Rosetta Code:Village Pump/Lang-tag bot: Difference between revisions

← Older edit

Rosetta Code:Village Pump/Lang-tag bot (view source)

Revision as of 17:39, 28 November 2010

1,898 bytes added , 13 years ago

no edit summary

Anonymous user

rosettacode>Parsleyfirefly

Revision as of 17:22, 1 December 2009 (view source) rosettacode>Kevin Reid (TI-89 BASIC needs not-geshi. Please revert.) ← Older edit		Latest revision as of 17:39, 28 November 2010 (view source) rosettacode>Parsleyfirefly No edit summary
(3 intermediate revisions by 2 users not shown)
Line 1: {{Vptopic \|topic=Lang-tag bot \|summary=Program to add lang tags and correct them to use the right identifier }} ==Announcement== Line 44 ⟶ 48: Also, I think it's a bad idea in general for the bot to strip out tags with attributes it doesn't expect (pre style= here), since they indicate some special situation. —[[User:Kevin Reid\|Kevin Reid]] 17:22, 1 December 2009 (UTC) : I think it makes more sense for Short Circuit to either :* install a stub language definition for TI-89 BASIC, as I explained above (this would be extremely easy! I'd do it myself, but the permissions on the relevant directory won't let me), or :* fix [[Rosetta Code:Village Pump/Syntax Highlighting#Non-ASCII characters and unrecognized languages\|this bug]]. :Changing all that markup just to install a bug workaround which would then have to be reversed when the bug is fixed would be, in my humble opinion, completely boneheaded. —[[User:Underscore\|Underscore]] ([[User talk:Underscore\|Talk]]) 21:00, 1 December 2009 (UTC) ::Ah, I missed that GeSHI is itself Unicode-clean and it is the no-highlighting path at fault. Fixing the bug is of course preferable to preserving or generating nonstandard markup; I only claim that ''the bug should have been fixed first'', and we should now either fix the bug or have the bot change the markup, whichever will happen sooner, since the affected examples are currently unreadable. ::However, I do feel that your bot did two, ah, boneheaded things: ::* The Unicode problem is known. The bot's table of lang codes should not include languages which use non-ASCII characters and do not yet have definitions, until the bug is fixed. ::* The bot should not discard markup (style=) it does not understand; it should assume that there is some special situation and not make any changes to that code block. ::(Note that even if GeSHI gets TI-89 support there would still need to be a special case (whether in RC/GeSHI or in the markup) to specify the font.) --[[User:Kevin Reid\|Kevin Reid]] 21:59, 1 December 2009 (UTC) :::Short Circuit fixed the bug. —[[User:Underscore\|Underscore]] ([[User talk:Underscore\|Talk]]) 01:24, 2 December 2009 (UTC)