Better Hyphenation

[PunkyKoopa]

In order to have hyphens more to my liking, could I unzip the .tgz file, change the values of LEFTHYPHENMIN 2 and RIGHTHYPHENMIN 3 to LEFTHYPHENMIN 3 and RIGHTHYPHENMIN 4, and then create a .tgz again?

[JSWolf]

Originally Posted by PunkyKoopa

In order to have hyphens more to my liking, could I unzip the .tgz file, change the values of LEFTHYPHENMIN 2 and RIGHTHYPHENMIN 3 to LEFTHYPHENMIN 3 and RIGHTHYPHENMIN 4, and then create a .tgz again?

I would change the LEFTHYPHENMIN to 3 and leave the other as is and see how you get on.

[PunkyKoopa]

I will try with 3 and 3 then.

The process to do it is as I said, right? Unzip, go to the dictionary, change the values, and create the .tgz again, right?

[JSWolf]

Originally Posted by PunkyKoopa

I will try with 3 and 3 then.

The process to do it is as I said, right? Unzip, go to the dictionary, change the values, and create the .tgz again, right?

Or you can extract the dictionary, edit it, and put it back over the existing dictionary.

[PunkyKoopa]

Do you mean extract from the Kobo?

I was talking about the download from the first page of the topic.

[JSWolf]

Originally Posted by PunkyKoopa

Do you mean extract from the Kobo?

I was talking about the download from the first page of the topic.

So was I. Extract from the KoboRoot.tgz, make the edit and put the file back. That makes less chance of an error in building the KoboRoot.tgz.

[PunkyKoopa]

I see, but I can’t see the difference. Maybe I am missing something but I think that we are saying the same thing.

[JSWolf]

Originally Posted by PunkyKoopa

I see, but I can’t see the difference. Maybe I am missing something but I think that we are saying the same thing.

Maybe.

Just edit the dictionary with a program that will not change the line endings such as Notepd++ (if on Windows) and rebuild KoboRoot.tgz so it's the same as it was and you're good to go.

[.tycho]

Originally Posted by .tycho

Hi everyone,

I was hoping to get some help, please. I'm reading KePUBs on my Libra 2. I've installed JSWolf's dictionary from the first page, and in Calibre, I have KTE set to:

Hyphenate files: On
Min word length: 6 characters
Min characters before hyphen: 2 characters
Min characters after hyphen: 3 characters
Max consecutive: 2 lines

I sent my book to the device, and here's an example screen:
Attachment 208424

I'm a bit confused. Why is "th-ing" being hyphenated? It's only 5 characters. Any thoughts? Also, do the settings in KTE override the LEFTHYPHENMIN and RIGHTHYPHENMIN in the hyphenation dictionary?

Thx!

Ok, so I did some more testing and I think I figured out what's going on. It looks like kepubs do NOT respect the minimum word length setting. Hence, why "thing" became "th-ing", because it met the 2 characters before and 3 characters after the hyphen setting.

So what I've done is change the setting in KTE to 3 characters before and 4 characters after, so that only words 7 characters or more get hyphenated. Now, things look MUCH better in my kepubs.

I'm curious what you think about my approach, and thanks for your help!

[JSWolf]

Originally Posted by .tycho

Ok, so I did some more testing and I think I figured out what's going on. It looks like kepubs do NOT respect the minimum word length setting. Hence, why "thing" became "th-ing", because it met the 2 characters before and 3 characters after the hyphen setting.

So what I've done is change the setting in KTE to 3 characters before and 4 characters after, so that only words 7 characters or more get hyphenated. Now, things look MUCH better in my kepubs.

I'm curious what you think about my approach, and thanks for your help!

That should work. It's a good solution.

[Usered]

What do the 1!1 1,1 etc at the beginning of the english hyphenation dictionary mean? I tried using them on the italian one but that resulted in words being splitted just a letter before the punctuation, like this: "Isabe-l." Probably the dot is counted as part of the word? I resolved that by eliminating the first 1 from the rules in the first section of the hyph file, but I don't know how they work or how that resolved the issue.
Also, in my italian file there's a rule like this one: l'2, which is intended to impede the hyphenation right after the apostrophe for word like l'hai. But as I was reading I found precisely that horrible instance of hyphenation: l'-hai. Does anyone know how to counter this by adding or modifying the rules?

[JSWolf]

Originally Posted by Usered

What do the 1!1 1,1 etc at the beginning of the english hyphenation dictionary mean? I tried using them on the italian one but that resulted in words being splitted just a letter before the punctuation, like this: "Isabe-l." Probably the dot is counted as part of the word? I resolved that by eliminating the first 1 from the rules in the first section of the hyph file, but I don't know how they work or how that resolved the issue.
Also, in my italian file there's a rule like this one: l'2, which is intended to impede the hyphenation right after the apostrophe for word like l'hai. But as I was reading I found precisely that horrible instance of hyphenation: l'-hai. Does anyone know how to counter this by adding or modifying the rules?

I have no idea. But as it's in the English hyphenation dictionary, I'd just leave it there. All I know is it works for English.

[Usered]

Pity
I thought the rules in the first section could signal the punctuation marks as independent and not to be counted for hyphenation, but it wasn't working for me as it was and I'm not sure that just because they're there they're correct (someone could have added them tentatively?). I gathered that the 1 means a possible hyphenation point but it wouldn't make sense there.
Is there some provedly good italian hyph file? Maybe mine is not the best. Thanks for answering anyhow.

[JSWolf]

Originally Posted by Usered

Pity
I thought the rules in the first section could signal the punctuation marks as independent and not to be counted for hyphenation, but it wasn't working for me as it was and I'm not sure that just because they're there they're correct (someone could have added them tentatively?). I gathered that the 1 means a possible hyphenation point but it wouldn't make sense there.
Is there some provedly good italian hyph file? Maybe mine is not the best. Thanks for answering anyhow.

Thing is, I've never seen how a hyphenation dictionary is made. If you could find a site that gives the rules, that would be good.

[hengyu]

How do these hyphenation dictionaries persist in the device?

Do I need to re-install it when I update the firmware, or when I do a logout-login DB rebuild, or when I do a factory reset?