User talk:MerlLinkBot/Archives/2010

yu list
Can you update User:MerlLinkBot/yu please? :) --Joy &#91;shallot&#93; (talk) 11:03, 3 June 2010 (UTC)
 * done. Merlissimo 17:42, 8 June 2010 (UTC)

Links to cbc.ca
Would it be possible for you to correct links to news articles at cbc.ca from a few years ago which have now gone dead. The urls begin with "http://www.cbc.ca/story/*" or "http://cbc.ca/story/*" and they need to be reworked as demonstrated in this edit. Thanks very much if you can do this, because there are many Wikipedia articles about Canadian subjects that use cbc.ca as a source for news coverage. --Mathew5000 (talk) 18:18, 5 August 2010 (UTC)
 * I haven't found any link without www and 1621 links starting with http://www.cbc.ca/story/. But your rewriting schema does not always work.
 * http://www.cbc.ca/story/canada/national/2005/11/28/noconfidencevote051128.html -> http://www.cbc.ca/canada/story/2005/11/28/noconfidencevote051128.html
 * http://www.cbc.ca/story/world/national/2005/02/01/newdarfur-report050201.html -> http://www.cbc.ca/world/story/2005/02/01/newdarfur-report050201.html
 * http://www.cbc.ca/story/arts/national/2005/10/31/Arts/aussielaw_051031.html -> http://www.cbc.ca/arts/story/2005/10/31/Arts/aussielaw_051031.html
 * http://www.cbc.ca/story/science/national/2005/12/15/Wikipedia-review051215.html -> http://www.cbc.ca/science/story/2005/12/15/Wikipedia-review051215.html
 * Also links like http://www.cbc.ca/storyview/* are dead. Do you also know the rewriting schema for the last examples? Otherwise i can only correct the working ones and not all. Merlissimo 22:29, 5 August 2010 (UTC)


 * Taking your last question first, some of the dead urls with "storyview" can be changed to a live link, for example:
 * http://www.cbc.ca/storyview/AOL/arts/national/2006/08/28/hongkong-photos-tabloid.html ->
 * http://www.cbc.ca/arts/story/2006/08/28/hongkong-photos-tabloid.html
 * The other two examples you gave can be reworked this way:
 * http://www.cbc.ca/story/arts/national/2005/10/31/Arts/aussielaw_051031.html ->
 * http://www.cbc.ca/arts/story/2005/10/31/aussielaw_051031.html
 * and
 * http://www.cbc.ca/story/science/national/2005/12/15/Wikipedia-review051215.html ->
 * http://www.cbc.ca/health/story/2005/12/15/Wikipedia-review051215.html
 * this last one is a bit weird because the item got reclassified from "science" to "health". There might also be some items where they reclassified an item from "science" to "technology".
 * Thanks very much for helping with this! Mathew5000 (talk) 23:52, 5 August 2010 (UTC)
 * And http://www.cbc.ca/story/business/national/2005/12/14/trade-051214.html is now http://www.cbc.ca/news/story/2005/12/14/trade-051214.html.
 * I'll start my bot handling all these cases. Later we can have a look which urls are skipped and try to fix them on a second run. Merlissimo 13:11, 6 August 2010 (UTC)


 * Hi there Merlissimo, thanks for doing this work. However, I noticed a problem with the way the bot is updating links. See for an illustration of the issue - the bot updated an old dead link to a working copy, which is good, but it also overwrote a working archiveurl parameter (which happens to contain the original URL as a substring) with a modified URL that may or may not work. Existing archiveurl parameters probably shouldn't be changed, unless the old URL doesn't work, the new one does, and you verify this consistently. Please take a look at this issue, but in the meantime thank you for doing the updates of the main link URls. — Gavia immer (talk) 02:51, 7 August 2010 (UTC)
 * fixed. I had to change to lookbehind two jobs ago because of strange use in templates. But that caused this bug. Thanks. Merlissimo 19:06, 7 August 2010 (UTC)
 * I posted the error log for enwiki here. There still some not working urls which maybe have another rewriting schema:
 * http://www.cbc.ca/story/sports/national/2005/03/13/Sports/brier_final050313.xhtml
 * http://www.cbc.ca/storyview/CBC/2001/02/13/ambassador_010213
 * http://www.cbc.ca/story/news/?/news/2001/12/14/carignan_011214
 * http://www.cbc.ca/story/canada/national/2005/11/07/omarkhadr051107.html&cid=1102222795
 * http://www.cbc.ca/story/canada/national/2005/08/15/Taber_killer_walks_away_from_halfway_house20050815.html


 * http://www.cbc.ca/story/canadavotes2006/national/2006/01/18/elxn-harper-toronto.html
 * Merlissimo 01:38, 8 August 2010 (UTC)


 * The 2005 Omar Khadr article is at http://www.cbc.ca/canada/story/2005/11/07/omarkhadr051107.html. The last one in your list,, seems to be working fine now. The others, I don't know, the CBC might have deleted them. Mathew5000 (talk) 08:27, 9 August 2010 (UTC)

NPS.gov
A word of caution, you may be replacing a dead link with one that is not yet functioning. clariosophic (talk) 00:24, 11 August 2010 (UTC)
 * Do you have an example? I am only replacing the old multiples. While testing i found existing pdfs without the real content like only for nominations. Merlissimo 01:33, 11 August 2010 (UTC)
 * One of your most recent examples is at U.S. Post Office (Northport, New York). DanTD (talk) 01:31, 12 August 2010 (UTC)
 * this one? http://pdfhost.focus.nps.gov/docs/NRHP/Text/64000597.pdf works for me. Merlissimo 02:25, 12 August 2010 (UTC)
 * Yes it works fine now, but for some reason, none of the links that replaced the old NRHP site were working at the time. The feds have always had technical problems with this site. DanTD (talk) 13:13, 12 August 2010 (UTC)
 * Sure because my bot runs a DoS attack by sending one http-head-request every minute to this domain :D. But normally that shouldn't affect any server. Merlissimo 13:29, 12 August 2010 (UTC)

Washington Metro
http://wmata.com/about/met_news/* has become http://wmata.com/about_metro/news/* --Bsherr (talk) 22:10, 15 October 2010 (UTC)
 * I only count four links in Washington Metro rolling stock, Federal Triangle (WMATA station), Navy Yard (WMATA station) and zh:聯邦三角站. It's much faster if you fix these few links manually. Merlissimo 01:51, 16 October 2010 (UTC)
 * Ah, no problem. --Bsherr (talk) 01:58, 16 October 2010 (UTC)

.yu again
Hi, can you update User:MerlLinkBot/yu and can you also make it not actually link those old broken domains, so that it doesn't clutter Special:LinkSearch output? TIA. --Joy &#91;shallot&#93; (talk) 11:01, 11 October 2010 (UTC)


 * I've taken the liberty of editing the old output to unlink those URLs. That's what the new output should do, too. --Joy &#91;shallot&#93; (talk) 08:18, 13 October 2010 (UTC)
 * Sure, but i am not able to update the page before next week (this script is saved on another computer). Merlissimo 22:58, 13 October 2010 (UTC)


 * ? :) --Joy &#91;shallot&#93; (talk) 15:36, 28 November 2010 (UTC)