Plugboard, template, and custom composite column recipes

[ACGAuthor]

Originally Posted by eschwartz

Sure thing, then.

BTW, I created that template by opening the text editor, copying your list, and adding ",SFF/UF/PNR," to the end of each line. Then I did a search and replace for "\n" replaced by "" to remove line breaks. Nice and fast. You might have to use "\r\n" on a windows computer to match a line break, though.

Mechanizing stuff is fun.

Okay. I'm going to go through and add my various tags for M/M Romance. I think it might end up being a bit tricky figuring out how to separate out M/F romance, actually. Which seems weird, because it's so common you'd think it would be easy, but it's the fact that it's common that makes it tough.

Unlike M/M Romance, which usually has a tag like "Gay," "Gay & Lesbian" or "Gay Fiction" or "LGBT" or something, M/F romance will generally just have "Romance." But since M/M Romance ALSO often has "romance", I can't use that as a tag for sorting by genre.

Unless there's a way to find the ones tagged "romance" and then filter the ones with the various M/M sort of tags out from the ones without any such distinguishing tags.

I could always separate M/M Romance into it's own library and use a different template for it, but by that logic, I could side-step ALL of this by sorting these into different libraries, and then what would we do for fun?

[unboggling]

Originally Posted by ACGAuthor

My "High" Genres are as follows:

Family-Friendly (Childrens or Young Adult type stuff)
M/F Romance
M/M Romance
SFF/UF/PNR

If I made the top-tier genres any more specific, then I'd have many more top tiers than I really want....

I see.

Originally Posted by eschwartz

...Mechanizing stuff is fun.

Originally Posted by ACGAuthor

LOL well sure, if you're gonna go about being LOGICAL about it and all!

No, but seriously, I think we both realize this. The point here is that getting the program to do the work for us is a challenge and a puzzle. Well, I'm presenting the challenge, eschwartz is solving it. It's fun!

Well, I'm enjoying it. And learning stuff too.

[ACGAuthor]

Originally Posted by eschwartz

I know what genre each book is, and I don't care about it either; I choose based on series.

LOL I would love to be able to do this, but I have so many books that belong to series that if I had everything sorted by series, I'd have 400+ collections, plus another 800 or so books that don't belong in any series.

I have a rather large library.

[eschwartz]

Originally Posted by ACGAuthor

Okay. I'm going to go through and add my various tags for M/M Romance. I think it might end up being a bit tricky figuring out how to separate out M/F romance, actually. Which seems weird, because it's so common you'd think it would be easy, but it's the fact that it's common that makes it tough.

Unlike M/M Romance, which usually has a tag like "Gay," "Gay & Lesbian" or "Gay Fiction" or "LGBT" or something, M/F romance will generally just have "Romance." But since M/M Romance ALSO often has "romance", I can't use that as a tag for sorting by genre.

Unless there's a way to find the ones tagged "romance" and then filter the ones with the various M/M sort of tags out from the ones without any such distinguishing tags.

I could always separate M/M Romance into it's own library and use a different template for it, but by that logic, I could side-step ALL of this by sorting these into different libraries, and then what would we do for fun?

Maybe this will help: only the FIRST match will be used. If there is an extra "romance" tag, then as long as you list it later on in switch() it will be ignored, since a previous tag has already been matched.

It follows a general concept of switch(pattern,value,else_if_pattern,value,else_if_pattern,value,else_value)

Originally Posted by ACGAuthor

LOL I would love to be able to do this, but I have so many books that belong to series that if I had everything sorted by series, I'd have 400+ collections, plus another 800 or so books that don't belong in any series.

I have a rather large library.

Fair enough, I have "only" 796 books. For now.
And a very good memory.

[unboggling]

Originally Posted by eschwartz

Hey, I'm just a facilitator. ...I know what genre each book is, and I don't care about it either; I choose based on series.

That's a good point. I've been conducting some experiments. Almost always, a "contains" search of combination (Comments, Series, Title) for keyword (such as magic, vampire, or etc) produces more results than a search of #genres for secondary genre (such as magic, vampire, or etc). So I'm wondering what good my secondary genres do me at all. Edit: Particularly since usually I originally determined the secondary genres by looking for those keywords in Comments, Series, or Title in the first place.

Edit: Now I remember: I track secondary genres because keyword search results can't be listed in a catalog, keywords in Comments can't be searched quickly in calibre large library and can't be included in a simple way (not needing constant maintenance) in a metadata plugboard.

[ACGAuthor]

Originally Posted by eschwartz

Maybe this will help: only the FIRST match will be used. If there is an extra "romance" tag, then as long as you list it later on in switch() it will be ignored, since a previous tag has already been matched.

It follows a general concept of switch(pattern,value,else_if_pattern,value,else_if_pattern,value,else_value)

Okay, so I would list the various Gay/LGBT/etc tags first to switch those to MM Romance, and then do a general "Romance" tag later and switch that to M/F romance? Yeah I can do that.

Fair enough, I have "only" 796 books. For now.
And a very good memory.

Again, for me it's a matter of being lazy. Well, not really lazy, but finding something generally involves a lot more button pushing than I want to do on my PW.

I hate the search function on the Kindle--a lot of the time if I'm searching for a specific title or author, I can't find it because something in the metadata subtly differs from what I'm searching for. A good example of this is when a book is written by two authors. You have to search by the one listed first, and a LOT of authors in my genre co-write.

Then, also, a lot of these are actually books I haven't read yet that I've received from other authors in my genre or that I've bought to catch up with the various authors in my genre (so I can be conversant with my peers about their books, or just to keep an eye on what is going on in the genre.)

So a lot of the time, choosing a book to read involves going down a list of authors I'm not actually all that familiar with, doing an eenie-meenie-miney-mo and saying, "I'll read THIS one!" Unless I have them sorted into easily digestible chunks, I would be scrolling endlessly though page after page after page (1800 books and they're listed at, what, 8 titles per page in list view?) That's a lot of paging forward.

[unboggling]

Originally Posted by ACGAuthor

...Again, for me it's a matter of being lazy. Well, not really lazy, but finding something generally involves a lot more button pushing than I want to do on my PW.

I hate the search function on the Kindle.... I would be scrolling endlessly though page after page after page (1800 books and they're listed at, what, 8 titles per page in list view?) That's a lot of paging forward.

Since you're talking about collections on devices, I'll throw in my other 2 cents:

I don't do collections on devices at all, for similar reasons you mentioned. Lately in metadata plugboard I've been appending series, series index, and #genres to author_sort to show in Sony PRS-T2 list. That [edit: along with title] is really all I need on the device. Why do collections on devices when everything is already organized in calibre? Makes more sense to me to do all the searching in calibre, not on device. Speaking of simplifying, not doing collections at all is simple.

[ACGAuthor]

Originally Posted by unboggling

Since you're talking about collections on devices, I'll throw in my other 2 cents:

I don't do collections on devices at all, for similar reasons you mentioned. Lately in metadata plugboard I've been appending series, series index, and #genres to author_sort to show in Sony PRS-T2 list. That is really all I need on the device. Why do collections on devices when everything is already organized in calibre? Makes more sense to me to do all the searching in calibre, not on device. Speaking of simplifying, not doing collections at all is simple.

You mean just transfer the books from Calibre when I decide I want to read them? Again, definitely a more basic solution. I've been going back and forth on that with myself. The OCD part of me is convinced that if I don't keep my whole library on my Kindle, then I will of course not have the exact book I want when I want it and when I'm unable to get to my computer. The rational part of me knows this isn't true, but... It is definitely an option I consider from time to time.

[unboggling]

Originally Posted by ACGAuthor

A good example of this is when a book is written by two authors. You have to search by the one listed first, and a LOT of authors in my genre co-write.

In calibre, for identical sets of co-authors, I always put those co-authors in the same order. One reason is because first author in the Authors field list is important for sorting on device, as you mentioned.

Originally Posted by ACGAuthor

You mean just transfer the books from Calibre when I decide I want to read them? ....

Yup. That's what I do. Load a few books at a time to device then remove them after reading them before loading any more books. The only time I might load more than a few is just before a trip out of town, but that isn't really necessary. While traveling, from content server I can download anything to device from calibre library at home.

[ACGAuthor]

Originally Posted by unboggling

In calibre, for identical sets of co-authors, I always put those co-authors in the same order. One reason is because first author in the Authors field list is important for sorting on device, as you mentioned.

Yeah, I do too, when I realize the metadata has them in other orders.

Yup. That's what I do. Load a few books at a time to device then remove them after reading them before loading any more books. The only time I might load more than a few is just before a trip out of town, but that isn't really necessary. While traveling, from content server I can download anything to device from calibre library at home.

I haven't tried playing around with content server yet. Is it basically just a way to send the books to your device via wifi rather than USB?

[unboggling]

Originally Posted by ACGAuthor

...I haven't tried playing around with content server yet. Is it basically just a way to send the books to your device via wifi rather than USB?

Yes, that is one way. Also content server can be set up to be accessible from the internet from anywhere. Content server can be accessed and searched using simple browser on device, download book from there to device.

Another alternative is using Dropbox. You can access books through Dropbox from device, if you can install mobile version of Dropbox on device. Not sure about Kindle PW.

And there are other alternatives you can read about in Related Tools, Calibre Sub-Forum. See stickies at top for specific tools.

But since we are digressing off the thread topic, I'll shut up now.

[eschwartz]

Originally Posted by eschwartz

Problem from before:





Stupid me, I know what the problem is here. I created a new capture group at the beginning to remove leading spaces, but left the replace code replacing the nonexistent capture group. Properly, it should've been this (fix in red):

Name: #initials. Template: {#stripped_series:re(([\s])?([^\s])[^\s]+(\s|$),\2)}

Or really this:

Name: #initials. Template: {#stripped_series:re([\s]?([^\s])[^\s]+(\s|$),\1)}

since I can't think why I created a capture group in the first place -- it does nothing.

And without the formatting error, it seems to work perfectly. Try this again?

EDIT: @chaley, perhaps this should be updated in the first post? This is derived from what Kovid used in his general program mode guide. I used [^\s] instead of [\w], following your lead, but added the inverse at the beginning, like Kovid's example.

EDIT by chaley: done.

Thanks, chaley!

[Gary_M_Mugford]

Originally Posted by unboggling

In calibre, for identical sets of co-authors, I always put those co-authors in the same order. One reason is because first author in the Authors field list is important for sorting on device, as you mentioned.

Yup. That's what I do. Load a few books at a time to device then remove them after reading them before loading any more books. The only time I might load more than a few is just before a trip out of town, but that isn't really necessary. While traveling, from content server I can download anything to device from calibre library at home.

I'm of the same mind. Part of the issue is my determination not to go singularly on any series. Well, anything more than three books. I very diligently switch genres with each reading experience. I don't have a set pattern. It doesn't always go Mystery, Sci-Fi, Non-Fiction, Thriller, Humour/Sports Slot, repeat rinse and cycle again. But that's close to what turns up. I also try to follow something gut-wrenching with something in a funny vein. As long as I have a half-dozen books of each genre (sorted by desirability, genre, author and then series/series Index) then my Kindle is fully equipped. It won't stand up necessarily to my being kidnapped by gangsters and held for ransom. But I suspect the charge will run out long before somebody ransoms me anyways

[Gary_M_Mugford]

One of the reasons I still monitor this thread, long after Chaley provided me with the solution that I use to organize my Kindle books, is because of the enthusiasm and ideas that it is bringing forth. Particularly of late ESchwartz. Thank you for proving even the best solution can be improved upon.

Frankly, what little I have left on my desire list (some way to optionally cut off any incoming book's title at the first bracket/brace/parenthesis or colon) is to handle those tags a lot better as I transition them into my few primary tags (#tagsgm). As described earlier, I have saved searches that looks for uncategorized books and looks for terms inside tags. I then select all and edit in the tag category I believe them all to be. Running through my set of tagsGM takes me five minutes. But it leaves a LOT of books uncategorized, which I then manually assess and assign. THAT takes the time.

The issue is handling the tags terms. Biographies is different from Biographies & Memoirs is different from Biographies And Memoirs. And so on. I have the main terms covered. Fine-tuning it has proven beyond my capabilities and/or my time allowances. Still, this thread has re-invigorated that desire and I am more actively interested. I like the idea of a program that digests tags sets and produces a value in a created column. In delphi, I'd just split the tags into a stringlist, all caps. Then I would search each item for whether pos(searchterm, item.text) > 0 and then edit the created column for the pre-determined value. I'd be able to nest the searches in the order I WANT, so that Mysteries and Thrillers falls into the Mystery category rather than the Thriller category, no matter where that individual tag lands in the tag list. There's also the problem of Science Fiction Mysteries (which I would prefer to be in SF, rather than Mystery) having to be assigned in such a way (either by going first or last, depending on the methodology), so that I get the assignment I desire.

But Python is not Delphi and my knowledge is ... lacking. Doing the above, splitting the tag set, capitalizing the results and then comparing it to my tagsGM list, sub-term by sub-term is beyond me. But I'm watching your examples, playing with your programs. So keep up the work. Please. You'll be helping more than ACGAuthor.

Thanks.

[Gary_M_Mugford]

I should also mention one other wrinkle that would crop up. Sports Biographies. I TEND to manually categorize them as Sports, but that really didn't work all that well in practice. So, now they are BSports, leaving me with Bios that are not sports, and Sports for books that aren't Bios. The issue is writing that search. And thus far, I have failed due to the lack of understanding of negation in the search line.

Food for extra thought.