Re T18854: <ref></ref> without <reference/> comment 23: "we need to untangle cite.php's error tags so people can actually control them individually without inavdertantly borking the entire set in one." Also see T24323: Tag cite error revisions.
Description
Details
- Reference
- bz28843
Status | Subtype | Assigned | Task | |
---|---|---|---|---|
· · · | ||||
Invalid | None | T40638 [DO NOT USE] Interface messages needing rewording or documentation and other issues with existing messages [superseded by #Voice_&_Tone] | ||
Open | None | T353695 Remove meaningless "Cite error: …" prefix in favor of independent error messages | ||
Invalid | None | T30843 Cite error uses lego messages breaking localisation and customisation, should be deprecated | ||
· · · |
Event Timeline
happy.melon.wiki wrote:
I'm a little rusty on the details after over two years, but as I remember it, the main problem is that it uses lego messages: each error is defined separately, and then is injected into [[MediaWiki:Cite error]] ("Cite error: $1"). That is a problem for localisation and customisation; but combined with bug 17865, the cite_error message is called with different parsing rules in different contexts, which is bad.
The "Cite error: " prefix should be added to each cite error message individually, and the wrapper message should be deprecated.
Adding many blockers of bug 38638 to the list of "easy" bugs, to mark them as candidates for [[mw:Google Code-in]] tasks (gci2013). If you think this bug is not suitable, remove the keyword.
The only two affected messages appear to be "Cite error: $1" and "Cite warning: $1". We, the WMDE-TechWish team, actively work on Cite for several years now, but still don't really see how these messages are a problem. They are more technical error messages, not necessarily meant to form valid sentences. While a prefix is not terribly helpful, it's not wrong either. As of now it can't be removed because local wiki communities misuse it for all kinds of creative things (tracking categories, hiding certain messages, even replacing the entire error system with a custom one). See https://global-search.toolforge.org/?q=.®ex=1&namespaces=8&title=Cite+%28warning%7Cerror%29.
Cite has much bigger problems with messages, see T321217.
We are also working on T238061, T238188, and other tickets where we try to make Cite's confusing and misplaced error messages much easier to understand.