Is there a quick ref for KOReader data format?

keshlam · 08-20-2023, 01:49 AM

Apologies if this is a FAQ... Just acquired a (used) reMarkable for possible use as a songbook, and my main gripe about it so far is that when I have a long list of PDFs it takes forever to scroll to the one I'm looking for. I've gone through and Tagged everything with the first letter of its name to reduce that problem, but what I really want is to be able to assign tags programmatically. (I'd say I'd want to be able to search for a document containing a phrase, but I'm guessing there's no way the putt-putt processor on the tablet could do that fast enough unless I do the search in advance and store the results.)

So: Is there a document somewhere that would briefly get me started on understanding how KOReader binds tags to PDF files? Or do I just have to wade into the code?

(If the latter, it'll be a while; I've got too many projects already in progress.)

pazos · 08-20-2023, 08:45 AM

If you manage your pdfs with calibre you can add tags there, push your documents to your ereader (via wifi only on the remarkable).

Then bind "Browse all calibre tags" to your favorite gesture. It will show you a list of all your tags and you could navigate easily.

You can also bind the more generic "Calibre metadata search" in case you want to filter out the search by authors, title, series or do a filtered tag search.

Otherwise I'm not sure how to proceed. KOReader has a file browser, not sure if it understands any kind of tags. Maybe I'm wrong. After all I'm using less than a 5% of what KO offers.

keshlam · 08-20-2023, 03:52 PM

Interesting, and worth knowing about, but not what I'm looking for. I've got about 200 sets of lyrics. I can write some code on the PC to scan through them for keywords, and note which song contains which keyword. What I'm hoping for is the next step, to automagically turn those keywords into tags and associate the right songs with the tags.

Example: The tag for "rocket" would be attached to _The Star Spangled Banner_ ("And the rocket's red glare"), David Bowie's _Rocket Man_, etc.

I know, hundreds of one-page documents is not the most common use of e-readers. But that's my use, and the reMarkable's battery life (and ability to access its files) make it seem a good starting point. But I really do need that keyword filter, and while I could tag things myself it'd be a lot of tags. Automation is desirable.

(A way to write a page with hyperlinks to the other documents on the reader would also work.)

NiLuJe · 08-21-2023, 09:23 AM

What @pazos described is the *only* metadata-based approach you'll get.

The filemanager is, well, a filemanager, its source of truth is filesystem metadata, not document metadata (which, if any, is only parsed and shown as a *display* nicety). If you want to deal with that in a FM-friendly way, you'll have to play with your folder structure, which doesn't look like quite a good fit for what I think you're trying to do.

08-20-2023, 01:49 AM	#1
keshlam Junior Member Posts: 2 Karma: 10 Join Date: Aug 2023 Device: reMarkable	Is there a quick ref for KOReader data format? Apologies if this is a FAQ... Just acquired a (used) reMarkable for possible use as a songbook, and my main gripe about it so far is that when I have a long list of PDFs it takes forever to scroll to the one I'm looking for. I've gone through and Tagged everything with the first letter of its name to reduce that problem, but what I really want is to be able to assign tags programmatically. (I'd say I'd want to be able to search for a document containing a phrase, but I'm guessing there's no way the putt-putt processor on the tablet could do that fast enough unless I do the search in advance and store the results.) So: Is there a document somewhere that would briefly get me started on understanding how KOReader binds tags to PDF files? Or do I just have to wade into the code? (If the latter, it'll be a while; I've got too many projects already in progress.)

08-20-2023, 08:45 AM	#2
pazos cosiñeiro Posts: 1,331 Karma: 2200073 Join Date: Apr 2014 Device: BQ Cervantes 4	If you manage your pdfs with calibre you can add tags there, push your documents to your ereader (via wifi only on the remarkable). Then bind "Browse all calibre tags" to your favorite gesture. It will show you a list of all your tags and you could navigate easily. You can also bind the more generic "Calibre metadata search" in case you want to filter out the search by authors, title, series or do a filtered tag search. Otherwise I'm not sure how to proceed. KOReader has a file browser, not sure if it understands any kind of tags. Maybe I'm wrong. After all I'm using less than a 5% of what KO offers. Last edited by pazos; 08-20-2023 at 08:47 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Manually update KOreader without losing data	lucasrato	Kindle Developer's Corner	25	01-20-2024 04:42 PM
PW3 Quick help with debrick, format partition 4?	schuhumi	Kindle Developer's Corner	2	07-25-2018 01:54 PM
Got JB Voyage? User required for quick SSH test of KOReader	TheNeedforRead	Kindle Developer's Corner	0	05-30-2016 11:18 AM
Collections data format in 2.5 firmware	ngukho	Kindle Developer's Corner	18	10-30-2010 12:19 AM
Quick Format Question	MarcusStringer	General Discussions	9	07-13-2010 05:51 AM

08-20-2023, 03:52 PM	#3
keshlam Junior Member Posts: 2 Karma: 10 Join Date: Aug 2023 Device: reMarkable	Interesting, and worth knowing about, but not what I'm looking for. I've got about 200 sets of lyrics. I can write some code on the PC to scan through them for keywords, and note which song contains which keyword. What I'm hoping for is the next step, to automagically turn those keywords into tags and associate the right songs with the tags. Example: The tag for "rocket" would be attached to _The Star Spangled Banner_ ("And the rocket's red glare"), David Bowie's _Rocket Man_, etc. I know, hundreds of one-page documents is not the most common use of e-readers. But that's my use, and the reMarkable's battery life (and ability to access its files) make it seem a good starting point. But I really do need that keyword filter, and while I could tag things myself it'd be a lot of tags. Automation is desirable. (A way to write a page with hyperlinks to the other documents on the reader would also work.)

08-21-2023, 09:23 AM	#4
NiLuJe BLAM! Posts: 13,497 Karma: 26047188 Join Date: Jun 2010 Location: Paris, France Device: Kindle 2i, 3g, 4, 5w, PW, PW2, PW5; Kobo H2O, Forma, Elipsa, Sage, C2E	What @pazos described is the only metadata-based approach you'll get. The filemanager is, well, a filemanager, its source of truth is filesystem metadata, not document metadata (which, if any, is only parsed and shown as a display nicety). If you want to deal with that in a FM-friendly way, you'll have to play with your folder structure, which doesn't look like quite a good fit for what I think you're trying to do.