User talk:Ahasuerus

From ISFDB
Jump to navigation Jump to search

See User talk:Ahasuerus/Archive for discussions prior to 2021.

PLEASE NOTE:

If you're writing to inform me that you've either added a COVER IMAGE or NOTES to any of my VERIFIED PUBS, please follow THIS LINK and add it to the bottom of the list. A link to the pub record would be appreciated. Once the pub has been reviewed, I'll remove your note from the list. Thanks!

Weird link renaming

On this artist entry, the link to this site is renamed "A. E. van Vogt Cover Art Gallery". Will you figure out why, because it's definitely not anything to do with van Vogt. :) ···日本穣 · 投稿 · Talk to Nihonjoe 15:36, 22 January 2021 (EST)

Let's see...
The base URL of the "A. E. van Vogt Cover Art Gallery" is www.asahi-net.or.jp/~we8y-mrt/vanvogt/. The base URL of "Suhano Tobe's Web Site" is www.asahi-net.or.jp/~jh7a-sezk/. Our software assumes that anything hosted by "www.asahi-net.or.jp" must be a part of "A. E. van Vogt Cover Art Gallery", which is apparently not true. Let me see what I can do... Ahasuerus 15:54, 22 January 2021 (EST)
And fixed! Ahasuerus 16:22, 22 January 2021 (EST)
Thanks! Yeah, "www.asahi-net.or.jp" is an ISP, so there are likely many thousands of sites on it. ···日本穣 · 投稿 · Talk to Nihonjoe 13:49, 25 January 2021 (EST)

Man from U.N.C.L.E. publication dates.

You may have noticed that I took the Reginald number off of two publications in The Man from U.N.C.L.E. series which you have verified. I did this because the number in Reginald refers to the UK edition for these two publications which appeared before the US Ace editions. While looking at these two records, I noticed that they both have a note that states that the pub date is from the copyright and that the date is likely incorrect because of the catalog number. Reginald, in the Series Index, lists both the UK and the US ordering of The Man from U.N.C.L.E. and while the main section gives the date and publisher of the first edition (UK), the listing for the US series does give the later publication date. This is all a long way saying that Reginald indicates that the US edition of The Power Cube Affair was published in 1969 and The Corfu Affair was published in 1970. I checked the rest of the series and Reginald has the US editions of The Unfair Fare Affair published in 1969; The Thinking Machine Affair in 1970; and The Finger In The Sky Affair in 1971. The others all appear to have the correct date for the Ace printings. You're the only active verifier for these publications. Thanks. --Ron ~ RtraceTalk 14:22, 23 January 2021 (EST)

The 5 linked pubs have been updated. Many thanks! Ahasuerus 15:46, 23 January 2021 (EST)

The Compleat Werewolf and Other Stories of Fantasy and Science Fiction

You're the PV on this one. Is the submitted cover correct? The submitter is just guessing, based on their comments. ···日本穣 · 投稿 · Talk to Nihonjoe 14:09, 14 February 2021 (EST)

After checking my copy I can confirm that it is indeed the correct cover. Lucky us! :) Ahasuerus 21:23, 14 February 2021 (EST)
It was me who submitted that cover. --Username 12:09, 15 February 2021 (EST)
Ahasuerus, I left it in the queue for you, as the submission has a comment in the note-to-the-moderator about dates, and I see more info has been provided here, too. You're the only PV, so I think it would be best if you reviewed and processed it. --MartyD 12:26, 15 February 2021 (EST)
Thanks, I'll take a look. Ahasuerus 16:32, 15 February 2021 (EST)
I have approved the submission, updated Notes and changed the date from 1990 to 0000. Thanks! Ahasuerus 11:28, 16 February 2021 (EST)

Bill Breedlove and Publishers Export Co.

I noticed a 1967 novel on Bill Breedlove's ISFDB page, Devil Sex, which is by a man who wrote articles for magazines under the name William Breedlove, and I don't think he's related to the Bill Breedlove who started publishing stories 20 years later and runs Dark Arts Books. [--Username 12:09, 15 February 2021 (EST)

The author's Web page has his photos as of 2005-2006. He certainly doesn't look like he was born in the 1940s or earlier, which he would have to be in order to have published a book in 1967. Besides, it was probably a pseudonym anyway. I have disambiguated the name and added a note. Thanks! Ahasuerus 12:22, 16 February 2021 (EST)

While deciding what to do about this, I noticed Devil Sex didn't have a price or cover so I checked and only Goodreads had one, which had the price clearly displayed, so I entered it. Someone scrawled a big number on the cover so a better one is needed for ISFDB. [--Username 12:09, 15 February 2021 (EST)

I have approved the submission and added a note. Thanks! Ahasuerus 12:23, 16 February 2021 (EST)

Then I noticed the publisher, PEC, and saw online it stood for Publishers Export Corporation (or Company depending on where you look). There was 1 book by that publisher on ISFDB, Swap Academy, [1], which you PV'd, which has no cover and a date of 1970. However, [2] has it as 1969 with a different catalog ID. I nearly didn't find that page because they misspelled "swap". Also, I found exactly 1 book on picclick, Historic Orgies (1968), which shows a copyright page for this publisher and it says "Publishers Export Co., Inc.", which would agree with the note on ISFDB that says P.E.C, Inc. So I know Swap Academy is very rare but date and publisher name might need fixing, and maybe Devil Sex and Swap Academy should be on the same publisher page. Also, I found an archived 2008 discussion on ISFDB about not safe for work images which includes the cover for Swap Academy, [3]. [--Username 12:09, 15 February 2021 (EST)

After checking my copy, I updated the ISFDB record with more details. The stated copyright date is 1970, but it's conceivable that the book was published in late 1969. I'll leave it as 1970 until we find more information. The stated catalog ID is G-1201; it's possible that a later printing appeared as G-1205, but we'll need more data before we can create a separate publication record for it.
I have standardized the publisher name ("Publisher's Export Company") of Swap Academy and Devil Sex and added a note and a Web page link to the publisher page. Thanks! Ahasuerus 13:01, 16 February 2021 (EST)

Found this, [4], and copyright page has same publisher name but different address than the one in Swap Academy. Guess they moved. If only someone could find a hint of genre content in this book so PANDA BEAR PASSION can be entered on ISFDB! --Username 13:00, 16 February 2021 (EST)

Found this, [5], which is related to the above. --Username 01:30, 27 February 2021 (EST)

I have merged the publisher records and updated the affected pubs' Notes to clarify what different secondary sources state. Thanks! Ahasuerus 20:13, 27 February 2021 (EST)

I've recently started using Bookscans.com and have added a ton of covers here, including Devil Sex and Swap Academy. However, their site says the ID for Devil Sex is SP13, not PEC 13, --Username 00:33, 6 April 2021 (EDT)

Re: "Devil Sex", the cover that we currently link to has "sPECial 13" -- with "PEC" highlighted -- in the top right corner. I suspect that the person who created the Bookscans.com record assumed that the first two letters were the catalog ID, but it's possible that there is another ID on the spine. I have updated the record. Ahasuerus 10:24, 6 April 2021 (EDT)

and their cover for Swap Academy is censored with black bars over the woman's breasts and butt. Is your copy censored and, if not, is there a way for you to upload your cover? --Username 00:33, 6 April 2021 (EDT)

Since the 2008 discussion about linking explicit images was inconclusive, I have updated the record with the scan that I uploaded to the Wiki back then. Thanks for the heads up! Ahasuerus 10:29, 6 April 2021 (EDT)

Leinster's The Last Space Ship

Hi

We have two publication records for the Fell edition of The Last Space Ship, one of which you've verified. That record has a note that there is no stated publication date. The other record notes a stated publication date of November 1949. Happily, I also found a scan that verifies the date appears on the copyright page. Could you double check that your copy doesn't have a stated date. If so, I think we should probably change that publication to an unknown date and move the many of the secondary verifications to the other copy. I'm happy to handle that, but you've done the Tuck verification, so I'll leave that one to you. I'm also checking with Bob as he has also verified the copy. Of course, if both your copies actually have the publication date, we just need to combine the two records. Thanks for checking. --Ron ~ RtraceTalk 09:27, 28 February 2021 (EST)

Unfortunately, my copy of The Last Space Ship is in the part of my collection which was boxed in 2015 and I can't get to it :-( I really need to do something about that... Ahasuerus 10:26, 28 February 2021 (EST)
I see Bob found his copy. I've moved the appropriate verifications, except for Tuck. I've left that slot open on the first printing. Thanks. --Ron ~ RtraceTalk 12:39, 28 February 2021 (EST)

Rip Van Winkle

There is something wrong associated with the yellow warning message for the cover on this import content submission. It says the publication tag should be GRLSGDTVMP2003. However, I added the image from the publication upload link. And if you go to publication and click the "Upload new cover scan", the resultant file is RPVNWNKLTX1894.jpg. So something is not consistent. Thanks. -- JLaTondre (talk) 17:50, 3 March 2021 (EST)

Possibly related: I've seen this with a few author photos I've uploaded, even though there isn't a link to upload an author photo on the author page itself, and even though (as far as I know) there's no way to even know what the internal tag is for an author. It's only started happening in the last several days for me. I wonder if some recent update tweaked something? ···日本穣 · 投稿 · Talk to Nihonjoe 19:24, 3 March 2021 (EST)
It's a recently discovered issue with this type of yellow warnings -- see Bug 765, "Entering an author image URL can generate a spurious yellow warning". The way the submission display software is designed, there is no easy way for the code to distinguish between author images and publication images. I may need to remove this warning until the software is made smart enough to distinguish between the two types of images. Ahasuerus 21:58, 3 March 2021 (EST)
My issue was a publication image and not an author image. The import screen is using the wrong publication ID in its validation logic. -- JLaTondre (talk) 16:35, 4 March 2021 (EST)
Oh, I see. Sorry about the confusion! I'll take a closer look. Ahasuerus 17:17, 4 March 2021 (EST)
You may have already figured this out, but it occurred to me what is happening in this case. GRLSGDTVMP2003 (the publication id it is checking against) equates to this this pub which has a pub id of 91841. That number is the also title id of the short story being imported in this edit. So the verification check is incorrectly using the short story title id instead of the chapbook pub id. -- JLaTondre (talk) 09:05, 6 March 2021 (EST)
Thanks! The logic appears to be working correctly for NewPub submissions, but it fails for Clone/Import/Export submissions. The software checks the value of the 'Record' element in the XML payload, but, as luck would have it, 'Record' has a different meaning in Clone/Import/Export submissions. Working on it... Ahasuerus 10:32, 6 March 2021 (EST)
Fixed -- see the original example. Thanks for spotting the problem! Ahasuerus 10:54, 6 March 2021 (EST)

Magazines with no Fiction Titles

When marking a title as ignored on the Magazines with no Fiction Titles cleanup report, can it take you back to the subpage you were on instead of the "main" page? When ignoring multiple issues within a year, the current behavior is a pain. Thanks. -- JLaTondre (talk) 12:10, 26 March 2021 (EDT)

Yes, please! That is annoying. Annie 12:19, 26 March 2021 (EDT)
Done! Ahasuerus 14:22, 26 March 2021 (EDT)
Thank you! -- JLaTondre (talk) 08:24, 27 March 2021 (EDT)

Invalid Prices

Is it possible to put a yellow warning on invalid prices at submission? Something that would catch cases like this one. That way they would be more likely to get fixed at submission instead of waiting on the cleanup report. Thanks. -- JLaTondre (talk) 08:23, 27 March 2021 (EDT)

Good point -- FR 1406, "Add a yellow warning for invalid prices", has been created. Ahasuerus 18:26, 27 March 2021 (EDT)
Done -- see this Community Portal announcement. Ahasuerus 20:39, 26 May 2021 (EDT)

Series Names That May Need Disambiguation

When the series is a subseries of the reported match, I'm not sure we need it to show up on the "Series Names That May Need Disambiguation" report. Example: Crónica de Tinieblas and Crónica de Tinieblas (short fiction) were on the report (I ignored them). Since this is a pretty typical case, it could be skipped (though it is easy enough to ignore so low priority suggested change). -- JLaTondre (talk) 08:42, 27 March 2021 (EDT)

Another fair point. FR 1407 has been created. Ahasuerus 18:43, 27 March 2021 (EDT)
Done. Ahasuerus 22:03, 26 May 2021 (EDT)

Security Warnings

Please see this discussion. As browser security warnings have gotten more stringent, I have to suspect that is impacting new user signups. -- JLaTondre (talk) 08:45, 28 March 2021 (EDT)

Thanks for the heads-up! When Al last surfaced a few months ago, he said that he was planning to concentrate on the HTTPS upgrade and the Wiki upgrade. Unfortunately, he hasn't been seen since then. I'll see if I can ping him. Ahasuerus 09:27, 28 March 2021 (EDT)
I have received an update from Al. He has been investigating, but it's more complicated than it looks. Ahasuerus 12:52, 5 April 2021 (EDT)

Advanced Search Request

Greetings, Is it possible to add more lines in the selection criteria screen? There have been times I have needed to merge more than six titles. Since I am not a moderator, I have to wait for approval of each merge before I can submit the next one. Most elegant solution would be for the user to be able to continuously add search lines as needed. Thanks for your consideration. John, Scifibones 10:50, 27 April 2021 (EDT)

Thanks for reporting the problem! Originally, Advanced Search pages were limited to three rows and supported a small number of searchable fields. When I last redesigned Advanced Search, I thought that six rows whould be sufficient in the vast majority of cases, but it looks like it's not always true.
As you said, the most elegant way to address the issue would be to allow adding an unlimited number of search lines. Unfortunately, it wouldn't be viable because:
  • each search value is added to the resulting URL, and
  • URLs with more than 2,000 characters can't be correctly processed by some older browsers and even by some servers.
Bumping the number of rows to 10 is probably the easiest and least impactful compromise solution, at least until older browsers with their 2K limitations are a distant memory. Let me ask on the Community Portal.
Oh, and are you familiar with "Show All Titles" in the navigation bar displayed on Summary pages? It lets you merge a bunch of titles for a single author, which may help depending on your use case. Ahasuerus 21:24, 27 April 2021 (EDT)
Advanced Search is really the best tool for the purpose I have in mind. Many times we have myriad title records for the same map or drawing, when it occurs in multiple volumes in a series. I try to clean this up as I encounter them. I'll post a comment on the Community Portal. Thank you John, Scifibones 07:18, 28 April 2021 (EDT)
Done -- please see my last Community Portal response. Ahasuerus 15:18, 3 May 2021 (EDT)

WatchDate report not working?

This pub has the { { WatchDate } } template in its Notes. However, it doesn't show up in the report.

Querying Saturday's copy of the database: 

   select * from notes where note_note like '%{ {WatchDate} }%'; -- (remove the spaces between the curly parens)

finds 7 matches, but only 4 records are shown in the report. I note that of the 7 matches, 4 are "watchDate", 1 is "watchdate" and 2 (mine) are "WatchDate", which makes me wonder if it's maybe a case-sensitivity thing? I've not looked at the code, but the ticket implies that "WatchDate" is the "official" form: 

   Create a 'WatchDate' template and a corresponding cleanup report.

Or is the report maybe showing old data? ErsatzCulture 12:42, 28 April 2021 (EDT)

EDIT: That the template as I've used it in the linked pub gets replaced by the text message, would indicate that I've used it correctly, so maybe it's just the report not picking things up properly, or being updated? ErsatzCulture 12:59, 28 April 2021 (EDT)

Thanks, I'll take a look... Ahasuerus 14:03, 28 April 2021 (EDT)
It turns out that it was a typo in the display code. Thanks for spotting the issue!
P.S. Notes templates are case-insensitive, so we should be good on that front. Ahasuerus 14:42, 28 April 2021 (EDT)

Contento Link

It appears that the link to Contento1 has changed. The link we do have in the secondary verifications list, http://www.philsp.com/homeville/ISFAC/0start.htm, now returns a 404. The new link is http://www.philsp.com/resources/ISFAC/0start.htm. I assume it's either a software or properties change. I've gone ahead and changed it here. Thanks. --Ron ~ RtraceTalk 21:17, 3 May 2021 (EDT)

I'll take a look, thanks! Ahasuerus 21:41, 3 May 2021 (EDT)
Fixed! Ahasuerus 13:32, 6 May 2021 (EDT)
Thanks! I meant to comment on the Community Portal discussion, but I was kind of ambivalent as to which way we went. This certainly does have the virtue of being easier to update when the external urls change. Anyway, it looks good. --Ron ~ RtraceTalk 18:05, 6 May 2021 (EDT)

Cleanup report count mismatch

I just noticed that on the cleanup report page, we have "Publications with Invalid Prices (4)", but if you click through to the report, only 3 rows are shown.

Maybe this is just a caching issue where one of the errors has been fixed, but thought I should let you know in case it's something else? ErsatzCulture 13:12, 22 May 2021 (EDT)

Yes, it's a timing issue. As the "Legend" section at the top of the Cleanup Reports list says:
  • The numbers in parentheses show how many problem records were found when each report was last regenerated; the current numbers may be lower.
That being said, sometimes the report generation logic and the report display logic get out of sync when I tweak the code, as was the case when you reported the previous issue.
Ideally, each cleanup report would use the same code base when generating/displaying reports, but that would require a major rewrite. In addition, it may not be feasible in certain cases due to performance issues. While it's not a big deal if a report takes 15 seconds to compile (as long as it doesn't affect other users), the display side is more demanding. Ahasuerus 13:30, 22 May 2021 (EDT)
Ah, apologies for failing to notice that text, and consequently wasting your time :-( ErsatzCulture 13:41, 22 May 2021 (EDT)
No worries! :) Ahasuerus 13:43, 22 May 2021 (EDT)

Serial vs. Shortfiction

Ran across this record which caused me to do this search. Seems like the need for a cleanup report that find titles with " (Part" but are not SERIAL type. There may be some valid cases, but many of those look like they should be serials. -- JLaTondre (talk) 14:04, 22 May 2021 (EDT)

Oh boy! I didn't realize how many problematic records we had. FR 1417 has been created. Thanks! Ahasuerus 18:06, 22 May 2021 (EDT)
Done -- see the Community Portal announcement. Ahasuerus 12:46, 26 May 2021 (EDT)

Transmaniacon

Cover artist for this confirmed as Les Edwards via email from the artist. --Mavmaramis 14:38, 10 June 2021 (EDT)

Approved and tweaked, thanks! Ahasuerus 15:29, 10 June 2021 (EDT)

A serious problem with submitting

Hi, Ahasuerus! Is there some general problem going on or is it just my special status? I tried to submit an update for a publication (this one) and got the message / answer 'Moderator privileges are required for this option'. Christian Stonecreek 12:58, 15 June 2021 (EDT)

Not that I know of. Let me take a look... Ahasuerus 13:07, 15 June 2021 (EDT)
After setting up a self-approver user on the development server and creating a new submission, I received the same error message. It looks like one of the recent software updates may have had an unintended side-effect. Investigating... Ahasuerus 13:15, 15 June 2021 (EDT)
Yup, it was an unintended side effect of the patch which I installed earlier today. Sorry about that! Fixing it now... Ahasuerus 13:24, 15 June 2021 (EDT)
The fix has been installed. Could you please try it again and let me know if it's back to normal? Ahasuerus 13:28, 15 June 2021 (EDT)
Many thanks for the quick help! All seems to be back as it was a few hours earlier. Christian Stonecreek 23:25, 15 June 2021 (EDT)
Excellent! :) Ahasuerus 23:53, 15 June 2021 (EDT)

Top Contributors and similar reports

I'm suspicious that the nightly update of the Top Contributors and similar reports may be failing. My evidence for this is that the last user update for me on verified publications is showing as 2021-06-17 which is correct. However, my date on the reports is stuck at 2021-06-14. Not an urgent issue, but I wanted to bring it to your attention. Thanks. --Ron ~ RtraceTalk 06:52, 18 June 2021 (EDT)

You are right -- the Facebook-related changes that I have been working on caused the nightly reports to fail starting earlier this week. Thanks for identifying the issue! Ahasuerus 09:14, 18 June 2021 (EDT)
Fixed. The data will become available tomorrow morning. Thanks again! Ahasuerus 10:33, 18 June 2021 (EDT)
Alas, the reports don't appear to be updated this morning. --Ron ~ RtraceTalk 06:26, 19 June 2021 (EDT)
Well, the good news is that the fix that I installed yesterday worked as expected. The bad news is that one of the subsequent patches broke the nightly job again -- I keep forgetting to test it every time I update the core ISFDB module invoked by all Web pages and by standalone processes :-(
I have fixed the nightly job yet again and, hopefully, nothing else will get broken as I pump out more patches. Thanks for your patience! Ahasuerus 13:38, 19 June 2021 (EDT)
No worries. I'm a java programmer, so I know how things go. Hopefully, it will back tomorrow. Thanks again. --Ron ~ RtraceTalk 17:07, 19 June 2021 (EDT)
Everything looks great this morning. Thanks again. --Ron ~ RtraceTalk 06:32, 20 June 2021 (EDT)
Excellent! :) Ahasuerus 07:16, 20 June 2021 (EDT)
Looks like this glitch has returned. John Scifibones 08:09, 21 July 2021 (EDT)
The nightly reports didn't even try to run at 1am. It looks like the server crashed again around midnight. Investigating... Ahasuerus 12:03, 21 July 2021 (EDT)
The nightly reports have been rerun, so the numbers should be correct now. Ahasuerus 13:14, 21 July 2021 (EDT)

(unindent) I think this problem has cropped up again. Thanks. --Ron ~ RtraceTalk 07:20, 5 September 2021 (EDT)

It looks like the server was rebooted overnight, which interrupted nightly processing. Hopefully, things will go back to normal tomorrow morning. Ahasuerus 12:05, 5 September 2021 (EDT)
It would appear that everything was back to normal as of this morning. Ahasuerus 20:04, 6 September 2021 (EDT)

Notes/changes needed to Ryk E. Spoor entry

Hi, this is Ryk himself with a few notes on my page. Hopefully this works, I have never worked on a site that functions quite like this one

There's a few additions/changes that need to be done to my entry, as follows:

1: Jamaica Blue Magic is the second in Fall of Veils, so should be under that heading.

2: _Shadows of Hyperion_ is now out, and is the fourth in the Grand Central Arena, AKA Arenaverse, series.

3: On-Site for the Apocalypse is in the same setting as Princess Holy Aura (setting is called “The Ethical Magical Girl” and I’ll be starting the second book in the series very soon)

4: Training and Truth is in the Zarathan/Zahralandar universe (it’s part of the first book in the forthcoming “Spirit Warriors” trilogy)

5: Preparations and Alliances is part of the Arenaverse series.


Thanks, whenever you (or someone else) could do these. And thanks for maintaining this awesome database! —The preceding unsigned comment was added by Seawasp (talk • contribs) .

Done, done and done -- thanks for the updates! :-) Ahasuerus 13:19, 19 June 2021 (EDT)

A new bug

I noticed the following problem yesterday. If you are not logged in and access an author's page, the option button "Never display translations" is active. If you select it, you get the following error message. "Author not found: XXX+None" John Scifibones 08:35, 23 June 2021 (EDT)

Thanks, I'll take a look. Ahasuerus 10:07, 23 June 2021 (EDT)
Fixed! Ahasuerus 10:46, 23 June 2021 (EDT)

Problem logging in

I've been trying to log in for a week. While I could log in as Biomassbob, I couldn't make any changes, including replying to messages on my talk page or editing existing pubs. I thought the website was dead. But today I created a new account, new username and password, and using that identity, I can at least do those things. I did get a new computer, but don't know if that caused the problem. I'd like to be able to use my old identity if possible. What do I need to do? Balboabeach 10:36, 2 July 2021 (EDT)

Answered on User talk:Balboabeach. Ahasuerus 11:33, 2 July 2021 (EDT)

Марс пробуждается

Is Марс пробуждается an English work or was that a data entry mistake? Thanks. -- JLaTondre (talk) 17:56, 5 July 2021 (EDT)

Sorry, my bad! The records have been fixed. Thanks for the heads-up! Ahasuerus 18:23, 5 July 2021 (EDT)
Parents also need fixing in such cases (done now) :) Annie 20:38, 5 July 2021 (EDT)
Thanks! Ahasuerus 20:56, 5 July 2021 (EDT)

Commune 2000 A.D. / Mack Reynolds

Please check conversation at [[6]]. Thanks! --GlennMcG 02:28, 6 July 2021 (EDT)

Incident on Ath cover art issue.

It might be easier to contnue the discussion of this specific issue here as opposed to the other message board. Thanks again for your patience with this. It must be very frustrating and damn annoying. I'm guessing no one knows what caused "The Incident" in the first place ? --Mavmaramis 15:55, 22 July 2021 (EDT)

I am afraid I haven't heard anything from the hosting company. A hardware failure would be my first guess, but it could really be any number of things. I'd have to ping Al and he'd have to ping the hosting company and even then we may not learn much. Ultimately, the impact on our databases was the same whether they lost power or whether something more involved happened.
Going back to the immediate issue, do you happen to have your new scan -- which I assume is an improved version of www.isfdb.org/wiki/index.php/Image:NCDNTNTHVT1981.jpg -- stored somewhere where I could access it? Google Drive, Drop Box, etc? If you could send me a link, I could then download it to the development server and try to upload it to the Wiki using various tricks. It would be much faster than trying to walk you through the same tricks remotely. Ahasuerus 16:02, 22 July 2021 (EDT)
Quick question - are you absolutely sure that it is not uploading? According to the history page, you uploaded it 3 times today. Keep in mind that the caches can be very annoying sometimes - you WILL see the old image for awhile even though the image was swapped... So can you look at it from a different browser/device and make sure it is not just the cache thing? Annie 16:14, 22 July 2021 (EDT)
That's a very good point. Doing a full page reload -- Control-F5 under most browsers -- would be another way to make sure that your browser is not using an old version of the image. Ahasuerus 16:32, 22 July 2021 (EDT)
I am well aware of the CTRL F5 thing. Anyway the full cover now displays so everything is tickety boo. As for "The Incident" - ISFDB can't be the only site the hosting company have and they can't exactly be happy bunnies about it any more than you (and others). You've done a great job under what must have been difficult circumstances and hope you (or anyone else) hasn't had to deal with any unpleasant criciticisms. --Mavmaramis 22:04, 22 July 2021 (EDT)
Thanks for the kind words! Glad to hear things are back to normal :-) Ahasuerus 22:08, 22 July 2021 (EDT)

Henri Vernes' deathdate

You beat me to it :o) ! Linguist 07:42, 27 July 2021 (EDT).

It was a tight race! :-) Ahasuerus 10:15, 27 July 2021 (EDT)

MediaWiki

Hi. Based on your user page I assume you are also maintaining the wiki? I was just wondering if there was a hardware/stack/local modifications reason for the fairly old version of MediaWiki? Wondering since I assume there is a fair number of known security issues which have not been patched since the version hasn't been supported for years. Not a criticism, just general interest/Lokal_Profil 03:11, 19 August 2021 (EDT)

I am currently responsible for keeping the server, including the Wiki, running, but, unfortunately, my knowledge of PHP/MediWiki is virtually non-existent. Sadly, learning new things gets harder as we get older plus my health hasn't been that great the last few years :-(
For this reason, Al von Ruff, the other ISFDB administrator with developer level access to the server and some PHP experience, has taken the lead on the Wiki side -- see SR 51 and SR 184 for recent developments on that front. Unfortunately, Al has been very busy with non-ISFDB things lately, so this mini-project is currently on hiatus. Ahasuerus 12:20, 19 August 2021 (EDT)

Make Alternate names: possible bug?

Hello Ahasuerus, it seems to be possible to create multiple links between primary and alternate author names - have a look here: http://www.isfdb.org/wiki/index.php/ISFDB:Moderator_noticeboard#Alternates. Regards, MagicUnk 04:56, 19 August 2021 (EDT)

Thanks for identifying the problem! I have responded on the Moderator Noticeboard and will do more digging on the development server. Ahasuerus 12:23, 19 August 2021 (EDT)

New award

Ping to remind you (or notify you, if you missed it) about this post. I know you have a lot of things to do, and it's easy to miss things on some of the boards these days due to so many posts being made. TIA! ···日本穣 · 投稿 · Talk to Nihonjoe 19:54, 19 August 2021 (EDT)

Oops, I missed it! Response posted. Sorry about that! Ahasuerus 22:06, 19 August 2021 (EDT)
No worries. Thanks! ···日本穣 · 投稿 · Talk to Nihonjoe 11:26, 20 August 2021 (EDT)

Accidentally edited the ISFDB_Downloads page

and put an old version into it. I confused two Firefox tabs while trying to enter the data in my local installation of ISFDB.

Hit Undo button and restored your version. Hope, this was successful. Nevertheless I would like to leave a note. elsbernd 07:25, 22 August 2021 (EDT)

No problem, everything looks good. Thanks for letting me know! Ahasuerus 11:59, 22 August 2021 (EDT)

Missing link to 'moderator' page on Duplicate finder page.

Hello Ahasuerus, I think I found a bug. When on the Duplicate finder page (after having clicked on 'Check for duplicate titles' when on the title screen), and then clicking on the Moderator menu, I got a 'Not found' message. The URL behind the 'Moderator' menu-item reads 'http://www.isfdb.org/cgi-bin/list.cgi?N'. Regards, MagicUnk 15:19, 26 August 2021 (EDT)

Fixed. Thanks for reporting the issue! Ahasuerus 15:34, 26 August 2021 (EDT)

Minor software change request

Ahasuerus, A somewhat unusual situation arose yesterday while working with an awards page. I made a minor edit to a title record which already had an award linked. I noticed that the award page did not re-sort to reflect the change. The change was visible, just had no effect on sort order. I asked Rtrace about it. He explained how the software makes separate titles for award purposes. It uses those for sorting, but, displays the regular title. Is it possible to unlock the title field in the award edit screen? This seems like a better solution than deleting an award and then entering it again, just to correct the sort order. John Scifibones 18:18, 6 September 2021 (EDT)

Thanks for identifying the issue! I have responded on Ron's Talk page. Ahasuerus 20:03, 6 September 2021 (EDT)

Strange translation

Hello! I found something strange. When I enter the title page of this pub O homem que foi para Marte porque queria ficar sozinho the complete page is translated into German. Look: Der Mann, der zum Mars ging, weil er allein sein wollte Regards Rudolf Rudam 12:19, 7 September 2021 (EDT)

That's odd. When I open the linked URLs, the metadata (like field names, Notes, etc) appears in English and the titles are in Portuguese.
Thinking some more about it, Google and some browsers have features which can be configured to translate Web pages to your preferred language automatically. Is it possible that that is what's happening here? Ahasuerus 13:25, 7 September 2021 (EDT)
With my browser, it's still Portuguese. Christian Stonecreek 13:55, 7 September 2021 (EDT)
First I had used Portuguese and English as preferred language and after that I tried it in German. Each time the same result. I have observed that for a milisecond it first appears in Portuguese and then it changes into German. Other Portuguese titles appear as usual. It's the only page. Rudolf Rudam 14:13, 7 September 2021 (EDT)
Sounds like you had told your browser to always translate this page for some reason - I know I've done that by mistake a few times because the button/link for it was just there. Try another browser (and share what browser you are using so someone can try to help you unstick the translation). Annie 14:21, 7 September 2021 (EDT)
That's right -- "for a milisecond it first appears in Portuguese and then it changes into German" proves it conclusively. The ISFDB software is unable to translate titles on the fly, but some browsers can do it for you. Ahasuerus 14:29, 7 September 2021 (EDT)
Problem solved! The browser was told to translate this page from Portuguese into German. After I withdrew it, everything was ok. Thanks for your help. Regards Rudolf Rudam 15:42, 7 September 2021 (EDT)
Excellent! :-) Ahasuerus 16:40, 7 September 2021 (EDT)

Jules Verne - genre vs. non-genre

I've collected most of his works in the course of my project, but only recently read some of them. Many seem to be non-genre. A good example is Around the World in Eighty Days. The Help suggests it's hard to make the distinction prior to 1800, but this is 1860's and on. I'd guess about 10-20% are genre. The concept of threshold will kick in if we start marking them. It's a can of worms, but it might be time to go fishing. Any suggestions? ../Doug H 16:33, 15 September 2021 (EDT)

Was if possible with the technology of 1860? That’s what will define if it is genre or not, especially that early on. Eighty days is definitely genre IMO. :) Annie 16:53, 15 September 2021 (EDT)
Doing it in 80 days would have been extremely difficult -- but probably not impossible -- for a single adventurer in 1872, when the novel was published. It became a well-known challenge and people successfully circumnavigated the globe in less than 80 days just a few years later.
(after edit conflict) I am not a Jules Verne expert, but I did read a lot of his books back in the day. I agree that many were "marginally SF" at best. For example, Verne was very fond of the "stuck on a desert/tropical island" plot device to the point of writing a sequel to "The Swiss Family Robinson" (which was marginally speculative as we note in our record.) Fair enough. However, consider his other, better known, "stuck on a tropical island" novel -- "The Mysterious Island". Like "The Swiss Family Robinson", it's marginally speculative, but it also serves as a bridge between "Twenty Thousand Leagues Under the Seas", which is clearly SF, and "In Search of the Castaways", which has hardly any speculative elements. Everything is so intertwined that it's hard to draw a definitive line.
I would suggest asking on the Community Portal to see if other editors have thought about this issue. Ahasuerus 17:03, 15 September 2021 (EDT)
I tried starting a discussion, but it doesn't seem to be of much concern. I was looking into who the top verifiers for JV were, in hopes that a more direct appeal might work better. It turns out the top 8 account for 80% of the verifications. The second highest contributor dealt in German translations and two of the others in Dutch. Which made me realize that I was dealing in English translations which were notoriously bad and known for dropping slow/boring content. So before getting bogged down in how to assess the genre-ness of a work, I thought it better to go back to what I really wanted.
I have access to a reference work from which hundreds of publication entries might be generated for ISFDB. A future evaluation of any given title as non-genre would reduce the value of such effort, and if a threshold were to be applied and not reached, then much work would be for naught. So before investing time, I'd like to resolve the non-genre/threshold issues. The Policy is, of necessity, somewhat vague about about border-line genre and thresholds. What I'm looking for is something much more concrete in process, that is a way to address a particular title or author and document the result. The latter could be as simple as genre and threshold templates that could be added to titles and authors. Moderators would control / monitor their use, ensuring the appropriate discussion / documentation were included/referenced before allowing and to check prior to deletion if a future contrary opinion were raised. Defining such a control should help structure the process for actually making assessments of genre and thresholds. I've less in the way of ideas for holding the discussions - maybe the ISFDB wiki needs polls?
My plan is to propose such a 'change' in the Community Forum, as it seems easier to get people involved in discussions if there are technical details and basic principles involved rather than particular authors or titles. Any suggestions regarding points to include/exclude to frame/contain the discussion? - I'm assuming you'll weigh in on the actual proposals in the posting. ../Doug H 10:53, 23 September 2021 (EDT)
Sorry about the delay in responding! My brain is fried tonight, but I hope to answer tomorrow. Ahasuerus 21:34, 25 September 2021 (EDT)
No problem. I've been thinking about some of the background information that might be required (e.g. how many people make a quorum, based on previous decisions) and how to get it. ../Doug H 23:44, 25 September 2021 (EDT)

(unindent) One thing to keep in mind is that, as per ISFDB:Policy, we include:

  • Works (both fiction and non-fiction) which are not related to speculative fiction, but were produced by authors who have otherwise published works either of or about speculative fiction over a certain threshold (see below).

Since Jules Verne is clearly "above the threshold", this means that all of his works are eligible for inclusion even if we decide that some are not SF. The only difference is whether we flip the "non-genre" flag for some works, which will make them appear at the bottom of his Author Bibliography page. Ahasuerus 20:21, 26 September 2021 (EDT)

A lot of them won’t even fall down to the bottom for him though - as a lot of them are in the same series, they will stay above the fold. Annie 20:42, 26 September 2021 (EDT)
We haven't counted, but as few as 18 of his 80 novels may qualify as genre. Does that warrant a 'Clearly'? My second question (assuming the first is that we do classify all of them) is whether that level might put him below the threshold? As for being in the same series, I don't think it has been resolved whether Voyages extraordinaires is a title series or publisher series, but regardless, why does being in the same series make them 'genre' (above the fold)? An interesting point about the series, if a non-genre work is in a title series, does it appear in the series or in the non-genre section of the bibliography (or is that what you were getting at when you said 'above the fold'? ../Doug H 21:59, 26 September 2021 (EDT)
It does not make them genre. But the way we show series will show them above the fold on the page even if you mark them as non-genre because the series has a genre work. If Jules Verne is not a genre author, we’ve veered way too away in how we define genre authors. The whole above the threshold thing had always been about genre authors vs authors writing genre occasionally. Are you seriously proposing that the father of science fiction is actually not a genre author and should be treated as someone who occasionally wrote speculative fiction? If it bothers you so much we can add a special exception that will keep him as above threshold. But removing the adventure stories of Verne will make the DB poorer. We are not accountants - numbers are not everything sometimes. Annie 22:09, 26 September 2021 (EDT)
After all the works and work I've put in, I'm hardly likely to say he is not a genre author. I am, however, a bit afraid that someone might. All I'm looking for is somewhere official to record it, not really for Jules Verne, but for every time there's an argument (1a) that achieves a momentary consensus. There's a change log for modifications to the policy and help entries, can we get something comparable for genre/threshold and such like decisions that are considered consensual? ../Doug H 22:40, 26 September 2021 (EDT)
So we're having this conversation here in Ahasuerus Talk page and asserting that Jules Verne is 'above the threshold', but what does that mean years from now when someone points out that Mathias Sandorf is non-genre and should be removed and a moderator agrees? I still think some formalization of these decisions is a good idea. Jules Verne might make a good example or test case. ../Doug H 21:59, 26 September 2021 (EDT)
A solution could be to clearly state in the Author record Note field why a certain author is considered 'above the threshold' even if the majority of his/her work cannot be considered spec fic. For the Jules Verne case, it can be argued that his spec fic titles have been so influential, and that he is such an iconic author for the field, that that alone warrants his inclusion in the DB as an 'above the threshold' author. When doing so there's less chance (not zero though) that an unsuspecting editor and/or moderator screws up authors categorized such... MagicUnk 14:50, 27 September 2021 (EDT)
That's true, but keep in mind that we also have FR 860, "Add a 'non-genre' field to Author records". I suppose we could use this proposed field as a "Yes/No" flag while specifying the rationale in the Notes field.
Let me also point something else out. We have discussed two different approaches to the "threshold" issue. The first one would use "genre significance" as the main criterion when deciding whether an author is "above the threshold". The second one would use "proportion of genre works" as the main criterion, the idea being that we don't want our users to wonder when they come across an unlisted book by an otherwise genre author -- how would they know if it was excluded on purpose as opposed to missed?
In many cases there is no conflict between these approaches. A "significant" genre author like Heinlein or Norton typically has a high "proportion of genre works". However, there are exceptions. For example, consider George Orwell. He wrote two very influential genre works, but most of his novels (4 out of 6) were non-genre and we do not list them.
We were never able to decide one way or the other or to find a compromise solution. Ahasuerus 15:18, 27 September 2021 (EDT)

Web API

There is a line in Web API that says there are two ways to retrieve publication data, but you seem to have added a third. ../Doug H 23:52, 27 September 2021 (EDT) (No, I'm not stalking you, I've been reviewing the Wiki documentation from the top down. This was recent enough to bring to your attention.)

Updated, thanks! Ahasuerus 09:11, 28 September 2021 (EDT)
Actually, you missed the spot I'd noticed under Publication Lookups: "At this time there are two ways to retrieve publication data from the ISFDB database". :-( ../Doug H 11:51, 28 September 2021 (EDT)
Oops, missed that! Thanks again! Ahasuerus 13:12, 28 September 2021 (EDT)

Bibliographic Comment

I ran across a few Wiki pages that were part of a bibliographic verification project. While a dog's breakfast and not to be approached (yet), my approach to it was through a publication page showing Bibliographic Comments field with a link (e.g. this). I didn't recall having ever seen the field when editing a pub. Sure enough, if you edit this particular publication, the field does not exist. I suspect the field was added to support the project and when it didn't go anywhere it was taken out of the editor, but left in the display. That would mean that the value(s) cannot be removed either. Unless it is a restricted field for project members? No current problem, but a bit of back story might help me understand what would be involved in any clean-up. ../Doug H 10:56, 29 September 2021 (EDT)

There was a time when some ISFDB records didn't have "Notes" fields and Wiki pages were used instead. Wiki pages were also used to record more detailed information about certain publications even though Publication records had a Notes field. This caused a number of problems, mostly because the database and the Wiki diverged over time.
Once we added Notes fields to all ISFDB records, we started the process of migrating these Wiki-based pages to the database proper -- see the "Wiki Cleanup" section of the Cleanup Reports menu. Pages which cannot be easily migrated to the database -- typically because they are very long, tangential or contain graphics -- can be left in the Wiki and linked from the database side by entering the URL of the Wiki page as a "Web Page".
Once everything has been cleaned up, we will remove the "Bibliographic Comments" link from Publication pages. Ahasuerus 12:48, 29 September 2021 (EDT)
So the intent is to drop the database field, rather than 'edit' the publications. I'll synch my investigations up to the reports. ../Doug H 13:28, 29 September 2021 (EDT)
No, the intent is to move the information INTO the publication Notes field when possible - so editing the publication is guaranteed - either the notes OR the Web pages links (unless the data is already in both places). The only case where we should still have the wiki page is when the Notes field cannot hold the information due to its complexity. Wiki cleanup is one of the project I had been chipping at for years :) Probably need to give it a final push and clear these... Annie 13:33, 29 September 2021 (EDT)
(after edit conflict) I am not sure I am parsing your response correctly, so let me outline the usual sequence of events:
  1. Use the "Publications with Wiki pages" cleanup report to identify Publication records with matching Wiki pages. To use the first record displayed on the report as an example, the ISFDB record 2001: A Space Odyssey matches this Wiki page.
  2. Determine whether the Wiki page should be migrated to the Publication record, then:
    1. In the case linked above, the Wiki page is too long and too complex to migrate to the Note field of the publication record, so we leave it alone. We then enter its URL in the "Web Page" field of the Note.
    2. If the Wiki page is short, e.g. see this Wiki page for an example, we copy the data to the Publication record's Note field. We then delete the Wiki page.
Hope this makes sense! Ahasuerus 13:40, 29 September 2021 (EDT)
One more note: As non-moderators cannot delete pages, when you are done with a page, add this template on top of it. That puts in a list for the moderators to zap them later. I think I've seen Doug doing that before but just wanted to make sure the instructions are complete :) Annie 13:44, 29 September 2021 (EDT)
Your process is clear as far as it goes. Is there a way to mark the Wiki pages that are not to be migrated so they aren't checked repeatedly? But the conversation drifted from the original point. In my example above, the publication displays a field called "Bibliographic Comments" in bold, just after the Notes and before the link to Upload new cover scan. Given the field does not appear on the EditPub page, it is not possible to clear the field. My question about dropping the field was to determine if such values exist in the database. I still don't know if it does and would be cleared, does, but will just be ignored by changing the interface or doesn't exist in the database and is generated automatically by the interface based on the wiki. But if there are no other cleanup steps other than those listed above, I'm okay. ../Doug H 20:05, 2 October 2021 (EDT)
For the ones that get deleted, it will disappear as soon as the page is deleted.
For the ones staying - once we finish the cleanup, the whole thing will be removed by code base and it will disappear. We already did that for a few types of pages. Annie 20:25, 2 October 2021 (EDT)
Is there a list of which one's are staying or does anyone trying this start from scratch? ../Doug H 22:46, 2 October 2021 (EDT)
Wiki-specific cleanup reports ignore Wiki pages whose database counterpart has a matching "Web Page" URL. All you have to do is add the Wiki URL to the database record as a "Web Page" and it should disappear from the cleanup report. Ahasuerus 23:18, 2 October 2021 (EDT)

Handling ASINs that are no longer valid, and any impact on Fixer

Hi, just FYI, I dunno if you have any thoughts on User_talk:Anniemod#No_longer_valid_.28.3F.29_ASIN_-_what_to_do - specifically if there's any issues with Fixer and its ASIN database? Thanks ErsatzCulture 18:25, 2 October 2021 (EDT)

Please do not remove ASINs which had been checked to exist at one point post publication even when one of the places which uses them stops recognizing them. Amazon may have created them but they are identifiers elsewhere as well so Amazon forgetting them is NOT a reason to delete them from our data. Annie 19:17, 2 October 2021 (EDT)

Artwork on the interior of a dust jacket

Will you comment here, please. Thanks! ···日本穣 · 投稿 · Talk to Nihonjoe 12:08, 11 October 2021 (EDT)

Author match in the DB

Hi. I was trying to help with ISFDB:Help_desk#Schwarze_Messen:_Dichtungen_und_Dokumente. Would you take a look at this? It's not finding the existing Gustav Meyrink and adds a new one instead, so we end up with to having seemingly identical canonical names. I assume something's up with the original that I can't see, although I freely admit I could be looking directly at something obvious and am not seeing it.... Tyvm! --MartyD 14:59, 13 October 2021 (EDT)

It may be related to the hidden/unexpected Unicode characters in certain fields found by this cleanup report. Let me take a closer look... Ahasuerus 17:20, 13 October 2021 (EDT)
Preliminary results: the author name in this submission has two hidden Unicode characters, presumably a copy-and-paste artifact. Continuing the investigation. 17:24, 13 October 2021 (EDT)
Final results: the culprit is the Soft Hyphen character in the Latin-1 character set, hex value "AD". I am going to add it to standard ISFDB filter later today. Thanks for helping identify the problem! Ahasuerus 17:49, 13 October 2021 (EDT)
Done. Ahasuerus 19:46, 13 October 2021 (EDT)

Amazon image link rule

I processed this submission, which flagged the image URL 

https://m.media-amazon.com/images/M/MV5BMjA5NzkxMjI2NF5BMl5BanBnXkFtZTcwNDczOTA1Nw@@.jpg

perhaps due to the "@@"? I couldn't find anything wrong with the URL, nor did I find any alternative, so I let it through. It seems to be working properly. I am mentioning it in case I should not have let it through and need to be set straight or there's a regex/pattern analysis that needs tweaking.... --MartyD 08:27, 16 October 2021 (EDT)

That was me. Normally if I can only find an Amazon photo which is on Amazon Prime I don't use it because of that yellow warning; I decided to let this one stand because I'd fixed a lot of info for Hardy's record and wanted a photo to go along with it. The photo is in his record now so that warning doesn't seem to be a problem. Also, it took a while for someone to approve this edit; was info I changed verified by someone or was it just approved to get it out of the way? --Username 08:41, 16 October 2021 (EDT)
We currently recognize the following letters after "/images/" in Amazon URLs:
  • G
  • I
  • P
  • S
Only "I" images are believed to be stable; the rest generate the following yellow warning: Note that Amazon URLs which do not start with "/images/I/" may not be stable.
In this case, it was the fact that the image contained an "M" after "/images/" that generated a warning. I don't know if it's a recent addition to Amazon's menagerie which is supposed to be stable. If we can determine that "M" images are stable, I can add "M" to the list of recognized letters. Ahasuerus 10:38, 16 October 2021 (EDT)
@Ahasuerus: Yes, I have seen that warning, but the warning on this one is different: Unsupported formatting in an Amazon URL. Only properly structured _CR formatting codes are currently allowed., which is what made me suspect the "@@". FWIW, on https://www.amazon.com/prime-video/actor/Robin-Hardy/nm0362736 the link to that image does include formatting (...Nw@@._V1_SX300_.jpg). --MartyD 11:31, 16 October 2021 (EDT)
The "Unsupported formatting" warning is triggered by any Amazon-hosted image whose URL doesn't start with a "G", an "I", a "P" or an "S". Since this is an "M" URL, it triggered the warning.
The other warning is for "G", "P" and "S" URLs -- they are grandfathered by our software, but may not be stable in the long run and it may be best to replace them with equivalent "I" images. Ahasuerus 11:51, 16 October 2021 (EDT)
Let me make one of these much stronger - all /P/ images need to be replaced, hopefully while most of the PVs that have books with them are around to verify the covers again. These are ISBN-based links and whatever Amazon sticks on this ISBN is what we get - new printings, an image from a secondary vendor, an image based on a typo in the ISBN of a new book - whatever takes their fancy sometimes. Annie 18:24, 16 October 2021 (EDT)
@Username: I did review the information and decided the edit was ok to accept. I suspect the submission sat at least in part due to the image link warning (which required research and potentially Wiki talk page communication if the URL needed to be removed or replaced with something else) and more likely due to the change in the birth date and birth place with no explanation, requiring a moderator to go look at the available sources and see if those changes were appropriate. Submissions requiring more than a minute or two to process often get left for someone with time and inclination. --MartyD 11:31, 16 October 2021 (EDT)

Upgrading ISFDB and Testing

If you need any help (for testing or alike), do not hesitate to leave a note.

I recently read about your problems with python 2.5/openssl. Well, currently run python 2.7 with no problems I'm aware of. And openssl 1.02u or 1.1.1k elsbernd 12:58, 17 October 2021 (EDT)

Thanks for volunteering and for sharing your experience with Python 2.7!
Al von Ruff is currently looking into upgrading our version of OpenSSL since he was the one who last upgraded it a few years ago and has the experience.
On the Python side... I have been delaying upgrading from 2.5 to 2.7 because it will require testing every feature in the system, but the issue with SFE suggests that we may have to bite the bullet and do it sooner rather than later. Probably right after we finish the HTTPS changes. Ahasuerus 13:04, 17 October 2021 (EDT)
Given any thought to setting up a test suite and automating some of the testing? There's lots of developers on board, surely we can find a tester to help organize such an effort. ../Doug H 16:49, 17 October 2021 (EDT)
Not to overstate the obvious but I kinda do not see anyone stepping up and offering their help about almost anything. If there is someone who wants to assist in any way or form, I am pretty sure Ahasuerus will be more than happy to find something for them to do (or a way for their pet project to be incorporated). Annie 20:23, 17 October 2021 (EDT)
Let me first make sure that I understand what "setting up a test suite" means in this context. Does it refer to a set of automated test cases (using something like Selenium) which could then be run on test systems hosted and maintained by individual testers? Ahasuerus 09:55, 18 October 2021 (EDT)
Let me reply by asking if there is a list of every feature to be tested. And a way to test each feature. If these don't exist, there's no point to the automation. If they do, then test cases could be generated and organized into a suite of tests to cover different testing requirements. ../Doug H 23:40, 18 October 2021 (EDT)
sourceforge user ErsatzCulture has been working on some scripts using unittest (thread https://sourceforge.net/p/isfdb/feature-requests/1305/ (2019-11-03)). Maybe a little bit outdated. I haven't read the complete discussion but Ahasuerus will remember it. elsbernd 08:23, 19 October 2021 (EDT)
From the sound of it, this is unit/module style testing. I was wondering about more of a user acceptance type of test (e.g. create a pub, change the contents, variant to something, find it in various lists, unvariant, add a transliterated title, more checking in another browser, delete the pub, check that the (misspelled) illustrator disappears from the database ...). Sounds like another interesting project to put on my list. ../Doug H 15:45, 19 October 2021 (EDT)
That would be nice, but it would require a separate test copy of the ISFDB database/software. We can't realistically set one up on the main server, but it may be possible to do so within a third-party environment like Python Anywhere. The publicly available version of the ISFDB database should probably work OK with their $12/mo plan, although we would need to confirm it.
Also, ISFDB_Downloads#Installation_and_Setup describes how to set up a clone of the ISFDB database on an arbitrary computer; some of our editors have done it at home. If you want to give it a try, please let me know how it goes and whether the instructions work for you. Ahasuerus 17:22, 21 October 2021 (EDT)

Nimrod Hunt edit

Please weigh in at [7]. Thanks. --GlennMcG 16:33, 21 October 2021 (EDT)

Done. Ahasuerus 17:10, 21 October 2021 (EDT)

Author Image Not Matching Publication Tag Warning

Please see the warning message on this edit and this conversation. -- JLaTondre (talk) 16:43, 30 October 2021 (EDT)

Missing Pub Creation History

Can you please see this conversion? A publication approved today has no edit history and is not showing up in recent edits. Thanks. -- JLaTondre (talk) 10:42, 31 October 2021 (EDT)

Response posted. Ahasuerus 15:21, 31 October 2021 (EDT)

Recently Removed Secondary Verifications

Recently Removed Secondary Verifications is giving an error message. See here. -- JLaTondre (talk) 11:28, 31 October 2021 (EDT)

Fixed. It was a regression issue introduced as part of the HTTPS migration project. Sorry about that! Ahasuerus 14:49, 31 October 2021 (EDT)

12.5 Cents

There are three pubs with a $0.125 price showing up on the Publications with Invalid Prices report. This is actually a valid price for these pubs. See [8] and [9] for example. I'm not sure how it worked in practice as the bit was no longer in use by that time as far as I know, but it was the price on the cover. Could the report ignore that price? I think it's likely to be the only odd ball US one and I've seen plenty of valid errors with three decimals so wouldn't want to relax the checking to allow three decimals. -- JLaTondre (talk) 11:44, 31 October 2021 (EDT)

Originally, my devious plan was to wait until all the bad record were fixed and then add the ability to "ignore" false positives. At the rate we are going -- 600 records fixed in 1 month -- we'll probably get to that point by the end of the year. Ahasuerus 15:09, 31 October 2021 (EDT)

Author name oddity

Can you check this and make sure that this won't cause other issues (and if it does, that loophole needs closing). Annie 19:58, 1 November 2021 (EDT)

Done. Ahasuerus 11:32, 14 November 2021 (EST)

Relative URLs in notes

Hi, could you take a look at http://www.isfdb.org/wiki/index.php/User_talk:ErsatzCulture#World.27s_Edge and add any insight? That talk item came up when I submitted an edit to try to fix an alleged bad URL that came up in http://www.isfdb.org/cgi-bin/edit/cleanup_report.cgi?82 , which seems to object to relative URLs (NB: I haven't checked the code to see if there's more to it than that) - but that seems to be a bad idea IMHO. Thanks.

Done. Ahasuerus 11:31, 14 November 2021 (EST)

Interesting double author issue

Please wander by this discussion. Thanks! ···日本穣 · 投稿 · Talk to Nihonjoe 14:08, 15 November 2021 (EST)

Altov & Zhuravlyova, Ballad of the Stars

Hi Ahasuerus, according to our Publication Type rules under ANTHOLOGY, should this collection be classified an anthology? It has two authors writing separately, with only one story a joint effort. What do you think? PeteYoung 13:27, 16 November 2021 (EST)

For what it is worth, I still think that this is a collection under our rules. Annie 13:51, 16 November 2021 (EST)
Let's see... Template:PublicationFields:PubType says:
  • ANTHOLOGY. A publication containing fiction by more than one author, not written in collaboration, should be typed as an ANTHOLOGY. For example, "Late Knight Edition" contains stories by both Damon Knight and Kate Wilhelm, individually; this is an anthology, not a collection. If a book of Conan stories contains stories which are all partly or wholly by Robert E. Howard, it is a collection; if one or more of the stories is by Lin Carter or L. Sprague de Camp, not in collaboration with Howard, then the book is an anthology.
In this case we have 5 stories by Altov, 3 stories by Zhuravlyova and one co-written story. If I am reading the bolded section above correctly, even a single non-collaborative story makes a book an anthology under the current rules, which would make it an anthology.
If Annie's interpretation is different, perhaps we could take this discussion to the Community Portal to see how other editors read this part of Help? Ahasuerus 15:52, 16 November 2021 (EST)
I could have sworn that the bolded page is not there. Nah - with it there, it is indeed an anthology. I know what happened - I was looking at just the collection definition - and it is not there. Annie 16:03, 16 November 2021 (EST)
With everyone on the same page, I have changed the title/publication type from COLLECTION to ANTHOLOGY. We may also want to review the Help language to ensure that it's self-consistent. Ahasuerus 16:12, 16 November 2021 (EST)
Possibly. I'd say to have the same example in both type descriptions... The way the collection one reads, one would not even read the rest as it matches. Manages to trip me now and again (I have a deja vu here) :) Annie 16:28, 16 November 2021 (EST)

Another non-printable character

See the price Die metallenen Herrscher. It looks fine but it pings on the report. Any time I try to change it, it complains for more than one space. Feels like a weird non-printable space character in the DB. Can you check what is going on? Thanks! Annie 23:02, 16 November 2021 (EST)

And another one with the same malady: Nie wieder Krieg.Annie 00:30, 17 November 2021 (EST)
Will do. Ahasuerus 09:11, 17 November 2021 (EST)
These two prices had adjacent space characters between "DM" and the numerical value. Fixed now. Ahasuerus 09:18, 17 November 2021 (EST)
Two more down, 11 to go :) Thanks! Annie 14:40, 17 November 2021 (EST)

Editors vs Authors

Can the new pub and edit pub forms be changed to display Editor instead of Author when it is an anthology or magazine? Seem like it should be pretty straight forward to have JavaScript update the label based on the pub type. The use of Author for these pub types continually confuses new editors. -- JLaTondre (talk) 09:17, 25 November 2021 (EST)

Let me take a look... Ahasuerus 10:22, 25 November 2021 (EST)
It looks doable. FR 1469 "Change 'author' to 'editor' for anthologies, magazines and fanzines" has been created. I'll make the necessary changes it once I am feeling better. Ahasuerus 11:59, 25 November 2021 (EST)

Ongoing work on HTTPS-Support of ISFDB

I have moved the discussion to Development/HTTPS Ahasuerus 16:27, 28 November 2021 (EST)

Title Web Page Field Error

I was trying to add the URL in the moderator note of this submission into that record's web page field. I received a "URL must not contain angle brackets, double quotes or spaces" pop-up error and it won't let me submit. Something is inadvertently triggering as the URL is valid and doesn't contain any of those three. Thanks. -- JLaTondre (talk) 19:27, 22 December 2021 (EST)

Let me take a look... Ahasuerus 19:57, 22 December 2021 (EST)
I was able to add the URL by copy-pasting it in Firefox. I wonder if it's a browser issue. Which browser did you use? Ahasuerus 20:01, 22 December 2021 (EST)
Non-printable characters (and regular spaces) had been a common problem in that field for me when I copy from text and not from an address bar (I tend to do the latter while verifying the URL anyway to make sure it does not block me again) - both on title pages and on author pages (not sure about pub pages but it is possible that we lack it there as well). Can we add "trim()" to that field as we do for others? Annie 20:25, 22 December 2021 (EST)
Oh, I see. It looks like the JavaScript code behind most pop-ups strips leading and trailing commas, but the code that handles URLs doesn't. I have created FR 1472 -- thanks for the heads-up! Ahasuerus 23:22, 22 December 2021 (EST)
Done. It may require a full page reload (Control-F5 under most browsers) to take effect. Ahasuerus 14:47, 26 December 2021 (EST)

Ha ar brezoneg ?

Hello Ahasuerus, long time no see… Would you be able to add Breton to our language list ? I have just fallen upon a French SF novel translated from it, and discovered its absence in the list at the same time. Thanks a lot (and best wishes for the end of the year, by the way…) ! Linguist 05:30, 28 December 2021 (EST).

Breton is an ISO 639-2-recognized language, so that shouldn't be a problem. Give me a few minutes... Ahasuerus 13:59, 28 December 2021 (EST)
Done -- happy holidays! :-) Ahasuerus 14:01, 28 December 2021 (EST)
Thanks a lot (trugarez) ;o) ! Linguist 15:43, 28 December 2021 (EST).

Damon Knight's Orbit 15 - Edward Wellen stories

Hi,

James Davis Nicoll has reviewed an archive.org scan of this anthology, which he says lacks the 2 Edward Wellen stories in the pub you verified. I wouldn't have followed up on this, except for the fact that the recorded page numbers seem a bit unlikely, unless these stories are *very* short indeed. That isn't outside the realms of possibility, looking at a translated pub that has these and 7 other stories apparently all sharing the same page (?!?).

I did have a quick look at archive.org, and those Wellen stories aren't mentioned on the scanned contents page. (I haven't "borrowed" the book from them to verify beyond the small handful of pages they make available to anyone, as I don't agree with their cavalier approach to book copyright.) Both the pub here and the scan on archive.org are the first edition.

I did notice in the edit history there's an edit that only references these 2 stories, but that seems to be making a slight tweak to existing data, so I don't think it's relevant.

Any chance you or the other PV can double-check your copies? ErsatzCulture 11:20, 28 December 2021 (EST)

Never mind - the mystery has been solved. ErsatzCulture 14:08, 28 December 2021 (EST)
I was just about to post the same thing -- "*very* short indeed" :-) Ahasuerus 14:12, 28 December 2021 (EST)

obj2xml

Hi. I've been looking a bit at how the api could be extended to return xml data for other objects than publications. While doing this I came across the obj2xml function present in some of the clssses under commmon/. I had two questions regarding these which might help me understand the codebase better.

  1. What is the main usecase of the obj2xml functions today (which would limit how they can be reused/extended)?
  2. Why was the choice made to add a custom xml outputter in rest/pub_output.py:onePubOutput()rather than using common/pubClass.py:pubs.obj2xml()?

I havent submitted any patches (e.g. for the cleanup report tweak) for the simpel reason that I haven't worked with svn for years. But if this api extension is possible in the way I've thought of it then that might make it qorth setting all of that up. /Lokal_Profil 14:43, 29 December 2021 (EST)

Here is how a typical submission is created and processed:
  1. A submission type-specific HTML data entry form is displayed and the user populates it.
  2. When the user clicks "Submit", JavaScript-based popup validation is called.
  3. If popup validation is successful, a submission type-specific script in the "edit" subdirectory is called.
  4. The script validates the submitted values by calling the "cgi2obj" method in the class that matches the submission type.
  5. If validation passes, the script creates an XML payload and files it in the "submissions" table.
  6. A moderator pulls up the submission and displays it using a submission type-specific function in "common/viewers.py".
  7. When the reviewing moderator clicks "Approve", the link calls a submission type-specific filing script in the "mod" subdirectory.
SUBMAP, which is defined in common/isfdb.py, links these steps in a somewhat messy (but documented) way.
If memory serves, "obj2xml" was originally supposed to be called in step 5, but it didn't work out because different submission types create different XML payload. For example, NewPub and EditPub create different payloads even though they both work with the same record type, "publication".
In addition, please note that, for historical reasons, the function "XMLescape" in common/library uses a non-standard escaped value for apostrophes when internal XML payloads are created. The XML emitted by the REST API uses the standard escape value instead.
Hope this makes sense. I am once again under the weather at the moment, so I am not in a good position to implement the recently created FRs with requested enhancements, but I agree that they are desirable. Ahasuerus 17:57, 29 December 2021 (EST)
Many thanks for the explanation, background and pointers for further info. Sounds like what I was thinking ahould be possible so will try to set up svn and a local isfdb dev environment for testing.
No worries about either of the FRs. Take care and get well soon! /Lokal_Profil 14:39, 30 December 2021 (EST)

Author Merges & Alternate Names

I merged "J. C. H. Rigby" (148444) and "JCH Rigby" (287823) resulting in the "new" JCH Rigby. One of them (I'm not certain which at this point, but think it was the 287823 entry) had an alternate name of Charlie Rigby. When I did the merge, the alternate name was lost. That caught me by surprise. When a merge is done, the normal expectation would all the information is merged. Title merges handle variants so I would have expected author names to handle alternate names. Ideally, it should handle the alternate name. If not, there need to be a big warning that further action is required. I wonder if we have lost alternate names because of this... -- JLaTondre (talk) 09:06, 1 January 2022 (EST)

Retrieved from "https://isfdb.org/wiki/index.php?title=User_talk:Ahasuerus&oldid=621445"