Plugboard, template, and custom composite column recipes

[Gary_M_Mugford]

Originally Posted by speakingtohe

@Gary_M_Mugford
Kindle sucks at a lot of things. TTS is especially unpleasant. Web browsing... I have yet to get interested enough to try it, and I am generally curious about these things.

I got the kindle 3 for my mother (85)and suspect that it was a mistake. However she can have my Sony if she wants it instead. Next week is her birthday so we shall see

I do use the kindle collection manager while putting books on for her as Kindle menus seem a bit haphazard without it.

Helen

Helen,

I got my mom, who's just a little younger than your mom, a Kobo on Mother's Day this year. Despite the fact that the first one died two weeks later (putting in a too-large SD Card MIGHT have had something to do with it) and the replacement died in June after a Chapters store clerk bricked it when trying to 'fix' a page-skipping issue, Mom loves her basic little reader. The first replacement took three weeks, the second one about a week. So, given the slothliness of the Kobo Classic in turning on and turning pages, you'd think she'd at least LOOK at the Kindle3 when she was here this week. Nope. And now that she has the routine of adding books from Calibre to her Kobo down pat, I suspect she'll be blissfully finishing off the last of the Fern Michaels and Beverly Barton books she hasn't read. I figure that'll keep her contented until about next September.

Like you, I'm using the Kindle3 as a reader. I can gripe about the title browser, but honestly, Chaley has made a major in-roads on that area as far as I'm concerned. I'd like 'this and that' to be better, but can live with 'this and that' being just what they are right now. I knew I wasn't getting a device that would read 8x10 PDFs (An Android Tablet of some sort will eventually scratch that itch). For me, the Kindle3 is doing exactly what I want and I am contented.

GM

[speakingtohe]

Yes Gary, I was a little too disparaging about the kindle They really are quite nice and have lots of features that Sony lacks. And vice versa.

I was talking to a guy on the seabus last week who hadn't seen the kindle and was wanting to get one for himself and his mother. I told him you can't go wrong with either

And chaley is an awesome guy with amazing abilities and patience with us all. He has helped a lot with the kindle and doesn't even own one

Helen

[Mriswith2k]

Originally Posted by chaley

General Metadata
Metadata: Put series into the title, using either initials or a shortened form. Strip leading articles from the series name (any) (inspired by Gary_M_Mugford)

Spoiler:

The solution requires creating three composite columns. The first column is used to remove the leading articles. The second is used to compute the 'shorten' form. The third is to compute the 'initials' form. Once you have these columns, the plugboard selects between them. You can hide any or all of the three columns on the library view.

First column:

Name: #stripped_series. Template: {series:re(^(A|The|An)\s+,)||}

Second column (the shortened form):

Name: #shortened. Template: {#stripped_series:shorten(4,-,4)}

Third column (the initials form):

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

Plugboard expression:

Template:{#composite:lookup(.\s,#initials,.,#shortened,series)}{series_index:0>2s| [|] }{title}
Destination field: title

This set of fields and plugboard produces:
Series: The Lord of the Rings
Series index: 2
Title: The Two Towers
Output: LotR [02] The Two Towers

Series: Dahak
Series index: 1
Title: Mutineers Moon
Output: Dahak [01] Mutineers Moon

Series: Berserkers
Series Index: 4
Title: Berserker Throne
Output: Bers-kers [04] Berserker Throne

Series: Meg Langslow Mysteries
Series Index: 3
Title: Revenge of the Wrought-Iron Flamingos
Output: MLM [03] Revenge of the Wrought-Iron Flamingos

After trying to get that template setup to work for the last hour or three , I finally found what's broken in the template, the:

Template:{#composite:lookup(.\s,#initials,.,#shortened,series)}{series _index:0>2s| [|] }{title}
Destination field: title

should instead be

Template:{#stripped_series:lookup(.\s,#initials,.,#shortened,series)}{series _index:0>2s| [|] }{title}
Destination field: title

I assume you used '#composite' to test out the the original 'stripped_series' output and forgot about it since your output would still be coming out correct , cheers for making / collecting all these awesome templates as well, saved me from continuing renaming all my titles manually before sending them to my kindle >_>

[speakingtohe]

Create a plugboard
Format: device_db
Device: PRS505
Source template: {series:re(([^\s])[^\s]+(\s|$),\1)}{series_index:0<2s| - | - }{title}
Destination: Title

This would produce for your example of a few posts back:

Series: Meg Langslow Mysteries
Series index:3
Title: Revenge of the Wrought-Iron Flamingos

Title on Sony: MLM - 03 - Revenge of the Wrought-Iron Flamingos

I spoke a little too soon
Instead of
Title on Sony: MLM - 03 - Revenge of the Wrought-Iron Flamingos
I get
Title on Sony: MLM - 30 - Revenge of the Wrought-Iron Flamingos

And no leading zeros anywhere. Didn't sort too good. 2 and 20, 3 and 30 were the same series number
I tried every which way to figure it out, used kindle, custom column template etc. and same result.

Form the first post in this thread I managed to derive

{series:re(([^\s])[^\s]+(\s|$),\1)}{series_index:0>2s|| - }{title} - {series}

Which gives me
MLM03 - Revenge of the Wrought-Iron Flamingos - Meg Langslow Mysteries

So far I am happy with that and would never have arrived at it by myself.

Thanks
Helen

[chaley]

@Mriswith2k: sorry about that. My test setup has 8 columns named #composite, #oomp2, #comp3, etc. I wanted reasonable names in the post, but was too lazy to redo my columns. Clearly I didn't get it right.

@speakingtohe: again, sorry about that. There is a typo in the template. The < character near series_index should be a > character. Not sure how this happened, because it is correct in my test column.

The correct string is:
Source template: {series:re(([^\s])[^\s]+(\s|$),\1)}{series_index:0>2s| - | - }{title}

[Gary_M_Mugford]

Been playing with the plugboard for the smart series prefix. Now I tried to add a suffix that would show me a brief shortening of the tag info. I really only use tags on a very basic level. In fact, most of the tags are blank, indicating a SF book. So I created a composite genre column with this template:
{tags:.2:ifempty(SF)}

Works. Mostly. Except in situations where the tag function has two entries. Say Bio,Humour. In ALL such cases, the result is the first two letters of the last tag entry. Hu in the cited case.

Now, in some cases, this actually has a desired affect. I have tags that read AJ,YA and Angela,YA for my nephew and niece (I try to pre-read whatever I give them). The genre is showing up as YA. And I can certainly live with that.

Ultimately, I LIKE the overall result of the last tag being the one that is used. But can I count on that behaviour going forward, as it would seem to be a sort of bug?

Regards, GM

[chaley]

Originally Posted by Gary_M_Mugford

Ultimately, I LIKE the overall result of the last tag being the one that is used. But can I count on that behaviour going forward, as it would seem to be a sort of bug?

No, not a bug, and no, you can't count on it.

The order of individual tags in tags fields is not defined. My suspicion is that the order is the order of initial creation, but I haven't taken the time to verify the supposition. However, being undefined means that the order can be considered accidental, and could change.

In your case, what you are seeing as the 'last' tag is really the first. The different views arise because calibre sorts the tags when they are displayed on the library view.

If you want the first two letters of each tag, try
{tags:re( *(.[^\,]?)[^\,]*?((\,)|$),\1\2)}
If you also want the isempty part, then you will need to create a composite column containing the above :re, then apply the isempty test to that column.

[Sweetpea]

I've to ask, what does the "|$" do?

[chaley]

Originally Posted by Sweetpea

I've to ask, what does the "|$" do?

Before explaining, I note that the expression can be simplified to
{tags:re( *(.[^\,]?)[^\,]*?(\,|$),\1\2)}
The extra parentheses were left over from an experiment.

The expression (\,|$) means (comma OR end_of_string). It causes the previous [^\,]*? to stop matching, and consumes the comma if there is one. As a side benefit, because it is parenthesized, the matched value can be used in the replacement to separate the two-letter shortened tags by that value, which is either a comma or nothing.

As long as I am in explain mode , the explanation of the rest of the expression is:

re( *(.[^\,]?)[^\,]*?(\,|$),\1\2)
    | |    |     | |   |    |_ Replace with first paren group value, followed by second
    | |    |     | |   |_Match comma or end of tags string. In parens, so remember as \2
    | |    |     | |_ As many non-comma chars as needed
    | |    |     |_Any character not a comma
    | |    |_zero or one non-comma. Still in the parens, so remembered
    | |_ Any character. It is in parens, so remember it as \1
    |_eat any leading blanks. They would be there because commas are followed by blanks.

[Gary_M_Mugford]

Originally Posted by chaley

No, not a bug, and no, you can't count on it.

The order of individual tags in tags fields is not defined. My suspicion is that the order is the order of initial creation, but I haven't taken the time to verify the supposition. However, being undefined means that the order can be considered accidental, and could change.

In your case, what you are seeing as the 'last' tag is really the first. The different views arise because calibre sorts the tags when they are displayed on the library view.

If you want the first two letters of each tag, try
{tags:re( *(.[^\,]?)[^\,]*?((\,)|$),\1\2)}
If you also want the isempty part, then you will need to create a composite column containing the above :re, then apply the isempty test to that column.

Charles,

Hmmmm. Interesting. Since I use so few tags and I tend to put them in basically in the order they naturally occur, I could also very much live with first-in as the default single tag to shorten. In fact, I'd VERY MUCH like that to be the default case.

What I'm doing, of course, is to cheat and not use categories on the Kindle. I don't have the desire nor the time to handle categories as per Kindle's methodology. But I wouldn't mind the books being sorted BY category (i.e. tag). So I am going to use the 'current' behaviour until it changes. The template I am using for the plugboard is:

[{#genre:}] {#stripped_series:lookup(.\s,#initials,.,#shortened,series){series_index:|-|-}{title}

And yeah, I'm willing to accept not having leading zeroes in the index. Since I delete books as I read them, I tend not to have xxxx10 and xxxx9 in that order for too long. Just a personal preference.

Last aside, using this, it was interesting to see that I had not tagged a couple of humourous biographies as such. Sooo, I ended up with both a Stephen Fry bio and Anna Gurwitch's Fired being listed as SF. I was able to correct both post-haste and re-export to the Kindle.

At any rate, I will be happy with the current state of the plugboard until the behaviour for genre creation changes. In the meantime, thanks so very much for your efforts.

GM

EDIT: I just noticed something taking a quick re-read. I copied and pasted the plugboard template into this post. It doesn't have a space before the last d in shortened in the template. Yet, when I pasted it here, there WAS a space. And now I remember that was the case with a couple of other posts with code in them. In my editor here, it took two lines in the editor and NOW I wonder if the space was inserted there to break up the code. At any rate, might be worth being on the lookout for. I have corrected the code to remove the inserted space. Or, at least I tried. Still seems to be there. By the way, message #78 has the same issue. Solved by pasting it into code wrapper. Live and learn.

[Gary_M_Mugford]

Charles,

One last aside to the previous last aside. Did discover one little issue with the shorten function for series. It doesn't handle Frank Stewart (Columns), which, admittedly, could be Frank Stewart Columns. (There is a Frank Stewart Books, but I wanted the columns to come BEFORE the books in my export list. So I used the bracketing to accomplish that). What shorten outputs to the Kindle is FS(. I doubt very much that there is anything that can be done about it within the regex, but I thought I should point it out, just in case.

In reality, I think this is probably a very much improbable situation and no time at all should be wasted on it.

GM

[chaley]

Originally Posted by Gary_M_Mugford

What shorten outputs to the Kindle is FS(. I doubt very much that there is anything that can be done about it within the regex, but I thought I should point it out, just in case.

Try using the regexp
re([^\w]*(\w?)[^\s]+(\s|$),\1)
in the #initials column. This one will put only alphabetic characters in the output sequence.

[speakingtohe]

@speakingtohe: again, sorry about that. There is a typo in the template. The < character near series_index should be a > character. Not sure how this happened, because it is correct in my test column.

The correct string is:
Source template: {series:re(([^\s])[^\s]+(\s|$),\1)}{series_index:0>2s| - | - }{title}

Stupid of me not to have noticed the </> but I didn't and changed everything else to try to get it to work

I haven't got it to work onthe Sony PRS-600 yet but on the kindle it is pretty good.

Shipped my mothers kindle off to her today and have reasonable hopes of her being able to use it without too much pain to me.
Thanks again for the help, above and beyond....
Helen

[Sweetpea]

Would it be possible to make a custom composite column that limits the text you can add?

Example: I now use the "tri-state boolean" to keep track of read/to-be-read/reading books. But, I'd like to use the normal boolean for other columns. But I can't have both (or can I?).

[chaley]

Originally Posted by Sweetpea

Would it be possible to make a custom composite column that limits the text you can add?

Yes, it would be possible to create a text column type that is actually an enumeration. There are, of course, complexities. The biggest is ensuring that all the places that can change the contents respect the enumeration values. This will touch metadata editing, drag & drop, the CLI (I think), and the library GUI.

Example: I now use the "tri-state boolean" to keep track of read/to-be-read/reading books. But, I'd like to use the normal boolean for other columns. But I can't have both (or can I?).

Nope. T'is all or nothing.

Thinking about it, you can get a poor-man's enumeration type using a normal and a composite column. The normal column would be used to enter the information. The composite column would use a switch statement to limit the text. For example, if the first column is integer, then the switch would test for 1, 2, etc. If the value is not one of those, it takes the default.