[GUI Plugin] Reading List

[aleyx]

I've tested, and indeed, a revert to the full library then a resync before ejecting corrects the column. So next step would be to a test in SmartEject: "You have loaded a virtual library, and you have a list populated from the books on your device. You should close the virtual library then resync before ejecting." I'll have to look into that sometime.

[Jellby]

After setting a list to sync to device and clear after syncing, if there is some error during the sync, the list is cleared anyway (or so it seemed to happen to me once). This would be a big problem when the list has been populated during several session.

[sengian]

Hello Kiwidude,
I am using this pluggin for a while and now that I have moved to a NAS storage (I know I am a sinner, but for me this is working quite well), a problem I had before with removing and adding books is amplified.
Indeed, it takes a very long time to do such a simple operation, so I timed the code and the problem is mostly in action.py, with the update_series_custom_column function and especially the following loop timed up to 261.895619221s in my case:

for idx, book_id in enumerate(book_ids):
                series_idx = idx + 1
                existing_series_name = db.get_custom(book_id, label=label, index_is_id=True)
                existing_series_idx = db.get_custom_extra(book_id, label=label, index_is_id=True)
                if series_name != existing_series_name or series_idx != existing_series_idx:
                    db.set_custom(book_id, series_name, label=label, commit=False, extra=series_idx)
                    changed_series_book_ids.append(book_id)
                if book_id in existing_series_book_ids:
                    existing_series_book_ids.remove(book_id)

I thought this loop would hugely benefit from the new database optimizations which you are not using currently.

I have modified it a little bit to use the new db api and I have a significant improvement, at least 3 times faster. You will find attached the modified file if you want to include the modifications in the official release.

If you are interested and you think of other parts of the code that could benefits from those improvements and you want me to have a look, please let me know.

I have tested this modification, but mainly for adding/removing a book and not extensively.

[sengian]

I am wondering as I am making new changes to the code, don't you think this could benefit from a git repository or some sort of code management?

[aleyx]

It is my professional opinion as a programmer that revision control is always a good thing to have. It is my personal opinion as an open source supporter that Git would be a good choice ^_^

[kiwidude]

Sengian - thanks for that, I will take a look when I get a chance.

As for the source control, just to kill this subject off again as it has been discussed a couple of times before. In over two years of developing several dozen plugins the number of code contributions from the community can be counted on one hand. There is absolutely no justifiable need for the extra overhead in the monitoring and release process - you have the source code for the plugin within the plugin zip file etc being uncompiled python.

And no I dont buy the argument that putting them in Git will magically increase the number of contributions. Something like Git just adds another complication. The subset of calibre users who are technically inclined in Python and care enough to want to make a change to a plugin who also happen to be familiar with Git is so small it is ridiculous.

As I have said before it may happen one day, and when I one day stop maintaining these plugins completely then that may be as good a time as any. In the meantime attaching a zip like sengian did makes it easier for other users to try it out and test it before I find time to look, which isnt going to happen if instead people had to work out how to connect to git and produce their own deployable zip...

[Orangeredman]

Hi, is there a way to get the Reading list feature to sync to my Kobo Glo card rather than main memory?

[davidfor]

Originally Posted by Orangeredman

Hi, is there a way to get the Reading list feature to sync to my Kobo Glo card rather than main memory?

The drop-down list for the "Device to sync this list to" should have the main memory and the card listed separately. If the card is listed, you can select it as the destination.

If the card is missing, have a look at the devices tab to see what is listed there. If the card is missing from that list, connect the device with the card inserted. When calibre has finished processing, open the dialog and check again. On the devices tab, you might need to press the "Add connected devices" button to add the device memory and card to the list.

[aleyx]

Originally Posted by kiwidude

And no I dont buy the argument that putting them in Git will magically increase the number of contributions.

Well. For my part, I don't use my VCS for team dev (I'm the only dev) nor distribution (my pro work isn't distributed anywhere beyond my workplace, and the non-pro depends on the projects I contribute to). I use my VCS solely for source control. Meaning, version control and central management of my code. Because let's be honest, bad stuff can and will happen to any code, and a VCS is here to mitigate that. Also: code history, branch-and-forget for testing purposes, release tags, etc.

So, yeah. If you already have a VCS on hand, excellent, forget this post. If you don't, it would be a good idea to get one. It doesn't have to be public, and you don't have to share it, especially if you work alone (or close enough) on your projects. Just a [Subversion|Git] repo on your dev box would be more than enough.

Sure, it'll take a few hours to install, set up and learn to use, and your workflow changes a little, and you have to adjust your backup policy. But in the long term, it's worth the effort even if you code solo.

Wow. Juuuust a little OT, I guess. Sorry about that. It's just that with the advent of GitHub, even coders begin to think that a VCS's purpose is to share code. Yeah, it helps tremendously for that, but it's more than that, and it's not just for teams. I personally couldn't work (easily) on my projects without a VCS to back me up, even though I work alone.

[sengian]

@kiwidude: Wait a little bit to review this file. I am having some time so I am doing other thinks with this plugin mainly trying to go to the new database wherever it is useful, ie when when you are making a loop call to set_property. I also plan to add translation as I see it is now possible (I tried to implement it a few months ago and it was not so straightforward).

@aleyx, @kiwidude: Personnally, I am agreeing with aleyx point of view and I am using it for knowing what I changed, what I wanted to do and being able to revert.
I am also asking because I put these changes on my cloud Team Foundation Server, which is accepting Git, not as nicely as Github, but I am able to keep my projects private for free so this is the nice plus side. So if you want I can make you a member of this project, if you have a microsoft account and you are interested.

[kiwidude]

I do not want this thread to be sidelined into a source control discussion . Please understand I know exactly what source control can and cannot do - I use it all day every day for my day job as a developer which I have been doing commercially for over 25 years now. However my point is that *so few* people contribute to calibre plugins that the effort for me to move all of them to public repositories, change all my build and packaging procedures and introduce some sort of change management procedure/process around them far outweighs any benefits of doing so in terms of my time to bother doing it.

[riothamus]

So I've been trying for a few days to figure out how to auto populate my Collections custom column so when I add fanfics with the FanfictionDownloader plugin, "Fanfics" will automatically be added in the Collection custom column. I couldn't find anything other than Reading List. Apparently back in 2011 there were tag and custom column options that I don't see now. I tried adding a new list with only the purpose of updating the Custom column, but since FanficDL doesn't have the option to call it, no go. Any help would be appreciated.

[davidfor]

Originally Posted by riothamus

So I've been trying for a few days to figure out how to auto populate my Collections custom column so when I add fanfics with the FanfictionDownloader plugin, "Fanfics" will automatically be added in the Collection custom column. I couldn't find anything other than Reading List. Apparently back in 2011 there were tag and custom column options that I don't see now. I tried adding a new list with only the purpose of updating the Custom column, but since FanficDL doesn't have the option to call it, no go. Any help would be appreciated.

You need to check your installation. The FFDL options has a tab called "Reading Lists". If it is missing, either you have old versions for the one or both of the plugins, or there is an error in their installation. Try removing FFDL and then installing it again.

[riothamus]

Oh, FFDL has a reading list tab. It works perfectly for updating the newread column and on_kindle column. There's just no option for the collection list I tried to create in the reading list plugin. So FFDL will call reading list correctly. I just can't figure out how to make reading list update other tags or custom columns.

[davidfor]

Reading List will set one column/value per list the book is put into. The FFDL options for Reading List is a comma separated list.

FFDL will also set values in custom columns from the metadata it generates. And I think you can set some in the personal.ini. I haven't played with these options enough to know exactly what you can do.