Difference between revisions of "ISFDB Bug List"

From ISFDB
Jump to navigation Jump to search
(Note location for bug reports)
(cleanup)
Line 1: Line 1:
 
{| width="100%" cellpadding="0" cellspacing="5" style="vertical-align:top; background-color:#F5FFFA; padding:0;"
 
{| width="100%" cellpadding="0" cellspacing="5" style="vertical-align:top; background-color:#F5FFFA; padding:0;"
−
! <div style="margin:0; background-color:#ff8080; font-family:sans-serif; font-size:120%; border:1px solid #084080; text-align:left; padding-left:0.4em; padding-top:0.2em; padding-bottom:0.2em;">This page has been deprecated and is here for archival purposes only. Please file new Editing Bug reports at: [http://sourceforge.net/projects/isfdb/ http://sourceforge.net/projects/isfdb/].</div>
+
! <div style="margin:0; background-color:#ff8080; font-family:sans-serif; font-size:120%; border:1px solid #084080; text-align:left; padding-left:0.4em; padding-top:0.2em; padding-bottom:0.2em;">This page has been deprecated and is here for archival purposes only. Please file new Bug reports at: [http://sourceforge.net/projects/isfdb/ http://sourceforge.net/projects/isfdb/].</div>
 
|}
 
|}
  

Revision as of 17:49, 31 August 2008

This page has been deprecated and is here for archival purposes only. Please file new Bug reports at: http://sourceforge.net/projects/isfdb/.

Open Bug List

Note that there are separate pages for reporting bugs within ISFDB.

Fixed Bugs

  • EditBug:10093 Page numbers are not saved when cloning novels OPEN I normally enter the starting page number for novels so that we can later see which books have unnumbered pages before the start of the novel also to help accurately determine the novel's length. It turns out for clone-pub with novels that the parent title entry is available for editing, the page number is included in the XML blob, but it’s not integrated. Marc Kupper 16:06, 27 Jan 2007 (CST)
Well, NONGENRE is not a valid publication type (Soapbox warning: In fact, I greatly dislike the nongenre title type, as it leads to a bunch of special circumstances, just like this one, and we'll have more for NONGENRE collections, and NONGENRE anthologies, etc... At some point we'll have a teetering tower of special cases to handle nongenre types, when it's really an attribute of a NOVEL). The real problem is that you can't clone a NONGENRE novel. The code has been hacked so that nongenre novels are cloneable.
  • EditBug:10095 8888-88-88 is not a valid MySql date NAB Template:PublicationFields:Year includes "8888-88-88 means that the book has been announced but not published and is reserved for well known examples like Last Dangerous Visions." Apparently if you set the title_copyright to '8888-88-88' MySql does not recognize it and stores 0000-00-00 in the field meaning we should pick another code to flag publications that were announced but apparently never published. Marc Kupper (talk) 18:49, 1 Feb 2007 (CST)
There's been a transcription error someplace. '8888-88-88' has a bad month and day, so it's not a valid SQL date. We've been using '8888-00-00' as the unpublished designator. I've changed Template:PublicationFields:Year to reflect that. Alvonruff 06:09, 3 Feb 2007 (CST)
This has now occurred on another merge: the Pratchett merge in the queue now. I'll keep both on Hold till the bug is fixed. Mike Christie (talk) 09:42, 21 Jan 2007 (CST)
  • 10064 FIXED Make variant allows you to point a record at a non-existent parent, by entering a record number that doesn't exist. The approval screen warns you of this, so this isn't that serious an issue, but the tool really shouldn't let you enter an invalid record. Mike Christie (talk) 10:40, 23 Dec 2006 (CST)
  • EditBug:10085 “Remove Titles From This Pub” should not allow deleting parent FIXED When you select “Remove Titles From This Pub” for ANTHOLOGY, COLLECTION, FANZINE, MAGAZINE, or OMNIBUS only content items other than the parent are listed and available to be deleted. If the publication type is NOVEL, NONFICTION, CHAPTERBOOK the parent title is available and editors can end up orphaning publications. I suspect the parent title should never be listed for remove-titles regardless of the publication type. Marc Kupper 12:56, 10 Jan 2007 (CST)
  • EditBug:10089 Python error adding essay with two authors FIXED There was an existing anthology that has two editors and was listed in ISFDB with no contents. I got a copy today and entered the contents into ISFDB. On hitting [Submit Data] I got the attached Python error. As I had quite a bit of contents data I switched to another window and first edited/saved the pub-metadata, no problem, and then added the introduction, and replicated the python error. I then backed up, removed the second author from the introduction, was able to submit/approve and did a second edit cycle to add the second author. After that I went back to the first window, deleted the introduction, and was able to save/approve the remainder of the contents. I did not do any further testing to see if it’s related to that the parent publication has two authors or anything similar. Marc Kupper 00:07, 18 Jan 2007 (CST)
As an editor also ran into this tonight I did some more testing and the bug only seems to apply to the first title record in the contents and hit me because my anthology had two editors and the first record was the introduction. It seems I can add-title and add the introduction with both authors as a second or later title. Marc Kupper 03:41, 18 Jan 2007 (CST)
Someone deleted the parent of the variant. I've added some error handling for that case, and also pointed the variant at the corrent title. Alvonruff 13:01, 14 Jan 2007 (CST)
At present we're preventing the submission of blank TAGS. Later, when the editing is beefed up, we'll give feedback to the user, and allow resubmission. Alvonruff 18:55, 9 Jan 2007 (CST)
  • update pubs set pub_pages='387' where pub_id=12803
  • delete from notes where note_id=21331
  • update pubs set note_id=NULL where pub_id=12803
  • update submissions set sub_state='I' where sub_id='182761'
  • update submissions set sub_reviewer='757' where sub_id='182761'
  • update submissions set sub_reviewed=NOW() where sub_id='182761'
The page count is updated correctly but the Notes field was silently deleted. Fortunately I caught this as an editor and I were discussing what the notes should say and I put the notes back in though have no idea how many other notes have been lost recently. Marc Kupper 13:04, 9 Jan 2007 (CST)
Update on this - Al has removed the code to deal with allowing editors to blank out the notes field and that fixed the silent delete of the notes. Here's the updated SQL where you can see the stuff to delete the notes no longer happens: Marc Kupper 13:20, 9 Jan 2007 (CST)
  • update pubs set pub_pages='386' where pub_id=12803
  • update submissions set sub_state='I' where sub_id='182931'
  • update submissions set sub_reviewer='757' where sub_id='182931'
  • update submissions set sub_reviewed=NOW() where sub_id='182931'
  • 10061 WORKSFORME Adding or deleting Web page URLs in an Author's record causes the following error:
 
AttributeError Python 2.4.2: /usr/local/bin/python
Thu Dec 21 14:17:40 2006 

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  
   30         
 
   31         old = authors(db)
 
   32         old.load(int(new.author_id))
 
   33         
 
   34         submitter = getSubmitter()
 
old = <authorClass.authors instance>, old.load = <bound method authors.load of <authorClass.authors instance>>, builtin int = <type 'int'>, new = <authorClass.authors instance>, new.author_id undefined 

AttributeError: authors instance has no attribute 'author_id' 
      args = ("authors instance has no attribute 'author_id'",)

Hitting the "Back" button and immediately resubmitting the page works fine with no error reported. Ahasuerus 14:21, 21 Dec 2006 (CST)

Look's funky. Maybe it was a one-time server problem. I can't get it to fail. Alvonruff 20:41, 21 Dec 2006 (CST)
  • EditBug:10067 Date width data entry OPEN The date fields in the New Collection/Anthology/Magazine edit forms and are displayed inconsistently. The original date fields only display 8 characters, while the date fields that are added when you hit the "Add Title" button are wider and let you see the full YYYY-MM-DD format. This has already confused one new user, who thought that he only had enough space to enter YYYY. I suggest we change them all to be big enough to see YYYY-MM-DD. Ahasuerus 11:18, 25 Dec 2006 (CST)
  • EditBug:10077 ISFDB login name is case sensitive FIXED An issue that came up with one of the new editors is he entered his user name in all lower case. MediaWiki apparently converts the first letter to upper case before adding it to the user table and its login is case insensitive for the user name. The issue is that ISFDB does a case-sensitive compare and the editor could not log in using the all lower-case name he had entered initially. Marc Kupper 04:54, 5 Jan 2007 (CST)
I've looked through the mw_users table of MediaWiki, and the user names are not stored in lower case (only the case of the first character is changed), and it does not perform a strictly caseless search (I can not log in as "AlvonRuff", for instance). Rather than try to guess the MediaWiki algorithm for user name munging, I'm instead put a note on the login page that the first letter of their login name will be capitalized. Alvonruff 05:29, 9 Jan 2007 (CST)
  • EditBug:10074 Pub Editing, New Authors FIXED I selected "New Novel", then realized that I was entering a Magazine instead and changed its type to MAGAZINE. I was a little concerned that it may not create the EDITOR Title correctly, but I could see EDITOR in the SQL statements, so I assumed everything was fine. I then pulled up the newly entered magazine issue via Edit Pub to enter its contents. After going through the motions and hitting Submit, I encountered the following Python error:
  • EditBug:10079 ISBN display/validation issue FIXED This is both a display and edit bug but I'll file just this report. There was a publication record with ISBN 0932322240 which is displayed as 0-932322-24-0. Someone edited it and changed other fields but not the ISBN. When they saved submitted the record the ISBN validator did not recognize it because the checksum is wrong (it should be 7) and so the record was saved as 0-932322-24-0. I saw the pub-update in the queue, spotted the hyphenated ISBN, and decided to follow up on this. The display logic displays “0-932322-24-0” as “ISBN-13: 978-2-322-240-” and there’s no ISBN: line. I hit edit-pub again and the logic gave me “978-2-322-240-” in the edit field! Marc Kupper 01:15, 6 Jan 2007 (CST)
  • EditBug:10081 ISBN display/validation issue FIXED It's impossible to edit this title as it has a # at the start of the ISBN and it makes it throws a Python error. (Found by User:Unapersson. This is kind of a display bug, but since it stops you editing I'm reporting it here. The workaround would be to clone the pub and delete the offending information, then delete the old pub. However, I am sure I have seen # working successfully in the past -- has there been a recent change to the parsing of this field? Mike Christie (talk) 14:33, 6 Jan 2007 (CST)
Only about abzillion lines have been added recently to get all of the ISBN requirements working. Fixed.
  • EditBug:10075 ISBN13 truncation FIXED An editor submitted a hyphenated ISBN13 which looks ok through the entire queue but gets silently truncated when the publication record was created as 978-1-55143-481-0 is 17 characters long and the pub_isbn field is 16 characters wide. I ended up doing a work-around of accepting the submission and filling in the ISBN with the ISBN10 value. Marc Kupper 14:13, 29 Dec 2006 (CST)
The ISBN field is now 32 bytes. Alvonruff 11:00, 6 Jan 2007 (CST)
  • EditBug:10080 Unmerge on short story has unexpected results NAB I am pretty sure this is a dataloss bug, rather than as designed. I had edited a collection of Aldiss's to have "Brian Aldiss" throughout, rather than Brian W. Aldiss; then I realized I'd made the same mistake we've been discussing since the other pubs for those short story titles probably used "Brian W. Aldiss". So I unmerged a short story, "Out of Reach". It had been in three pubs: this one, this one, and this one. After the unmerge it disappeared from two of them; it's only in the second one of the three above now. Prior to the unmerge three pubs were listed; I was expecting that after the unmerge, two more title records would be generated, each pointing to one of the three pubs. Mike Christie (talk) 09:58, 6 Jan 2007 (CST)
Moved to new unmerge tool feature request 90097. Alvonruff 10:36, 6 Jan 2007 (CST)
  • 10076 FIXED After submitting and approving a "Delete Pub" request, I alt-tabbed to a different window and spent some time there. I then went back to the ISFDB window and used the back button to bactrack to the original, now deleted, publication. After hitting "Back" a few times, I received the following Python error:
TypeError Python 2.4.2: /usr/local/bin/python
Sun Dec 31 13:09:42 2006 

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/deletepub.cgi  
   44         
 
   45         publication = SQLGetPubById( int(sys.argv[1]) )
 
   46         print "<b>Request to Delete:</b> <i>%s</i>" % publication[PUB_TITLE]
 
   47         print "<p />"
 
   48 
 
publication = 0, PUB_TITLE = 1 

TypeError: unsubscriptable object 
      args = ('unsubscriptable object',)

Ahasuerus 13:13, 31 Dec 2006 (CST)

  • 20070 OPEN Adding a shortfiction title to a series drops it from the shortfiction list. If a short story is included in a series it’s shown with [SF] (short for shortfiction I assume) and is no longer listed in the shortfiction list. I suspect it should be both in the regular series list and listed as shortfiction. I can’t remember if the SF work was also dropped when we had the “Shortfiction Series” mechanism. See Owl Goingback for example. Marc Kupper 18:34, 26 Dec 2006 (CST)
My assumption is that it should not be displayed twice. Alvonruff 18:35, 26 Dec 2006 (CST)
Typically, encyclopedias list titles only once unless they do some kind of elaborate indexing a la Tuck. In out case, we already have Alphabetical and Chronological listings for those who want to see everything with no regard for series info. I don't think the current behavior is a bug, although it could possibly be made into a feature request. Ahasuerus 19:22, 26 Dec 2006 (CST)
Ok – Let’s leave that one alone. It’s gone both ways with me because sometimes when the vt Titles are not duplicated I’ve spent time staring at the list wondering why I can’t see the title and other times when the vt Titles have been duplicated I wondered about the extra listing. :-) Marc Kupper 23:25, 26 Dec 2006 (CST)

12 • The Lord of Tranerica (Complete Novel) • (1939) • serial by Stanton A. Coblentz (aka Lord of Tranerica) [as by Stanton A. Coblentz ]

The display logic correctly finds the "aka" data, but also adds the unnecessary "as by" data. Ahasuerus 19:43, 23 Dec 2006 (CST)

Yeah. I've spruced up the part that notes a title difference (hence the aka clause), but noticing an author difference is non-trivial. Okay, it's trivial for works by one author, but gets complicated for works with two or more authors, considering order (none guaranteed) and at least one of the authors may be a pseudonym. And since it's merely annoying... Alvonruff 20:20, 25 Dec 2006 (CST)
Well, the change is clearly a net plus, so I guess we can just call it a "low priority display issue" and move to the end of the queue. I hesitate to close it because it's something that out users may inquire about and it's probably better to have it here where it can be found than buried in the archives. Ahasuerus 21:20, 25 Dec 2006 (CST)
  • 20059 NAB If an Author has written a Serial, but that Serial hasn't been reprinted in book form, the Serial will not appear in Long or Short Works. For example, see Victor Rousseau's biblios and then pull it up via the Title merge screen. The latter finds The Beetle Horde and The Phantom Hand, which are not accessible i the regular Long/Short Works pages. This is likely a superset of Bug 20052. Ahasuerus 01:38, 23 Dec 2006 (CST)
The display logic works correctly, but the page header was confusing since it said "Summary Bibliography" and not "Short Works Bibliography", so I thought I had landed in a different place. Closing the bug. Ahasuerus 23:22, 27 Dec 2006 (CST)
  • 20047 CLOSED From the "Find Stray Titles" screen, clicking on "Long Works" leads to a "Mangled Argument" error; "Short Works" produces an "Internal Server Error", as do the other bibliographies. Mike Christie 08:38, 25 Nov 2006 (CST)
"Find Stray Titles" was zapped a while back. Closing the bug. Ahasuerus 23:12, 27 Dec 2006 (CST)
  • 20036 CLOSED There is a problem with either the data or the display code for some magazines/editors. For all issues of Galaxy after July/69, the editor shown in the publication page gets no credit in his author biblio. This affects Ejler Jakobsson, James Baen, John J. Pierce, etc. For issues of Imagination, William L. Hamling was the only editor but his biblio lists all the issues as Stray Publications. For issues of If, Mar-Sep/52 is OK but Nov/53 through Feb/59 don't credit the editors in their biblios. For Amazing Stories, the problem starts with Jul/39 when the editor chages to R. A. Palmer and his biblio lists the issues he edited as Stray Publications but works fine when the editor changes again in Jan/40. I haven't checked beyond that or the other periodicals. PortForlorn 23:38, 17 Oct 2006 (CDT)
    • We have Ray Palmer listed under four different names:
      1. R. A. Palmer
      2. Ray Palmer
      3. Raymond A. Palmer
      4. Raymond Palmer
    • I wonder if the ISFDB1-to-ISFDB2 conversion didn't get all the pseudonym/variant name associations and created "stray records"? Ahasuerus 15:22, 18 Oct 2006 (CDT)
      • I checked some of the other editors and many of the problem ones also had probable pseudonyms (e.g., Baen, Pierce, Hamling, Gold) but neither Jakobsson nor Damon Knight (If, Feb 59) suffer from this condition. PortForlorn 21:10, 18 Oct 2006 (CDT)
Turns out that this is being caused by missing EDITOR records for many magazine Publications. The current workaround is to fix it manually. Not a display bug, so I am moving it to the Projects page. Ahasuerus 20:08, 27 Dec 2006 (CST)
Not a display bug but rather a data bug. I will add a request for a database scan for commas in URLs to the "General Cleanup" project. Ahasuerus 19:05, 27 Dec 2006 (CST)
I deleted it a few seconds after integration. There is a more serious editing problem with adding new magazines. I suspect an XML impedance mismatch that's easily fixed. Alvonruff 15:44, 10 Jul 2006 (CDT)
  • 20021 FIXED Very long titles such as 28367 (which should be deleted, but I'm leaving it there for illustration) don't display properly; the third line of the title gets hidden behind the large white box containing the publication list. View at some resolution like 800x600 to see what I mean. grendel|khan 21:06, 10 Jun 2006 (CDT)
Is grendel|khan still around? The reason I ask is that I’m confused by the bug report. When looking at pages I don’t think I have seen ISFDB use any CSS absolute positioning or Z-layering that would cause stuff to get hidden. I viewed the test page at 640x480 and 800x600 with both FireFox and IE and also in a 25x80 text window using Lynx. I did not see anything to that looked like a problem and also did not see anything in the HTML that could potentially cause a problem. Marc Kupper 14:04, 27 Dec 2006 (CST)
  • 20020 FIXED The regular (non-advanced) title search only shows the first author for each returned title. For instance, search for "Good Omens", and it appears to be written by Neil Gaiman, as opposed to Neil Gaiman and Terry Pratchett, which can be a bit confusing. grendel|khan 09:23, 9 Jun 2006 (CDT)
  • 20019 FIXED When reviews come up in the regular (non-advanced) search, they don't appear any differently from other titles, which can make a confusing mass of titles where there's really only one. Cover art is prefixed with "Cover: "; perhaps the search results could return "Good Omens (Review)" or "Review of Good Omens". grendel|khan 09:23, 9 Jun 2006 (CDT)
  • 20015 FIXED When viewing the "Chronological" version of an Author's bibliography, e.g.

H. G. Wells', clicking on "Editing Tools: Titles" results in an empty form with only the "Merge Selected Records" button displayed. Ahasuerus 18:39, 25 May 2006 (CDT)

  • 20011 FIXED Publications not of type ANTHOLOGY, COLLECTION or OMNIBUS still show a "Contents" header. Unless there's a plan for something to be eventually entered there, it probably shouldn't display for types like NOVEL, where it will always be empty. grendel|khan 19:26, 16 May 2006 (CDT)
    • Actually, novels can have "forewords" (3 in that recent edition of "Silverlock"), "afterwords", etc. It's not terribly common, but definitely possible. Ahasuerus 19:30, 16 May 2006 (CDT)
      • Check out publication CWLSXFRVSH2005; it includes the title which references it. Perhaps it would be useful if by default, when adding a new publication to a NOVEL, its contents were set to that title. I'll make a note up in the Editing section. grendel|khan 22:09, 16 May 2006 (CDT)
  • 10069 WONTFIX Changing "Brett Mcbean" to "Brett McBean" in Pub Edit (both the Pub Author and the Title Author) seemed to go through, but when you look it up, the case never changed. Ahasuerus 19:49, 26 Dec 2006 (CST)
You can change which author you want from the pub edit app, but you cannot (and should not) be able to modify the author's canonical name from the publication editor. That can only be done from the author editor. The pub edit app can select a new author (for instance if you changed from Brett Mcbean to Robert A. Heinlein), and if the new author doen't exist it will create one (for instance if you changed from Brett Mcbean to Xvstrqc). When you integrated, an SQL search for "Brett McBean" found "Brett Mcbean", so it didn't add a new author since it already existed. We do NOT want to do a caseless search, because once the author name has been changed to "Brett McBean" and someone later enters "Brett Mcbean" we don't want to add the errant author to the database, so we need "Brett Mcbean" to match on "Brett McBean". Once the author names are correct, doing a caseless search saves us work later when people get the author case wrong, as they are automatically corrected. Alvonruff 07:51, 27 Dec 2006 (CST)
As a FYI on this author-case thing – an editor just submitted this magazine with the note that a particular’s authors name was spelled in lower case but on approval I saw that ISFDB kicked it to standard case as the author already had a record... Marc Kupper 04:04, 30 Dec 2006 (CST)
  • EditBug:10070 Add-pub gets No-Tag error FIXED I was on a title, did an add-publication, and on submit got “*** ERROR: No tag specified.” I ended up adding the publication via bew-pub and merge-title though this title still gets “*** ERROR: No tag specified” if you try to add other pubs for testing. Marc Kupper 22:31, 26 Dec 2006 (CST)
I think I'm going to have a moratorium on addressing "it would be nice" bugs for a while, as we're getting to the point where fixing them actually causes more problems than the original problem. 10070 stems from the "prevent users from blanking the tag" field, and it was a complete bitch to detect the difference between an automerge edit (which doesn't have a tag) and a user edit (which should have a tag), requiring the hack of adding extra hidden variables to the editing apps. Alvonruff 07:41, 27 Dec 2006 (CST)
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>) 
  686                 if self.form.has_key('pub_isbn'):
 
  687                         self.pub_isbn = self.form['pub_isbn'].value
 
  688                         if validISBN(self.pub_isbn):
 
  689                                 self.pub_isbn = string.replace(self.pub_isbn, '-', '')
 
  690                                 self.pub_isbn = string.replace(self.pub_isbn, ' ', '')
 
global validISBN = <function validISBN>, self = <pubClass.pubs instance>, self.pub_isbn = '2-266-11140-X' 
 /www/isfdb.tamu.edu/root/cgi-bin/edit/pubClass.py in validISBN(isbn='226611140X') 
   34         remain = sum % 11
 
   35         if remain == 10:
 
   36                 if int(isbn[9]) != 'X':
 
   37                         return 0
 
   38         else:
 
builtin int = <type 'int'>, isbn = '226611140X' 

ValueError: invalid literal for int(): X 
      args = ('invalid literal for int(): X',)
  • EditBug:10072 ISBN validator blow up FIXED I wanted to change this pub from a Novel to Anthology and got the following error. As a temporary fix I changed the ISBN from 075640410X to 075640410X-xxxx so the validator code would not try to process it. See the bug notes for the python details. Marc Kupper 03:43, 27 Dec 2006 (CST)
  • 20004 OPEN Generic "Author" searches apparently search all Work and Publication author fields. If there are no matching Work Authors to display, but there are matching Publication Authors, then the Author's name is listed in the search results list, but when you click on it, it comes up empty. This may be related to the previously reported Hartmann the Anarchist problem. Ahasuerus 20:10, 3 May 2006 (CDT)
  • 20010 FIXED Variant titles display in the <title> tag with HTML tags (as a <b> tag inside the title is not evaluated). See The Rolling Stones (vt Space Family Stone) for an example. The generated HTML is 
    <title>Bibliography: The Rolling Stones [<b>vt</b> Space Family Stone]</title>
    but should probably be 
    <title>Bibliography: The Rolling Stones [vt Space Family Stone]</title>
    grendel|khan 16:07, 16 May 2006 (CDT)
  • 20035 OPEN Eric Kotani is a pseudonym for Yoji Kondo. The display for Kotani's long works shows the novel Supernova (which was coauthored with Roger McBride Allen) as "Supernova (1991) with Roger MacBride Allen [as by Roger MacBride Allen and Eric Kotani ]" This seems to indicate that both Allen and Kotani are pseudonyms. Is this the expected behavior of the isfdb? Also, the long works listing for Yoji Kondo shows several other collaborative works with the same format (i.e., ... with other author [as by other author and Eric Kotani]) but Supernova doesn't show in this list and those other titles don't show in Kotani's list. PortForlorn 0:35 12 Oct 2006 (EST)
    • In this case, http://www.isfdb.org/cgi-bin/ea.cgi?Yoji%20Kondo and http://www.isfdb.org/cgi-bin/ea.cgi?Eric_Kotani need to be merged by making "Kotani" the primary name and making "Kondo" (which he has used on a few works) an alternate name. I'll leave it as is for now and make the change in a couple of days once we are all on the same page. As far as the issue of displaying pseudonyms goes, yes, that's how it currently works. It can look misleading, but there are complex cases involving non-trivial pseudonyms (e.g. somebody writing as two people or v.v.) and that's the best we have been able to come up with so far :( Ahasuerus 01:25, 12 Oct 2006 (CDT)
OK, sorry 'bout that report then since I'm trying NOT to post database errors here - only functional ones. Although it did look a bit strange to see Kotani shown in the AS BY clause of Kotani's listing as if Kotani were Kotani's pseudonym. PortForlorn 1:16 14 Oct 2006 (EST)
  • 20067 FIXED Serials are now occasionally displayd twice, e.g. see Charles Dye's bibliography. BTW, Bug 20059 is still there, but I am not sure if it's related. Ahasuerus 21:24, 25 Dec 2006 (CST)
  • 20066 FIXED AFAICT, if there are no Long Works to display, Shortfiction series titles are now displayed after all other Shortfiction Titles and without any indication that they belong to a series. For example, Noel Loomis has two shortfiction series consisting of 2 and 4 Titles respectively, which now show up at the bottom. On the other hand, T. Jackson King's The Forty-Seventh Florescence is displayed correctly since (?) it has a Long Work to anchor it.
This is probably the last long/short issue left from the split bibliographies. I have have put the series display back to the way it was in ISFDB1 - one fiction series area, mixing long and short titles. Collections and short fiction are annotated. There is no specific short fiction series ary any longer. This works well for authors who write mostly short fiction or mostly long fiction, but is "interesting" for those who write both within the same series (The ISFDB2 way would split titles in the same series, depending on if they were long or short, which seems bad. The ISFDB1 way further complicates series numbering). Let's wear this for a while and see how it fits. Alvonruff 03:47, 26 Dec 2006 (CST)
  • * 20065 FIXED The Summary Bibliography display logic correctly displays reviews of collaborations in the reviewER's biblios, but gets it wrong in the reviewEE's biblios. For example, in C. M. Kornbluth's bibliography the "unknown"'s review of Kornbluth/Pohl's Space Merchants is listed as:
The Space Merchants (1975) by unknown and Frederik Pohl

Ahasuerus 19:54, 25 Dec 2006 (CST)

  • 20062 FIXED Variant Titles in Long Works are now displayed twice, once by themselves and once under their parent title. See Dinosaur Park, a Variant Title of The Thirteenth Majestral, on Hayford Peirce's Summary page. This used to happen with Short Works only, but I suspect that the current Long Works/Short Works integration project gave this bug a chance to spread. This is causing user confusion, so we may want to zap it earlier rather than later. Ahasuerus 13:08, 25 Dec 2006 (CST)
I had seen those before and was undecided on if it’s a bug or feature. Something that may help is when the vt is displayed in the main list to show
Dinosaur Park (1994) Variant Title of The Thirteenth Majestral (1989)
This would let people know that the system is deliberately showing the title twice and the linkages between titles. Marc Kupper 13:27, 25 Dec 2006 (CST)
  • 10066 FIXED An attempt to approve a submission (138911 -- still there) that Cloned a Publication without an associated Title record resulted in the following Python error:
IndexError Python 2.4.2: /usr/local/bin/python
Sun Dec 24 17:23:05 2006 

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/cv_new.cgi  
  138                 if TagPresent(merge, 'Parent'):
 
  139                         title_data = SQLloadTitle(int(value))
 
  140                         print 'Automerge with title <a href="http:/%s/title.cgi?%s">%s</a>' % (HTFAKE, value, title_data[TITLE_TITLE])
 
  141 
 
  142 
 
HTFAKE = '/www.isfdb.org/cgi-bin', value = '0', title_data = [], TITLE_TITLE = 1 

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

Ahasuerus 17:25, 24 Dec 2006 (CST)

  • 10062 NOTABUG Dates are zeroed (00-00-00) when adding titles to a publication.

You can see it in the cDate field under the ContentTitle node in the XML shown on submit. Note the stories with a date of (0000) on the Ramsey Campbell entry.

--Unapersson 10:28, 22 Dec 2006 (CST)

Retrieved from "https://isfdb.org/wiki/index.php?title=ISFDB_Bug_List&oldid=126619"