Problem with cyrilic file names

[ReaLx3m]

I have bunch of comics that i transfer to my original H2O using calibre with file names in cyrilic, the titles change to latin letters when transfering is done.
The peculiar thing is that it doesnt just change the cyrilic letters to latin analogues, but it does it by changing them by adding additional latin letters so the titles are phonetically correctly pronounced when you read them.
Is this a kobo issue and it doesnt support cyrilic titles and does this conversion, or is it some setting i need to adjust in calibre?
Using 4.6.10075 firmware and latest calibre version

[embryo]

No, this is a Calibre thing.
Only Latin characters.
What I do is to "Save to disk" and then manually copy them over to reader.

[ReaLx3m]

Thats a bummer, as i do shelf management through calibre, and as i suppose with just copying them i wouldnt have it on the correct shelf, or am i wrong?

[davidfor]

As stated, when calibre sends a book to the device, the filenames are massaged to be "safe". That means stripping out or replacing characters that could cause problems. Calibre has a set of rules, and the KoboTouch driver adds a couple. And the driver makes sure the full path of the file is not to long for the device. Kobo devices use the file name in several places in the database and there have been some problems cause by unexpected characters.

Then, when the device sees the book, it imports it into an internal database. If it can, it reads the metadata from the book and uses that. But, for comics and some other formats, it cannot do this. For these formats, the file name is used as the title of the book and the author is "Unknown Author". There won't be anything in the synopsis or other metadata fields.

The way around this is my Kobo Utilities plugin. It can directly update the metadata in the database for a book on the device. For comics, this is the only way I know of to get the correct metadata on the device.

If the filename is important to you, then you could manually copy them to the device and then connect calibre after the device has imported the books. Calibre will attempt to match the books on the device to the library. For Kobo devices, this is done based on the database, not the file names. From experience with English titles, this seems to be accurate. For the rest, I need to manually match the books. I don't know how other languages fare.

[ReaLx3m]

As i understand if i use your plugin i would need to manually edit each title so i get the correct title, right?
If thats the case then it seems pretty cumbersome.
As for the filenames, yep id say theyre important as it contains the number of the issue at the begining for sorting purposes, and the name of the story.
So from this i guess the second option would be my best bet, copy them manually to the reader and then let calibre sync and assign the correct shelf if i understood correctly.
By going this route, would i need to unplug the reader after manually copying the files so the reader does its indexing, or running calibre straight away would work too?

Sent from my Nexus 4 using Tapatalk

[davidfor]

Originally Posted by ReaLx3m

As i understand if i use your plugin i would need to manually edit each title so i get the correct title, right?
If thats the case then it seems pretty cumbersome.

The plugin will update the metadata on the device with the metadata in the calibre library. You select the books you want to update, choose some options and then press a button. Assuming your calibre library has correct metadata, then updating the device is quick.

As for the filenames, yep id say theyre important as it contains the number of the issue at the begining for sorting purposes, and the name of the story.

From what you said, I wouldn't have expected those parts of the file name to be changed. The title and series name could be, but I would expect them to be consistent.

But, when using the Kobo device, the file name is immaterial. You don't get to see it (except where it is used as the title) and if the metadata is set correctly, either from the book or the utilities plugin, then you will see the title and have sorting working in the library lists. That means you could randomly generate a file name and it wouldn't affect your usage once the device had imported the book.

So from this i guess the second option would be my best bet, copy them manually to the reader and then let calibre sync and assign the correct shelf if i understood correctly.
By going this route, would i need to unplug the reader after manually copying the files so the reader does its indexing, or running calibre straight away would work too?

After putting the books on the device, you need to unplug it, let the device import the books and then connect again. The driver is looking at the database, not the files, so the import has to be done.

[ReaLx3m]

Ok, will try next time im on my pc and see what makes most sense for me.
Thank you both for provided input.

Sent from my Nexus 4 using Tapatalk