Difference between revisions of "ISFDB talk:Community Portal"

From ISFDB
Jump to navigation Jump to search
Line 58: Line 58:
  
 
:Thank you, this is much appreciated!  {{Unsigned|Fangles|04:56, 1 June 2008}}
 
:Thank you, this is much appreciated!  {{Unsigned|Fangles|04:56, 1 June 2008}}
 +
 +
Is there any chance that this script or a similar script could be extended to extract title information?  My use case is as follows:  When I get a new book, I scan in the ISBN and run a script that extracts the publication information for my local database of what I own.  I would like to be able to include a series and seriesnum field in this db.
 +
 +
Maybe it could be a different script, accepting the record number of a publication as an index.  Or the <publication> record returned by getpub.cgi could include a title field, with the index of the title information. [[User:Md5i|Md5i]] 01:21, 18 June 2008 (UTC)
  
 
== New WorldCat Tool  ==
 
== New WorldCat Tool  ==

Revision as of 21:21, 17 June 2008

To leave a comment or question click the "+" that's the right of "edit" above. You will be prompted to enter a subject and your message. Please sign the end of your message with four tildes, ~~~~, which will generate a name/date signature.

For old discussions, see the archives


Wikiadmins and password changes

Two of the robots are hitting pages that should not be protected

Does anyone know how to change someone else's wiki-password? Marc Kupper (talk) 18:42, 19 Mar 2008 (CDT)

From times when corrupted accounts showed up at wikipedia, that probably requires direct access to the wiki db. I know an ordinary admin can't do it. -DES Talk 06:02, 20 Mar 2008 (CDT)
Once we have a newer version of media-wiki, we should be able to block the robot accounts, and perhaps semi-protect such pages if need be. (semi-protection prevents an account from editing a page unlesss that account has exsited for a certian minimum period (4 days by default) and has made a minimum number of edits (I think 25 or 50 by default). Both numbers can be configured. -DES Talk 08:57, 20 Mar 2008 (CDT)
According to http://www.mediawiki.org/wiki/Manual:FAQ#How_do_I_reset_a_password.3F there is a script to do this. From what Al has said in the past about such scripts, he can't run them on the TAMU server, but he may be able to extract from the script the SQL code needed to do the same thing directly. -DES Talk 10:40, 20 Mar 2008 (CDT)
The mediawiki site also says:
You can also use the old way, by modifying the database directly. Assuming that $wgPasswordSalt is set to true (the default), you can use the following SQL query for MySQL:

UPDATE user SET user_password = MD5(CONCAT(user_id, '-', 
  MD5('somepass'))) WHERE user_name = 'someuser';

Where obviously "somepass" is changed to the password you want to set and "someuser" is changed to the user name as it is listed in the table "user".
Note: The user_id in the CONCAT string is a column name and is not meant to be replaced with 'someuser'
Note: If you obtain a 'dbname.user table does not exist' error, please check the LocalSettings.php file, and double-check the value for the $wgDBprefix variable. If that variable is not empty, try repeating the command, replacing $wgDBPrefix_user instead of user in the UPDATE clause of the SQL statement.
If using PostGreSQL, use this query instead:

 UPDATE mwuser SET user_password =
   md5(user_id || '-' || md5('somepass')) WHERE user_name='someuser';

You can also try the Extension:Password Reset
I hope this is helpful. -DES Talk 10:51, 20 Mar 2008 (CDT)

Reviews linking to shortfiction

Amongst all the verbiage I did not find a mention of this - in at least one case I found reviews linked to a shortfiction piece rather than a novel of the same name by the same author.--swfritter 22:17, 2 June 2008 (UTC)

I have been entering reviews (for example those in The Issue at Hand that are explicitly of shortfiction works, and linking them appropriately. i don't see any problem with this. Of course, in soem cases, mistakes will happen. -DES Talk 23:20, 2 June 2008 (UTC)
Just something to keep an eye on and perhaps projectize. And there are probably cases where there are reviews of both a short story and a novel with the same title. It also means it's a good idea to check any new reviews we put in.--swfritter 00:17, 3 June 2008 (UTC)

getpub.cgi shows less info than expected.

I've developed a simple book getter using the API but noticed that the getpub.cgi proc returns less data in the query than the example for submission shows.

Notably, the image link is left out (and there may be others that I am not aware of).

Will the image link be returned in the API eventually? I realise the links are quite often old or misleading but they would be useful.

Thanks, Gary.

P.s. I tested the getpub.cgi using the isbn from the submission example (which did have an image link) and getpub returned no image link. —The preceding unsigned comment was added by Fangles (talkcontribs) 01:56, 31 May 2008.

Image and tag info are now returned. Alvonruff 11:52, 31 May 2008 (UTC)
Thank you, this is much appreciated! —The preceding unsigned comment was added by Fangles (talkcontribs) 04:56, 1 June 2008.

Is there any chance that this script or a similar script could be extended to extract title information? My use case is as follows: When I get a new book, I scan in the ISBN and run a script that extracts the publication information for my local database of what I own. I would like to be able to include a series and seriesnum field in this db.

Maybe it could be a different script, accepting the record number of a publication as an index. Or the <publication> record returned by getpub.cgi could include a title field, with the index of the title information. Md5i 01:21, 18 June 2008 (UTC)

New WorldCat Tool

Has anyone else noticed the wonderful tool that Al's added to the menu on the publication listing page? If you click on the WorldCat link, you're taken directly to the WorldCat page for this pub based on its ISBN. Great tool! Thanks, Al. MHHutchins 00:22, 4 June 2008 (UTC)

Oh yes, I was going to add it to "What's New" once we reach 300,000 titles and have another reason to celebrate :) Ahasuerus 03:59, 4 June 2008 (UTC)
Yes, looks good to me too. Ahasuerus, which title TYPES are you counting? As we're at 315,635 titles in general already... BLongley 18:09, 4 June 2008 (UTC)
The ISFDB Statistics page has 299,570 non-variant Title records at the moment. We have about 15,000 variant titles, which presumably accounts for the difference :) Ahasuerus 19:47, 4 June 2008 (UTC)
Ah, OK. I don't find that page useful myself. Well, except that "None: 0" and "CHAPTERBOOK: 5" are good signs of Entropy. BLongley 21:57, 4 June 2008 (UTC)
I'm not interested in "COVERART: 42856" as it seems people are unwilling to merge them and I'm sure we really don't have that many actual artwork differences. "INTERIORART: 23619" - well, that's not Speculative FICTION, is it? Nor are "NONGENRE: 1018", by definition. "ESSAY: 41632" might be useful for SF purposes, but might not be fiction-related, as might "NONFICTION: 3989" - I know some are Bibliographies or Encyclopedias of SF and good things like that. "INTERVIEW: 1174" - good background material I presume, but not Fiction. (OK, I suppose it depends who you interview.) "SERIAL: 2739" - I suspect most of those are duplicated by actual Novels or Novellas at least, and even if they aren't, it probably takes two or three of them to make one proper SF work. "OMNIBUS: 1908" - well, those are just confirmation of proper titles mostly. Same with "REVIEW: 26343". "EDITOR: 2004" just confirms magazines, not contents, and I know we have loads of missing Editor Records. "COLLECTION: 4807" and "ANTHOLOGY: 4703" cover shortfiction (hopefully, although I appreciate they may cover shortfiction we haven't got yet). "POEM: 8985" - not my cup of tea really, but they count for short-fiction. BLongley 21:57, 4 June 2008 (UTC)
So it comes down to "NOVEL: 48423" and "SHORTFICTION: 85364". I keep finding Dups of shortfiction whenever I try to find the Dups I've just created by adding another anthology or collection, so that's probably overstated as well. Still, the figures suggest there's enough SF already created that I can stay happy for the rest of my expected lifespan, even if we don't separate out the Fantasy and Horror I'm unlikely to read, so don't be disheartened with these stats! I think I'll be happier with seeing "edit number 1,000,000" though, rather than "300,000 titles". BLongley 21:57, 4 June 2008 (UTC)

Section creation link

I have added the "+" link to create a new section to the top of ISFDB :Community Portal. This can be done for any page by placing the "magic word" __NEWSECTIONLINK__ on the page, preferably at the top, but anywhere that it will not be erased. -DES Talk 23:59, 17 June 2008 (UTC)

Very nifty, thanks! :) Ahasuerus 01:09, 18 June 2008 (UTC)