Date Read in Your Books diffesr from Actual Date Read

I just spent several hours chasing what I thought was an error in my newly imported date read values. I decided to go to a Years only for my books. When I imported the values for my whole library, I was shocked to see the dates were 1 year less than the correct values.

After chasing this for a while I imported a complete year-month-day for the books. All with the month and day being January 03. I imported this new set overnight and found my dates were all shown as January 02.

Equally mysterious is that I found the correct January 03 in my full library export for the books.

I want to emphasize that this bug sounds innocuous if you lose a day in a year-month-day format, but if you lose a whole year while using the year only format it is very significant!

Please provide a solution so I can go to my preferred yearly format.

IS ANYONE IN THE BUG FIXING STAFF LISTENING?

I have worked around this problem for a while by adding 2 days to my Date Read field but stumbled into it again today when I inadvertently imported all of my dates read in year-month-day format where the day in every record was the first day of the first month for each year the book was read.

Guess what! all dates were placed as the 31st of December for the prior year.

So, what is displayed in my catalog are dates read for the year preceding the proper year.
This totally messes up any analysis of books read by year.

As previously, when I export my complete library, the proper dates are in the export.

Please look into this, so I don't have to do the + 2 days add to all of my dates and re-import.

So is this happening during IMPORT or during EDITING of the books individually? It sounds like it may be during import but I want to be sure before I chase it down.

>4 conceptDawg: I imported all of my books and created the complete library with these new "date read' dates. The import went totally fine. I can export the total library, and everything is fine.

The problem shows up when I go to the Your Books of my library. Every book has a finished date which is 2 days less than proper. So, a date of 2000-01-01 shows up as 1999-12-31.

Just go to my library and there shouldn't be any books with a date less than 2000. Similarly, there shouldn't be any with 2004. This problem isn't just in the library alone. If you go to look at the tag cloud and look for books read in 2000 there aren't any.

In my earlier posting I didn't wait for a solution from you, and I just created all dates as being in 2000-01-03, etc. and went on with life. My guess is that likely any date read for anyone stands the chance of being 2 days wrong.

I just checked this last observation by looking at books that actually have the date read as 2022-08-30. They are showing up in my library at 08-29.

So, there must be a 1-day subtraction for every date read, and not just for 01-01 dates.

>5 JoeB1934: Ok. Thanks. That helps a lot.

I have been working around this problem by simply making my YYYY-MM-DD format for setting Date Read on import by making the day field 1 day longer than actual. However, not hearing about any solution I decided that all I really care about is the year and that I would truncate all dates to year only.
Wouldn't you know .... the imported years all had one year less than I wanted, so I added a year artificially to the real year and re-imported.

Got the correct year but I am disgusted that 1.5 months after reporting the bug LibraryThing still requires tricking the system to get the correct value.

>8 JoeB1934: LibraryThing is a small company with very few developers. Bugs that are not urgent can take time to get fixed, because there are other priorities. Some get fixed right away, some take years. Some never get fixed.

The status of this bug was listed as "needs discussion", so it was not showing up on the list of active bugs. I've changed it to "reopened".

Hollow laugh at the idea of a minor bug taking more than a month being "disgusting".

>10 lorax: :-)

It took me a while to understand the bug report. It looks to me like it's just a display bug in what shows up in "Your books" and "Edit book" where Date_Started and Date_Read are shown as "a day early" compared to what ends up in the export file.

I.e. one of my books exports as:

Date_Started: 2022-09-27
Date_Read: 2022-09-27

but when viewed here
https://www.librarything.com/work/5484227/details/31050338

it's displayed as

Date_Started: 2022-09-26
Date_Read: 2022-09-26

Anyone seeing other bugs here?

Fun fact: I've never noticed this bug, since I normally just enter "Today" in the two fields and since LT runs a different time zone than me, I've never given it a thought to check the date.

What happens if we input a Start or Finished date in "Edit book" and export the book. Is a day added?

PS Kudos to >1 JoeB1934: for discovering this bug. I can see how irritating it is when you import books.

>11 bnielsen: I have considered this as more of an irritant when I was just losing a day, but when I discovered that I lose a YEAR when I decided to make my Date Read just be the year then I got more agitated.

Am I the only member that just enters the year that a book was read? If so, they might be unaware of the bug.

>12 JoeB1934: I have a script that looks very carefully at the TSV export tile. Most of the time it finds errors or omissions of my own fault, but it also repairs the TSV in a few places. I think I may have noticed this bug once or twice before without realising what was going on. I.e. comparing "review date" with "date read" and wondering why "review date" was a day off. I blamed it on "review date" or time zone or something, without investingating further). "Review date" isn't part of the TSV export, so most people (as in 99.999%) won't see that something is off by a day. ETA: Rubbish. I checked my code and it seems that I'm pushing "Review date" a day forward in my own database, because some books had "entry date" set after "Review date". Oh well. I'll just leave my code alone.

So, yes, you could very well be the only one to be hit by the bug hard enough to notice.

I wonder if this has always been how things worked or if something changed (python2 / python3 / a new version of a date/time library etc)? I might have a couple of older versions of an export file lying around somewhere.

Nice. This is weird. I went to the Edit book page:
https://www.librarything.com/work/5484227/details/31050338 and change Start and Read from 2022-09-26 to 2022-09-26, i.e. no change. But now the export shows the correct date.

Date_Started: 2022-09-26
Date_Read: 2022-09-26

I have enough experience developing commercial software that I fully recognize that there needs to be recognition that there isn't an immediate fix to a bug when it is reported. What I find especially unacceptable is the lack of feedback from LT staff about the status of a bug fix, even as little as "yes there is a bug, and we will work it into our schedule."

I don't expect a promised fix date, just some sort of acknowledgment. Currently it seems like dropping a stone into a deep well and hoping to hear a splash.

>15 JoeB1934: I have enough experience developing commercial software that I fully recognize that there needs to be recognition that there isn't an immediate fix to a bug when it is reported. What I find especially unacceptable is the lack of feedback from LT staff about the status of a bug fix, even as little as "yes there is a bug, and we will work it into our schedule."

Remember, this is a free service. LibraryThing staff usually acknowledge and try to replicate bugs, so that the developers can look at them -- but it's a lean organization, and the more time spent interacting with users, the less time spent fixing the code! And what may seem like a "simple" bug to us sometimes turns out to be incredibly difficult to fix.

Are there things I wish LibraryThing would fix? Sure, I have a long list, starting with the complete and utter disaster that is their handling of editions of pre-copyright works, which is so bad that it makes the site almost useless with regard to those books (The Variorum Chaucer version of A Treatise on the Astrolabe is not A Treatise on the Astrolabe, and the facsimile edition of the second quarto of Hamlet is not Hamlet!). I'd love it if there were a way to vote on which bugs we'd try to fix first, too. But it's free. If it truly doesn't suit your needs, possibly you should be looking at another service. For something free, I think LibraryThing is amazing -- both in what it does and in how it responds to users.

>15 JoeB1934: You got that at >7 conceptDawg:. See the LT icon in front of the name? That’s someone from LT. They are aware of the bug.

If all you want is an acknowledgement, not sure what else you want. Your “being disgusted” comment contradicts a bit your later statement that all you want is an acknowledgement though - it sounds like you expected it to be fixed asap.

If you look at all the announcements and bugs being fixed, LT are doing a lot of work with very limited staff - and a lot of what they do is happening behind the scenes (developing their library product and tinycat and all the prep work for LT2 and so on). There is just as much as they can get to at a day.

>16 waltzmn: I agree that LibraryThing is an amazing world class product, and I will continue to use it because of the incredible features. I am simply pointing out that there is a bug that might affect a few of the millions of users.

Recently I reported a bug in the Tag Mirror which was very important, and the response was terrific with a lot of discussion about that feature and acknowledgment with a bug fix. I sent a message complimenting the staff member for his help.

In this case I will just move on with my work around and let the LT staff do what they want to do with the issue.

>16 waltzmn: The issue of not all editions of Hamlet being the same isn't a bug but a combining issue and by LT combining rules if the only differences are a few words here and there then they are probably the same work. If whole scenes are left out that would probably make them different. If you think they should be separated then you are free to do so, but someone else could come along and combine them again

>19 jjwilson61: The issue of not all editions of Hamlet being the same isn't a bug but a combining issue

I never said it was a bug. It is not a bug in the sense that it doesn't work. It is a defect, in that it makes the data very difficult to use. I did not bring it up as an example of a bug; I brought it up as an example of not always being able to get what one wants. I use LibraryThing despite what is, for a student of old manuscripts and prints, a terrible problem.

if the only differences are a few words here and there then they are probably the same work.

It's a facsimile. I don't know the process used, but think of it as photograph of the Huntington Library copy of the Second Quarto. Completely unedited. There's an ink splot on the next to last page. Page H.2 has a name that was blurred by a printing defect. The catchwords and quire numbers are preserved on several pages.

A photograph of a person is not the person. I suppose you could argue that if someone transcribed that facsimile of the second quarto, that would be a copy of Hamlet, but the facsimile is not. It's a photograph. What's more, the purpose of that facsimile is not to perform or read Hamlet. It's to help textual critics edit Hamlet.

I'll give you an example from one of the most famous lines in the play. The edition I cited, the second quarto, reads
     O that this too too ſallied fleſh would melt,
The first folio (as found in my facsimile of the first folio) reads
     Oh that this too too ſolid Fleſh, would melt
Neither one marks the line number or any such thing! The quarto doesn't so much as mark acts.

Yet in the Signet Shakespeare, this line (I.ii.129) reads
    O that this too too sullied flesh would melt

(This is Shakespeare's actual reading, I think, although I could defend "sallied" -- i.e. assailed, assaulted. "Solid" is patently wrong.)

So why does the Signet read "sullied"? Because, by comparing the folio's stupid reading "solid," and the second quarto's better "sallied," and the (extremely corrupt) first quarto's "grieu'd and sallied," they concluded that an original reading of "sullied" was the one most likely to explain the errors of the three surviving witnesses. "Solid," for instance, is a case of the extremely common phenomenon of a copyist (in this case, the typesetter) not understanding the text at a quick glance and producing an easier, more obvious, wrong reading.

To reconstruct what Shakespeare wrote, which is Hamlet, we can only compare the three surviving witnesses, the first quarto, the second quarto, and the folio, none of which is Hamlet but all of which are corrupted versions of it. What is more, even if you think the text of, say, the second quarto is Hamlet, a facsimile of one copy of it is not the second quarto, it's just a facsimile of one copy. To textual critics, the people who worry about reconstructing the text, lumping the witnesses with the edited text is perverse.

but someone else could come along and combine them again

Yes. It has happened to me in other cases. It's why I wish the philosophy of LT were different, because this really and truly does make life difficult for bibliographer types and textual critics! But I agree it's not a bug.

>20 waltzmn: Nope, it is a design decision you disagree with. :) that does not make a defect. You are trying to use the site for something it is not designed for (tracking separate editions). It is not a defect - in the same way in which the fact that you cannot drive your car through the ocean like a boat is not a defect.

>20 waltzmn: On the other end of the spectrum one could argue that every book is a bit different from every other book. I.e. I have some books with dedications from the author. Maybe they should be considered different from the un-dedicated books? (I'm just saying that the lines between works / editions / books are blurry.)

>22 bnielsen: On the other end of the spectrum one could argue that every book is a bit different from every other book. I.e. I have some books with dedications from the author. Maybe they should be considered different from the un-dedicated books? (I'm just saying that the lines between works / editions / books are blurry.)

I think you're missing the key factor here, and that is the difference between mass and individual production -- and the differences which arise as a result.

Every modern individual copy of a particular edition of a particular book is individual in the sense that it's made of different atoms, but the goal of production (usually) is to make them as alike as possible. And even different editions are probably trying to reproduce the same original text although perhaps in a different format.

That's not so in earlier books, particularly before the age of printing. Forget Hamlet. It apparently became the basis for people's response because people know what it is. Let's instead take the Canterbury Tales. There are roughly eighty manuscript copies of all or part of the tales. If you want to check my data, I am using A Catalogue of Chaucer Manuscripts, Volume II by M. C. Seymour.

The first manuscript listed is National Library of Wales (Aberystyth), Peniarth 392D. Generally known as the Hengwrt Manuscript. It contains only the Tales. It was written by Chaucer's scribe Adam Pynkhurst probably shortly after Chaucer's death. It lacks the end of the Parson's Tale. And it never included the Canon's Yeoman's Tale. (The order is also strange, but I won't go into that.)

Second is Alnwick, Duke of Northumberland MS. 455. It lack the beginning, it lacks many leaves throughout -- and it adds a new tale, a second Merchant's Tale, "The Tale of Beryn."

I'll skip most of them to get to the very last, Columbia University (New York) MS. Plimpton 253, described as "One parchment bifolium (1 and 8 of a quire of 8) containing the Merchant's endlink, the prologue to the Franklin's Tale... and lines 709-52 and 1198-272 of the Franklin's Tale."

Some editions of the Canterbury Tales include the "Tale of Gamelyn," which is not by Chaucer, in place of the Cook's Tale, which Chaucer never finished.

No major printed edition of Chaucer follows any of these. All include the Canon's Yeoman's Tale, none includes Beryn, and if they include Gamelyn, it is with some sort of special treatment. (The Riverside Chaucer, the standard edition, omits both Beryn and Gamelyn.) Certainly none of them contain only the few lines found in Plimpton! And photographic facsimiles exist of some of them (e.g. Hengwrt; none of Plimpton that I know of).

No two major manuscripts of the Canterbury Tales contain the same text, even in the limited way that the Second Quarto and First Folio of Hamlet do (i.e. they both intend to convey Shakespeare's text, although they fail). The tales included are different, the order of tales is different, they have different links. You cannot approximate the text of Northumberland from Hengwrt, or Hengwrt from Northumberland, and no decent scholar would try. In the volumes of the Variorum Chaucer, the book prints the text of Hengwrt (usually, and for the tales where it exists) -- and devotes half the page, or more, to the readings of other manuscripts -- precisely because they are different copies.

If I go to a library and ask for the Donaldson edition of Chaucer, and they give me the Benson, I may ask if they have Donaldson but will probably accept Benson if they don't; I agree that they're basically the same book. But if I ask for the Hengwrt photo facsimile, and they give me the Ellesmere facsimile, they are not the same book and will not be used for the same purpose.

I know I'm going to lose this fight, because LibraryThing is not organized by manuscript scholars -- but manuscript scholars will agree with me.

>23 waltzmn: What you want and what LT is is so far apart that there really isn't anything to fight about. What makes LT unique is the ability to have your individual book records but to also be able to merge them into a work record so that people can compare their libraries. If each book is unique, that destroys what makes LibraryThing LibraryThing

>23 waltzmn: LibraryThing can handle all of that though, can't it? If someone has catalogued The Canterbury Tales (Hengwrt Manuscript) and you know that doesn't contain the Yeoman's Tale, then it rightfully wants separating out from the 'complete/unabridged' versions, add a work-relationship (has/is expanded version) and a disambiguation note. Repeat for other manuscripts wrongly combined (your expertise comes into play). Just because we are ignorant doesn't make the site broken :-P

>23 waltzmn:, >24 jjwilson61: and >25 Nevov: I agree with all of you. Some of the books in my library are different translations of books, that I also have in the original untranslated version and comparing them is often quite interesting. I like the fact that I can rather quickly find all reviews on LT on the same work. (Most of the time there are not very many reviews, so browsing through all of them for an interesting angle or observation is easy).

>23 waltzmn:
Thanks for the detailed information about The Canterbury Tales. I have a couple of translated versions which just say: "Translated from The Canterbury Tales" or "Translated from "The Wife of Bath's Tale" but doesn't say anything about which version. I can see that I've noted one of them lacking "The Prioress' Prologue and Tale" and "The Tale of Melibee", but that might be the editors choice.

Thanks to all the people who didn't get mad at me here. :-) This has been very disturbing for me. I'm going to make a few (well, more than a few :-) comments, but the goal is not to fight.

>24 jjwilson61: What you want and what LT is is so far apart that there really isn't anything to fight about. What makes LT unique is the ability to have your individual book records but to also be able to merge them into a work record so that people can compare their libraries. If each book is unique, that destroys what makes LibraryThing LibraryThing.

This is not at all what I am saying. I would agree, e.g., that every modern edition of Hamlet should file together. In the case of Hamlet, all I was proposing is that source texts -- facsimiles of the first quarto, second quarto, and first folio -- be separated from resultant texts (Signet Shakespeare edition or whatever).

A pretty good rule of thumb is that if one would not accept one edition as a substitute for another, it's not the same book. If I go to a library to seek an edition of Hamlet to read, or to act, or for a class, and they give me a copy of the facsimile of the Second Quarto, I will not take it, because it's not an edited text and is not really suitable for performance. But if I planning to edit an edition of Hamlet, then I would need the second quarto text and would find a modern edition completely useless. Since no one who wants one can use the other for that purpose, they are separate works.

>23 waltzmn: LibraryThing can handle all of that though, can't it? If someone has catalogued The Canterbury Tales (Hengwrt Manuscript) and you know that doesn't contain the Yeoman's Tale, then it rightfully wants separating out from the 'complete/unabridged' versions, add a work-relationship (has/is expanded version) and a disambiguation note. Repeat for other manuscripts wrongly combined (your expertise comes into play). Just because we are ignorant doesn't make the site broken :-P

You're right -- in theory. LibraryThing has the mechanisms (although another tier in the hierarchy series/work would make this easier -- something that would combine all Canterbury Tales editions as a "group" or something, while not combining them as a "work," would be really nice). In practice, it doesn't happen. When I managed to get the facsimile edition of the Hengwrt manuscript of the Canterbury Tales (which, as noted above, doesn't have the Canon's Yeoman's Tale and has a strange order and has lost its ending), it showed up as just another Canterbury Tales edition -- no way to tell it from, say Neville Coghill's translation. I split it; someone else promptly recombined it. Eventually, after much pleading, the combiners people let me have my way. But that took days.

My Variorum Chaucer edition of The General Prologue still lumps with the "general" General Prologue. So does the Wife of Bath's Tale. (Oddly, the Manciple's Tale seems to be correctly split; either that, or else no one has published it separately.) I have, I think, fourteen Variorum Chaucer volumes. I have facsimiles of dozens of other books, such as Piers Plowman and Shakespeare's First Folio (not the same as a "Complete Works of Shakespeare," because it lacks Pericles and The Two Noble Kinsmen). Based on trying to get Hengwrt split from the Canterbury Tales, I wouldn't live long enough to convince people to split them all.

>26 bnielsen: I have a couple of translated versions which just say: "Translated from The Canterbury Tales" or "Translated from "The Wife of Bath's Tale" but doesn't say anything about which version. I can see that I've noted one of them lacking "The Prioress' Prologue and Tale" and "The Tale of Melibee", but that might be the editors choice.

It probably is the editor's choice. The Neville Coghill translation mentioned above, for instance, omits Melibee because it's in prose, and The Parson's Tale because it's in prose and deadly dull :-). (It was the Official Boring Sermon That Medieval Books Had to Have. :-) The Prioress's Tale was doubtless omitted from your edition because of its anti-Semitic traits. (Keep in mind that Edward I had expelled all Jews from England; what Chaucer knew about Jews approximated what a twentieth century American playing "Cowboys and Indians" knew about Navajo civilization.... Compared to the people of his time, Chaucer was very modern and universalist -- e.g. he a lot of people at the time despised him because he was so pro-Feminist. But the prejudices of the time still applied sometimes.)

If your translation is from about 1870 to 1935, it was probably taken from Skeat's edition of the Canterbury Tales, which was mostly based on the Ellesmere Manuscript (which, like Hengwrt, was written by Chaucer's scribe Adam Pinkhurst, but copied several years later, after someone had gotten Chaucer's papers in better order). From about 1935 to 1987, a translation would be based on F. N. Robinson's edition of Chaucer, which used the Ellesmere order order but with a lot of individual readings from Hengwrt. And if it's after 1987, the text is probably that of the Riverside Chaucer, which still uses the Ellesmere order but has even more readings from Hengwrt and other manuscripts.

This, again, is why I want to separate these things. The facsimiles of Hengwrt, and Ellesmere (which I don't have), and other Chaucer manuscripts are not used to print readers' editions, or for translation; they are used as the basis for edited texts which someone then translates.

>27 waltzmn: Ah, nice information about what the translated version may be based on. In a foreword it says "based on newer editions of The Canterbury Tales". And the translation is from 1978, so Robinson's is probably the source or at least the main source. Thanks!

After reading the recent messages I had to re-confirm that I hadn't wandered into a different reading universe. Best of luck to you in resolving these quite complex issues.

Now if you wanted to discuss extending the standard genre list to some meaningful categories, I'm all in for that.

>29 JoeB1934: Now if you wanted to discuss extending the standard genre list to some meaningful categories, I'm all in for that.

I already lost that fight. :-) When genres were proposed, I wanted several genres that were helpful for scholars, and got none of them. :-) They eventually added "music," which is something, but no one would go near my other categories.

And having read my posts here, you probably can guess why. :-)

>27 waltzmn: I don't have a problem with that. You should argue for those separations and LT will be better for it. The problems may lie in distinguishing the different books based on just the title and author.

Bump.

Back to my original issue I would like to add:

I also want to mention when I do an export the date read field has the +1 added to the correct year.

>33 JoeB1934: Based on the fact that the import and export dates match, this does seem like it's a display issue. Can you answer a couple questions:
- Are you still seeing this; can you give me an example of a book that you're seeing the wrong read date on.
- What timezone are you set to (settings -> account settings), and have you recently/ever changed it?

>34 knerd.knitter: My location has always been Denver. ALL books are subject to this effect. To give you an example, the import of some books I read in 2023 are below.

TITLE AUTHOR (L,F) ISBN TAGS RATING DATE READ
Lessons: A novel McEwan, Ian 593535200 5Star,Literature,Mystery,Literary Fiction,Literary Mystery,Historical Mystery,Historical Fiction,Magical Realism,Coming of Age,Books about Books,Espionage,LGBTQ+,Relationships,Family,Romance,Non British Mysteries 4.1 2024
Our Missing Hearts: A Novel Ng, Celeste 593492544 5Star,Literature,Mystery,Literary Fiction,Literary Mystery,Historical Fiction,Coming of Age,Speculative Fiction,Dystopian,Racism,Art,Relationships,Family,Non British Mysteries,Science Fiction 3.9 2024
Tomorrow, and Tomorrow, and Tomorrow: A novel Zevin, Gabrielle 593321200 5Star,Literature,Mystery,Literary Fiction,Literary Mystery,Historical Fiction,Coming of Age,Books about Books,Fantasy,Art,Relationships,Family,Romance,Non British Mysteries,Science Fiction 4.1 2024
Lessons in Chemistry: A Novel Garmus, Bonnie 038554734X 5Star,Literature,Mystery,Literary Fiction,Historical Fiction,Magical Realism,Womens Fiction,Fantasy,Relationships,Family,Romance,Non British Mysteries,Religion,Science 4.2 2024
Dandelion Wine: A Novel (Grand Master Editions) Bradbury, Ray 553277537 5Star,Literature,Mystery,Literary Fiction,Historical Fiction,Magical Realism,Coming of Age,Fantasy,Horror,Speculative Fiction,Family,Non British Mysteries,Biography & Memoir,Science Fiction 4 2024
The Maid: A Novel Prose, Nita 593356152 5Star,Literature,Mystery,Literary Fiction,Historical Mystery,Historical Fiction,Coming of Age,Relationships,Family,Romance,Non British Mysteries,Crime Fiction,Suspense 3.8 2024
Once There Were Wolves McConaghy, Charlotte 1250244153 5Star,Literature,Mystery,Literary Fiction,Magical Realism,Relationships,Romance,Scotland,Non British Mysteries,Crime Fiction,Suspense,Science 4.1 2024

Notice that the date read in every case is 2024.

If you go to my library, look at the Collection 'Books in 2023'. You will see these books, as well as others, all with the proper date read, and not 2024 from the import.

Now, I will display the export file of the Collection 2023 books.

Book ID Title Sort Character Primary Author Rating Date Started Date Read
235588595 Out Here on Our Own: An Oral History of an American Boomtown 1 Anselmi, J. J. 4 2024
235588596 All the Broken Places: A Novel 1 Boyne, John 4.5 2024
235588597 The Librarian Spy: A Novel of World War II 5 Martin, Madeline 4 2024
235588598 A Sliver of Darkness: Stories 3 Tudor, C. J. 4 2024
235588599 The Twist of a Knife: A Novel 5 Horowitz, Anthony 4 2024
235588601 A World of Curiosities: A Novel (Chief Inspector Gamache Novel, 18) 3 Penny, Louise 4.5 2024
235588602 The Magnolia Palace: A Novel 5 Davis, Fiona 4 2024
235588603 The Whalebone Theatre: A novel 5 Quinn, Joanna 4 2024
235588605 Case Study 1 Burnet, Graeme Macrae 4 2024
235588606 We Begin at the End: Crime Novel of the Year Award Winner 2021 1 Whitaker, Chris 4 2024
235588607 Companion Piece: A Novel 1 Smith, Ali 4 2024
235588609 Once There Were Wolves 1 McConaghy, Charlotte 4 2024
235588610 The Maid: A Novel 5 Prose, Nita 4 2024
235588611 Lessons in Chemistry: A Novel 1 Garmus, Bonnie 4 2024
235588613 Tomorrow, and Tomorrow, and Tomorrow: A novel 1 Zevin, Gabrielle 4 2024
235588614 Dandelion Wine: A Novel (Grand Master Editions) 1 Bradbury, Ray 4 2024
235588615 Our Missing Hearts: A Novel 1 Ng, Celeste 4 2024
235588619 Lessons: A novel 1 McEwan, Ian 4 2024
235588621 Wednesday's Child (Inspector Banks series Book 6) 1 Robinson, Peter 3.5 2023
235590028 A Death in Tokyo: A Mystery 3 Higashino, Keigo 4 2023

Notee the date read here are all 2024.

I should mention that last year when I discovered this it was at a time where I was providing date read in YYYY-MM-DD format, but the month was always 01 and the day was always 01. The result was that after import all dates read were in the previous year.

It has always seemed to me that you could replicate this yourself by using any test library and to do a universal import such as I have done.

>35 JoeB1934: Yes, I was able to replicate it; I just like to get an idea of exactly where you were seeing it, i.e., for a specific book, what you put on the import, what you see on the catalog page, and what you see on the export.

As far as your time zone setting, I have no way of knowing whether it changed between the time you reported it and now, which is why I was asking that.

>36 knerd.knitter: Do you need anything else from me?

>37 JoeB1934: Not at this time. I will let you know.

Verified that this issue occurs with year only read dates.

>39 knerd.knitter: So an import with yyyy-mm-dd date read doesn't lose one day in import? Of course, for most people losing one day isn't significant.
So, if I make make all my dates read at least beyond jan 1 I will be OK?

>40 JoeB1934: I don't believe that's what she said.

>40 JoeB1934: No, I believe all dates will lose a day, at least on our side of the international date line.

>42 knerd.knitter: Thanks

I believe this is resolved. But timezones scare me, so I'm not going to swear to it. Please let me know JoeB1934

>44 knerd.knitter: It is fixed for complete y-m-d format. Not losing a day. You can close it from my standpoint. Thanks.

Hi, I live in Spain (for the timezones consideration) and the problem is not fixed.

Days ago I wrote this in other report, now closed:

"When editing an owned book in the app, it is impossible to record any date properly. If you click in the section "reading dates" and try to select a date in the calendar that pops up, the recorded date is the day before you clicked, every time. If you try to write down the date in the box when you abandon the box the date goes back two or three days (I mean, for example, you write 2023-03-10 and when you exit the box you can see how the date you have written backs off three days as in a countdown, is really weird)

I've also noticed that if you click on an already recorded date in the app, when the calendar pops up, the date marked on the calendar is the day after."

Today I wanted to record the "reading date" 2023-03-14, I have to check the day 17th, and when I exit the date box, the date marked goes back to the day 14th, one by one as in a countdown. then I save the edition, and check the information: the date recorded is the 13th.

So I open the website and the "reading date" recorded is 2023-03-14

>46 raulvilar: This bug report is regarding the import, not the read date field on the book edit page.

>47 knerd.knitter: funny thing, because someone closed my bug report because it was discussed in this threat ...

>48 raulvilar: It's actually related to another thread... https://www.librarything.com/topic/343567