Wikipedia:Bots/Requests for approval/FrescoBot 7


 * The following discussion is an archived debate. Please do not modify it. To request review of this BRFA, please start a new section at WT:BRFA. The result of the discussion was Symbol keep vote.svg Approved.

FrescoBot 7
Operator:

Automatic or Manually assisted: auto

Programming language(s): Python

Source code available:

Function overview: Additional fixes for the task Bots/Requests for approval/FrescoBot 2

Links to relevant discussions (where appropriate): Wikipedia:Bot owners' noticeboard#Wikilink simplification

Edit period(s): montly or less

Estimated number of pages affected: 35k?

Exclusion compliant (Y/N): Y

Already has a bot flag (Y/N): Y

Function details: Bots/Requests for approval/FrescoBot 2 already fixes a wide range of wikilink syntax problems and some redundancies. I plan to extend it a bit with: Please note, these fixes will be added to this collection.
 * 1) Simplification of selected piped wikilinks in the "shortened sort-of-piped link". This substitution has to improve the intuitiveness and readability of the wikitext, for this reason I will be pretty conservative and I will convert only few popular suffixes:
 * 2) *suffix "s",  architects --> architects 
 * 3) *suffix "es",  viruses --> viruses 
 * 4) *suffix "n",  Croatiann --> Croatian 
 * 5) *suffix "ns",  Sardinians --> Sardinians 
 * 6) *suffix "an",  European --> European 
 * 7) *suffix "ian",  Egyptian --> Egyptian 
 * 8) *suffix "ic",  logarithmic --> logarithmic 
 * 9) *suffix "ist",  steel guitarist --> steel guitarist </tt>
 * 10) *suffix "ern", <tt> Middle Eastern --> Middle Eastern </tt>
 * 11) *In this way I will avoid to create not-so-intuitive shortcuts, eg:
 * 12) ** I will not convert Mi-171Sh --> Mi-171Sh
 * 13) ** I will not convert Rødbyhavn --> Rødbyhavn
 * 14) ** I will not convert compositions --> compositions
 * 15) ** I will not convert Windows 2.03 --> Windows 2.03
 * 16) ** I will not convert Siddhar --> Siddhar
 * 17) Title linked in text (Check Wikipedia #48). Wikilinks to the current page will be de-linked (example).
 * 18) *<tt> pagename --> pagename</tt>
 * 19) *<tt> label --> label</tt>
 * 20) Additional syntax fixes:
 * 21) * Brahin --> Brahin = Brahin --> Brahin

Discussion
Comments? Is there any exception for #2? Are there other link fixes that I should fix? -- Basilicofresco  (msg) 06:15, 20 September 2010 (UTC) Magioladit asked my help for #1 because FrescoBot is faster than AWB bots. After catching up with the backlong he doesn't expect many links created per week. -- Basilicofresco  (msg) 10:31, 20 September 2010 (UTC)
 * For additional syntax fixes, you can add Brahin --> Brahin since that is a plausible typo. — HELL KNOWZ  ▎TALK 18:19, 22 September 2010 (UTC)
 * Yep, good idea. -- Basilicofresco  (msg) 20:48, 22 September 2010 (UTC)


 * I see little benefit in part 1; those are minor problems that hardly merit their own bot looking after them. They could be integrated into AWB's general fixes, though (perhaps they already are). Ucucha 20:52, 22 September 2010 (UTC)
 * Please, take a look at Bot owners' noticeboard. The main reason for this request is that AWB is not fast enought to reduce the whole number of redundant links. -- Basilicofresco  (msg) 21:14, 22 September 2010 (UTC)
 * Why is this urgent? Ucucha 21:18, 22 September 2010 (UTC)
 * It is not actually urgent, but speed matters. Here are some numbers:
 * within enwiki-20100730 file dump there were 25142 articles in need of #1 replacement;
 * within enwiki-20100916 file dump there are 24156 articles in need of #1 replacement.
 * It means -4% in about 6 weeks. How many years will take to AWB users to reduce the number of these articles by 50%? -- Basilicofresco  (msg) 16:51, 23 September 2010 (UTC)
 * And in that time SmackBot was AWBing about 3-600 articles a day, with Smack sleeping I would bet the number is actually going up. Rich Farmbrough, 16:17, 3 October 2010 (UTC).


 * So what? If this were a major problem, I could see your point, but I don't see how it is one. Ucucha 17:00, 23 September 2010 (UTC)
 * I do not think that this must be a major problem just to warrant a bot operation. There are many small tedious fixes that can be performed as well. Besides, these additions build up with previous tasks and eventually we may have a bot that does a long list of corrections. I have nothing against small steps when they lead to overall improvement. — HELL KNOWZ  ▎TALK 18:02, 23 September 2010 (UTC)
 * This task is just an enhancement of what FrescoBot is already doing. foo is already fixed by FrescoBot and it's an approved task. I agree that FrescoBot could do much other tasks at the same time but at some point we have to find a balance of what is "too much" and what is "too few". -- Magioladitis (talk) 20:20, 23 September 2010 (UTC)
 * For me code cleanup is an important matter. Many other errors (unbalanced brackets, etc.) are caused because sometimes wikicode isn't easy to check with naked eye. -- Magioladitis (talk) 22:45, 23 September 2010 (UTC)


 * AWB does already do some or most of it. If FrescoBot was going to do clean up run specifically on the non-broken but improvable links, then I would say use Awb with genfixes, tagging and unicodifying. For broken tags, approval for a trial run is a no-brainer. Rich Farmbrough, 16:17, 3 October 2010 (UTC).


 * Yes, AWB does most of it. The good thing with FrescoBot is that is faster and it already has approval to do most of the job already. I thought it would be a speedy approval but it takes some time. -- Magioladitis (talk) 00:32, 4 October 2010 (UTC)

So if there are no objections we can go for a non-BAG close? :) -- Magioladitis (talk) 08:26, 13 October 2010 (UTC)


 *  MBisanz  talk 23:47, 14 October 2010 (UTC)
 * The above discussion is preserved as an archive of the debate. Please do not modify it. To request review of this BRFA, please start a new section at WT:BRFA.