Export

Not to an alarmist, but a thousand books into inputting my home collection I've begun to wonder. Should LibraryThing ever shut down, is there a way to export one's personal collection to, I don't know, Excel or some such?

fg

Top menu =More
Side menu = Import/Export

:)

But, probably not after they shut down (perish he thought, of course, but still). So, make a backup export every once in a while.

Make sure you export the tab-delimited version; it is much more complete than the CSV.

#1 by ovevs>

Yes. The export options are not perfect but you can certainly get most of your data in either CVS (fewer fields) or tab-delimited (more fields) text file exports. These can be loaded into Excel or even one of the mobile phone apps for Android and iOS.

In the "chocobar" click on the "More" link and from that page select "Import/Export". Alternatively, you can go directly to this link: Import/Export

Some things that you will notice in the export. "Other Authors" data and Common Knowledge (CK) data like series are not part of the export. Tags and Collections do export.

Here is a list of fields from the tab-delimited export (recommended):

book id
title
author (last, first)
author (first, last)
other authors
publication
date
ISBNs
series
source
language 1
language 2
original language
LCC
DDC
BCID
date entered
date acquired
date started
date ended
stars
collections
your tags
review
summary
comments
private comments
your copies
encoding

Of these, Tags and Collections are comma-separated values in a single field. For example, it might look like:

Drupal,Content Management System,computers,nonfiction,pb

for tags of a programming book if those are the tags you have applied to it.

Again, it's not perfect. Getting cover images is nearly impossible in this way. But, it does let you have an archive of your data (especially ISBNs, comments, etc.) in case you have to use another system, roll your own database, or wish to use a mobile device to view your inventory.

James

Thanks, all.

fg

Being fairly new to LibraryThing, I have downloaded in both formats, and figured out that tab-delimited had more info. My big problem is that a majority of the Dewey Decimal numbers are missing. I asked about this in the wrong forum a couple of days ago, and was politely told my question didn't belong there. They did say, however, that the green numbers generated from LibraryThing don't show up in a download. They said to go into Edit and simply click in the number box, save, and the number would show up. This didn't work. I then started trying retyping numbers. They turn black, but if they're compatible with LibraryThing numbers, they turn green again and won't download.
Any Suggestions? I have about 2,000 items cataloged now, and would like more complete backups.

Huh. It works for me - if I have a green number, I double-click (on the field in Your Books) and hit Save and I don't have a green number any more. If I refresh, it goes to blue, but it's still in my records (as opposed to being in LT's records). Hmmm, I haven't actually tried the Edit page (I go there as seldom as possible - it's so much easier to edit direct from Your Books).

Well, isn't that interesting. No, it doesn't work from the Edit page - not even if I retype it. That's DUMB. In fact, I think that's a bug. To get it to stick, I have to type in something else (from 813.6 to 813.65 - I just typed the 5 on the green number), then save, then go back to the Edit page and remove the extra number. Putting in something that isn't a number (like a slash) works, too. But that's really annoying! And I think it was bugged, many many years ago - I'm going to post it as a bug, if I can't find the old one.

Anyway, the solution is to have DDC showing in Your Books, double-click the field and hit Save. That will set it to black/blue and make it show up in the export.

Thank you so much! That's much faster than going into Edit, anyway. Whew!

Yes, that's why I avoid the Edit page - unless I'm making a lot of edits on one book, double-clicking in Your Books is so much faster and easier...

I bugged it - couldn't find the old bug, which was probably before the Bug Tracking system anyway. It's clear why it happens, but there should be a solution - not just a silent fail.

http://www.librarything.com/topic/157832

Last night I was exporting my books again, and noticed that the series field, despite being there, is empty.
Then I realised that all CK fields don't get exported.
Any chance of adding them to the tab-delimited export option in the future?

#11

I think in the past Tim has asked about the future of export - http://www.librarything.com/topic/101108

Hmm quite some time ago.

Yes, a long time ago hehe
This all came up when we had that downtime blip, and then I thought, what if all the records disappear? Better export!
And then I discovered lots of data I've been including aren't there :(

>13 ScarletBea:

You can copy from the Your Books page and paste to a spreadsheet. I've configured several of my styles to show CK data and have considered exporting the information. It would be a time-intensive process for my library, but you could probably manage it with yours. Be careful how your spreadsheet merges cells for fields that carry multiple values and click the "(# more)" to show all the values.

For now, I'm relying on LT being available for most of the CK, and use exportable fields for information I do want to have. For instance, I'm using the Summary field for series information—"Liaden Universe ; Bk. 17 Trade secret by Sharon Lee and Steve Miller" (it also has the advantage of being sortable)—and the Comments field for Omnibus/Anthology details.

Yes, CK doesn't export at all (and I really wish Tim would get that XML export going...). The Series field actually predates CK, but it's always been empty. I have no idea what it was supposed to be.

I put my series info in Comments, then (for no particular reason, just my tweak of OCD), when I export I copy the Comments field into that empty Series field (and then import the modified - that and other things - xsl file into a mobile database).

The problem with CK exporting is that it's (as mentioned in the thread andyl linked to) mostly multi-valued fields - so which one exports, or how are the multiple values shown? Of course, Other Authors is multi-valued too (though not the same way). And so is Reading Dates/Start Date/Finished Date, but then that doesn't export either (still messed up).

Worst case scenario, you could set up a display with only the fields you want to export but won't, then print it. You could probably print it as a pdf to your drive. At least if something happened, you'd have documentation that could be used to reconstruct.

#16 by PhaedraB>

If I wanted to do that, I would probably set up the columns as you suggest but use the browser's "Save Page As" to record the HTML. This has tags and a program could be written to parse this text file far easier than a PDF.

It is important to note that most larger collections would not fit on a single page so it would be necessary to repeat the process for every page. Mine would require 40 pages, for example—a tedious process which might be able to be scripted. For example, the URL for page 40 is:

http://www.librarything.com/catalog.php?offset=7800&view=Keeline&collect...

The "offset" value is the key of course. My default is 200 (maximum value) per screen. A shorter URL seems to work just as well by using defaults:

http://www.librarything.com/catalog.php?offset=7800&view=Keeline

As I've mentioned elsewhere, the ability for LT users to completely export their data is critical to confidence in the site. This was discussed a bit when GoodReads members came to LT after the Amazon buy out of that site. TimSpalding seemed to acknowledge some value in this but we have not seen a fuller export even though it was discussed way back in 2010.

Our ability to fully export data is also important for being able to use it in other programs, including desktop and web databases and even mobile phone apps.

We have seen a couple of outages, including the big one, and any promises of lifetime accounts and data integrity are only as good as the company's ability to stay in business. Sure there may be no plans to sell LT. That doesn't mean it will never happen, however. If a company was going to sell, they'd deny it vociferously until the day of the announcement.

Our financial contribution to the site is negligible. Even at $30 or $40 for a lifetime account it would not be. However, most of us have contributed thousands of hours to update our own catalogs and some of that data benefits other users and LT in general. On that basis we have a "stake" in the future of LT.

I think it is only fair that we be given proper tools to export our own catalog data, including the Common Knowledge values such as series, other authors, original publication date, etc.

I hope this functionality is addressed in the near future.

James

17 > It is important to note that most larger collections would not fit on a single page so it would be necessary to repeat the process for every page.

Not really. If you click on the printer icon, you can print-as-pdf one long multi-page document, same as if you were using a physical printer.

You can also just copy the lines from your catalogue page and paste it in an excel sheet. It will also copy the bit with the icons on the right, but you can delete those columns, then, after selecting all cells, in Cell Properties untick the 'wrap text' and 'merge cells' boxes and then sort to remove empty lines.

That's the easiest and fastest way to copy data not in export I believe.

eta: Excel help.

@19: That method also gets you the cover thumbnails (if your catalog shows them) which is neat.

19> Wrap text and merge cells, is the English for what you're saying, I think. That might be worth trying, especially since it gets covers (ok, tiny ones) as well, as saltmanz says.

Not in all browsers though, so try IE if something doesn't work as expected.

In the past with other tables I've had trouble getting rid of unwanted icons with IE, so test other browsers as well to see which one performs the best for you.

Is it possible to export your library in csv with cover image URLs?

No, but that is a good suggestion, it could be a way to more easily save all the covers you contributed or carefully chosen.

LT has its own library of cover images, so it would be really nice to have an export that includes a URL for LT covers at least - whether uploaded yourself or selected from those that others had uploaded. (In fact, it'd be wonderful if information about whether the cover you're using was one that you uploaded yourself was part of the export!)

>26 felius: Yes! (I've tried to do something like this on my own computer by scanning almost all covers myself and saving them with names derived from the title). This also allows me to check cover image dimensions against the physical dimensions of the book (catching a few errors caused by ratty amazon data).

It would also be nice if you could get a link to the original uploaded picture, but I'm not sure it would be a good thing for LTs servers :-)

Take a look here if you're curious :-)
http://wiki.librarything.com/index.php/User:Bnielsen/CoverNames

Does anyone know if this export can be imported to Zotero?

>28 lfriedla: IF it can accept a CSV file or JSON format, or MARC format, or EXCEL file, then it can be imported into Zotero.

I normally export my book catalog in Excel and Tab Delimited once a week as a back up. I tried yesterday but it wasn't working. I decided to try again today and it still seems to be not working, I'm stuck at "0 books processed". Should this be reported somewhere or is it due to the holiday weekend?

Bug has been reported. https://www.librarything.com/topic/334983

>31 MarthaJeanne: Thank you for letting me know. I had half convinced myself that the problem was something I was doing wrong ;)

The Excel export seems to be working nicely (it was down last week). What, exactly, are "Book ID" and "Sort Character"? Are they unique to LibraryThing, or are they some external reference I've never become aware of in nearly 60 years of reading?

I'm trying to decide if I need to keep them in my spreadsheet backup.

>33 steve.florman: Book ID is the ID of your book in the LT DB. It is LT specific and if you add a book again, you get a new Book ID and these are always unique. It is very much LT thing (other dbs will have similar concept, calling it how they want). Unless you want to know how to look up the book in LT (your specific record), you don’t need to keep that usually.

Sort character tells LT how to sort the title when you look at the titles in alphabetical order. So for “The New…”, you need 5 if you want to sort with the N-s. While English only libraries can be automated easily, LT is international so take for example the word Die at the start of a title. In English title you want this book sorted with the Ds. But in German titles, you want to skip it as it is the definitive article. So LT allows a user to sort their libraries any way they want. It gets assigned automatically for English when you add books but if you edit the title, it may need adjusting.

BookId is a unique number for your book on LT. They are unique to LibraryThing.
Sort character tells LT where to start sorting the title. So if the title is "The Bell Tower", sort character is typically set to 5, so the title sorts like "Bell Tower". This is something you can set on each book in your catalogue. I.e. you can set it to 3 and the book will sort like "e Bell Tower" (which doesn't make a lot of sense).

I'd keep them in the backup, unless they annoy you in some way :-)

Just to say that you find the sort character on the edit page right after the title. It's just a little unlabelled box that people often ask about.