Feature Request: preserving original filename

[Giuseppe Chillem]

Hi Kovid,
I have a feature request for you. It should be simple but when we talk about programming nothing is "simple" as it seems.

I add books reading metadata from the file or parsing the file name using regular expression.

In both scenarios I find really usefull to maintain somewhere the original file name.

For example:

1) You have read metadata but they are not complete and the original filename contains valuable informations like the ISBN.

2) The metadata read are partially complete and the filename has still some informatios

3) The regolar expression rule is good for 90% of the books but some other have a different format and you have not acknoledget this.

What I ask is a column "original filename" and a an option in preference to copy the original filename to this comumn (independent from the Metadata/Regular Expression choice)

Hope you can help.

[kovidgoyal]

This isn't something I care to implement, personally, patches are welcome.

[Giuseppe Chillem]

Originally Posted by kovidgoyal

This isn't something I care to implement, personally, patches are welcome.

Kovid, I am not a Phyton programmer. I kindly request if you reconsider this feature in the future. I think many people would welcome this option.

Thanks anyway.

[cybmole]

[QUOTE=Giuseppe Chillem;2378803]. I think many people would welcome this option.

/QUOTE]

& this assertion is based on what evidence ???

personally I have zero interest in this suggestion

[theducks]

[QUOTE=cybmole;2378898]

Originally Posted by Giuseppe Chillem

. I think many people would welcome this option.

/QUOTE]

& this assertion is based on what evidence ???

personally I have zero interest in this suggestion

Agreed, why bother if Calibre manages your Library.
Just look at the mess with a Project Gutenberg download pg12345.epub

[Giuseppe Chillem]

[QUOTE=cybmole;2378898]

Originally Posted by Giuseppe Chillem

. I think many people would welcome this option.

/QUOTE]

& this assertion is based on what evidence ???

personally I have zero interest in this suggestion

Assertion ? Opinion !

I find it a lot useful in many scenarios. I have also explained the reasons in the first message.

Try to bulk add books with different naming, extracting metadata and then you discorver some are incomplete BUT the original filename was IXN RANDOM 019238485.pdf and another 81309876 RANDOM ABC.pdf
You would prey to have again this 019238485 and 81309876 somewhere.

First try to walk in my shoes for few days then reconsider your point of view !

[jgaiser]

The assertion/opinion is "many people".

At this point it appears there is "one people" that is interested. *I* personally have no interest in your feature request.

[Giuseppe Chillem]

Originally Posted by jgaiser

The assertion/opinion is "many people".

At this point it appears there is "one people" that is interested. *I* personally have no interest in your feature request.

At this point there are only 3 users Vs 1. Not really significant if we consider calibre has over hundred thousand of users...

Also, there are no studies about usage habits and since I am an active user with a common usage path I can infer that my opinion is more common than you think.
However it is a personal feeling, an opinion, which is different than an assertion/statement.

Instead of writing "I won't use" why don't you think if the other hundred thousand users would like this inprovement ?

[jgaiser]

So you infer, because you think it's important *and* because there are thousands of users, obviously *many other somebodies* must think it's important.

Ok, that makes sense.

[Sabardeyn]

He's not that far wrong. I have found it worthwhile to go back to the original file name to recover information I didn't originally think I would need.

However, I also keep two complete copies of my ebooks. One "my way" and one "calibre's way" - just in case there are any issues I've got a backup. Despite the above, I don't think it makes sense for calibre to remember the original file name because for most users this would be unwanted data.

[Adoby]

Actually I could see that it might be useful to add a uri to the file used when doing the import. It could be used by first copying the books to some backup unit and then doing the import from there. That way you could easily retrieve the original file later, if needed. Also the filenames and elements in the path would be available for future reference.

I might even use that feature. But I don't have any pressing need for it. I achieve almost the same using folders with date of addition in my backup storage. On the rare occasions that I need to restore a file I can retrieve it that way.

[Sabardeyn]

Yes, that is basically my point. Within about 45 days (I would guess) of adding a book you're most likely to need to refer back to the original file as you attempt to enter and normalize the book's information into calibre. Or if you discover some kind of issue with the book which leads to re-adding, re-building or re-structuring the database. Beyond that there would be a substantial period where the filename is of no concern to the user at all.

The next time you might consider the filename is when your needs and expectations from the calibre Library change. Re-focusing your usage/intent will require changing your data and possibly your database. Which means you might need to go back to your source and see if any newly sought metadata is available therein (probably not - ebook standards are too minimal as yet).

With a well organized Library and applying "best practices" (whatever those are - I'm still working on trying to acquire them!) to data entry, there is little or no need for that filename beyond that period. Regular additions to your Library will mean that you're going to forget that a certain book still needs more work - but that is why certain plugins are so popular.


PS: Here's a possible "best practices" discussion thread: How I Manage ebooks with calibre. I haven't done more than glanced at it since it is very highly detailed. I've flagged it for future reading after working out some database structure issues.

[cybmole]

Originally Posted by Giuseppe Chillem

At this point there are only 3 users Vs 1. Not really significant if we consider calibre has over hundred thousand of users...

true - but there's a "poll" option somewhere here that you can enable.

calibre metadata already caters for ISBN - & once you have captured that, what else is in original filename that's useful & that can't easliy be added to a custom column?

NB creation of custom column & custom add books rules doe not require python skills; it does require careful study of available options within calibre documentation though. I know, for example, that you can turn off use of metadata for add books, & force use of filename content instead- that is just a one-tick change to preferences

[Agama]

Originally Posted by jgaiser

The assertion/opinion is "many people".

I personally have no interest in your feature request.

+1

I have no interest in the original filename cluttering my calibre database.

[Agama]

Originally Posted by victthoe

The next time you might consider the filename is when your needs and expectations from the calibre Library change.

You could of course be right, although I have used calibre for some time now with various changes of requirement and have not found a use for the original filename.

Do you have any specific scenarios in mind? Perhaps you have found a use for it.