Optimize Kobo span generation about calibre-kobo-driver HOT 3 CLOSED

In order to split only on ".", "!" and "?", and to avoid adding a span around single puctuation marks, i.e.

<span class="koboSpan" id="kobo.x.y">.</span>

you could chage the regex in line 468 to look like this:

groups = re.split(ur'(.*?[\.!?][\'"\u201d\u2019]?)', node, flags = re.UNICODE | re.MULTILINE)

I've tested it and it works, but then the conversion eats a space before other tags like <em> or <b>, i.e.

<p>Some text. <em>Italic text.</em></p>

i.e.
Some text. Italic text.

is changed to

<p><span class="koboSpan" id="kobo.1.1"><span class="koboSpan" id="kobo.1.2">Some text.</span></span><em><span class="koboSpan" id="kobo.1.3"><span class="koboSpan" id="kobo.1.4">Italic text.</span></span></em></p>

i.e.

Some text.Italic text.

with the space removed...

from calibre-kobo-driver.

Awesome, thanks! I'll play with this when I get some time.

from calibre-kobo-driver.

I have some good news and some bad news: the good news is that the new regex works fine, the bad news is that the space before the italics is probably lost by fix #47 ...
I had the same results (i.e. missing spaces between plain and italic texts) also with the vanilla driver as downloaded from github (version 1.3.3) :-(

from calibre-kobo-driver.