Tag troubles

[ccowie]

I had been using an old version of Calibre for about 9 months and hadn't bothered to upgrade. Yesterday I uninstalled the old version and installed version 7.

I used it to upload a few books to my 505, but afterwards I noticed that it had done something funny with my tags. I now have 45 collections when previously I only had 11, which is the number I want.

In Calibre my books still show the tags I had set, but in my reader they are a mess. I thought I'd fix simply double clicking on the tags while in Calibre, but that function doesn't seem to work. Is there a reason that I can't edit by double clicking as I was able to do before?

[kovidgoyal]

Its a bug will be fixed in 0.7.2

[ccowie]

Thanks Kovid.
I've said it before, but I will make a contribution when I have a bit more to give. Thanks so much for all your work!

[ccowie]

Is there any other way to change the tags in the mean time?

[itimpi]

You can edit them on the copy of the book held in the calibre library, and then resend the book (ideally deleting the existing copy first) to the device.

[ccowie]

I'm using 0.7.2 now and I still can't edit the tags on my reader by double clicking on them the way I could in the old version.

[ccowie]

I'm using 0.7.2 now and I still can't edit the tags on my reader by double clicking on them the way I could in the old version.

[kovidgoyal]

yes you cannot edit collections anymore. This is a side effect of the closer association between books in the claibre library and ont he device. To change tags, delete and resend the books to the device.

[ccowie]

Hmmm. Too bad. I liked being able to make adjustments that way, particularly because I still had a couple of occurences of all tags being wiped out inexplicably.

Is this also the reason why formatting always takes so long when I disconnect my sony from calibre now?

[chaley]

Originally Posted by ccowie

Is this also the reason why formatting always takes so long when I disconnect my sony from calibre now?

It is still taking as long, even with 0.7.2? 0.7.1 had a problem in that regard, but I thought we had taken care of it.

How long is 'so long'? Mine takes around 15 seconds with a library of around 140 books. It takes the same amount of time regardless of whether calibre touched the device or not.

[DoctorOhh]

Originally Posted by chaley

It is still taking as long, even with 0.7.2? 0.7.1 had a problem in that regard, but I thought we had taken care of it.

How long is 'so long'? Mine takes around 15 seconds with a library of around 140 books. It takes the same amount of time regardless of whether calibre touched the device or not.

Mine takes about a minute or so, but this is the same time it would take if I shut it down and restarted it.

I thought this was always the norm when disconnecting from the usb connection?

The good news though is that on connection to calibre it takes only a couple of seconds to show the list of books on the device.

[chaley]

Originally Posted by dwanthny

Mine takes about a minute or so, but this is the same time it would take if I shut it down and restarted it.

Based on experimentation and observation, the two processes (disconnect and restart) are similar.

When you disconnect (or restart) the Sony, its SW scans its database, doing several checks. The most important is verifying that the date of a book in its DB is the same as the date of the book. Cards in the device are subjected to the same tests. This check requires the SW to look at each book in its filesystem. It also (seems to) throw away fields it doesn't use. For example, my 300 removes 'title-sorter', but I understand that other models use it. Next, it verifies the collections. Finally, it next checks the filesystem for books not in the database. This operation requires a scan of the filesystem. It may do this operation and the date check at the same time.

If a date is different or a book has appeared, the SW extracts the metadata from the file (if it can) and rebuilds the database entry. This is an expensive operation.

Calibre 0.7.1 and Sony had a difference of opinion over what date one should use, and because Sony won, the time to disconnect went up (a lot) while Sony rebuilt its database. Starting with 0.7.2, calibre was changed, and I haven't been able to find any scenario where Sony disagreed with the dates. I am not saying that there isn't such a scenario, and want to hear about anomalous conditions.

Clearly the time taken will depend on the number of books the Sony must look at, the speed of the memory and memory cards, and the speed of the Sony's processor. I don't know whether the models differ in basic performance profiles.

Making a bald assertion: disconnecting after using calibre should take no approximately the same time as connecting and immediately disconnecting with calibre not started. This assertion is conditioned upon you doing nothing to the device with calibre during its connection. The first connect/disconnect is can be an exception because calibre builds its cache. If someone runs this experiment and experiences different times (more than a few seconds), I would like to hear about it.

The good news though is that on connection to calibre it takes only a couple of seconds to show the list of books on the device.

The cache does seem to work.

[JvdW]

Originally Posted by chaley

Based on experimentation and observation, the two processes (disconnect and restart) are similar.

Making a bald assertion: disconnecting after using calibre should take no approximately the same time as connecting and immediately disconnecting with calibre not started. This assertion is conditioned upon you doing nothing to the device with calibre during its connection. The first connect/disconnect is can be an exception because calibre builds its cache. If someone runs this experiment and experiences different times (more than a few seconds), I would like to hear about it.
The cache does seem to work.

Just tested this with Calibre 0.73.
Start Calibre, connected my Sony 505 (with 49 books on it), took some seconds the show the column OnDevice, great.
Disconnected throught the icon next to the library, took the usb cable out of the Reader and started timing when the text 'usb connected' disappeared--> 2:20
Closed Calibre, plugged my reader in, waited until the windows dialog appeared asking what todo with the new drive, close that window, ejected the reader and took it of the usb cable, started timing when 'usb connected' disappeared--> 17 secs

Just a wild guess, but I think people should tell whether they are running a plain reader or a reader with for example PRS+. I'm running PRS+ 1.1.3 WITH a date setting of MM-DD-YY which is NOT the DEFAULT and might explain why it takes so long.

Is ther anything I can do to help debug this?
For example take a snapshot of the files modified with and without Calibre active?


Regards,

Joop

[chaley]

Originally Posted by JvdW

J
Disconnected throught the icon next to the library, took the usb cable out of the Reader and started timing when the text 'usb connected' disappeared--> 2:20

This is completely consistent with the Sony and calibre fighting over dates.

Closed Calibre, plugged my reader in, waited until the windows dialog appeared asking what todo with the new drive, close that window, ejected the reader and took it of the usb cable, started timing when 'usb connected' disappeared--> 17 secs

And this is what it should be

Just a wild guess, but I think people should tell whether they are running a plain reader or a reader with for example PRS+. I'm running PRS+ 1.1.3 WITH a date setting of MM-DD-YY which is NOT the DEFAULT and might explain why it takes so long.

An excellent thought. Yes, it would cause the problems.

Is ther anything I can do to help debug this?
For example take a snapshot of the files modified with and without Calibre active?

The best debug information would be a copy of media.xml before and after calibre connects to the device. It is in database/cache/media.xml. Connect the device, make a copy of the file, start calibre and let it connect, stop calibre, and make another copy of the file. (I know that you know how to do this. I am just being compulsive. )

If there is nothing in the files that you mind people seeing, create a ticket with both files attached. If you would rather keep them private, PM me and I will give you my email address.

[JvdW]

Originally Posted by chaley

The best debug information would be a copy of media.xml before and after calibre connects to the device. It is in database/cache/media.xml. Connect the device, make a copy of the file, start calibre and let it connect, stop calibre, and make another copy of the file. (I know that you know how to do this. I am just being compulsive. )

If there is nothing in the files that you mind people seeing, create a ticket with both files attached. If you would rather keep them private, PM me and I will give you my email address.

Made a copy before and after and did a diff. The strange thing is it shows modified dates but no format changes. See below for the orignal part:

		<xs1:text author="Maria V. Snyder" page="0" pageOffset="0" part="0" scale="0" status="0" sourceid="1" id="2" date="Sun, 16 May 2010 09:41:46 GMT" mime="application/epub+zip" path="database/media/books/CBUK125EP000300F.epub" size="434352" title="Poison Study" titleSorter="Poison Study"/>
		<xs1:text author="Fiona Harper" page="0" pageOffset="0" part="0" scale="0" status="0" sourceid="1" id="3" date="Sun, 16 May 2010 09:41:46 GMT" mime="application/epub+zip" path="database/media/books/CBUK125EP000400G.epub" size="236956" title="English Lord, Ordinary Lady" titleSorter="English Lord, Ordinary Lady"/>
		<xs1:text author="Margaret McPhee" page="0" pageOffset="0" part="0" scale="0" status="0" sourceid="1" id="4" date="Sun, 16 May 2010 09:41:46 GMT" mime="application/epub+zip" path="database/media/books/CBUK125EP000500H.epub" size="291504" title="The Wicked Earl" titleSorter="Wicked Earl, The"/>
		<xs1:text author="Willy van der Heide" page="1" pageOffset="0" part="0" scale="0" status="0" sourceid="1" id="5" date="Sun, 16 May 2010 09:41:46 GMT" mime="application/epub+zip" path="database/media/books/Willy van der Heide/De strijd om het goudschip/De strijd om het goudschip - Willy van der Heide_41.epub" size="138034" title="De strijd om het goudschip" titleSorter="De strijd om het goudschip">

And this is the modified part:

		<xs1:text author="Maria V. Snyder" page="0" pageOffset="0" part="0" scale="0" status="0" sourceid="1" id="2" date="Tue, 01 Jan 2008 12:00:00 GMT" mime="application/epub+zip" path="database/media/books/CBUK125EP000300F.epub" size="434352" title="Poison Study" titleSorter="Poison Study"/>
		<xs1:text author="Fiona Harper" page="0" pageOffset="0" part="0" scale="0" status="0" sourceid="1" id="3" date="Tue, 01 Jan 2008 12:00:00 GMT" mime="application/epub+zip" path="database/media/books/CBUK125EP000400G.epub" size="236956" title="English Lord, Ordinary Lady" titleSorter="English Lord, Ordinary Lady"/>
		<xs1:text author="Margaret McPhee" page="0" pageOffset="0" part="0" scale="0" status="0" sourceid="1" id="4" date="Tue, 01 Jan 2008 12:00:00 GMT" mime="application/epub+zip" path="database/media/books/CBUK125EP000500H.epub" size="291504" title="The Wicked Earl" titleSorter="Wicked Earl, The"/>

Now I should tell you that I use Calibre on more than one computer to sync news, books etc. The first example is from a system at home (0.7.2?) and the second is from Calibre at work (0.7.2). I Think that might complicate things since at work I can plugin my Reader and have Calibre detect it, unplug it and it scans rather quickly then (10secs). Only the first time it took a long time and only then are the dates changed.
The problem is I can't figure out where those dates are coming from since those books are on my Reader, not in the library at work but in my library at home. (which on second thoughts my not be true, will check and report if relevant)



Will update all systems used to Calibre-0.7.4 and see if it still happens.

Regards,

Joop