+2
Planned

Scan statistics

Elouan 4 years ago updated 4 years ago 6

This is an annoyance: when I stop then restart ubooquity [I use ubooquity on a synology NAS], it doesn't remember when the last scan occured. in admin page, the 'last scan' shows nothing, .

I think it would be nice it that this information was stored somewhere in the database so when I restart ubooquity, that information is not lost...

Even better: if you could store the scaning history information in a table, not just the last scan, but register in a table the time and statistics about the last n [n = a certain number] scans. This way, we may be able to extract nice graphics about collection evolution

This would be a great extra feature. If Tom were to externalize the web files and make the Ubooquity variables available through a simple API when building the html files we could make some spectacular graphs.

Planned

This would be nice indeed, but with my backlog of features growing faster than I can implement them, it'll have to wait a bit. ;)


What I can easily do, if you think you can exploit it, is generating a log line with scan results each time a scan is done. You could quite easily get statistics using a few shell commands.


What do you think ?

well, I'd have to think about it a little more. my first thought would be that ubooquity store the scan statistics in a table (one record per scan, one column per stat about the scan) in the existing database. I'm assuming this wouldn't be too difficult to build. Then depending on what format is the db, we may develop graphics using that table using other tools: standalone tools and database connection with like excel or access, or even build Html files with php database connection. this sounds like the easiest way... if the database is accessible of cours

The solution you have in mind would not be difficult to implement, agreed, but it still would require time that I prefer to spend on what I call "core" feature of Ubooquity (as opposed to "peripheral" or "nice to have" features).


More explicitly, I am currently working on the online epub reader feature, then I plan on adding server side bookmarks and read/unread status. And after that I'll try to replace the PDF renderer Ubooquity uses for books by the PDF.js library.


I try to estimate what features will be useful to the highest number of people (based on what I read here, and my personal needs) and work on them first.


That's why I was proposing to write a log line (for now), it takes literally 2 minutes to add.

+1

I was envisioning viewing the statistics through the Ubooquity admin page and this would require being able to write custom HTML files with replacement variables that Ubooquity.jar replaces with actual values (Like {scan_time} being replaced with '2015-12-31 23:11:23' in the .html).


We could certainly grab stuff from the log file, but that would require execution of a script outside of the ubooquity process and that will be too confusing for new users. I'd rather have the ability to modify the HTML template so this is as easy as installing a new theme than write a script that would require an user to download the script and manually run it just for statistics.


You're certainly busy writing important new features. I look forward to the time you put this up on Github so we can dabble, too.

Well, if it only takes 2 minutes of your time, then do the log thing... I'm not sure how I can use that in graphics and stuf, but it would still be an improvment from current situation: I could retrieve manually my scan history