) feature of MediaWiki itself.  — SMcCandlish ☏ ¢ 😼  22:05, 1 June 2018 (UTC)

RP is not working

I am trying to use the {{RP}} template in some new editing, but right now it is not working. It does not list the page numbers. My previous uses, several years ago, are still working just fine. The documentation is at https://en.wikipedia.org/wiki/Template:Rp. What is the problem and how can it be solved? Thank you. BeenAroundAWhile (talk) 10:50, 26 October 2018 (UTC)

It seems to be working as documented. You've got the ":5" after the ref number resulting from this edit. --David Biddulph (talk) 11:09, 26 October 2018 (UTC)

You are correct. Everything is OK. My panic is averted. Thanks. BeenAroundAWhile (talk) 11:21, 26 October 2018 (UTC)

RP can generate cite errors

If {{rp}} template is used after invoking ref by name, it will cause Cite error: The named reference "<foobaref>" was defined multiple times with different content (see the help page).. Several of these currently appear in the ref list of Second French intervention in Mexico. Dl2000 (talk) 15:24, 7 September 2019 (UTC)

Just for the record (this is old), that was not related to {{Rp}}. It was caused by the article transcluding {{Second French intervention in Mexico infobox}}, which named and defined references like "Niox". Those same references were again defined in the article and named the same, instead of just referring to them with, e.g., <ref name="Niox" />{{Rp|123}} or {{R|Niox|123}}. This is a general problem with transcluding something that has references – you have to make sure not to duplicate reference names. Of course, if you refer to those sources from the template and someone edits the template cites, they can break anything that transcludes them. (Einstein95 fixed the article in December 2019 in the way I noted.) —[AlanM1 (talk)]— 20:20, 21 May 2020 (UTC)

Dashes and en dashes

Is there some reason that dashes are not being converted to en dashes, as is done by the citation modules? It's certainly confusing, and makes articles harder to write in the VE. Please therefore kindly copy Template:rp/sandbox to Template:rp. I synced it beforehand. Test cases at User:Psiĥedelisto/sandbox § {{rp}}. Psiĥedelisto (talk • contribs) ^{please always ping!} 09:31, 25 July 2020 (UTC)

 Done Mdaniels5757 (talk) 20:05, 26 July 2020 (UTC)

FYI: ref extends= coming down the pipeline

Hello, I thought everyone who finds use for this template, and works on it, should know that an integrated alternative has been created within MediaWiki itself during the past several years, thanks to the stalwart and heroic developer team:

meta:WMDE Technical Wishes/Book referencing/Call for feedback (May 2018) § This feature can now be tested on the beta cluster

...and presumably, one glorious day, will be available here as an option on enwiki. --‿Ꞅtruthious 𝔹andersnatch ͡ |℡| 19:03, 9 December 2020 (UTC)

Broken

This template is broken: it is now appearing as "Template:Rp" in articles. See Latinx for an example. Crossroads ^-talk- 19:04, 29 August 2021 (UTC)

Seems fixed now somehow. Will comment at Template talk:R. Crossroads ^-talk- 19:18, 29 August 2021 (UTC)

This was unrelated to {{rp}} and was caused by a post-expand limit condition caused by a (former) bug in {{r}} which kept most templates from expanding on that page. Long fixed.

--Matthiaspaul (talk) 17:24, 4 September 2021 (UTC)

Request quote parameter

I'd like to be able to include a pull-quote, in the manner of {{cite}}'s quote parameter. This would probably require each such use of rp to generate a new-numbered entry in the citation list, with "ibid" referring back to the full entry at the beginning of the group. Jeh (talk) 05:10, 8 January 2015 (UTC)

The {{rp}} template has no connection with the reference list. Yes, it's normally placed after a </ref>, but there is no association between the two other than the coincidence of being physically adjacent. As regards ibid, please, no. See WP:IBID. --Redrose64 (talk) 10:21, 8 January 2015 (UTC)

Ok. It doesn't have to say "Ibid". It doesn't even have to be done by the Rp template. Where, then? Jeh (talk) 00:00, 9 January 2015 (UTC)

Try using {{sfn|Author|Year|p=Page|ps=Quoted text}}. Alternatively, <ref>{{harvnb|Author|Year|p=Page}} Quoted text</ref>. --Redrose64 (talk) 00:29, 10 January 2015 (UTC)

But be aware that doing so may be a citation style change, which per WP:CITEVAR requires consensus to implement, at any non-stub article.  — SMcCandlish ☏ ¢ 😼  22:06, 1 June 2018 (UTC)

You can use the optional |quote= (and |trans-quote=) parameter of {{rp}} (and {{r}}) to display a quotation from the source as a tooltip when hovering over the link to the reference, however, what you seem to prefer is to have the quote in the full citation. For this, you could either use the |quote= (and |trans-quote=) parameter of the CS1/CS2 citation templates, or, if you want to define the quote in the invocation of the reference rather than the full citation, you can use {{r}} to define the quote locally and not only make it available as a tooltip, but optionally to automatically append it to the full citation defined elsewhere. For example, the citation could be defined as:

{{r|n=Smith2021|r=MyCitation.}}

Lorem ipsum dolor sit amet, consectetur adipisici elit, ...

(The |r=/|reference= parameter can take raw citations just as well as, for example, a definition through a CS1/CS2 citation template. In addition to this, local page numbers etc. could be specified as well and would be displayed alongside the superscript link just as with {{rp}}.)

And now the quote would be appended to the full citation through the |a=/|annotation= parameter as:

{{r|n=Smith2021|p=42|q=This is [[quotation|quoted text]] from the source|a=q}}

... sed eiusmod tempor incidunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, ...

(The parameter takes free text input, but also recognizes a number of special tokens to take either the page or the quote defined in the other parameters, so you don't have to write it down twice in the template call.)

And this is how it looks in the References section:

Note, that in this simple example I deliberately did not specify a quote in the original definition of the citation, but in the References section it still appears appended to the full citation. This also works for page numbers, page numbers and quotes combined, or arbitrary commentary, even with wikilinks (including backlinks), further embedded citations (to bundle multiple citations under a single entry) or nested references.

--Matthiaspaul (talk) 18:17, 4 September 2021 (UTC)

Suggestion, state "p" or "page"

Would it be possible to have the template provide "p" with an optional "pp" or perhaps "page"/"pages", so it is more obvious that this is a page number? Jytdog (talk) 14:40, 6 June 2018 (UTC)

? Jytdog (talk) 22:58, 9 June 2018 (UTC)

I agree completely. I am a long-time Wikipedia user (and occasional editor) and I just recently came across a page that uses this rp template. It took far too long to decipher what it meant. There are many other voices on this Talk Page saying the same thing. Why can't this obscure code be made more transparent with Jytdog's suggestion? --seberle (talk) 13:32, 23 August 2018 (UTC)

Done. It now displays a tooltip containing this info. Or switch to |style=ama.

--Matthiaspaul (talk) 23:14, 4 September 2021 (UTC)

Too obscure

Shouldn't this template give readers a bit more of a hint as to what it's trying to convey? I've just encountered it on a page and assumed it was a formatting error. It wasn't until I went to edit the page that I saw it was actually a template rather than mangled text that I realised it was probably intentional, but still had no idea of the intent until I manually typed in the template's URL and found my way here. Okay, I know I often don't "do" obvious but it still strikes me as being less than ideal. --Vometia (talk) 17:43, 8 February 2020 (UTC)

The same thing just happened to me. I jsut stumbled upon a use of rp, and thought the extra numbers was garbage. How many people actually understands this reference method? If you are not used to editing on wikipedia, you ahve no way of finding this template information, and you will be stuck at not knowing what the numbers mean. Moberg (talk) 15:36, 13 April 2020 (UTC)

@Vometia and Moberg: (Late response) Perhaps the template could add a tooltip that explains the notation when you mouse over it? —[AlanM1 (talk)]— 20:04, 21 May 2020 (UTC)

I'm a huge fan of RP because it keeps the reference sections neat (I know others prefer SVN because it makes reference sections look more like printed bibliographies, and because the in-article citations are shorter). I agree that some sort of "page" indicator might be helpful. --Neopeius (talk) 18:57, 8 December 2020 (UTC)

Done. It now displays a tooltip containing this info. Or switch to |style=ama.

--Matthiaspaul (talk) 23:16, 4 September 2021 (UTC)

AMA-like mode: new parameters "at", "page", "pages", and "nopp"

Unresolved

 – Is the nopp parameter useful?

Use of these parameters trips the template into "AMA-like mode", a slight modification of the AMA style of citation that uses parentheses instead of the colon and includes a "p." or "pp." for clarity. The names of the parameters are intended for compatibility with {{citation}} so please don't change them (though you could add additional parameter names as alternatives if desired.) For usage examples see testcases. --❨Ṩtruthious ℬandersnatch❩ 08:52, 3 February 2009 (UTC)

In classic AMA style, there is no p. and space, instead there it looks like this^1(p30) or this.^2(pp30-40) I prefer to leave out the space and punctuation. Could we do that? II | (t - c) 23:26, 9 March 2009 (UTC)

It's been over 2 months now, and this code has not been documented at the /doc page. — SMcCandlish [talk] [cont] ‹(-¿-)› 20:23, 8 April 2009 (UTC)

I've documented the page, pages, and at parameters. It looks like the nopp=true parameter causes page and pages to function exactly like the at parameter. Is the metadata that would be generated by hidden page and pages parameters valuable for anything? If not, it seems simpler to drop the nopp parameter altogether. Lyrl^Talk _C 00:49, 14 August 2009 (UTC)

{{rp}} does not generate any metadata. Still, |no-pp= is provided for compatibility with CS1/CS2 (and other citation templates). --Matthiaspaul (talk) 23:20, 4 September 2021 (UTC)

AMA style and precedent

Resolved

 – AMA style supported.

A while back, I added mention of Template:Rp to WP:CITE. It was removed by User:SallyScot based on lack of precedent (WP:CITE thread). I objected and argued for reinstating it based on the AMA's precedent (second CITE thread), but never got around to doing so. The only major opponent to representing it on that page seemed to be SallyScot. Anyway, this argument of precedent may be a consistent voice against Template:Rp. The most similar citation practice comes from AMA style, which uses looks like^[1](p3) or^[1](pp3-30) compared to the current default of.^[1]:3 Personally I kinda prefer the current default, but switching the default to AMA would overcome any arguments that it's nonstandard and thus inadmissible. II | (t - c) 08:11, 28 April 2009 (UTC)

I see no opposition, and your arguments seem convincing. Let's try it out. —TheDJ (talk • contribs) 19:34, 7 May 2009 (UTC)

On second thought, the current practice of :page seems to provide some visual connection to the citation it applies to. Perhaps ^[1]:(pp3-30) ? —TheDJ (talk • contribs) 19:36, 7 May 2009 (UTC)

Template now supports AMA style as an option. — SMcCandlish [talk] [cont] ‹(-¿-)› 00:30, 27 November 2009 (UTC)

And now also a |style=ama parameter to select it (so the |page=/|pages=/|at= etc. named parameters can also be used with AMA style). --Matthiaspaul (talk) 23:26, 4 September 2021 (UTC)

Odd behavior

I often use cites like {{Harvnb|Dyal1996|p=[http://books.google.com/books?id=PvxFKPI6q_oC&pg=PA19 19]}}, which renders as

Dyal1996, p. 19

I see that <ref>{{Harvnb|Dyal1996}}</ref>{{Rp|19}} renders as

However, <ref>{{Harvnb|Dyal1996}}</ref>{{Rp|[http://books.google.com/books?id=PvxFKPI6q_oC&pg=PA19 19]}} renders as

Can this be fixed?

As a workaround, <ref>{{Harvnb|Dyal1996}}</ref>{{Rp|1=[http://books.google.com/books?id=PvxFKPI6q_oC&pg=PA19 19]}}, naming parameter one explicitly, renders as

as desired, but that is ugly and confusing. Wtmitchell (talk) (earlier Boracay Bill) 03:23, 16 February 2010 (UTC)

well, the problem is the equals sign in the links you're using. for instance, is you change http://books.google.com/books?id=PvxFKPI6q_oC&pg=PA19 19 into http://books.google.com/books?i{{=}}PvxFKPI6q_oC&pg{{=}}PA19 19 - changed = into the template call {{=}} - the template works as expected. I don't think this is the way the Rp template was designed to work, however. I think it just wants page numbers, and trying to feed it entire links confuses it. at any rate, using the 1= notation is probably the simplest way around the problem short of changing what you're doing. --Ludwigs2 06:22, 16 February 2010 (UTC)

That explains it. Thanks. Wtmitchell (talk) (earlier Boracay Bill) 20:29, 16 February 2010 (UTC)

This trick is now no longer necessary as the named |page=/|pages=/|at= etc. parameters can be used also for the default ":" display format.

See also: now

--Matthiaspaul (talk) 23:44, 4 September 2021 (UTC)

Request url parameter

A regular reference to a book page can be externally linked to an external display of the page. That capability should be extended to this template via a url parameter. The usage would be {{rp|nn|url=link}}, where nn is the page number and link is the url of the desired page. It would display the page number with a standard external link.

Sample usage

The book ''Tom Sawyer'' was mainly the truth.<ref name=Huck>{{cite book |title=Huckleberry Finn |author=Mark Twain}}</ref>{{rp|15|url=https://books.google.com/books?id=S3JHAAAAYAAJ&dq=huckleberry%20finn&pg=PA15#v=onepage&q=huckleberry%20finn&f=false}} The widow called Huckleberry a poor lost lamb.<ref name=Huck/>{{rp|16|url=https://books.google.com/books?id=S3JHAAAAYAAJ&dq=huckleberry%20finn&pg=PA16#v=onepage&q=huckleberry%20finn&f=false}} Huckleberry went to his room with a piece of candle and put it on the table.<ref name=Huck/>{{rp|17|url=https://books.google.com/books?id=S3JHAAAAYAAJ&dq=huckleberry%20finn&pg=PA17#v=onepage&q=huckleberry%20finn&f=false}}

Display (current)

The book Tom Sawyer was mainly the truth. The widow called Huckleberry a poor lost lamb. Huckleberry went to his room with a piece of candle and put it on the table.

How it would work if the url parameter were enabled

If the url parameter were enabled, "15", "16", and "17" above would display something like this:

15
16
17

— Anomalocaris (talk) 09:55, 18 December 2014 (UTC)

I would like to second this request! Also I'll mention that "sfn" makes for a decent alternative; see here: Template:Sfn#Adding a URL for the page or location — Dranorter (talk) 11:43, 9 June 2016 (UTC)

Huh? You don't need {{sfn}} for this as you can do exactly the same with {{rp}}:

The book ''Tom Sawyer'' was mainly the truth.<ref name=Huck>{{cite book |title=Huckleberry Finn |author=Mark Twain}}</ref>{{rp|1=[https://books.google.com/books?id=S3JHAAAAYAAJ&dq=huckleberry%20finn&pg=PA15#v=onepage&q=huckleberry%20finn&f=false 15]}} The widow called Huckleberry a poor lost lamb.<ref name=Huck/>{{rp|1=[https://books.google.com/books?id=S3JHAAAAYAAJ&dq=huckleberry%20finn&pg=PA16#v=onepage&q=huckleberry%20finn&f=false 16]}} Huckleberry went to his room with a piece of candle and put it on the table.<ref name=Huck/>{{rp|1=[https://books.google.com/books?id=S3JHAAAAYAAJ&dq=huckleberry%20finn&pg=PA17#v=onepage&q=huckleberry%20finn&f=false 17]}}

Display:

The book Tom Sawyer was mainly the truth. The widow called Huckleberry a poor lost lamb. Huckleberry went to his room with a piece of candle and put it on the table.

The trick is that you must replace the first use of the anonymous parameter (for pages) by the numbered parameter |1= because anonymous parameters don't accept "=" which is quite common in URLs. Then you can use the normal wikitext syntax to define links (as you would have to do in {{sfn}}'s |loc= as well). Of course, this also works within |page=, |pages= and |at= if you preper AMA style.

--Matthiaspaul (talk) 13:31, 30 August 2021 (UTC)

References

This trick is no longer necessary as the named |page=/|pages=/|at= etc. parameters can now be used also for the default ":" display format.

See also: now

--Matthiaspaul (talk) 23:44, 4 September 2021 (UTC)

New |style=ama parameter

I was bold and switched the default display format of this template to use the ":" notation even when the named |page=/|pages=/|at= etc. parameters are used. This is more consistent with the logic used in other templates, and there also have been some sort-of-complaints about this in the past, see:

• Template_talk:Rp#Suggestion, state "p" or "page"
• Template_talk:Rp#Too_obscure

The tooltip switches between "Page", "Pages" and "Location" depending on if a singular, a plural, or an in-source-location parameter has been used.

There is a new |style=ama parameter to reenable the supposed-to-be AMA-style display with parentheses. Likewise, with |style=ama, this also gives AMA-style output when the page information was given as unnamed parameter 1.

This change also means that editors, who want to link page numbers to external sites, do not have to use the |1= trick any more and can just use the named parameters to provide the link (as with {{sfn}} etc.). See also:

• Template_talk:Rp#Odd_behavior
• Template_talk:Rp#Request_url_parameter

--Matthiaspaul (talk) 23:01, 4 September 2021 (UTC)

@Matthiaspaul: can you explain why these IP edits [2][3] are claiming the existing format of the rp template is broken and that |style= is now required. I've reverted it as it changes the visible appearance of the refs, adding p. or pp. which to me is just clutter on the page. That should not be compulsory and I'm hoping you are going to say it isn't. SpinningSpark 17:52, 10 September 2021 (UTC)

Hi Colin, they aren't "broken" at all. They have been enhanced. It is just that the behaviour has changed. Previously, this template displayed the page info in "superscript :" format when the page info was given as unnamed parameter 1, then someone added the alternative AMA format using the parameters |page= and |pages= and |at= not considering that these parameter are used for the "superscript :" format in other related templates (and are not associated with any particular output format in other citation templates). This not only led to inconsistencies in the parameter format between the templates but also made it difficult to further enhance the handling for the original ": format", i.e. requiring this odd "1=" trick to use page links in conjunction with the ": format" (which incorrectly made people believe RP would not support linked pages at all), or to not be able to display the proper numerus in the tooltip. Since I was now merging the redundant parts of some related templates to add new features and ease future maintenance, I now changed the interface so that the |page=/|pages=/|at= parameters can now also be used for the default ": format" (as in the other templates) as well, and at the same time introduced a new |style= parameter to select the AMA output format (and potentially some more, if desired) where this is actually desired. This gives a more logical parameter interface for this template, achieves consistency with other related templates and allows to further enhance the scheme in the future.

So, in your example cases, the IP obviously wanted to switch the display format (back) to AMA style. I don't think this was/is a necessary change, but on the other hand, if they prefer this format, that's the reason why we have the |style= parameter now, so your reversion was not necessary, either... ;-) They were certainly acting in good faith.

--Matthiaspaul (talk) 18:26, 10 September 2021 (UTC)

So is it the case that I can no longer see in the revision history what these templates used to render as? That makes the history very confusing. SpinningSpark 18:35, 10 September 2021 (UTC)

Well, while the edit summaries do not (can not) always mention all implications when doing a major rewrite moving functionality into subtemplates and sharing some code among several related templates, this change was documented quite well (to the extent possible in a summary). It may still not be easy to follow the changes when multiple files are involved, I agree. Hence, for the discussion of the reasoning and the implications I created this thread.

In regard to {{rp}} specifically, if, before the change, the template was called with one of the page parameters, you can be sure the output was in AMA style; if it instead used the first unnamed parameter to supply the page information, the output used the default "superscript :" style. However, this holds true only for {{rp}}: {{ran}} and {{r}} always gave "superscript :" style output with the page parameters (and still do, unless |style= would be used to choose a different format). Now, after the change, they are consistent in regard to parameters and behaviour.

--Matthiaspaul (talk) 20:19, 10 September 2021 (UTC)

The article in question quite clearly was not intending to use AMA style because colons before page numbers had been explicitly included. These are not used in AMA style per AMA Manual of Style (10th edition) page 44. If these have previously been interpreted as AMA style then that is a mistake of the template and I stand by my reversion. My point about edit history is precisely that I cannot be sure what was intended because the way the template renders has changed (very possibly more than once). Remember, I may not be looking at the edit that put in the ref, so its edit summary does not immediately help, even if useful. I'm not actually opposing what has been done here, I was just asking for clarification, but we don't want to end up like Wiktionary where if you go back in the history a few years almost every template is either redlinked or littered with syntax error messages. SpinningSpark 11:03, 11 September 2021 (UTC)

I now had another look at the article you linked to. Unless I overlooked something they did not use colons in front of the page numbers. I saw they were using unnecessary narrow non-breaking spaces ( ), and lots of thin spaces between the refs. To me, it looks as if they were, in fact, using AMA style before, so, if they really want it back, I propose to just let them add |style=AMA and be happy (although I, like you, prefer the superscript : format). After all, it's just cosmetics. This answer is only to, hopefully, provide you with the clarification you were asking for - I have no investment in that article and therefore this is a complete non-issue for me. What is important is that the information gets displayed correctly, and it does (and did) regardless of the output format. What's also important is to keep those who are actually working on the article happy, because otherwise it will be detrimal to further article development. So, if they want AMA, let's give it to them. If they don't care, at least I do not care as well.

Regarding ending up like Wiktionary, I don't see this situation arising here at all, as none of the changes broke anything (the new implementation correctly displays the provided information just like the old one did, in both formats). While some cleanup work in articles can help to further improve the output, no fixups are actually necessary.

I think it is always good to invest some time in a good design (like choosing a future-compatible parameter like |style=ama rather than the originally proposed |ama=yes or even just "occupying" the |page=/|pages= parameters), as this will likely help to avoid breaking changes in the future, but often enough templates are developed ad hoc with a narrow focus on solving immediate issues only. For example, the way AMA style was originally added to this template, was an example for a crude ad hoc solution without thinking about the future, actually without even thinking about the other format already supported at that time - so we can be happy that it was possible to finally rectify this in a compatible manner.

The scenario you describe is what I have seen in the development of other templates, but I have no idea how to avoid this completely unless we would want to stop any further development, which would be the end of the project.

--Matthiaspaul (talk) 11:38, 12 September 2021 (UTC)

Yes, I was assuming the html was for the colon, rather than the thin space it actually was. But if it was |pages= that removed the colon and turned into AMA style, then that was done by IP edit immediately before the explicit |style=AMA. So thanks for the explanation of where I went wrong, but I still think my reversion as an undiscussed style change stands. SpinningSpark 14:28, 13 September 2021 (UTC)

New hyphen2dash code breaks semicolons?

The new hyphen2dash code appears to convert semicolons to commas, breaking HTML entity code used in |1=. – Jonesey95 (talk) 22:24, 1 September 2021 (UTC)

@Jonesey95 Should be fixed now. The code from Module:Citation/CS1 had the same limitation since it only dealt with certain defined HTML entities, but I changed the version in String2 so that it uses mw.text.decode to decode all HTML entities, splits the list, and then uses mw.text.encode to re-encode it. --Ahecht (TALK
PAGE) 01:40, 2 September 2021 (UTC)

Well done. Thanks. And, for your reward, I will ask you to take a look at the degenerate case shown under "Missing page numbers" in the documentation. It looks like the semicolon at the end of the style declaration is also being converted to a comma, breaking the display. This case is showing what not to do, of course, but it might be able to degrade more gracefully. – Jonesey95 (talk) 03:41, 2 September 2021 (UTC)

Ahecht: More seriously, the new code is breaking usages in article space. See Novalis#Birth and early background, where plaintext span tags are being displayed. Also see St. Casimir Church (Cleveland, Ohio). I think the changes might need to be reverted, with test cases from these articles added to the testcases page. – Jonesey95 (talk) 04:07, 2 September 2021 (UTC)

@Jonesey95  Fixed. --Ahecht (TALK
PAGE) 15:24, 2 September 2021 (UTC)

You do good work. I'll post here if I come across any more. – Jonesey95 (talk) 15:44, 2 September 2021 (UTC)

See also:

• Template_talk:Rp/Archive_1#Dashes_and_en_dashes
• Template_talk:R#Hyphens_to_dashes
• Help_talk:Citation_Style_1#{{hyphen}}_converted_to_dash
• Talk:Antoine_Ephrem_Cartier/GA1
• Template_talk:Sfn#Add_automatic_endash_for_page_number_range?

--Matthiaspaul (talk) 10:14, 12 September 2021 (UTC) (updated 12:53, 13 September 2021 (UTC))

Novalis is broken again, showing half-formed span tags in the article prose. Pinging Matthiaspaul and Ahecht. – Jonesey95 (talk) 16:13, 15 September 2021 (UTC)

This was caused by the use of {{plainlink}} in the |page= parameter to suppress the external link icon normally displayed for external links. While I consider this to be some form of "abuse", I have to admit that it looks nicer with the external link icon suppressed in the superscripts.

However, it collided with our code which reduces spaces to thinspaces in the superscript. Since this also broke two other test cases (with stripmarkers and cat links), I have temporarily commented this out until we have a better solution for this.

--Matthiaspaul (talk) 17:11, 15 September 2021 (UTC)

New code appears to have broken existing usage of asterisk

Please see Wolk v. Kodak Imaging Network, Inc.. I tried restoring the literal asterisk character, but neither that nor the template appear to work now. I put test cases on the testcases page. – Jonesey95 (talk) 01:43, 4 September 2021 (UTC)

{{rp|*1}} →

{{rp|{{Asterisk}}1}} →

I also observed a number of other problems (like invalid encoding of HTML entities) and completely reworked the code (regarding the display of the superscripts it is now almost the same as in {{r}}).

It would now be easy to make the superscript ":" appearance the default format (also when using the various named page parameters) and introduce a new |style=ama parameter to switch to the AMA format. This would improve consistency with how the parameters are used in various other templates.

The hyphen-2-dash code (which has not yet been incorporated into {{r}}) might need a bit more tweaking, but I have left it enabled so we can study possible effects (it only works when using the plural page parameters or the unnamed parameter, not when using the singular forms or the in-source-location parameters).

I also added the |trans-quote= (for translations) and the |wrap= parameters from {{r}}.

The display of translations requires the definition of a quotation, and the display of quotations requires the specification of a page or location information in the tooltip. Since there have been so many complaints in the past that the {{rp}} notation would not be self-descriptive, even when no quote is defined, a tooltip is silently available anyway to display the page information in more verbose form. I hope this will mute those complaints and increase the acceptance of this format.

The default for the new wrapping feature is auto-wrapping enabled. This will make sure that wrapping can occur after each superscript link. In the rare cases, where this is not desired, |wrap=n can be used to disable this (and reenable the previous behaviour), so that sequences of superscript links do not wrap. Alternatively, to allow wrapping inside of (possibly long) page values, wrapping can be enforced using |wrap=f, but the actual behaviour may depend on the browser and CSS (and thereby the skin used). In rare cases, the auto-wrapping can cause errors when {{rp}} is included in other templates, which check the parameter input for invalid input, see: Template_talk:R#Problems_with_number_conversions. In such cases, |wrap=n can be used to disable the feature. We could also change the default to off, but so far only a few occurences of this issue could be found, so the benefit of keeping it enabled might be higher.

If you spot problems, please report them here.

--Matthiaspaul (talk) 16:36, 4 September 2021 (UTC)

Using Plain_text.main() parameter encode=true worked with " < > but broke &, whereas using Plain_text.main() parameter encode=false with String.replace() " into " worked with " and &, but not with < or >. I have therefore incorporated EncodeDecode.encode() with a reduced custom charset of only " < and >. Now § works in the visible superscript page info as well as in the tooltip, but < and > work as well:

{{rp|§A}} →

{{rp|§A}} →

{{rp|"A}} →

{{rp|"A}} →

{{rp|<1}} →

{{rp|<1}} →

{{rp|>2}} →

{{rp|>2}} →

Hopefully, all possible cases are covered now.

--Matthiaspaul (talk) 12:39, 13 September 2021 (UTC)

I believe that the above markup was working for a while, but the last five testcases are broken again. – Jonesey95 (talk) 21:16, 14 September 2021 (UTC)

Matthiaspaul, thanks for the quick response above. The asterisk testcases are still broken. They were working a few days ago. – Jonesey95 (talk) 17:26, 15 September 2021 (UTC)

(edit-conflict) That's right. The original buggy display you reported was different from the interpretation as "start bulleted list in new line" now (IIRC the asterisk was just suppressed), which I fixed. I also encountered similar bugs for |page=<1 and |page=>1, which displayed as half-formed span tags (but different from those you reported in the other thread above:  <span title="Page / location: <1"><1  instead of just <1), which I fixed as well. Testing this now, there are two more characters producing strange results when used as first character in the page argument, : and # (probably ; as well, but this gets changed to a , as part of the hyphen-to-dash conversion and processing of lists and can therefore be considered okay).

{{rp|pages=*1}} → (starts new line with bullet point *1 now)

{{rp|page=*1}} → (starts new line with bullet point *1 now)

{{rp|*1}} → (starts new line with bullet point *1 now)

{{rp|pages=:1}} → (starts new line with 1 now)

{{rp|page=:1}} → (starts new line with 1 now)

{{rp|:1}} → (starts new line with 1 now)

{{rp|pages=#1}} → (starts new line with 1.1 now)

{{rp|page=#1}} → (starts new line with 1.1 now)

{{rp|#1}} → (starts new line with 1.1 now)

{{rp|pages=;1}} → (converted to comma, okay)

{{rp|page=;1}} → (converted to new line in bold)

{{rp|;1}} → (converted to comma, okay)

{{rp|pages=,1}} → (okay)

{{rp|page=,1}} → (okay)

{{rp|,1}} → (okay)

{{rp|pages=.1}} → (okay)

{{rp|page=.1}} → (okay)

{{rp|.1}} → (okay)

They used to work at some point. I know that MediaWiki treats some of these characters specially as first character (Help:Template#Problems_and_workarounds), but there must be a way to workaround (without reenabling the other problems, that is). Still investigating...

--Matthiaspaul (talk) 17:52, 15 September 2021 (UTC)

This is fixed now. It was caused by an {{#if:...|...|...}} clause I had added recently to distinguish between two internal parameters holding the in-source-location information, the normal one (used in the tooltip and elsewhere) and one with optimized contents for the superscript which had hyphens replaced by endashes for plural pages and the space following a comma list separator reduced to a hairspace. I first tried to evaluate the first letter of these strings and replace :,*# by their HTML entities but while this worked in principle the invokes introduced new trigger points where MediaWiki's special treatment of :,*# as a first character occured, so it didn't solve the underlying problem. Using nowiki was also no option here as it would have broken embedded HTML tags and similar stuff. So, just removing the extra {{#if:...|...|...}} and making 'sup-where' an internally required parameter seemed to be the easiest fix.

--Matthiaspaul (talk) 02:04, 16 September 2021 (UTC)

Thanks. Please make sure to use the sandboxes and the testcases page to test future modifications. – Jonesey95 (talk) 04:03, 16 September 2021 (UTC)