Quote:
Originally Posted by Nick Payne
In the third line of the para that begins "He walked round my office", the justification is bad, and the em dash is hard up against the word to its left with all the spacing on the right. When I load the book into Sigil and have a look at the text, there is no space on either side of the em dash, and there is nothing I can see that should prevent the justification from being correct.
|
I have noticed your problem too.
- (arbitrary?) remove a whitespace character (which is present in "code view")
- at the end of a line there is a space (about the size of one or two characters; the last word is or seems not aligned to the right page margin).
This happens if the problematic line has too few characters. And the "hyphenation routine" can not hyphenate the first word on the next line ("Minister").
There is IMO no solution to fix this permanently.
To reduce this justification issue you can edit, the hypenation rules of your language.
See also this thread
Hyphenation
Code:
UTF-8
LEFTHYPHENMIN 2
RIGHTHYPHENMIN 3
COMPOUNDLEFTHYPHENMIN 2
COMPOUNDRIGHTHYPHENMIN 3
Lower values for LEFTHYPHENMIN and RIGHTHYPHENMIN means MORE aggressive hypenation.
Build a patch for the hyphenation dictionary on to the Kobo. Containing only:
Code:
.\usr\local\Kobo\hyphenDicts\hyph_en.dic
.\usr\share\hyphen\hyph_en_EN.dic
The easiest way to this is to download the firmware for your device (see the direct firmware links section). Then open the firmare archive with 7-Zip and
delete everything except the hyphenation dictionary files of your langage (you should keep the English dictionaries). Then close the archive and save the modified archive file by prompting "yes" when asked.
Then open the archive again and edit both "hyph_en.dic" and "hyph_en_EN.dic" by reducing the values for LEFTHYPHENMIN and RIGHTHYPHENMIN.
This is a trade off for your issue as it can create an other ugly issue.
If both values are set to four:
Mini-ster.
Suppose you set RIGHTHYPHENMIN to two. Then it can look like this:
Ministe-r. (as the dot is also count as a character; as is a comma question mark, etc.).
It might be that the COMPOUNDLEFTHYPHENMIN and COMPOUNDRIGHTHYPHENMIN fixes this issue. I don't know. I use the Dutch hyphenation dictionaries and they do not contain these parameters.