View Single Post
Old 02-20-2013, 01:49 AM   #36
roger64
Wizard
roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.roger64 ought to be getting tired of karma fortunes by now.
 
Posts: 2,608
Karma: 3000161
Join Date: Jan 2009
Device: Kindle PW3 (wifi)
ABOUT SPELLCHECKING

Quote:
Originally Posted by meme View Post
@roger64
And since you mention Spellchecking - how is this version working for you. The issue with apostrophes should be sorted out - well, at least better than it was hopefully.
To say it in a few words: it was a real pleasure to test it. The "list of mispelled words" is a killer feature which makes v. 0.7 a worthy upgrade and a reward for hard-building Linux users (and others).
- The processing is very quick, very convenient.
- All the previous apostrophe problems (curly vs straight) are solved.
- Adding the words in the user dictionary is a also very handy feature as we can do it selecting one, many or all words of the list.

Congratulations.

Proposals

I would suggest some small things in no particular order for next version:

- Propose (preferences) a comfortable "spellcheck display mode ", separating the text and the list (like the one in the screenshot below).

- The user should have (preferences) a choice of seeing the list
- either your way: with UPPER characters first and then lower ones
- or in strict alphabetical order.
There are advantages and drawbacks for both.

- To say it precisely, the words on the list are either unknown and/or mispelled. A vast majority of them are even perfectly correct. They are just "un-hunspelled", if I may. This should be corrected.

- The spellcheck window is to be translated.

- I would propose to add the number of lines included in the list (as a bottom line info?).

- Some few words are truncated by code (dropcap, sup, span, i or b tags) and just appear on the list for this reason. It's not a real problem but maybe it would be good to make hunspell ignore all this kind of code and plug the letters on both sides if there is no white space of any kind in between.

A question to finish:

when you have a list of elided forms, which precise command do you use to make hunspell analyse it? If I use the plain
Code:
-l
argument, it declares all of them to be unknown...

AN OLD QT BUG SQUASHED!! (at last)

The old one, long overdue, about italics not being displayed with embedded italic fonts. I just noticed that, with Sigil 0.7 (and Qt5) t's gone!!!
Attached Thumbnails
Click image for larger version

Name:	capture-screen-2013-02-20-122115.png
Views:	229
Size:	178.8 KB
ID:	101589  

Last edited by roger64; 02-20-2013 at 06:55 AM. Reason: Bug squashed
roger64 is offline