User talk:Alvonruff/Archive02

From ISFDB

Jump to: navigation, search

Contents

Austin

Al, are you likely to move to Austin, as a recent post of yours implies? If so, let me know, as that's my home town -- it would be good to put a face to the name. Mike Christie (talk) 07:02, 26 Jun 2007 (CDT)

I'll be down in a couple of weeks for house hunting, and should be moved in (northwest or southwest side) by the end of August. I imagine a meeting for all ISFDB moderators living in the Austin area will be in order... Alvonruff 13:04, 26 Jun 2007 (CDT)
Enjoy Austin - it's a nice little town. Or rather it used to be a nice little town :) There is a decent specfiction/mystery bookstore in the bar district and the owner, Willie, is a knowledgeable fan. Last I checked, they had quite a few British books, which may save you a few bucks on shipping :) Ahasuerus 14:23, 26 Jun 2007 (CDT)
Sadly, that closed a few years ago, though as you can tell the website is still running. The lease ran out and the owner wouldn't renew; I guess that part of town was too valuable to lease to a second-hand book store. Mike Christie (talk) 17:25, 26 Jun 2007 (CDT)
We're losing a LOT of second-hand book stores. :-( There used to be 3 or 4 within walking distance in town when I moved here. Nowadays finding 3 or 4 means visiting 3 or 4 towns within driving distance. BLongley 17:48, 26 Jun 2007 (CDT)
I haven't been to Austin since the last millennium, but even then the store looked a bit out of place in that part of town. Which is too bad, it was a nice place :( Ahasuerus 17:51, 26 Jun 2007 (CDT)
Maybe we should be promoting them a bit more here? It doesn't really fit in with the bibliographical stuff, but I'd like to recommend the few remaining ones near me rather than have them turn into Starbucks or suchlike. Says the guy that has over 300 pub images on Amazon now, none of which say you should support your local dealer instead... :-(
Actually, just saying where "here" is might be a start. Anyone feel like creating/pointing me at a http://www.frappr.com/ map for good bookshops? BLongley 18:07, 26 Jun 2007 (CDT)
Well, there is Evelyn Leeper's list. Overall, I am very happy with the impact that online commerce has had on the book selling business. Many strange and wonderful things have come out of it, e.g. Harry Stephen Keeler's resurrection (it doesn't get much stranger than that!), but on the other hand some brick and mortar bookstores had/have tons of personality. Oh well, I guess you can't have everything... Ahasuerus 18:21, 26 Jun 2007 (CDT)

Invalid submission with invalid character in page #

Submission record 686357 is

  • Causing the following to appear at the top of http://www.isfdb.org/cgi-bin/mod/list.cgi?N - parse error on: 23758 Hall3730 New Spring: The Novel 2004-01-00 334 New Spring: The Novel listed as title on dust jacket, title page, & copyright page. Darrell K. Sweet 29831 New Spring: The Novel New Spring: The Novel Matthew C, Nielsen+Ellisa Mitchell 2004-01-00 INTERIORART New Spring: The Novel endpaper maps Ellisa Mitchell 2004-01-00 fep & bep INTERIORART
  • The record is displayed in the detail list as 686357 N PubUpdate 2007-07-09 21:50:14 unknown unknown
  • There is a Python error when you click on the link to approve/reject. This seems to be cause because the page # for one of the items is "<cPage>fep & bep</cPage>"

I suspect the blob can be fixed to change the & into an &amp; Marc Kupper (talk) 23:09, 9 Jul 2007 (CDT)

Entry is fixed and integrated. Haven't fixed the code that allowed it yet. Alvonruff 08:48, 11 Aug 2007 (CDT)

Proposed policy changes re: e-publications and XML submissions

When you get a chance, could you please take a look at the current discussion page at ISFDB talk:Policy? Aside from the issue of including stable e-publications in the database, there is some talk about XML submissions that you would be uniquely qualified to comment on :) Ahasuerus 12:48, 11 Jul 2007 (CDT)

Please DO consider those as two separate issues: the "stable e-publications" issue seems to have a lot of support, in this one example. I'm afraid I raised the XML submission idea as well, and realise that a code change is far more difficult here than a policy change, (which merely requires cat-herding abilities, rather than teaching the loose cats XML-skills as well). BLongley 15:43, 11 Jul 2007 (CDT)

The Postman

Al, you verified this edition of Brin's The Postman back in November. Looking at this Title's history and especially its "price evolution" ($3.95 -> $4.95 -> $5.95), it would appear that your copy may be a later printing from ca. 1994, possibly the tenth printing?. Could you please double check the number line when you get a chance? Thanks! Ahasuerus 11:54, 26 Jul 2007 (CDT)

9th printing. No date clues on the page. Alvonruff 08:19, 11 Aug 2007 (CDT)
Thanks, fixed. (And welcome back!) Ahasuerus 18:38, 11 Aug 2007 (CDT)
Good to be back. Should be able to contribute for about 2 weeks - then I'll be moving into the new house in Austin, and won't have Internet connectivity for a few days. Alvonruff 19:39, 11 Aug 2007 (CDT)
To paraphrase Forry Ackerman, two weeks are better than none! :) Good luck with the Austin move, it's a long drive if you decide to do it yourself.
AFAIK, there is only one "active" Python bug at the moment, but there are probably others hovering in the background. There has been a fair amount of talk about moving to another host because of intermittent downtimes, slowdowns, lack of anti-spam support, etc. I'd be happy to provide the funding, but obviously nothing that major can happen without your input. Other than that, certain parts of the application could use a face lift, but there are no show stoppers. Ahasuerus 20:55, 11 Aug 2007 (CDT)

Barrayar

Added a cover image to Barrayar, is this the correct one for your edition? Dana Carson 04:50, 19 Sep 2007 (CDT)

Yep. Alvonruff 08:16, 11 Aug 2007 (CDT)

MediaWiki update

You wrote at What's New #What's New - 1 July 2007 that you tried to update to 1.10 but that turned out to be impossible as it requires PHP 5 while the hosting runs on 4. Judging by the date, any change of that doesn't seem forthcoming. So, could you at least upgrade to 1.6 which according to http://www.mediawiki.org/wiki/Important_Release_Notes is the last PHP4-compliant version? AFAICR it's a pretty decent version already (frex it has template parameter defaults), it might even have finer-tuned spammer-blocking options, and anyway pretty much anything would be improvement compared to the current mess. Thanks, --JVjr 06:18, 13 Aug 2007 (CDT)

OK - sounds like a good idea. I'll check it out over the next few days. Alvonruff 12:12, 13 Aug 2007 (CDT)

I see you've also mentioned it at ISFDB:Moderator noticeboard #WebBot Attack. As for the database transferability issues discussed there, http://www.mediawiki.org/wiki/Release_notes/1.6#Upgrading warns "If you are upgrading from MediaWiki 1.4.x or earlier, some major database changes are made, and there is a slightly higher chance that things could break." But it would be nice if you could use the upgrade to scrub the DB of all spammer accounts (and ideally also all their edits - together with all immediately following edits with summaries including "spam"? but that is getting too complicated to the possibility of breaking someting, I guess). --JVjr 07:46, 14 Aug 2007 (CDT)

Hmmmm.... Thinking out of the box a bit: although upgrading to the latest MediaWiki is desirable, what we really want is spam control. The MediaWiki database is already intermixed with the ISFDB database, so there's nothing that prevents us from making a standalone spam deletion tool, which given a particular user, deletes the user and everything they've ever done. That'll require some excellent understanding of the MediaWiki schema. Alvonruff 14:36, 14 Aug 2007 (CDT)

There might already be an extension written and usable for similar purpose, but I know nothing of such things and a brief browse through http://www.mediawiki.org/wiki/Category:Extensions discovered only some forms of spam blacklisting, restricting of access to pages for finer-tuned user groups, and approving creation of new accounts; still, even some of that might be useful. And there's http://www.mediawiki.org/wiki/Extension:Deleteuser but it seems to delete just the account and not zir edits. --JVjr 06:17, 15 Aug 2007 (CDT)


Any progress? Also, it might be a good idea to update the links to the wiki from ISFDB which have tamu.edu domain hardcoded, even if the current failure is fixed soon. --JVjr 08:26, 29 Nov 2007 (CST)

I was thinking along similar lines the other day. The inability to access the Wiki and especially the Help pages seems to be having a considerable impact on new editors, who figure that they are simply unavailable. I added a note to the Main Page, but I am not sure how helpful it will be. And welcome back, Jan! :) Ahasuerus 11:35, 29 Nov 2007 (CST)

Bad submission

Al, could you please blow away this bad submission when you get a chance? TIA! Ahasuerus 01:28, 19 Aug 2007 (CDT)

Done. Alvonruff 08:53, 20 Aug 2007 (CDT)
Thanks! :) Ahasuerus 18:46, 20 Aug 2007 (CDT)

Witch Bank

Al, could you remove this submission [1]. Thanks :-)Kraang 22:24, 27 Aug 2007 (CDT) And another one "The Unteleported Man / The Mind Monster"[2], this time one of mine. :-( Thanks!Kraang 21:55, 7 Sep 2007 (CDT)

Ok. I *finally* have Internet access at the new house, so I'll start to look at these. Alvonruff 19:40, 18 Sep 2007 (CDT)
And there was much rejoicing!! :-) Ahasuerus 19:59, 18 Sep 2007 (CDT)
I have Internet access (satellite), but no furniture (not until next week), so there's only so much time I can spend sitting on the floor hunched over this tiny Fujitsu laptop. Funny, the idea of beaming packets up to an orbital platform from a laptop on a planet's surface was really cool in the movie Aliens, but now that it's really posible... Alvonruff 20:31, 18 Sep 2007 (CDT)
Commuting between the two coasts sounded quite exciting back in the day as well... Ahasuerus 22:26, 18 Sep 2007 (CDT)
P.S. Do you think you could upload the latest backup file when you get a chance? Now that more of us are writing data cleanup/analysis scripts, there appears to be more demand for fresh juicy data :) Ahasuerus 23:47, 18 Sep 2007 (CDT)

A problem with Notes

Al, we may have found either a case of minor database corruption or a potential software problem -- see thsi discussion for details. When you get a chance, could you please remove the suspicious single quotes from the record and see if it may solve the problem? Thanks! Ahasuerus 23:50, 10 Sep 2007 (CDT)

This is EditBug:10103 Title notes-id set to 0. I've already fixed the affected record. Marc Kupper (talk) 02:16, 11 Sep 2007 (CDT)


Amazon UK link

I THINK I asked for this before, but can't remember where, but it should be a quick change I think. Can you please add "£" as well as "L" to the price-prefix logic used when determining whether to show an Amazon link next to a pub with a good ISBN? A lot of editors are now using the correct symbol, but this actually breaks some fairly useful functionality. BLongley 16:33, 16 Sep 2007 (CDT)

I'll check for/make a proper feature request at some point, but the full request will probably ask for Euro-support, and maybe add Amazon.DE or .FR or suchlike. But I only own one German title (bought by accident) so I'm probably not well-equipped with examples for those, but I have hundreds of British ones that could do with the quick "£/L" fix, please? Pretty please? BLongley 16:33, 16 Sep 2007 (CDT)

Database degradation?

The Publication record linked from Silverberg's Between Worlds has an empty tag, which makes the application rather unhappy. Ahasuerus 21:56, 20 Sep 2007 (CDT)

I fixed that one (sorry if I destroyed any evidence in the process, but I described all I did to it.) I went and checked for similar problems in the last backup and they all seem to be Doctor Who ones, e.g. this REALLY vacant one. I know I fixed some problems with non-unique tags for Doctor Who pubs (see here) but if I've actually CAUSED any, please DO let me know. BLongley 15:12, 21 Sep 2007 (CDT)

Duff record to zap

Sorry, forgot a mid-sequence approval and left this to be zapped when you get a chance. BLongley 14:04, 28 Sep 2007 (CDT)

Carrion Comfort

I see that you verified the Warner paperback edition of Carrion Comfort in October 2006. Is it really 1989-10-00 or is it 1990-10-00 as per the Locus Index and the copyright page of my fourth printing? Also, when you get back, could you please post the latest ISFDB backup? There are a couple of cleanup projects that are currently in progress and that could benefit from a fresh backup. TIA! Ahasuerus 12:59, 7 Oct 2007 (CDT)

Carrion Comfort: Stuck away in a box, stacked with all my other books (except for the reference books, which I did find), so I can't answer that one anytime soon. Backup: I'll try to put one up tomorrow morning. Alvonruff 10:24, 12 Oct 2007 (CDT)
Welcome back and no hurry on the Simmons! Ahasuerus 10:48, 12 Oct 2007 (CDT)
It was indeed 1990-10-00. Changed. Alvonruff 01:53, 25 May 2008 (UTC)
Glad to see that you seem to have defied the "three moves = one fire" rule! :) Ahasuerus 02:18, 25 May 2008 (UTC)

October 2007 Backups

Do you know what your schedule will look like for the rest of the month yet? The last few months worth of discussions have resulted in an informal list of priorities and if you expect to be available the next few weeks, I can write them down. Ahasuerus 11:51, 12 Oct 2007 (CDT)

I'll probably be on a normal schedule (whatever that means) in about 2 weeks. Also no backup this week - tried the download twice, and it failed both times - after eating almost half of my satellite bandwidth budget for the week (the raw backup file is about 320 Meg compressed). Alvonruff 07:29, 13 Oct 2007 (CDT)
320Mb? The final uncompressed file that I have for August 11 is only 77Mb, so either the raw backup file is much larger than the final backup file or the database has grown tremendously in the last 2 months. I wonder if it's all the new versions of various Wiki pages that were attacked during the last vandal assault? Ahasuerus 11:21, 13 Oct 2007 (CDT)
The version I put online is post-processed with the removal of all wiki tables, so it's considerably smaller. Alvonruff 11:47, 13 Oct 2007 (CDT)
Oh, I see! I guess I looked at the contents of the mw_cur table and figure it must have included all of our Wiki stuff. I should have realized that we have much more than 10Mb worth of discussions by now :)
I wonder if your satellite connection is capable of supporting large downloads on a consistent basis? If not, we will need some other way of accessing the data, both for backup purposes and to facilitate data cleanup. Is it something that I may be able to help with? I have a bunch of random tools like Perl, Python and gcc that I have installed on this laptop over the last year to support data mining. I guess I could always download Cygwin and anything else that may be needed.Ahasuerus 15:30, 13 Oct 2007 (CDT)

Change to Analog, January 1965

I changed the pub format (binding) on Analog, January 1965 from "large" to "bedsheet". You are listed as verifying this issue. Most of these issues in the March 1963-March 1965 range said "bedsheet"; I changed the ones that said "large" to match. If I'm missing something, let me know & I'll change them back. Thanks. -- Dave davecat 15:20, 27 Oct 2007 (CDT)

Forever Shores

Al could you have a look at the Forever Shores submission. There is an error in the record # and I can't reject or approve it. Thanks :-)Kraang 09:52, 18 Nov 2007 (CST)

Dissembler and Hugh Holton

Checking our forthcoming books, I see that Dissembler has imported Hugh Holton's latest (and probably final, since he has been dead for years) mystery, Criminal Element. I wonder if Dissembler is currently set to grab anything by TOR, which seems to be branching out into non-SF areas? Ahasuerus 12:51, 29 Nov 2007 (CST)

Nope, not set to pull in everything by Tor. The subject annotation at some sites can be peculiar (like a certain amount of gay fiction that show up under the subject heading of Fantasy), so dissembler has some heuristics to catch those (typically by looking at the publisher name). But if it's from Tor (which does publish a ton of SF), and somebody say's it's SF, and there's no picture or book information as a last check, then it's going to slip through. Overall the false positive rate is pretty low now, I just need to find more sources for subjects and publication types. Alvonruff 12:40, 2 Dec 2007 (CST)
Yes, I can see how that would be a problem. Oh well, as long as it's a manageable number, a little human supervision will go a long way. I have deleted the offending pub/title for now.

Things to Fix in December

BTW, how is your ISFDB time looking through the end of the year? Should we try to prioritize the bug list or will your time be so limited that we should wait until next year? Ahasuerus 12:40, 3 Dec 2007 (CST)

I have the last two weeks in December off, so I should be able to look at a few bugs (been working on Forthcoming Books and catching up on awards the last couple of weeks). Alvonruff 13:25, 3 Dec 2007 (CST)
Sounds good! Let me post on the Community Portal and see if we can agree on a list of priorities. I suspect that accidental changes to Title records in magazine/collection/anthology Contents is the biggest thorn in our editors' side: once a mistake slips past the moderators, it can (and has been known to) mess up multiple Verified publications. We have been discussing various ways to address the problem, some of them seemingly easy to implement. Ahasuerus 00:44, 4 Dec 2007 (CST)
P.S. How difficult would it be to add full support for ISBN-13s? More and more books are now published in that format and it's getting to be more of a chore to manually convert ISBN-13s back to the ISBN-10 format. Ahasuerus 12:18, 4 Dec 2007 (CST)
There is a lively discussion of what we would like to see fixed/changed first over on the Community Portal. Ahasuerus 11:21, 13 Dec 2007 (CST)
I've done some completely inadequate testing of ISBN-13 (including submitting an ISBN13 value instead of an ISBN10 value) and it seems to work for me. What features are failing? Alvonruff 06:35, 19 Dec 2007 (CST)
I have entered a few new pubs with 13 digit ISBNs and changed the ISBNs in a couple of pre-existing 2007 pubs ones from 10 to 13 digits and everything looks good. In the past, when you entered a 13 digit ISBN, it would show up in the "ISBN-10" field instead and the "ISBN-13" field would have a "bad number" message displayed in red. I guess the problem was fixed a while back, but nobody noticed :)
The only (and rather minor) outstanding issue is that when a 13 digit ISBN is entered, the 10 digit version of the ISBN is no longer displayed. It may be useful to have the 10 digit ISBN displayed as well since some sources, including OCLC, show both ISBNs for each pub. Of course, at some point in the future it will become impossible to unambiguously convert a 13 digit ISBN to a 10 digit representation, but I don't think we are there yet. Thanks for all the fixes! :-) Ahasuerus 14:50, 19 Dec 2007 (CST)
I noticed one other oddity: this pub displays the 10-digit ISBN only. I believe this is because the ISBN is actually stored with spaces in rather than dashes. (The digits are all correct though.)BLongley 12:51, 23 Dec 2007 (CST)
It was the spaces. Fixed. Alvonruff 13:00, 23 Dec 2007 (CST)
By fixing the spaces or fixing the display code? If the latter, I won't bother hunting the rest. BLongley 13:14, 23 Dec 2007 (CST)


The isbn13 converter now strips out spaces before starting. Alvonruff 14:39, 23 Dec 2007 (CST)

Astonishing

Yes it's Astonishing that Al is actually in editing mode. But more to the point - the page counts on the Astonishing's you added. Is that including the covers? The Wiki page for Astonishing is actually a little misleading as it actually lists a 112 page count. The one issue that I own (February 1943) has 116 including the cover pages.--swfritter 19:29, 18 Dec 2007 (CST)

I'm getting the data from a secondary source, but it does show Astonishing had 112 pages through the April 1941 issue, then expanded to 116 pages in the September 1941 issue (where it stayed until its demise). So I presume that the count includes the covers, but am unable to verify that. Alvonruff 06:32, 19 Dec 2007 (CST)
If it's Contento then I think I am starting to understand his notations. The Contento/Locus CD database lists the page length as 112p+ through the April 1941 issue and 116p from there on. Some further investigation would probably reveal that Contento uses the first convention with the + when the front cover is not page 1 and the second convention when it is. Day seems to always use the inside page count and he lists all issues with 112 pages. Also, according to Gorgon776 the page front page was not used in the page count for the August 1940 issue. According to my 30 year old notations in Day I used to own seven issues of Astonishing but can now only find one of them. Guess it's time for a major rummage through the garage - where could they have gone during the seven moves I have made during that time?--swfritter 14:48, 19 Dec 2007 (CST)
They say "three moves equals one fire", so you might want to check the chimney? If they're not there, leave a note to Santa to bring some! ;-) BLongley 14:54, 19 Dec 2007 (CST)
Oops! I just sold my wood stove with the pipes.--swfritter 14:57, 19 Dec 2007 (CST)
I have a complete set of Astonishing in my collection, at least according to my catalog. Of course, in my capacity as the current incarnation of the Wandering Jew, I move dozens of time a year and my collection has been sadly neglected for well over a decade. Still, I can make an effort to find and verify my Astonishings when I am back at my base of operations between 2007-12-29 and 2008-01-06. Ahasuerus 15:10, 19 Dec 2007 (CST)

Page count for audio-books

I assumed the bibliographic warning would go away when I added a binding type here, but it seems even audio books need page counts? Or did I use an unapproved binding type? BLongley 11:12, 23 Dec 2007 (CST)

I'm open to suggestions on this one. On the one hand, an empty field implies a "Don't Know" situation, which isn't the case here. On the other hand, I could see an empty field meaning "not applicable", which would be the case here. The empty page count field warning is useful, but not very meaningful in the audio book case - but we haven't standardized categories for audio books, so there is no guaranteed string that the warning code can look for to suppress an empty page count warning. I put '0' in the field earlier today, which is essentially true, but we should decide what the official stance should be. Alvonruff 12:54, 23 Dec 2007 (CST)
Good workaround. I've been changing "unk" to "audio CD" or "audio cassette" in these cases (occasionally audio MP3 CD) but leaving page-count blank. It might be useful to use the field to store the number of CDs/Cassettes involved (if known) as that can show the difference between abridged and unabridged works. I'm open to suggestions too though - post on Rules and Standards? (Although we didn't get much response when Ahasuerus listed the binding problems before.) BLongley 13:02, 23 Dec 2007 (CST)

Author Directory problem

Kraang pointed out a problem with some of the Authors with distinguishing suffixes, e.g. "(US)" and "UK". I submitted a load of fixing updates, but they fail at the approval: e.g.

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/mod/aa_update.cgi
  130                         UpdateColumn(merge, 'Lastname',   'author_lastname',   Record)
  131                         if oldlastname[0:2] != newlastname[0:2]:
  132                                 update_directory(oldlastname[0:2])
  133                                 update_directory(newlastname[0:2])
  134 
update_directory = <function update_directory>, oldlastname = '(I)'
 /www/isfdb.tamu.edu/root/cgi-bin/mod/common.py in update_directory(section='(I')
  130         result = db.store_result()
  131         record = result.fetch_row()
  132         bitmap = record[0][0]
  133         print " Old bitmap: %08x" % bitmap
  134         bitmask = 1
bitmap undefined, record = ()

IndexError: tuple index out of range
      args = ('tuple index out of range',)

It seems a "(" in the first two characters isn't catered for? BLongley 13:15, 24 Dec 2007 (CST)

Try it now. Alvonruff 18:05, 24 Dec 2007 (CST)
Still not working, although I see some of the data has been updated behind the scenes and I can clear the queue a bit. BLongley 19:08, 24 Dec 2007 (CST)
OK, queue cleared and I can't think of any more suspicious areas to check at the moment. however, what do we do about:
Amelia G
Laurie O
Zeke K
Billy T
Doctor X
Ex-Private X
 ? BLongley 10:31, 25 Dec 2007 (CST)
Just thought of a few more problems: O'Neill, O'Brien, D'Arcy, etc... Maybe add an extra cell for apostrophe as second letter? BLongley 11:19, 25 Dec 2007 (CST)
I updated the cell matching to strip out period, comma, quote, double quote, left paren, right paren, and space. So O'Neill maps to ONeill. Lastname (US) then maps to LastnameUS, which is the desired effect. Haven't figured out the above examples yet. Alvonruff 17:08, 25 Dec 2007 (CST)
For authors with single-letter lastnames, append the letter 'a' to the lastname field. So the lastname for "Amelia G" becomes "Ga" instead of "G". The lastname is used only to sort the directory, and the directory pulls authors out by lastname (which would include "Amelia G" in the Ga section), sorts by lastname then canonical name (which would put "Amelia G" first), but prints the canonical name (so no one ever sees the "Ga", unless editing the field). Alvonruff 17:20, 25 Dec 2007 (CST)

Credits for data entry on wiki pages

Ahasuerus suggested that I bring this issue to your attention: A number of the wiki pages have names and email addresses of the editors who originally entered data. Would it be a good idea to give them credit for "initial data entry" and remove their email addresses? As Ahasuerus suggested most of these email addresses are probably defunct and in any case the former editors are not familiar with current issues and standards.--swfritter 12:14, 28 Dec 2007 (CST)

That would be fine. The email address came from the original ISFDB html pages, which were create in the mid 90's when no one cared about putting email addresses on a web page. (Of course, spam is now so pervasive that it doesn't matter whether you put your email address on a web page or not). Alvonruff 21:38, 28 Dec 2007 (CST)


False alert about Primary verification?

This submission gave me this warning:

Column 	Current [Record #35909] 	Proposed Changes
Title 	The Best Science Fiction of the Year, No 6 	-
Authors 	Terry Carr 	-
Tag 	THBSTSCNCF1977 	-
Year 	1977-00-00 	1977-07-00
Publisher 	Ballantine 	Ballantine Del Rey
Pages 	- 	388
Binding 	tp 	pb
PubType 	ANTHOLOGY 	-
Isbn 	0345257588 	-
Price 	$1.95 	-
Artists 	- 	-
Image 	0345257588.01.LZZZZZZZ.jpg 	
Note 	- 	1st printing.

WARNING: This publication has been verified against the primary source.

Submitted by: Don Erikson 

But I can't see it as being verified in ANY way. Is this an area you're working on at present? BLongley 14:18, 29 Dec 2007 (CST)

I have seen this before. I think that somebody may have unverified it but there is still a verification record linked to it.--swfritter 14:53, 29 Dec 2007 (CST)
Possible - there's a CoachPaul(?) verification record for it in the backup, ver_status 0. BLongley 15:30, 29 Dec 2007 (CST)
This is a problem when someone verify's a pub and then retracts it. Should be fixed now. Alvonruff 08:30, 30 Dec 2007 (CST)
Now this one gave a false reading, presumably because it's Tuck-verified? BLongley 13:45, 10 Jan 2008 (CST)

Authors of reviewed books get clobbered when the reviewer uses a pseudonym?

There is something wrong with McAuley's reviews in Interzone, August 2001. Paul J. McAuley reviewed 5 books as "Paul McAuley" and the authors of all 5 books are missing. The same pattern is evident in Interzone, June 2001.

Also, pseudonymous reviews can't be easily edited (e.g. to correct typos) since you can't edit Review records directly and the publication record that points to the Review record will only let you edit the pseudonymous record, not the canonical one. The only way that I know how to fix typos in pseudonymous reviews is by removing/deleting them and then re-adding the data. Is there an easier way or will it have to wait until the software is beefed up? Ahasuerus 16:39, 30 Dec 2007 (CST)

Editing reviews and interviews was already next on my list. Maybe this weekend. Alvonruff 15:57, 2 Jan 2008 (CST)
Great!! Ahasuerus 16:03, 2 Jan 2008 (CST)
First pass at editing reviews and interviews is now online. Post problems with them here on my talk page. Alvonruff 14:24, 5 Jan 2008 (CST)
Looks pretty good; we can now add Synopsis data to Reviews/Interviews, which is quite useful. One thing that I have noticed is that something appears to be broken with Notes, although I have no way of telling if it has anything to do with the Reviews/Interviews changes. When I try adding a Note to a title with a variant title, even something as simple as "Test Note", it doesn't take. Here are the SQL updates that are applied:
<code>
SQL Updates:

    * update notes set note_note='Testing again.' where note_id=0
    * update submissions set sub_state='I' where sub_id='909421'
    * update submissions set sub_reviewer='3' where sub_id='909421'
    * update submissions set sub_reviewed=NOW() where sub_id='909421' 
</code>
I assume the problem is with note_id getting set to 0. Could you please take a look at it before it gets out of hand? TIA! Ahasuerus 19:32, 5 Jan 2008 (CST)
Well, I've bullet-proofed the SQL query for the case when note_id gets set to zero, but haven't found why it would be that way in the first place (deleted notes have the note_id set to NULL). This was peculiar to this particular title; the general case of editing a variant (or title with a variant) worked. Alvonruff 05:21, 6 Jan 2008 (CST)
P.S. And I see that you have verified Dozois' latest brick. Is "Julian: A Christmas Story" really a novel or is it a novella as per the page count and the Hugo folks?
The issue was that it had been published first as a standalone "novel". It's been properly made a novella. Alvonruff 05:21, 6 Jan 2008 (CST)
P.P.S. I don't know how much of a Heinlein completist you are, but I generally advise against reading Glory Road until one is done with at least the top 50% of RAH's juveniles, his Hugo winners and the Future History stories. It's not bad, but it's not great either. Admittedly, it was nominated for the Hugo award in 1964, but it was a fairly weak year. Ahasuerus 19:43, 5 Jan 2008 (CST)
I just finished Hamilton's Pandora's Star / Judas Unchained, which amounted to about 1600 pages, and couldn't quite bring myself to dive into the Night's Dawn Trilogy (the first book is almost 800 pages alone), and needed a light sorbet to cleanse the palette. I was at a used book store looking for the second book in the Hamilton trilogy, and came across the Heinlein in a 1969 paperback. Since the list price was 75 cents, and the store is called Half-Priced Books, it cost me all of 37 cents. I'd seen Glory Road in some Best Of lists recently (probably listed by someone who read it when they were 12), so I was suckered into giving it a go. I intend to finish it even though I find it highly annoying. Then I'll probably return to the comfort and warmth of space opera. Alvonruff 05:21, 6 Jan 2008 (CST)
Speaking of Heinlein and other popular authors, at this point we have no way of linking books and articles about Authors to their Summary pages. I wonder if we could creatively abuse our Review records to help us out? For example, if there is no book title entered, then the software will assume that the review covers the author's work as a whole. Ahasuerus 19:57, 7 Jan 2008 (CST)
Going back to the McAuley reviews - it seems we can fill in the blank authors on the variant titles now, but do we have to do these one-by-one or is there a fix script coming that will do those anyway? BLongley 19:46, 5 Jan 2008 (CST)
Do we know how widespread the issue is? Alvonruff 05:21, 6 Jan 2008 (CST)
Not really - I can see only 16 McAuley reviews that might have problems, but if it's spread to the Frederick Patten's or suchlike then there could be far more. BLongley 09:06, 6 Jan 2008 (CST)


Strange data corruption in submissions

When I try adding "Dead Girls Don't Dance" by MaryJanice Davidson to Cravings, the submission reads "Dead s Don't Dance":

<code>
<?xml version="1.0" encoding="iso-8859-1" ?>
<IsfdbSubmission>
<PubUpdate>
<Record>253036</Record>
<Submitter>Ahasuerus</Submitter>
<Subject>Cravings</Subject>
<Content>
<ContentTitle>
<cTitle>Dead s Don’t Dance</cTitle>
<cAuthors>MaryJanice Davidson</cAuthors>
<cDate>2004-06-00</cDate>
<cType>SHORTFICTION</cType>
<cLength>nv</cLength>
</ContentTitle>
</Content>
</PubUpdate>
</IsfdbSubmission>
</code>

I have tried a few permutations of this submission and got the same result. The latest submission is currently on hold. Any ideas? Ahasuerus 10:39, 2 Jan 2008 (CST)

Maybe the system is merely politically correct. Have you tried "Dead Women Don't Dance"? :-) Mhhutchins 10:48, 2 Jan 2008 (CST)
Seems to work for me. Are you typing the submission in or copying and pasting from somewhere? BLongley 12:21, 2 Jan 2008 (CST)
The original title was cut-and-pasted from Amazon.com. When I noticed that it didn't get filed correctly, I removed the title from the publication, deleted it from the database and then re-entered it manually. However, I was using a public PC with babysitting software installed on it this morning, so let me try it again from a different one... Ahasuerus 12:33, 2 Jan 2008 (CST)
Yup, that was it! Michael was right, that public computer was apparently removing the word "girl" from submitted forms, presumably to "protect children". I should go back and see what else it may have corrupted in my submissions this morning. Ahasuerus 12:39, 2 Jan 2008 (CST)
I was honestly just trying to be funny. I had NO IDEA that something like this really existed! Mhhutchins 12:47, 2 Jan 2008 (CST)
There is a variety of "child protection" software suites available these days, some of them quite sophisticated. I didn't check what the software on that PC was doing, but I know that it was installed, so it's Occam's Razor time! (The child safe version, of course.) Ahasuerus 13:17, 2 Jan 2008 (CST)
There is probably a lot of artwork from the pulps that would also be rejected.--swfritter 14:44, 2 Jan 2008 (CST)

"New Worlds" Locus award

We don't have a publicly available award editor yet and I don't know how easy it would be to fix manually, but our New Worlds (1997) anthology record currently has two Locus Poll Awards pointing to it, one for 1992 and one for 1998. The latter is correct according to Locus, but the former should point to the New Worlds 1 record according to the same Locus Index. Ahasuerus 08:46, 4 Jan 2008 (CST)

Fixed. Alvonruff 11:24, 10 Feb 2008 (CST)


"I Remember Lemuria!"'s Retro Hugo

When you get a chance, could you please zap the duplicate Retro Hugo nomination for "I Remember Lemuria!"? A single nomination is bad enough, even if it was made in jest, but two of them is just too much! Ahasuerus 12:04, 6 Jan 2008 (CST)

Fixed. Alvonruff 11:24, 10 Feb 2008 (CST)
Thanks! Ahasuerus 13:10, 10 Feb 2008 (CST)

Analogs in 1966

Working backward through my collection, I reached August 1966. I see that this, & at least a couple of the issues preceding it, were verified by you about a year ago, & so I thought I'd better check with you before making changes. Here are changes I'd be automatically making if it weren't verified:

  • Science Fact entries: remove "Science Fact: " from the titles, set the series to Science Fact (Analog). (I'm afraid that I submitted such a change to the first one before I noticed that the pub was verified. Of course, someone may reject that submission seeing that the pub is verified.)
  • (Also, in this case, I'd change the second Science Fact article's title from "The Psychoceramic" to just "Psychoceramic. The former is how it's listed in table of contents, the latter how it appears in the article itself. I'd expect to put in a pub note about this.)
  • In Times to Come, The Analytical Laboratory, & Brass Tacks features are all listed as being by Stanley Schmidt. This is just plain wrong, of course. Looking back: July issue also shows this. But in June I see that they're attributed to John W. Campbell, Jr. (who's at least the right editor). I'd expect to change these to say "The Editor" if the magazine says that, or "uncredited" if there's not a credit (always for Brass Tacks), in any of these.
  • I'd make sure that any pseudonyms I'm aware of are accounted for. What I see off hand is illustrations by Kelly Freas, which is a pseudonym for Frank Kelly Freas. (swfritter grumbles about that every time it comes up but says it's established that way.)
  • I'd normally be expecting to add illustrations, reviews, etc., as needed, checking each page of the magazine. Looking at the entry (but not having gone through the magazine in detail) it looks like this issue (& at a glance July & June) entry was thorough. Should I just not bother, or go through & ask if I see anything to correct, or go through & correct anything without asking? (The title of that science fact article is very much in my mind as I ask this.)
  • Oh! The number of pages shows as 164, shd be 180 (178 plus 2 sides of back cover); I'd fix that.

I'd appreciate instruction on how you want me to handle any or all of this, on this issue & for similar questions about other issues you've verified. Sorry to dump in such a mass of tangled questions in one bunch.
Thanks! -- Dave (davecat) 11:53, 9 Jan 2008 (CST)

Some of the changes you're suggesting (such as artist pseudonyms, Science Fact, Brass Tacks, etc) are conventions that have evolved after the verification (these issues, going back to 1965 were some of the very first verified pubs), so feel free to change them. A second pair of eyes are always welcome. In the case of magazines, if you find an error in a verified issue, it's unlikely that it's due to variations in a 2nd printing, so feel free to make corrections. Alvonruff 15:41, 9 Jan 2008 (CST)
Thank you. I'll attempt to keep track of changes I make & leave notes of them as I go through. (I'd recognized that some of these were changes in standards. I'm constantly amazed at how fast these things are still evolving. Lots of data going in means lots of decisions on things people hadn't thought about.) -- Dave (davecat) 15:47, 9 Jan 2008 (CST)
<SIGH> Regarding "In the case of magazines, if you find an error in a verified issue, it's unlikely that it's due to variations in a 2nd printing": I was just reading a Christmas gift (all of the 1959 Astoundings) and caught a note from the editor announcing that the cover price was to increase; he said that they'd been test-marketing with the higher cover price in some regions. Before this, it never would have occurred to me to check with someone on a cover price when I had that issue in my hand with a different one. Bleah.
I foresee more data entry & eventual verification than I'd expected in Astounding. -- Dave (davecat) 09:55, 15 Jan 2008 (CST)
The data for Astounding was collected from numerous sources and was one of the earliest magazines we had "complete" coverage for, so I expect it to be in the roughest shape. Alvonruff 12:11, 15 Jan 2008 (CST)
I wasn't clear on that last comment I made. It's just that I now have these 12 additional issues (just finished reading the last of them while eating lunch, too). I'd previously had only 6 odd issues of Astounding (two grouped with Analog here—from the period when both names were on the cover); now I have 18 to enter. Someday. When I work my way back that far.
Based on what I've been doing, the older Analogs have mostly been drawn from indexes of one sort or another, with contents partial & not up to current standards in lots of areas. (Makes me feel doddering to call them "the older" ones.) -- Dave (davecat) 13:41, 15 Jan 2008 (CST)

Python error in submitpub.cgi

I'm trying to submit changes (new stories added) to a COLLECTION. I get a python error in http://www.isfdb.org/cgi-bin/edit/submitpub.cgi

The original pub is here. editpub

IndexError	Python 2.4.2: /usr/local/bin/python
Wed Jan 9 16:28:48 2008

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/edit/submitpub.cgi
   74 
   75         new = pubs(db)
   76         new.cgi2obj()
   77         
   78         old = pubs(db)
new = <pubClass.pubs instance>, new.cgi2obj = <bound method pubs.cgi2obj of <pubClass.pubs instance>>
 /www/isfdb.tamu.edu/root/cgi-bin/edit/pubClass.py in cgi2obj(self=<pubClass.pubs instance>)
  714                                         if int(newTitle.id) > 0:
  715                                                 oldTitle = titleEntry()
  716                                                 title_data = getTitle(newTitle.id)
  717                                                 oldTitle.setTitle(title_data[TITLE_TITLE])
  718                                                 oldTitle.setID(newTitle.id)
title_data = (844227L, 'February 1999: Ylla', None, None, None, None, None, '2005-04-05', 'ss', 'SHORTFICTION', None, 2L, 64825L, None, 2L, 0L), global getTitle = <function getTitle>, newTitle = <pubClass.titleEntry instance>, newTitle.id = 844542
 /www/isfdb.tamu.edu/root/cgi-bin/edit/pubClass.py in getTitle(title_id=844542)
   29         result = db.store_result()
   30         record = result.fetch_row()
   31         return record[0]
   32 
   33 def getPageNumber(title_id, pub_id):
record = ()

IndexError: tuple index out of range
      args = ('tuple index out of range',)

Oh, I see what happened: moderator accepted one of merges of one of the stories added earlier to this collection. So, while I was editing the contents (adding more stories), one of the earlier stories disappeared (i.e. a record with title_id=844542, title 'G.B.S.--Mark V', disappeared). So, I can't now submit another 20 stories :-( Too many operations at the same time. If only such 'missing ids' could be ignored... --Roglo 16:35, 9 Jan 2008 (CST)

I accepted your submission merging the two title records for "G.B.S.--Mark V". When the merge was accepted the system chose to keep the first record and delete the second. So it appears that the second record was part of the pub record which you were editing. This pub no longer matched the system's record (because of the missing title) so the python error occurred. (This can happen, though rarely, when two people are working on the same pub.) It might be a good idea to make all pub edits, submit and wait for a mod to approve. Only then merge any titles with existing records. Thanks. Mhhutchins 17:54, 9 Jan 2008 (CST)

Changes to your verified pubs - Analogs from 1966

August 1966:

  • "Science Fact: One MOL Step Forward" changed to "One MOL Step Forward", & series added.
  • "Science Fact: The Psychoceramic" changed to "Psychoceramic" & added to series. ", Jr." dropped from author (but will make variant to canonical name). (Hmm. That's not quite what I did. I'll fix it up right after what I've already done has been approved.)
  • "Brass Tacks" changed to "Brass Tacks (Analog, August 1966)", added to series, author changed to "uncredited".
  • "In Times to Come" added to series, author changed to "The Editor".
  • Page count -> 180
  • Volume number (in note) to Roman numerals. (This is what the magazine has, & seems to be more common in the entry these issues, though both exist in our data.)
  • Added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. Will make the former a variant of the latter, after these changes are approved.
  • Changed bracketed numbers in the last two illustrations for "Too Many Magicians" installment. Added illustration on p. 31. Changed page number of one from 47 to 46. (The illustration occupies maybe 1/3 of p. 46 & all of p. 47. If you think it should say p. 47, let me know & I'll change it back.)
  • Added pub note about illustration on p. 140. (It's in two pieces, separated by text, with the second piece on p. 141.)

-- Dave (davecat) 16:36, 9 Jan 2008 (CST)

July 1966:

  • volume number (in note) to Roman numeral
  • where artwork titles duplicate each other (multiple artwork per story), added bracketed numbers
  • changed author from Stanley Schmidt to The Editor (In Times to Come, The Analytical Laboratory) or uncredited (Brass Tacks), adding series as well; added subtitle for AnLab
  • removed "Science Fact: " from two articles & put them in series; added pub note that "This is English?" is credited only in table of contents.
  • added pub notes that "An Ounce of Dissension" is listed as a short story, about illustration on pp. 126-127, & about some reviews
  • added credited cartoons (in "Meaning Theory" article)
  • made Kelly Freas a pseudonym (all items)

-- Dave (davecat) 10:56, 10 Jan 2008 (CST)

June 1966:

  • volume number (in note) to Roman numeral
  • Added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved
  • changed author to The Editor (In Times to Come, The Analytical Laboratory) or uncredited (Brass Tacks)
  • removed "Science Fact: " from title & put in series
  • added pub note that "Robin S. Scott" is "Robert S. Scott" in table of contents
  • put The Reference Library in series

-- Dave (davecat) 16:26, 10 Jan 2008 (CST)

May 1966:

  • volume number (in note) to Roman numeral
  • Added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved
  • changed author to The Editor (In Times to Come, The Analytical Laboratory) or uncredited (Brass Tacks)
  • removed "Science Fact: " from two articles & put them in series; changed author of one ("Titanium - The Wonder Metal") to uncredited & added pub note about form of that title
  • added note about "The Alchemist"
  • put "The Reference Library" in series
  • made variants of Kelly Freas illustrations

-- Dave (davecat) 10:19, 17 Jan 2008 (CST)

April 1966:

  • volume number (in note) to Roman numeral
  • Added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved. (Also changed ellipsis in title to proper form (with spaces).)
  • changed author to The Editor (In Times to Come, The Analytical Laboratory) or uncredited (Brass Tacks)
  • removed "Science Fact: " from article & put it in series
  • put "The Reference Library" in series
  • made variants of Kelly Freas illustrations
  • chgd page # for illustration on "Moon Prospector" from 9 to 8 (all of page 9, but 1/3 or 1/2 of p. 8). If you think it should be 9, let me know & I'll change it back.
  • added pub note about statement by Edward E. Smith prefixed to "Moon Prospector"
  • A couple of non-changes I dithered over: the puzzle answer on p. 57 and the credited illustrations to the science fact article - if you want (or anyone wants) to tell me I should have added these, I'll do them. (I don't plan to enter the puzzle (p. 144 in the March issue), either, unless someone else thinks it should be entered.)

-- Dave (davecat) 14:20, 17 Jan 2008 (CST)

March 1966:

  • volume number (in note) to Roman numeral
  • Added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • changed author to The Editor (In Times to Come, The Analytical Laboratory) or uncredited (Brass Tacks)
  • added "Giant Meteor Impact" to Science Fact (Analog) series
  • added note about "The Ship Who Mourned" (listed as short story, ISFDB says novelette)
  • added note about "Operation Malacca" (listed as novelette, ISFDB says novella)
  • made variants of Kelly Freas illustration
  • added illustration credit (John Schoenherr!) for science fact article ("Giant Meteor Impact")
  • added pub note about diagrams in "Operation Malacca" (on p. 113)

-- Dave (davecat) 12:29, 19 Jan 2008 (CST)

February 1966:

  • volume number (in note) to Roman numeral
  • Added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved. (Also changed ellipsis in title to proper form (with spaces).)
  • changed author to The Editor (In Times to Come, The Analytical Laboratory) or uncredited (Brass Tacks)
  • added parent of science fact article to series, taking "Science Fact: " out of its title
  • added pub note about "The Searcher"'s listing
  • added pub note about "The Switcheroo Revisited"'s listing
  • changed page number for "The Searcher" art from p. 8 to p. 9
  • made variants of Kelly Freas illustrations
  • added pub note about review of "Masters of Science Fiction"
  • added pub note about review of "Galactic Diplomat"

-- Dave (davecat) 10:34, 21 Jan 2008 (CST)

January 1966:

  • volume number (in note) to Roman numeral
  • Added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • changed author to The Editor (The Analytical Laboratory) or uncredited (Brass Tacks)
  • added pub note about listing of "Untropy"
  • added series for science fact article
  • made variants of Kelly Freas illustrations
  • added pub note about review of The Mallot Diaries

-- Dave (davecat) 12:12, 21 Jan 2008 (CST)

Changes to your verified pubs - Analogs from 1965

Al, it looks like you verified most of 1965, too. As I go through, I'll continue to leave you similar notes as I make changes, unless you'd rather I didn't.

December 1965:

  • volume number (in note) to Roman numeral
  • Added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • changed author to The Editor (In Times to Come) or uncredited (Brass Tacks)
  • added pub note about listing of "Mission 'Red Clash'"
  • added series for science fact article
  • made variants of Kelly Freas illustrations
  • added pub note about review of 13 French Science-Fiction Stories
  • added pub note about review of Analog 3
  • changed interiorart on p. 125 from "Mission "Red Clash" [2]" to "Mission "Red Clash" [3]"; added interiorart on p. 98 as "[2]"
  • fixed double comma in title ("The Reference Library (Analog,, December 1965)")

-- Dave (davecat) 16:05, 21 Jan 2008 (CST)

November 1965:

  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • added series to (JWCJr) editorial title record
  • changed author to The Editor (In Times to Come) or uncredited (Brass Tacks)
  • made variants of Kelly Freas illustrations
  • added pub note about "Problem in Thermodynamics" (not in TOC)
  • added pub note about "Some Preliminary Notes on FASEG"
  • added pub note about author of Exiles of Time (missing middle initial in review)

-- Dave (davecat) 11:02, 22 Jan 2008 (CST)

October 1965:

  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Deleted from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • added series to (JWCJr) editorial title record
  • added series to The Reference Library
  • changed author to The Editor (In Times to Come, The Analytical Laboratory) or uncredited (Brass Tacks)
  • made variants of Kelly Freas illustrations
  • added pub note about art credit for "Snakebite!"

-- Dave (davecat) 10:49, 23 Jan 2008 (CST)

September 1965:

  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Removed from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • added series to (JWCJr) editorial title record
  • added series to science fact title
  • added series to The Reference Library
  • changed author to The Editor (In Times to Come, The Analytical Laboratory) or uncredited (Brass Tacks)
  • made variants of Kelly Freas illustrations
  • added pub note about "The Life of Your Time" (TOC says shortstory)
  • changed typo in review listing ("Desination: Amalthea" -> "Destination: Amalthea")
  • combined two review listings into one (The Dark Intruder & Other Stories / Falcons of Narabedla - Ace double) (chgd title in one & removed second review from pub - does this create an orphan review title? will follow up & see)

-- Dave (davecat) 16:15, 23 Jan 2008 (CST)

(The answer to the question at the end of that list was as expected "yes", & I've submitted the deletion.) Dave (davecat) 09:39, 24 Jan 2008 (CST)

August 1965:

  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Removed from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • added series to (JWCJr) editorial title record
  • added series to science fact title
  • added series to The Reference Library
  • changed author to The Editor (In Times to Come) or uncredited (Brass Tacks)
  • made variants of Kelly Freas illustrations
  • added pub note ("Beautiful Venus" not in TOC)
  • added pub note about art credit in "Half a Loaf"
  • changed title (added "A ") & form of author's name (W. R. Cole") for review of "A Checklist of Science Fiction Anthologies". (This now matches the review; both versions of the book's title are in the database.)
  • added pub note about reprint reviews

-- Dave (davecat) 13:36, 28 Jan 2008 (CST)

July 1965:

  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Removed from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • added series to (JWCJr) editorial title record
  • added series to science fact title
  • added series to The Reference Library
  • changed author to The Editor (In Times to Come, AnLab) or uncredited (Brass Tacks)
  • made variants of Kelly Freas illustrations
  • set length for "Though a Sparrow Fall" from sf (shortfiction) to ss (shortstory)
  • added review of The Issue at Hand & pub note
  • added pub note about reprint reviews

-- Dave (davecat) 16:47, 28 Jan 2008 (CST)

June 1965:

  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Removed from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • added series to (JWCJr) editorial title record
  • added series to science fact title
  • added series to The Reference Library
  • changed author to The Editor (In Times to Come, AnLab) or uncredited (Brass Tacks)
  • added pub note about "The Muddle of the Woad" (listed as novelette)
  • added pub note about reprint reviews

-- Dave (davecat) 12:06, 29 Jan 2008 (CST)

May 1965:

  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Removed from pub the one for John W. Campbell, Jr. will make the former a variant of the latter, after these changes are approved.
  • added series to (JWCJr) editorial title record
  • added series to science fact title
  • added series to The Reference Library
  • added review of The Physicists & pub note
  • added pub note about reprint reviews

-- Dave (davecat) 16:19, 29 Jan 2008 (CST)

April 1965:

  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Removed from pub the one for John W. Campbell, Jr. Made the former a variant of the latter, since mhhutchins approved all the rest before I got this entered here.
  • added series to (JWCJr) editorial title record
  • removed "Science Fact: " from title of parent to science fact title
  • added series to The Reference Library
  • changed author to The Editor (In Times to Come, AnLab) or uncredited (Brass Tacks)
  • added pub note about "Fad" (TOC says short story)
  • added pub note about "No Throne of His Own" (TOC says short story)
  • added illustration p. 25
  • changed review dates to include month
  • made variant of Kelly Freas illustration

-- Dave (davecat) 11:08, 30 Jan 2008 (CST)

January 1965:

  • changed page count to 100 (initial leaf, table of contents, pages numbered 4-96, plus inside & outside of front & back covers)
  • volume number (in note) to Roman numeral
  • added editorial entry by John W. Campbell. Removed from pub the one for John W. Campbell, Jr. Will make the former a variant of the latter after approval.
  • added series to (JWCJr) editorial title record
  • added series to science fact title
  • added series to The Reference Library
  • changed author to The Editor (In Times to Come, AnLab) or uncredited (Brass Tacks)
  • added pub note about illustration on p. 17 (reduced B&W version of cover)
  • combined separate reviews of Son of the Tree & The Houses of Iszm into one for the Ace double (changed one review entry, removed the other - will delete orphan after approval)
  • added pub note about reprint reviews
  • made variants of Kelly Freas illustrations

-- Dave (davecat) 10:45, 31 Jan 2008 (CST)


Alternate names glitch?

Leo Summers, a prolific illustrator, often signed his name "Summers". We have a few of his "Summers" interior art works set up as "Leo Summers as by Summers", but "Summers" (unlike "Leo Ramon Summers" and "Leo R. Summers") doesn't appear on the list of "Used These Alternate Names" at the top of the "Leo Summers" page. Conversely, "Leo Summers" doesn't appear as a "Used As Alternate Name By:" name at the top op the "Summers" page. This is recreatable as I confirmed a few minutes ago by making one of our "Summers" titles into a "Leo Summers" variant title with no effect on the "alternate names" situation. I wonder if it may have something to do with the fact that Summers' work was art rather than fiction and the logic may not be quite the same? Ahasuerus 21:59, 16 Jan 2008 (CST)

P.S. I set up another variant title between a novel published as by Robert Morgan and its C. J. Henderson parent and the "Alternate Names" designations didn't appear on either page until I set up an "Author Pseudonym", so it's not just interior art. Is this change in behavior intentional or is it a side effect of recent changes? Or has it always worked that way and I just need to remember to take my memory medication (always a challenge)? Ahasuerus 23:57, 16 Jan 2008 (CST)
It's always worked that way. Alvonruff 14:53, 17 Jan 2008 (CST)
Thanks! Now, where did I put my pills?.. Ahasuerus 15:11, 17 Jan 2008 (CST)
In the little box that looks like a Dried Frog? That's where I keep mine. You can have them actually, as my memory works well enough to remember it always worked that way as long as I've been here, which is... well, not as long as you... bit longer than my tea-breaks... (goes and checks) Ah! One year, 4 days. Damn, nobody remembered my anniversary. Not even me. :-( BLongley 17:08, 17 Jan 2008 (CST)
It's a good thing in a way since it means that we are too busy doing enjoyable work here (creating order out of chaos!) that we lose track of time :) Ahasuerus 17:18, 17 Jan 2008 (CST)

Fast Feature Request (FFR) - wiki server name

Maybe there should be a page for (hopefully) very fast to fix issues.

Somewhere in the wiki configuration is the string isfdb.tamu.edu. Can this be changed to www.isfdb.org?

I believe this is the file http://www.isfdb.org/wiki/LocalSettings.php and that you would edit one or more of

This will allow the site to keep working for the user after editing a page. It will also fix that author pages link to biographic and bibliographic comments using www.isfdb.org/... but the links leave the spaces in the author names. The wiki has code that redirects these to isfdb.tamu.edu/... with the spaces replaced by _. TIA. Marc Kupper (talk) 19:38, 22 Jan 2008 (CST)

Actually, I have tried editing LocalSettings.php in the past (over the Christmas break in fact). That string refers to the base directory the wiki code is stored at; changing it caused a complete wiki failure. I'll look at this one more time tonight, but I'm sure I only found one instance of "tamu" in all the php files. Alvonruff 12:42, 23 Jan 2008 (CST)
There was no configuration in the wiki telling it to use isfdb.tamu.edu, so it must be getting the information from reverse DNS. Adding $wgServerName caused a fatal wiki error, but adding $wgServer seemed to do the trick. Alvonruff 18:57, 23 Jan 2008 (CST)
I just tried it and seems to have worked! Thanks a lot! :)
P.S. On an unrelated note, at one point Bill Longley volunteered to help with approving/editing Dissembler submissions. Is this something that might help free you up for software work or do you need to review these submissions yourself so that you could tweak Dissembler? Ahasuerus 19:20, 23 Jan 2008 (CST)
Some of the Dissembler runs are updates to forthcoming books, others are needed to fill in blanks with the awards (which IS a test of new award editing tools). If people don't mind sifting through Dissembler submissions, I'll start doing regular runs and not bother with the moderation part. Alvonruff 07:21, 24 Jan 2008 (CST)
Woo Hoo! Thank you! I butchered through Dissembler submissions a while back - it seemed easy enough. Marc Kupper (talk) 01:59, 24 Jan 2008 (CST)


Python errors

I tried to approve this submission, but ran into the following Python error:

#  update notes set note_note='2nd edition, revised and enlarged.\n\n(4th tp printing 1974-11-00; page count matches that for inital printing)' where note_id=126633
# insert into titles(title_title) values('The Science Fiction Novel: Imagination and Social Criticism');
# insert into pub_content(pub_id, title_id) values(254714, 849483);
# update titles set title_ttype="REVIEW" where title_id=849483
# update titles set title_copyright="1967-03-00" where title_id=849483
# update pub_content set pubc_page='4' where title_id=849483 and pub_id=254714
# insert into canonical_author(title_id, author_id, ca_status) values('849483', '633', '1'); setReviewees: ['Basil Davenport ', ' Robert A. Heinlein ', 'C. M. Kornbluth ', 'Alfred Bester ', 'Robert Bloch']
# insert into canonical_author(title_id, author_id, ca_status) values('849483', '2160', '3');
# insert into authors(author_canonical) values(' Robert A. Heinlein ');
# update authors set author_lastname='' where author_id='109642'
# count 46185 for section [] --> -->
 
 
IndexError	Python 2.4.2: /usr/local/bin/python
Sat Jan 26 04:51:59 2008

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/mod/pa_update.cgi
  619                                                 mergeReview(record, title, authors, reviewers, date, page, doBookAuthors, doReviewers)
  620                                         else:
  621                                                 insertReview(title, authors, reviewers, date, page)
  622 
  623                         ###############################################
insertReview = <function insertReview>, title = 'The Science Fiction Novel: Imagination and Social Criticism', authors = 'Basil Davenport + Robert A. Heinlein +C. M. Kornbluth +Alfred Bester +Robert Bloch', reviewers = 'Damon Knight', date = '1967-03-00', page = '4'
 /www/isfdb.tamu.edu/root/cgi-bin/mod/pa_update.cgi in insertReview(title='The Science Fiction Novel: Imagination and Social Criticism', reviewees='Basil Davenport + Robert A. Heinlein +C. M. Kornbluth +Alfred Bester +Robert Bloch', reviewers='Damon Knight', date='1967-03-00', page='4')
  288         for reviewee in revieweelist:
  289                 NewReviewees.append(reviewee)
  290         setReviewees(record, NewReviewees)
  291 
  292 def mergeReview(record, title, reviewees, reviewers, date, page, doReviewees, doReviewers):
global setReviewees = <function setReviewees>, record = 849483L, NewReviewees = ['Basil Davenport ', ' Robert A. Heinlein ', 'C. M. Kornbluth ', 'Alfred Bester ', 'Robert Bloch']
 /www/isfdb.tamu.edu/root/cgi-bin/mod/common.py in setReviewees(Record=849483L, NewReviewees=['Basil Davenport ', ' Robert A. Heinlein ', 'C. M. Kornbluth ', 'Alfred Bester ', 'Robert Bloch'])
  236                 deleteTitleAuthor(author, Record)
  237         for author in NewReviewees:
  238                 addTitleAuthor(author, Record, 'REVIEWEE')
  239 
  240 def setInterviewees(Record, NewInterviewees):
global addTitleAuthor = <function addTitleAuthor>, author = ' Robert A. Heinlein ', Record = 849483L
 /www/isfdb.tamu.edu/root/cgi-bin/mod/common.py in addTitleAuthor(author=' Robert A. Heinlein ', title_id=849483L, status='REVIEWEE')
  101                 author_id = record[0][0]
  102         else:
  103                 author_id = insertAuthorCanonical(author)
  104 
  105         ##############################################
author_id undefined, global insertAuthorCanonical = <function insertAuthorCanonical>, author = ' Robert A. Heinlein '
 /www/isfdb.tamu.edu/root/cgi-bin/mod/common.py in insertAuthorCanonical(author=' Robert A. Heinlein ')
  273         section = lastname[0:2]
  274         #update_directory(section)
  275         update_directory(lastname)
  276 
  277         return author_id
global update_directory = <function update_directory>, lastname = ''
 /www/isfdb.tamu.edu/root/cgi-bin/mod/common.py in update_directory(lastname='')
  138         print "<li> count %d for section [%s]" % (count, section)
  139 
  140         query = "select directory_mask from directory where directory_index='%s'" % (section[0])
  141         db.query(query)
  142         result = db.store_result()
query = "select COUNT(author_lastname) from authors where author_lastname like '%'", section = ''

IndexError: string index out of range
      args = ('string index out of range',)

I suspect it had something to do with the fact that one of the new book review records had "Kingsley Amis," (with a trailing comma) listed as the author. Ahasuerus 04:55, 26 Jan 2008 (CST)

...and 3 minutes later I had another Python error when trying to merge "Robert A. Heinlein" and " Robert A. Heinlein" (with a leading space). Ahasuerus 05:00, 26 Jan 2008 (CST)

More directory-related errors. Instead of trying to make code that forecasts all possible future name manglings, there is now bullet-proofing code that skips the directory update if the name isn't within the define constraints. Alvonruff 07:18, 26 Jan 2008 (CST)

Python errors when displaying Titles with awards

I am getting Python errors when displaying Titles with associated awards, e.g. Farmer in the Sky. I see that you have edited the Database Schema page, so I assume that you are making software changes this morning. Would it be safer to take the ISFDB offline for a bit while the changes are being made or are these just momentary interruptions? Ahasuerus 07:13, 26 Jan 2008 (CST)

Who knew anyone was up? Sorry, made all the index-related changes to the Python code, but missed the one SQL query for display. All should be working now. Award editing isn't generally online (which is the only way to affect the contents of the awards table, merging only affects the award mapping table), so I'm not too worried about damage. Alvonruff 07:28, 26 Jan 2008 (CST)
Thanks, looks much better now! I didn't know what the scope of the problem was, so I wasn't sure what it was impacting. As far as editors being up at night goes, well, my hours can be irregular when I am on the road (I was editing between 2:30am and 6am ISFDB time this morning) and, besides, we now have active editors/moderators from the UK, Australia, Germany, Poland and the Czech Republic. The ISFDB is pretty much a 24 by 7 operation these days, which is a good thing, of course, but makes it harder to implement changes without affecting operations. Ahasuerus 13:28, 26 Jan 2008 (CST)
I'm actually quite glad of some reminders that the ISFDB isn't QUITE 24 by 7 yet - e.g. there's a short 11:30pm (my time) outage each day that reminds me I should go make a cup of cocoa, tidy up what I'm doing, and go to bed. Not so useful at weekends when I can stay up and chat a bit more interactively, but I presume something important is happening. (Is that the Backup time?) BLongley 14:36, 26 Jan 2008 (CST)
That's right, the backup file is always time stamped 6:30pm, which probably matches 23:30 GMT once you factor in various time zones, daylight savings time and the phase of the moon. Ahasuerus 15:09, 26 Jan 2008 (CST)

Awards

I knocked up a quick'n'dirty Awards page for now - I know you're working in this area and it'll probably be obsolete soon, but in the meantime are there any Awards that should really NOT be linked to? BLongley 08:04, 27 Jan 2008 (CST)

This is cool and will continue to work after the editing tools are online. It actually makes it easier to figure out which years I'm behind on, so I'll probably extend it. Alvonruff 18:49, 27 Jan 2008 (CST)

Big Screwup Help!

I've accidentally nested "Acorna Universe" series inside itself & now its mirroring the "Delete This Series Record", the result is the McCaffery & Scarborough summary page take forever to display. ;-/Kraang 08:53, 27 Jan 2008 (CST)

I figured out how to fix it but now we have some more "delete this series" series! :-)Kraang 10:39, 27 Jan 2008 (CST)
Or retitle and re-use.--swfritter 18:59, 27 Jan 2008 (CST)
It's possible to re-use empty Series records, but some of them are linked to super-series records, a relationship which we, unfortunately, can't break at this time, so we have to be careful. Ahasuerus 19:47, 27 Jan 2008 (CST)
I learned my lesson the hard way by buggering up three authors pages(all fixed now) with a series that copied and mirrored itself off the end of my screen and down a couple of floors! I wouldn't recommend this Delete This Series Record[3](view with caution!) or a couple of others be reused! :-)Kraang 20:15, 27 Jan 2008 (CST)
The Series Ourobouros - the head of the series consuming the tale of the series. Got a sub-series situtation out there myself. Now all I have to do is find a new super series that requires 5 sub-series - not! Have enough other things to think about.--swfritter 20:20, 27 Jan 2008 (CST)
Yup, that's what happens when a series is a sub-series of itself -- it would be appropriate if they wrote a series of novels based on Groundhog Day :) I am sure Al will add checks preventing duplicate series names and self-referential series pointers when he gets to our (long) list of series bugs. Ahasuerus 20:24, 27 Jan 2008 (CST)

Editing problem: removing page numbers

I added page numbers 1,2,3, . . . to a pub to keep the contents listing ordered. Now I'm trying to remove these page numbers (as required by a moderator) and I can't. Looks like the problem is due to 2 lines being commented out in cgi-bin/mod/pa_update.cgi :

def mergeTitle(record, title, authors, date, page, type, length, doAuthors):
 [. . .]
        if page:
                setTitlePage(record, page, pub_id)
        #else:
        #        setTitlePage(record, '', pub_id)

The offending pubs are: The Dog Said Bow-Wow and The Starry Rift: Tales of New Tomorrows. The note on 'fake' page numbers was deleted from the second pub by the same submission that failed to remove page numbers. Is there a way to remove page numbers other than 'clone and delete'? Thanks. --Roglo 14:06, 29 Jan 2008 (CST)

It looks like it's same problem as the one that I reported back in December and Al had trouble recreating. Ahasuerus 14:34, 29 Jan 2008 (CST)
There's a temporary workaround in that you can use a single space to blank out the page number though that's related to a bug in that the page # field is not trimmed and so you are actually setting the page # to blank. I suspect clone-delete is the best course for now and did exactly that for The Dog Said Bow-Wow as the blank page numbers did not look good. Marc Kupper (talk) 23:30, 29 Jan 2008 (CST)
The commented-out code is wrong in that if someone modifies a title field other than the page number, then the XML submission will have no cPage tag, and page will get set to NULL, causing setTitlePage() to delete an existing page number, even when you don't want it to.
I've put a real fix online and have testing both adding/deleting page numbers, and so far it works. Alvonruff 18:36, 30 Jan 2008 (CST)
Just entered a couple of novels in an omnibus[4] without page numbers and the space for the page number displays this "NoXmlPag". Is this suppose to be there now?Kraang 19:51, 30 Jan 2008 (CST)
It worked for The Starry Rift: Tales of New Tomorrows. Thanks! (And thank you for the explanation). --Roglo 12:47, 31 Jan 2008 (CST)
Yup, seems to work fine! Ahasuerus 21:49, 2 Feb 2008 (CST)

Submission display bug when one of the Authors has been deleted

An editor created a MakePseudonym submission earlier today, but on closer examination it turned out that the proposed pseudonym was just a misspelling of the author's name. I merged the misspelled version of the name with the correct version and everything is fine now, but when I went back to reject the original submission, it looked weird and the submitter's name appeared as "unknown". Would you like to take a look at it before I reject the submission? Ahasuerus 21:49, 2 Feb 2008 (CST)

Fixed. Alvonruff 05:06, 3 Feb 2008 (CST)
Thanks! :) Ahasuerus 10:57, 3 Feb 2008 (CST)


Dissembler-generated notes?

User:Roglo just pointed out that "it would be better if Dissembler added a note to its records, something like Pre-release data generated by Dissembler from Amazon.com website on YYYY-MM-DD so that editors could update the records later without assuming that someone knew better". I seem to recall that Dissembler used to create notes along the lines of "Amazon 11/01", but I don't think I have seen them lately. Given our current emphasis on differentiating "what we know from what we think we know" and on recording the sources of our information (primary vs. secondary etc), is this something that could be easily added? Ahasuerus 10:37, 6 Feb 2008 (CST)

No problem (remind me if I forget, and no notes are present in the next run). Alvonruff 13:20, 6 Feb 2008 (CST)
Will do, thanks! Ahasuerus 13:28, 6 Feb 2008 (CST)
They appeared (with notes), and I whacked the submissions about a bit. Well, a lot really. Most seemed reasonable starting-points, but to get better data I switched a few titles from vague Amazon.com titles to less vague Amazon.co.uk titles. (Removing blame for Dissembler when I did so.) Does Dissembler look at different sites in different ways for different authors/publishers? Does Dissembler actually do an Amazon.co.uk run? (Sorry, I should have kept better notes on how I adjusted them, but feel free to release another set, I enjoy them.) BLongley 19:09, 9 Feb 2008 (CST)

Review editing causing a Python error

Earlier today I noticed that Horror: 100 Best Books included a review record with a flip-flopped author/reviewer for the early 17th century novel The White Devil. The author was entered as Diana Wynne Jones and the reviewer was entered as John Webster, quite a feat for somebody who died ca. 1625! I pulled up the review and reversed the author and the reviewer. Now that I have approved my change, whenever I try to display the Novel Title, I get a Python error. Also, I can't display my submission via the pv_update.cgi script because I get the following Python error:

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/mod/pv_update.cgi
  404 
  405 
  406         if IsVerified(Record):
  407                 print '<p><div id="WarningBox">'
  408                 print '<b>WARNING:</b> This publication has been verified against the primary source.'
IsVerified = <function IsVerified>, Record undefined

NameError: name 'Record' is not defined
      args = ("name 'Record' is not defined",)

Ahasuerus 17:50, 6 Feb 2008 (CST)

Looks like the reviewer is missing in this review (23 • The White Devil • John Webster • book review by). --Roglo 03:39, 7 Feb 2008 (CST)
I suspect that Al may have changed something since the page no longer errors out. Ahasuerus 10:14, 7 Feb 2008 (CST)
I tried to do similar edit.

I start with:

61 • Vurt • Test BookAuthor • book review by Test Reviewer

I have:

mysql> select  author_id, author_canonical  from authors order by 1 desc limit 2 ;
+-----------+------------------+
| author_id | author_canonical |
+-----------+------------------+
|    109898 | Test BookAuthor  | 
|    109897 | Test Reviewer    | 
+-----------+------------------+

(both test authors have some other works, so they won't disappear)

I swap them. Expected:

61 • Vurt • Test Reviewer • book review by Test BookAuthor

Submission:

<?xml version="1.0" encoding="iso-8859-1" ?>
<IsfdbSubmission>
<PubUpdate>
<Record>58935</Record>
<Submitter>Demo</Submitter>
<Subject>Interzone, April 1994</Subject>
<Content>
<ContentReview>
<Record>563661</Record>
<cBookAuthors>Test Reviewer</cBookAuthors>
<cReviewers>Test BookAuthor</cReviewers>
</ContentReview>
</Content>
</PubUpdate>
</IsfdbSubmission>

Proposed Publication Update Submission Modified Book Reviews

              Current          Proposed
Title         Vurt             -
Book Authors  Test BookAuthor  Test Reviewer
Reviewers     Test Reviewer    Test BookAuthor
Year          1994-00-00       -
Page          61               -

So far, everything OK. Approval:

SQL Updates:

   * delete from canonical_author where author_id='109897' and title_id='563661'
   * select COUNT(author_id) from canonical_author where author_id=109897
   * insert into canonical_author(title_id, author_id, ca_status) values('563661', '109898', '1'); setReviewees: ['Test Reviewer']
   * delete from canonical_author where author_id='109898' and title_id='563661'    <====== reviewer removed
   * select COUNT(author_id) from canonical_author where author_id=109898
   * insert into canonical_author(title_id, author_id, ca_status) values('563661', '109897', '3');
   * update submissions set sub_state='I' where sub_id='924025'
   * update submissions set sub_reviewer='13399' where sub_id='924025'
   * update submissions set sub_reviewed=NOW() where sub_id='924025' 

Result:

61 • Vurt • Test Reviewer • book review by

So, because the new reviewer is also a 'no longer author', he is removed from canonical_author just after he was added there as a reviewer (line 4 in the SQL code above). Problem is that when author info is removed, ca_status is ignored, and reviews have two kinds of authors. --Roglo 13:31, 7 Feb 2008 (CST)

Good job, Detective Roglo! :) Ahasuerus 15:26, 7 Feb 2008 (CST)
Yep, nice work indeed! Now go file the Bug Report... or maybe fix it? I know the code is up at Sourceforge, I just haven't learnt anything more about the ISFDB code than the database schema, and not even all of that yet. BLongley 15:46, 7 Feb 2008 (CST)
Thanks! This all is copy and paste from the displayed messages, figuring out how to fix it without breaking other things is quite different matter. I'll check if something similar was reported earlier. --Roglo 15:58, 7 Feb 2008 (CST)
Good summary. Fixed. There is a centralized routine for adding authors which uses the ca_status field, but the centralized routine for deleting authors didn't. It now does. Alvonruff 05:54, 9 Feb 2008 (CST)

Heads up - robots.txt blocks Googlebot

I did a Google search for something on ISFDB and was surprised nothing was found.

It turns out robots.txt is dated April 25, 2007 and says

User-agent: *
Disallow: /

The net result is at the moment Google only has

  • 39 pages for www.isfdb.org
  • The same 39 pages for isfdb.org
  • 336 pages for isfdb.tamu.edu (wiki pages)

Even with the few pages that are left Google is not showing cached results meaning I can't see how old Google's copies are. Marc Kupper (talk) 02:33, 7 Feb 2008 (CST)

TAMU put the robots.txt here a couple of months ago. I can't decide if this is a good thing or a bad thing. On the down side it reduces the number of new users to the site, but that's not really a disaster since we're not planning on generating ad revenue. On the up side, no Google results (or other search engines that honor robots.txt) means that new spambots won't find us as easily. Alvonruff 05:59, 9 Feb 2008 (CST)
Thank you Al - I'd personally vote for allowing at least the Google bot as I believe new contributors show up faster than spammers. Marc Kupper (talk) 00:24, 10 Feb 2008 (CST)

Disssembler - missing titles

I wonder why Dissembler apparently missed Kelley Armstrong's Personal Demon and Jeanne C. Stein's The Becoming (which I have since added based on OCLC's data)? We were also missing Tempting Danger, Blood Lines and Night Season by Eileen Wilks, Lina Gadiner's Grave Illusions, Jeaniene Frost's Halfway to the Grave, Jenna Black's Secrets in the Shadows and the whole Gardella Vampire Chronicles series by Colleen Gleason. Is Dissembler checking "Romance > Fantasy, Futuristic & Ghost", "Romance > Vampires", "Romance - Time Travel" and "Romance - Fantasy"? Ahasuerus 02:29, 10 Feb 2008 (CST)

Dissembler currently rejects books under Romance (fallout from previous Rules of Inclusion battles). It'll need to be extended to include subject subtypes which should be accepted which happen to fall under subject types which should be rejected. Alvonruff 10:54, 10 Feb 2008 (CST)
I can see why we need a certain number of "exclusion" clauses, e.g. "if one of the categories matches the word Comic, then skip the publication regardless of any other qualifying categories". Which, I should add, often leaves out legitimate fiction about superheroes, but I don't think there is anything we can do about that unless we do exclusion strictly by publisher.
However, I am not sure why we would want to skip publications just because they are tagged "Romance" any more than we would exclude pubs that are tagged "Western" or "Mystery". As long as a pub is tagged with an eligible SFnal category, be it "Vampires", "Paranormal Mysteries", "Occult Detectives", "Romance - Time Travel" or "Space Flight - Fiction", it shouldn't matter that it also says "Romance" or "Mystery", should it? Or am I missing something obvious?
BTW, are the baseline URLs that Dissembler uses to find forthcoming books public? If they are, I may be able to whip up something quick and dirty independently for cross-checking purposes. Unfortunately, my To Do list is getting depressingly long :( Ahasuerus 15:35, 10 Feb 2008 (CST)
Well, since Dissembler started adding notes I can confirm that the run I dealt with on Saturday was (mostly?) from Amazon.com - however, several of the titles would have been better sourced from Amazon.co.uk - where I made such adjustments I removed the blame from Dissembler though. I don't know if Dissembler looks for new pubs site by site and processes them immediately - in some cases it might be better to assemble the list of potential additions first from multiple sites, then choose the best site for each publication. BLongley 16:23, 10 Feb 2008 (CST)

Python error in Make Pseudonym

I made "Barton Werper" into a pseudonym of Peter T. Scott and then submitted another MakePseudonym for "Barton Werper", this one for Peg O'Neill Scott (it was a collective pseudonym). I received the following Python error:


ProgrammingError	Python 2.4.2: /usr/local/bin/python
Mon Feb 18 20:27:27 2008

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/edit/submitmkpseudo.cgi
   50 
   51         if parent_id == 0:
   52                 parent_data = SQLgetAuthorData(parent_name)
   53                 if parent_data:
   54                         parent_id = parent_data[0]
parent_data undefined, SQLgetAuthorData = <function SQLgetAuthorData>, parent_name = "Peg O'Neill Scott"
 /www/isfdb.tamu.edu/root/cgi-bin/edit/SQLparsing.py in SQLgetAuthorData(author="Peg O'Neill Scott")
   94 def SQLgetAuthorData(author):
   95         query = "select * from authors where author_canonical='" +author+ "'"
   96         db.query(query)
   97         result = db.store_result()
   98         record = result.fetch_row()
global db = <_mysql.connection open to 'localhost' at 837500c>, db.query = <built-in method query of Connection object>, query = "select * from authors where author_canonical='Peg O'Neill Scott'"

ProgrammingError: (1064, "You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'Neill Scott'' at line 1")
      args = (1064, "You have an error in your SQL syntax; check the ...right syntax to use near 'Neill Scott'' at line 1")

A problem with escaping apostrophes, perhaps? Ahasuerus 20:31, 18 Feb 2008 (CST)

Looks like it. I'll fix it this evening. Alvonruff 09:07, 19 Feb 2008 (CST)
Thanks! In the meantime, it looks like title editing got broken overnight as reported by Davecat. Always something :) Ahasuerus 09:52, 20 Feb 2008 (CST)
All the above (apostrophes; editing) are now fixed. Alvonruff 18:53, 20 Feb 2008 (CST)
Apostrophes now work, thanks! :) Ahasuerus 12:12, 23 Feb 2008 (CST)


Outstanding Publisher issues

Here is a list of various outstanding Publisher-related issues that we came up with over on the Community Portal:

  1. Links from Publisher search results to Author bibliographies are broken if the author's name contains apostrophes, e.g. if you look up iUniverse - Books Published in 2004 and click on Robert 'Dean' D. Russell, you will get an "Author not found: Robert ’Dean’ D. Russell" error.
  2. The new "Publisher" and "ISBN" drop down choices appear on most pages, but not on the Moderator page.
  3. The link to the Wiki at the top of the main Publisher page, e.g. the "Tesseract - Bibliography" page, is to the "ISFDB" namespace. We would like it changed to the "Publisher" namespace to match the "Author" and "Series" namespaces. An eager editor has already moved all Wiki pages to the new namespace :)
These will be easy to fix, once my Internet link becomes reliable; otherwise I'm likely to truncate files. Alvonruff 12:32, 25 Feb 2008 (CST)
Fixed. Fixed. Fixed. Alvonruff 20:49, 26 Feb 2008 (CST)
Many thanks! (Three, to be precise :-) Ahasuerus 20:58, 26 Feb 2008 (CST)

Also, a new Publisher-specific Note field has been requested, but I am not sure how time consuming it would be to implement. Thanks! Ahasuerus 10:31, 24 Feb 2008 (CST)

Adding the note is easy. Adding the editing tools to actually add and edit the notes - that's a different story. At least it forms the first agreed-upon field for editing. Alvonruff 12:32, 25 Feb 2008 (CST)
Just spotted a problem with the Publisher Wiki Links for e.g. Ace SF Special 1, presumably because of the square brackets around the 1. Is this an easy fix or should we find a Volunteer Publisher-Editor? (I'm not working on those, just came across it by accident.) BLongley 13:17, 25 Feb 2008 (CST)
The wiki software does not permit a square bracket inside a page name unless it is escaped, as I understand it. I think that it would be best if we avoided such characters in strings intended to link to the wiki. It would be posisble to dispaly one string and link to another, of course -- for example "[" could be converted to "(". -DES Talk 13:23, 25 Feb 2008 (CST)

Publisher Namespace

After some (but perhaps not enough) discussion at ISFDB:Community Portal#Publisher Issues, I recently moved all the publisher pages to a "Publisher:" pesudo-namespace -- that is, to pages whose names start with "Publisher:". But it would probably be better if a truwe custom namespace ("Publisher") were created.

According to the mediaWiki help page custom namespaces may be added simply by editing the LocalSettings.php file. If pages already exist in a pesudo namespace that matches an about to be created custom namespace, those pages should be first moved to a new name, then moved back after the namespace hass been created. For example Publisher:DAW might be moved to Publisher2:DAW, and later, after the Publisher custom namespace has been added to the LocalSettings.php file, it would be moved back to Publisher:DAW. According to the MediaWiki help page, there is a script (maintance/namespaceDupes.php) avialable to do such moves as a batch operation.

In light of this, I may have jumped the gun by making the moves I did, but it doesn't appear that any major harm was done. It looks to me as if the only requirement for havign a custom namespace is a simple 1-line edit (per namespace) to a configuaration file, which should take no time at all for you or anyone else who has access to the server wiki config files (if there is anyone else who has access). But if there are existing pages in a pesudo-namespace, they must be moved before the namespace is added (or while it is commented out), or they will be effectivly hidden and inaccesable. See the MW help page linked above for more detaiul.

I will be happy to do the moves, if this will be helpful, unless you prefer to use the script mentioned above -- assuming that you agree that creating such a namespace is a good idea. -DES Talk 19:31, 26 Feb 2008 (CST)

I personally would love to see Author, Bio, Fanzine, Magazine, Publication, and Series name spaces too but hopefully the scripted method is available and works. Here's the copy paste. :) Marc Kupper (talk) 03:44, 27 Feb 2008 (CST)
$wgExtraNamespaces[100] = "Author";
$wgExtraNamespaces[101] = "Author_talk";
$wgExtraNamespaces[102] = "Bio";
$wgExtraNamespaces[103] = "Bio_talk";
$wgExtraNamespaces[104] = "Fanzine";
$wgExtraNamespaces[105] = "Fanzine_talk";
$wgExtraNamespaces[106] = "Magazine";
$wgExtraNamespaces[107] = "Magazine_talk";
$wgExtraNamespaces[108] = "Publication";
$wgExtraNamespaces[109] = "Publication_talk";
$wgExtraNamespaces[110] = "Publisher";
$wgExtraNamespaces[111] = "Publisher_talk";
$wgExtraNamespaces[112] = "Series";
$wgExtraNamespaces[113] = "Series_talk";


The above code is now present in LocalSettings.php, and nothing blew up. Alvonruff 07:41, 1 Mar 2008 (CST)
However, please note that Publisher:DAW is now a red link. The first part of the moves will have to be done while the new namespaces are commetned out, unless the maintainence script can be used. -DES Talk 09:41, 1 Mar 2008 (CST)
Creation of pages to replace any of the temporarily missing (due to the new namespaces) pages will cause problems reclaiming the old pages. This is already happening. See ISFDB:Community Portal#Namespace problem for more detail. -DES Talk 17:44, 1 Mar 2008 (CST)

Dissembler and The Titan's Curse

Just a heads up that Dissembler apparently missed The Titan's Curse (2007), book 3 in Rick Riordan's YA series "Percy Jackson and the Olympians". Ahasuerus 15:53, 2 Mar 2008 (CST)

I shall chastise the bugger. Alvonruff 15:54, 2 Mar 2008 (CST)

Apostrophes in Authors' name bugged?

I have just noticed that all links to Author records that contain apostrophes are suddenly broken. If I search on the apostrophe character, the application finds 320 matching author names starting with 'Scoop' Goodstuff, but the links don't work and I get an "Author not found: \'Scoop\' Goodstuff" error instead. And yes, I realize how shocked you must be to learn that we have a problem with apostrophes, of all things! Ahasuerus 18:25, 2 Mar 2008 (CST)

How does this stuff keep creeping back? Fixed. For now. Alvonruff 18:44, 2 Mar 2008 (CST)
"You can't keep a good bug down"? Ahasuerus 18:58, 2 Mar 2008 (CST)
P.S. Speaking of apostrophes, I added Comtese d'Aulnoy's legal name, "Le Jumel de Barneville, Marie-Catherine, Baronne d'Aulnoy" and birthplace ("Barneville-la-Bertran, Eure, France") and the submission errored out as follows:
Submitting the following changes:
Column 	Current [110753] 	Proposed
Canonical 	Comtese d'Aulnoy 	Comtese d'Aulnoy
Legalname 	- 	Le Jumel de Barneville, Marie-Catherine, Baronne d'Aulnoy
Lastname 	d'Aulnoy 	d'Aulnoy
Birthplace 	- 	Barneville-la-Bertran, Eure, France
Birthdate 	- 	1650-00-00
Deathdate 	- 	1705-01-04
Wikipedia 	- 	http://en.wikipedia.org/wiki/Madame_d'Aulnoy
Imdb 	- 	-
Image 	- 	-
Emails 	- 	-
Webpages 	- 	-
--> -->
 
 
ProgrammingError	Python 2.4.2: /usr/local/bin/python
Sun Mar 2 19:02:30 2008

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/edit/submitauth.cgi
  217         submitter_id = SQLgetSubmitterID(submitter)
  218         update = "insert into submissions(sub_state, sub_type, sub_data, sub_time, sub_submitter) values('N', %d, '%s', NOW(), %d)" % (MOD_AUTHOR_UPDATE, update_string, submitter_id)
  219         db.query(update)
  220         
  221         PrintPostSearch(0, 0, 0, 0, 0)
db = <_mysql.connection open to 'localhost' at 837380c>, db.query = <built-in method query of Connection object>, update = "insert into submissions(sub_state, sub_type, sub... </AuthorUpdate>\n</IsfdbSubmission>\n', NOW(), 3)"

ProgrammingError: (1064, "You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'Aulnoy</Wikipedia>\n <Birthplace>Barneville-la-Bertran, Eure, France</Birthpla' at line 1")
      args = (1064, "You have an error in your SQL syntax; check the ...lle-la-Bertran, Eure, France</Birthpla' at line 1")


Update: Submitting a Wikipedia link with an apostrophe in the URL causes an error as well:
Submitting the following changes:
Column 	Current [110753] 	Proposed
Canonical 	Comtese d'Aulnoy 	Comtese d'Aulnoy
Legalname 	- 	-
Lastname 	d'Aulnoy 	d'Aulnoy
Birthplace 	- 	-
Birthdate 	1650-00-00 	-
Deathdate 	1705-01-04 	-
Wikipedia 	- 	http://en.wikipedia.org/wiki/Madame_d'Aulnoy
Imdb 	- 	-
Image 	- 	-
Emails 	- 	-
Webpages 	- 	-
--> -->
 
 
ProgrammingError	Python 2.4.2: /usr/local/bin/python
Sun Mar 2 19:09:00 2008

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/edit/submitauth.cgi
  217         submitter_id = SQLgetSubmitterID(submitter)
  218         update = "insert into submissions(sub_state, sub_type, sub_data, sub_time, sub_submitter) values('N', %d, '%s', NOW(), %d)" % (MOD_AUTHOR_UPDATE, update_string, submitter_id)
  219         db.query(update)
  220         
  221         PrintPostSearch(0, 0, 0, 0, 0)
db = <_mysql.connection open to 'localhost' at 837380c>, db.query = <built-in method query of Connection object>, update = "insert into submissions(sub_state, sub_type, sub... </AuthorUpdate>\n</IsfdbSubmission>\n', NOW(), 3)"

ProgrammingError: (1064, "You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'Aulnoy</Wikipedia>\n </AuthorUpdate>\n</IsfdbSubmission>\n', NOW(), 3)' at line 1")
      args = (1064, "You have an error in your SQL syntax; check the ...pdate>\n</IsfdbSubmission>\n', NOW(), 3)' at line 1")

Ahasuerus 19:11, 2 Mar 2008 (CST)

Must have been the first-ever Wikipedia apostrophe. Fixed. For now. Alvonruff 20:15, 2 Mar 2008 (CST)
Looks much better, thanks! Ahasuerus 20:32, 2 Mar 2008 (CST)

Python error in advanced search

This popped up in advanced search, in the Publication Search Form, when I tried to find all publications by a specific author for a specific publisher. The "Unknown column 'pubs.pub_publisher'" sure sounds to me like a fairly straightforward bug.

OperationalError	Python 2.4.2: /usr/local/bin/python
Tue Mar 18 11:10:55 2008

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/edit/pp_search.cgi
  220         db.select_db(DBASE)
  221         #print query
  222         db.query(query)
  223         result = db.store_result()
  224         num = result.num_rows()
db = <_mysql.connection open to 'localhost' at 820440c>, db.query = <built-in method query of Connection object>, query = "select pubs.* from pubs,authors,pub_authors wher...e '%gutenberg%' order by pubs.pub_title limit 100"

OperationalError: (1054, "Unknown column 'pubs.pub_publisher' in 'where clause'")
      args = (1054, "Unknown column 'pubs.pub_publisher' in 'where clause'")

Thanks. Dave (davecat) 11:19, 18 Mar 2008 (CDT)

Python error in "Recent Rejects"

When displaying Recent Rejects, the script errors out with the following Python error:


 
UnboundLocalError	Python 2.4.2: /usr/local/bin/python
Fri Mar 21 21:21:04 2008

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /www/isfdb.tamu.edu/root/cgi-bin/mod/recent.cgi
  138         color = 0
  139         while record:
  140                 PrintRecord(record, color)
  141                 color = color ^ 1
  142                 record = result.fetch_row()
PrintRecord = <function PrintRecord>, record = ((945923L, 'R', 21L, '<?xml version="1.0" encoding="iso-8859-1" ?>\n<Is.../AwardLevel>\n </AwardUpdate>\n</IsfdbSubmission>\n', datetime.datetime(2008, 3, 21, 12, 28, 42), datetime.datetime(2008, 3, 21, 12, 29, 14), 2L, 2L, 'bad quotes', 0L),), color = 1
 /www/isfdb.tamu.edu/root/cgi-bin/mod/recent.cgi in PrintRecord(record=((945923L, 'R', 21L, '<?xml version="1.0" encoding="iso-8859-1" ?>\n<Is.../AwardLevel>\n </AwardUpdate>\n</IsfdbSubmission>\n', datetime.datetime(2008, 3, 21, 12, 28, 42), datetime.datetime(2008, 3, 21, 12, 29, 14), 2L, 2L, 'bad quotes', 0L),), eccolor=1)
   86                         submitter = 'unknown'
   87         approver = SQLgetUserName(record[0][7])
   88         print '<td><a href="http://isfdb.org/wiki/index.php/User:%s">%s</a></td>' % (submitter, submitter)
   89         print '<td><a href="http://isfdb.org/wiki/index.php/User:%s">%s</a></td>' % (approver, approver)
   90         print "<td><i>%s</i></td>" % (subject[:40])
submitter undefined

UnboundLocalError: local variable 'submitter' referenced before assignment
      args = ("local variable 'submitter' referenced before assignment",)

Ahasuerus 21:23, 21 Mar 2008 (CDT)

Web API?

In case you haven't checked the Community Portal lately, a new editor posted a question about creating a Web API for the ISFDB earlier and even volunteered to help with creating one. I wonder what the system load implications would be, especially after we move?.. Ahasuerus 23:01, 29 Mar 2008 (CDT)

Minor Display Bug?

I'm looking at one submission that wants to change Price "$27.95" to "�20.00". It looks right that the price should change to a UK one as it's a Gollancz book, and the US price we currently have appears to be a Price-Conversion rather than a price printed on the book, but it's going to be difficult to be sure it's changing to "£" if the intermediate display just shows "?". Please take pity on us British Mods and solve yet another bloody Unicode problem? BLongley

My first attempt at posting this revealed I'd called it "Unicide". How's that for a Freudian slip? ;-) BLongley 22:12, 13 April 2008 (UTC)
I've changed the apache default charset from UTF-8 to ISO-8859-1 (which is what the ISFDB puts out in the headers). That seems to have fixed the "£" problem. Alvonruff 19:56, 14 April 2008 (UTC)
Looks a lot better now - thanks! BLongley 21:49, 14 April 2008 (UTC)


Captcha to log in?

You wrote on What's New "One of the associated changes will be the use of a captcha while logging in to reduce spambot activity." Captcha is not needed to log in but can this be enabled? I'm thinking of the hundreds of existing spam-robot accounts which still exist but should not be allowed to log in without establishing some human characteristics.

I see that creating an account has a simple captcha and so if adding one for login is not practical then could we disable the existing spambot accounts to force the spammer to create new accounts? I'd like to be able to unprotect all of the ISFDB pages we locked down to deal with the spambots.

Maybe it's ok to allow the spambots to log in. I just tested with a non-sysop login and see that to add external links that I need to solve for "76 + 7". I guess we could test that by unprotecting pages to see what happens. Marc Kupper (talk) 02:00, 14 April 2008 (UTC)

One solution to the "existing spambot accounts" problem would be to unprotect everything and then, when they start spamming again, revert the spam and ban their accounts. Since they won't be able to create new accounts any more, they will be locked out. Ahasuerus 02:12, 14 April 2008 (UTC)
I've taken a stab an unprotecting - starting with what's currently using Template:Protected. I'll need to take a break in a bit to deal with some other things. Marc Kupper (talk) 02:15, 14 April 2008 (UTC)
The following triggers are supported in the ConfirmEdit extension:
   $wgCaptchaTriggers['edit']          = false; 
   $wgCaptchaTriggers['create']        = false; 
   $wgCaptchaTriggers['addurl']        = true; 
   $wgCaptchaTriggers['createaccount'] = true;
   $wgCaptchaTriggers['badlogin']      = true;
There isn't a trigger for simple login (only the failed case), and the edit/create case sounds a little severe. Alvonruff 19:42, 14 April 2008 (UTC)

Python error during submissioin approaval

This submission errors when I try to approve it. The error text is as follows:

SQL Updates:

    * insert into pubs(pub_title) values('xxx');
    * update pubs set pub_title='The Monsters and the Critics and Other Essays' where pub_id=260735
    * update pubs set pub_tag='THMNSTRSNH1983' where pub_id=260735
    * update pubs set pub_year='1983-00-00' where pub_id=260735
    * update pubs set pub_pages='240' where pub_id=260735
    * update pubs set pub_ptype='tp' where pub_id=260735
    * update pubs set pub_ctype='NONFICTION' where pub_id=260735
    * update pubs set pub_isbn='0048090190' where pub_id=260735
    * update pubs set pub_price='' where pub_id=260735
    * update pubs set pub_frontimage='' where pub_id=260735
    * update pubs set note_id='133258' where pub_id=260735
    * select publisher_id from publishers where publisher_name='Allen & Unwin';
    * update pubs set publisher_id='195' where pub_id=260735
    * insert into pub_authors(pub_id, author_id) values('260735', '302');
    * insert into pub_authors(pub_id, author_id) values('260735', '2328'); --> -->
       
       
      <type 'exceptions.AttributeError'>	Python 2.5: /usr/bin/python
      Sat Apr 19 23:07:31 2008

      A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
       /var/www/cgi-bin/mod/ca_new.cgi in ()
        492         print "<hr>"
        493 
        494         Record = DoSubmission(db, submission)
        495 
        496         print "<hr>"
      Record undefined, DoSubmission = <function DoSubmission at 0x826956c>, db = <_mysql.connection open to 'localhost' at 811a9c4>, submission = 960212
       /var/www/cgi-bin/mod/ca_new.cgi in DoSubmission(db=<_mysql.connection open to 'localhost' at 811a9c4>, submission=960212)
        371                         artists = doc.getElementsByTagName('Artist')
        372                         for artist in artists:
        373                                 data = XMLunescape(artist.firstChild.data.encode('iso-8859-1'))
        374                                 addPubArtist(data, Record)
        375 
      data = 'Christopher Tolkien', global XMLunescape = <function XMLunescape at 0x81fd294>, artist = <DOM Element: Artist at 0x8271c2c>, artist.firstChild = None, artist.firstChild.data undefined

      <type 'exceptions.AttributeError'>: 'NoneType' object has no attribute 'data'

Ahasuerus 04:09, 20 April 2008 (UTC)

Note that in the submission above, fields Artists and Image are green, and Image has '[]'. I get similar result when I clone a pub with (credited) cover art and after removing the content of Artist and Image fields, I put one space into each. Dumpxml shows
<Artists>
<Artist></Artist>
</Artists>

and a few other empty nodes. Perhaps it is a 'feature' of the browser used? --Roglo 17:33, 20 April 2008 (UTC)

Similar error: Python error approving an e-book. It has Artists field displayed in green and dumpxml has
<Artists>
<Artist></Artist>
</Artists>

And the same editor submitted it. --Roglo 18:40, 20 April 2008 (UTC)

I was the submitter in both cases. Both were cloned from an existing publication. In both i probably cleared an existing artist and in one an existing image field, and i may wellhave done this by hitting "space", asuming that an all-whitespace field would be treated as blank. -DES Talk 02:59, 21 April 2008 (UTC)
BTW, I am now using IE7 for all submisisons -- some of my earlier submisisons were with IE6. But surely both of those are very standard browsers that we support fully? -DES Talk 03:00, 21 April 2008 (UTC)
They should be treated as blank, and if you were hitting "space" than browser is not related to the problem. For now, I would recommend hitting backspace or delete - until the bug is fixed or other cause of the problem found. --Roglo 07:44, 21 April 2008 (UTC)
The space causes the entry to fall between gaps in the error checking code. If blank, the field isn't used. If an entry has preceeding or following spaces, those are removed. The space causes the field to be non-blank, then after the space is removed, to be blank. I'll have to bullet-proof both ends. Alvonruff 16:13, 21 April 2008 (UTC)
Fixed in ca_new.cgi. Alvonruff 11:20, 1 May 2008 (UTC)

Current policy discussions

Al, just a heads up that there are a couple of ongoing policy discussions that may benefit from your review:

Thanks! Ahasuerus 16:35, 30 April 2008 (UTC)

When you have a chance, you might also want to look at ISFDB:Community Portal#Update needed to Privicy Policy. Thanks. -DES Talk 19:15, 2 May 2008 (UTC)


Minor Display Problem With Publishers

You might want to exclude Year 8888 before generating publisher pages like this one. ;-) BLongley 21:06, 3 May 2008 (UTC)

Ha! That's a good one. Fixed. Alvonruff 22:26, 3 May 2008 (UTC)

Pythin error in "Recent Integrations"

Submission 968818 is erroring out in "Recent Integrations" with the following error:

2008-05-03 22:21:59 	968819 - PubUpdate 	Circeus 	Ahasuerus 	Le Jour des fourmis 	-
2008-05-03 22:20:21
 
 
<type 'exceptions.UnboundLocalError'>	Python 2.5: /usr/bin/python
Sat May 3 22:23:23 2008

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.
 /var/www/cgi-bin/mod/recent.cgi in ()
  140         color = 0
  141         while record:
  142                 PrintRecord(record, color)
  143                 color = color ^ 1
  144                 record = result.fetch_row()
PrintRecord = <function PrintRecord at 0x8243f44>, record = ((968818L, 'I', 23L, '<?xml version="1.0" encoding="iso-8859-1" ?>\n<Is...;</Note>\n </PublisherUpdate>\n</IsfdbSubmission>\n', datetime.datetime(2008, 5, 3, 22, 20, 7), datetime.datetime(2008, 5, 3, 22, 20, 21), 2921L, 2921L, None, 0L),), color = 0
 /var/www/cgi-bin/mod/recent.cgi in PrintRecord(record=((968818L, 'I', 23L, '<?xml version="1.0" encoding="iso-8859-1" ?>\n<Is...;</Note>\n </PublisherUpdate>\n</IsfdbSubmission>\n', datetime.datetime(2008, 5, 3, 22, 20, 7), datetime.datetime(2008, 5, 3, 22, 20, 21), 2921L, 2921L, None, 0L),), eccolor=0)
   88                         submitter = 'unknown'
   89         approver = SQLgetUserName(record[0][7])
   90         print '<td><a href="http://isfdb.org/wiki/index.php/User:%s">%s</a></td>' % (submitter, submitter)
   91         print '<td><a href="http://isfdb.org/wiki/index.php/User:%s">%s</a></td>' % (approver, approver)
   92         print "<td><i>%s</i></td>" % (subject[:40])
submitter undefined

<type 'exceptions.UnboundLocalError'>: local variable 'submitter' referenced before assignment

Ahasuerus 03:25, 4 May 2008 (UTC)

Fixed. Alvonruff 12:38, 4 May 2008 (UTC)
This error has turned up again but its only visible from recent integrations on the moderators page. It also appears to be associated with publisher updates. My recent Borgo Press update created another one.Kraang 12:49, 4 May 2008 (UTC)
Okay - that one's fixed too. Now I just have to remember why there are two recent.cgi scripts... Alvonruff 17:29, 4 May 2008 (UTC)
I note that the "recent edits" link from the main page brings up "recent.cgi", and the "recent integrations" link from the moderator page also brings up a "recent.cgi" but the displays are slightly different -- the one from the main page omits the "reason" column (which seems to be almost always empty in any case). Why this difference exists, and whether this corresponds to the difference in scripts, you would know better than I. -DES Talk 22:17, 4 May 2008 (UTC)

Deleting old versions of Wiki pages

FYI, there are some questions re: the feasibility of deleting all "ancient" versions of Wiki pages here. Ahasuerus 05:04, 7 May 2008 (UTC)

User:Hall3730's User rights

When you get a chance, could you please take a look at this discussion? There may be a problem with changing user rights within the Wiki. Ahasuerus 00:11, 14 May 2008 (UTC)

Publishers bibliography page "date unknown" not displayed

Al it appears that books with "date unknown" are not turning up on the publishers Bibliography page, for example Airmont Books[5], but if you link to this title "Hidden World"[6] you'll see two "date unknown" titles. Other then this everything else appears to work fine.Kraang 17:19, 19 May 2008 (UTC)

I'm syncing up the source on the ISFDB with the CVS repository on SourceForge, and this particular fix may not have been checked in (I've already fixed this once before), so it was lost. I'll fix it up this evening. Alvonruff 17:32, 19 May 2008 (UTC)
Ok, the unknowns are now back. Alvonruff 23:37, 19 May 2008 (UTC)

Display oddity -- pub not shown on editors biblio display

Why does Whispers #21-22 not show up in the summery (or any other) bibliography of Stuart_David_Schiff, although the "editor" link from the pub to the author record works? -DES Talk 18:13, 19 May 2008 (UTC)

There's no actual "EDITOR" record in that publication. Do an advanced search for title 'Whispers' AND title_ttype='EDITOR' and you'll see that there are some missing records - issue 17-18 is missing too. BLongley 19:15, 19 May 2008 (UTC)
I went ahead and added the missing EDITOR record for Whispers #21-22. David, you'll find this happened with magazine records that were created before 2007. No EDITOR record was automatically generated, as it is now when you create a new magazine. If you find this again just edit the mag and under contents add record with the title matching the magazine title (EXACTLY), choose EDITOR in the Entry Type menu, and record the editor in the Author1 field. Hope this helps. MHHutchins 20:39, 19 May 2008 (UTC)
The fact that we have two "magazine editor" fields, one at the Publication level and one as the infamous "EDITOR" Title record, is probably the most confusing part of the application. Did we ever add an explanation to Help? I remember doing a few writeups on this issue (as successive editors bumped into it head first), but I don't recall if we ever put them in a generally accessible place. Ahasuerus 02:10, 20 May 2008 (UTC)
I see. I had heard of the EDITOR record, but not fully understod what it meant -- indeed I'm not sure that i fully understand even yet. What is the function of the EDITOR record for a magazine? Is there a problem when/if a magazine's title is changed? -DES Talk 15:23, 20 May 2008 (UTC)
I did another (brief) writeup on Tpi's Talk page the other day, but I am still sick, so it wasn't very good. We'll need to find our old posts about this topic and add it to Help. Ahasuerus 20:39, 21 May 2008 (UTC)

Creative Commons Attribution License

I don't know if you've seen this conversation, but an Ebay seller has been lifting our data, hyperlinks and all - fair game, except that nothing in the viewable web page credits ISFDB. Should someone be having a quiet word with the seller about the ATTRIBUTION bit? BLongley 18:21, 19 May 2008 (UTC)

I don't have an ebay account, which makes it hard to communicate with the seller, but if someone else would like to send a reminder, that's fine. I've seen worse. Alvonruff 23:47, 19 May 2008 (UTC)

In related activity, I noticed when checking our CC License page that it hadn't been updated for recent changes - I've corrected the typo, added the new Backup directory, but stopped short of adding isfdb.sql as that's new and probably still copyrighted to you. Please review/fix as necessary. BLongley 18:21, 19 May 2008 (UTC)

This looks good, although I moved the application scripts into a new section. I went back to the BSD license on those a couple of years ago. Alvonruff 23:47, 19 May 2008 (UTC)

Reversed variants

Al, can you check the fix you put in for the reversed titles on the approval screen? I think the AUTHORs are still reversed. (For the by title_id change, haven't checked the other type.) BLongley 20:00, 21 May 2008 (UTC)

Interesting. That means that the authors were in the correct order all along, and got reversed by the fix. It is now fixed again. Alvonruff 23:36, 21 May 2008 (UTC)


Review linking "The Weans" approval problem

Al, I linked a review to the title "The Weans" but when I try to bring it up for approval I get the message "The page cannot be found"(HTTP 404 not found).Kraang 12:37, 25 May 2008 (UTC)

Same here - I haven't had an approval work yet. BLongley 13:07, 25 May 2008 (UTC)
Sorry. That was actually caused by a later cleanup fix. Working now - you'll need to reload the moderator page to picj up the right link. Alvonruff 14:04, 25 May 2008 (UTC)
I see it now. Thanks!Kraang 14:13, 25 May 2008 (UTC)
OK, working well now. One thing though - is it possible to put the Reviewee in the left column rather than the Reviewer? It'll make it easier to check other people's submissions. E.g. comparing "Charles N. Brown" with "Jack London" didn't help. BLongley 14:57, 25 May 2008 (UTC)
Done.
Thanks! Looking good now. I've managed to link reviews to empty titles, awards, variant spellings: create missing pubs and link to them: make a link then edit the review anyway and still have the review link: now I want to try finding some suspect links and breaking them, but I'm not sure how best to find such. BLongley 15:53, 25 May 2008 (UTC)

Python error when deleting a review

I tried to delete this review but got a python error after approving the submission. I'd replaced the review with another one with the correct spelling of the book under review. How can we delete this review and clear the submission from the queue? Thanks. MHHutchins 15:49, 25 May 2008 (UTC)

Looks like a simple syntax error: 'delete from title_relationships where where title_id=790829' BLongley 15:55, 25 May 2008 (UTC)
Fixed. Alvonruff 16:02, 25 May 2008 (UTC)


Publisher merging: Approval problem

Hi Al, I seem to be having the same problem as I had with the reviews "page not found". Thanks.Kraang 01:23, 26 May 2008 (UTC)

Fixed. Be sure to reload the moderator page to pick up the right link. Alvonruff 11:02, 26 May 2008 (UTC)

Review appearing twice?

I changed the author of Davy and the Goblin; or, What Followed Reading "Alice's Adventures in Wonderland" from "Charles Edward Carryl" to "Charles E. Carryl" (plus I changed the spelling of the title_title and pub_title), which, naturally, resulted in this review disappearing from the Title page. I then linked the review to the Title record explicitly and now it appears on the title page twice. I am not quite sure how this happened, but I assume it's not what we want, right? Ahasuerus 23:36, 29 May 2008 (UTC)

Old review links weren't deleted, so each new review link adds a new one - so it shows up multiple times. Fixed that problem, and also updated the linking form to show the existing link to the title, if one exists. Alvonruff 00:21, 30 May 2008 (UTC)
Very nice, thanks! :) Ahasuerus 00:42, 30 May 2008 (UTC)
Personal tools