ADD Books & extract tags from title?

I have a 20 year collection of 9000+ documents and ebooks in several formats. I'd like to use Calibre to order and serve them out to my users.

The titles are currently coded by filename as follows:

Last, First[,Last, First,...]!Title[,The]+AA.FORMAT

Where "AA" is a two character case sensitive alpha numeric "tag" code indicating subject matter.

I could bulk ADD the books using "regular expression" something like:

(?P<author>.+)!(?P<title>[^_]+)\x2B(?P<tag>.+)

But "tag" or "tags" is apparently not a defined field.

For testing purposes with PREFERENCES/ADDING BOOKS/REGULAR EXPRESSION and TEST I used:

(?P<author>.+)!(?P<title>[^_]+)\x2B(?P<series>.+)

which worked exactly as I would like, putting the code in "series". So I'm close.

I could do it this way if I could write a script that would move "series" to "tags" and blank series for each metadata record where "series" contains one of the codes and nothing else. That's another entire education since I haven't done anything like that for at least 10 years.

If I can't do that.... then perhaps bulk edit would permit me to select all titles in series "XX", change the "tag" to "whatever" and blank "series".

OR I suppose I could use WinXP SEARCH to select the titles by "tag" then ADD them group by group and then bulk edit the tags in each group.

Unfortunately there are 150+ "tags" which would be a time consuming process.

Still, Calibre seems to have lots of ways to do things and
maybe some guru out there has a suggestion for a one step automation for this process?

As a separate issue, I'm trying to wean my users from Win to Linux so I want them to have access to the Calibre database from both WinXP and Ubuntu 10.04. The latest Win version of Calibre is 0.7.30 while the Ubuntu repository version is still 0.7.18. I skimmed through the revision log but perhaps someone can confirm that there have been no changes in the databases between those versions that would cause problems.

Finally, just something I'd like to throw out: I suspect most ebook libraries today are mostly novels but non-fiction will soon follow. Wouldn't it be nice to be ahead of the curve and implement something like the Dewey System in Calibre?

TIA

[Manichean]

Originally Posted by johnb0647

I could do it this way if I could write a script that would move "series" to "tags" and blank series for each metadata record where "series" contains one of the codes and nothing else. That's another entire education since I haven't done anything like that for at least 10 years.

You seem to have all the rest figured out, so I'll concentrate on this step: In bulk metadata edit, go to search and replace. Using the regular expression mode and backreferences, it should be possible to bring the info from series to tags.

[caluser2011]

Originally Posted by johnb0647

Finally, just something I'd like to throw out: I suspect most ebook libraries today are mostly novels but non-fiction will soon follow. Wouldn't it be nice to be ahead of the curve and implement something like the Dewey System in Calibre?

TIA

Agreed... a Dewey Metadata field is an absolute requirement. I'm surprised it has not been added yet this far in development. It would be most useful to see this feild sortable in the list view. Perhaps a collapsible tree with subs, like when we create categories and sub-categories, would be most appropriate.

[tantrictami]

Does anyone know what if calibre will change a file into a format that Adobe indesign can open so I can edit an ebook?