As I've only been using Calibre for a few days, I'm far from an expert on its use. However I've been a programmer for 30 years.
1) I am surprised to discover that while Calibre has a robust plugin architecture, that architecture does not include a "Tools" or "Utilities" category. Currently, tools appear to be hardwired in to the right-click menu.
Say for example I wish to extract and set the default ebook cover from a selection of ebooks. I'd like to be able to select the books in the GUI, click on a Tools menu, and select a python plugin to perform my desired task upon all selected ebooks.
2) Another idea, though perhaps not architecturally feasible, would be a preference setting to save ebook series as series/author/.... (My motivation is based upon German pulp series extending to over a thousand issues, with many, many authors. )
3) Metadata has a field for ebook Language, I'd like to be able to access it.
4) Ebook imput. Here is where I think something really cool could be done. Again, I have in mind large existing ebook collections. I would like to see a preferences screen as follows.
- Each metadata item, Title, Series, Index, ISBN, etc are listed separately.
- To the right of each item is a pull-down menu allowing selection of which input plugin to use for that particular item, as well as the "filename regex" feature already in place.
In other words, instead of all/nothing filename or metadata, an item-by-item user selectable preference.
Now I have a hard decision to make. Perhaps it is time to dive in and add Python to my list and see what I can do to help make Calibre the best it can be.
EDIT: Since the developer gets so much flack over his directory structures, I'd like to clarify why I wish for series to be stored in a different manner. The reason is not because I want to surf the folders manually (which appears to be the most common reason). I accept that Calibre is a database, and it does not matter how the folder structure looks, it matters how it works.
Assume I have several pulp book series, numbering in the hundreds of issues each. (which I do)
Very often the author is unknown. Other times the author is immaterial, as pseudonyms are so often used. Other times I just don't care who the author is and don't want to take endless hours looking at each cover and entering the author.
So, in the Calibre database, I have thousands of entries under author "Unknown." In Windows, this is not the best for performance. It would be better to have a more balanced folder structure. Therefore if series could be stored series/author, this would be avoided.
I want to further mention that this is not a high priority to me because I have a simple work-around: I simply set the Author field to the name of the series, which _almost_ has the same effect (but does not allow me to set the author for those series where I _do_ care.
Thanks for staying with me this long!
Last edited by LessPaul; 09-28-2009 at 11:05 AM.
Reason: add rational