02-17-2013, 10:02 PM | #16 | |
What the Dog Saw
Posts: 311
Karma: 981684
Join Date: Jul 2008
Location: Dunn Loring
Device: Sony PRS-650, Surface3
|
Quote:
There must be something quirky on the first XP SP3 computer. I can re-install 0.6.2 and that version runs fine. |
|
02-18-2013, 02:06 PM | #17 |
Addict
Posts: 320
Karma: 56788
Join Date: Jun 2011
Device: Kindle
|
Absolutely beautiful work, guys. I echo the sentiment that the new spellcheck dialogue is better designed than spellcheck dialogues for dedicated word processing software. Huge kudos.
I've only run into one bug with the "Clips" window. Namely, it doesn't work within the ncx file. Clips are applied to the ncx file if they are accessed through the context menu, but not when you try to apply it from the panel. |
02-18-2013, 02:48 PM | #18 | |
Sigil developer
Posts: 1,275
Karma: 1101600
Join Date: Jan 2011
Location: UK
Device: Kindle PW, K4 NT, K3, Kobo Touch
|
Quote:
There are a few ways to apply clips. Can you test each one. For each one click in the NCX file, and then try Paste Clip in the Clips Editor; try Insert->Clip->clipname; try clicking on a button in the Clip Bar; try using the View->Clips window and selecting a clip; right-click on the NCX and select a Clip; and, almost forgot this one, click in the NCX and then use a clip shortcut (e.g. ctrl-alt-1). |
|
02-18-2013, 04:15 PM | #19 |
Addict
Posts: 320
Karma: 56788
Join Date: Jun 2011
Device: Kindle
|
|
02-18-2013, 09:57 PM | #20 |
Addict
Posts: 320
Karma: 56788
Join Date: Jun 2011
Device: Kindle
|
Hmmm... sorry for the double post, but I've run into something kind of funky. "Count" and "Go to next" are failing to recognize matches throughout the document if the search is begun on a page in which there's no match for search.
I've noticed this on two separate occasions, but it's probably happened more than that, and simply gone unnoticed. these are the two regex searches (loaded from my saved searches) that have failed. Code:
<p [^>]*>(.*?)<a [^>]*>(.*?)</a>(.*?)</p> Code:
<p class="brheavy"><br /></p>\s+<p class="calibre1"> Code:
<p class="center"><a href="../Text/File_01.htm">Some</a> text<a id="qewrqwe"></a></p> Code:
<p class="brheavy"><br /></p>
<p class="calibre1">some text</p>
The first search type sometimes works... I can't replicate the error consistently. But the second one always fails. |
02-18-2013, 10:54 PM | #21 |
Sigil & calibre developer
Posts: 2,488
Karma: 1063785
Join Date: Jan 2009
Location: Florida, USA
Device: Nook STR
|
@ElMiko, Could the searching issue be due to wrapping being disabled? If you put the mouse cursor at the top of the document does it match. If you put it at the bottom does it not match?. If you expand the F&R options (the arrow on the bottom left) is wrap checked?
I tried the test expressions and content you posted and I could only get it to not match if wrapping was disabled. Though it looks like Count All might need to be renamed to Count if it's honoring wrapping. |
02-18-2013, 11:11 PM | #22 |
Addict
Posts: 320
Karma: 56788
Join Date: Jun 2011
Device: Kindle
|
|
02-19-2013, 04:32 AM | #23 | |
Sigil developer
Posts: 1,275
Karma: 1101600
Join Date: Jan 2011
Location: UK
Device: Kindle PW, K4 NT, K3, Kobo Touch
|
Quote:
Also, check your setting is Regex and All HTML Files (if you are expecting to search across files). And check your Clean Source settings are Pretty Print Tidy for Save/Open/Replace All (or if not let us know). Count All and Replace All both honour the wrap setting. The do count or replace all occurrences from your position to the end if no wrap is set. |
|
02-19-2013, 06:22 AM | #24 |
Guru
Posts: 655
Karma: 64171
Join Date: Sep 2010
Location: Kent, England, Sol 3, ZZ9 plural Z Alpha
Device: Sony PRS-300, Kobo Aura HD, iPad (Marvin)
|
ElMiko, Not helping with current problem, but ... regarding your second search, you may want to change the calibre class match <p class="calibre1"> to <p class="calibre\d+"> as calibre rarely converts two files the same and the calibre# is usually a different number, that may match more incidents.
|
02-19-2013, 07:19 AM | #25 |
Well trained by Cats
Posts: 29,800
Karma: 54830978
Join Date: Aug 2009
Location: The Central Coast of California
Device: Kobo Libra2,Kobo Aura2v1, K4NT(Fixed: New Bat.), Galaxy Tab A
|
@user_none
I have been seeing some inconsistent (7.x) REGEX S&R recently I might be confused about the use of #1 (I am not a REGEX guru ) <b class="foo">words</b> <b class="foo">more words</b></p> 1)match Minimal ticked: Code:
(?U)<b class="foo">(.+?)</b>\s I did something else then came back: Then it just started working . I tried just the minimal tick, Just the (?U) , both. 2) wrap Not Ticked still wraps within the current file if All HTML selected and using Replace/Find (I was wrapping carefully chosen words in a tag pair and noticed OMG, I had already wrapped that word before, twice ) Wrap was not ticked |
02-19-2013, 09:19 AM | #26 |
Wizard
Posts: 2,608
Karma: 3000161
Join Date: Jan 2009
Device: Kindle PW3 (wifi)
|
Linux Install troubles
The good news is that they are solved - for me - and I have Sigil 0.7 running on LMDE 64 bits. But I found the install very uneasy. I do not know why, but some locked tmp folders have been created during the install, and I could not even launch Firefox except with sudo... I had to reinstall for this reason. Even now, I can launch Sigil only with a script giving him every time the location of the Qt5 library (this idea coming from Diap Dealer). I hope I am alone facing these hurdles, because the number of Linux users may drop dramatically if this install is not made clearer. Here is the install record if this can help the next one.. Beware of the paths, beware of everything... Spoiler:
And a screenshot of the- already famous - spellchecking novelty... |
02-19-2013, 10:37 AM | #27 | |
Grand Sorcerer
Posts: 27,549
Karma: 193191846
Join Date: Jan 2010
Device: Nexus 7, Kindle Fire HD
|
Quote:
The problem is... it's not really an install, it's a build. As such, it will never be completely easy/clear. That's just the nature of custom compilation—it's not for the "casual" Linux user. I agree with you—to some extent—about possibly losing Linux users. But at the same time... I also recognize the need for the Sigil devs to wash their hands of any Linux "install program." In the Linux tradition, it's the duty of repository maintainers to provide a simple binary installation package (ensuring all necessary dependencies) for open-source software and not the software developers. Unfortunately, this means that the easy/clear installations of your favorite open-source software will never be the most up to date versions. There's a lag. If you want the cutting edge versions, there will always be hoops to jump through in setting up a proper build environment. If you want more recent versions of Sigil to be a part of your flavor of Linux's repositories... then you need to hound the maintainers of those repositories (although in this case, even if a repo maintainer were willing to take up the cause, Sigil 7's Qt 5 dependency is going to cause some issues, since I don't believe there's any binary Qt 5 packages in any of the standard repos just yet). Some knowledgeable users might be able to work up some custom debian|rpm packages that included only the necessary Qt libs and a custom launch script if they were feeling generous. Might make it a bit easier on the non-technical Linux user. |
|
02-19-2013, 12:31 PM | #28 | ||
Sigil developer
Posts: 1,275
Karma: 1101600
Join Date: Jan 2011
Location: UK
Device: Kindle PW, K4 NT, K3, Kobo Touch
|
Quote:
Minimal match just prepends (?U) to your expression. So you don't need to do it twice. And I believe the ? in .+? means minimal match as well, so it isn't needed if you use one of the other methods. That said, it works for me as shown. However, if I change your line to: Code:
<b class="foo">words</b> <b class="foo">more words</b> </p> If I then remove the ? from the .+? it correctly just matches one at a time. I'm not a regex guru either, but the use of (?U) and ? in .+? seems to cause a conflict or tells it to reverse the minimal matching. And as far as it not working, then working, I suspect this is due to Pretty Print Tidy - if you switched away and back it may have removed that space at the end, giving only one possible match with \s Quote:
|
||
02-19-2013, 12:36 PM | #29 |
Sigil developer
Posts: 1,275
Karma: 1101600
Join Date: Jan 2011
Location: UK
Device: Kindle PW, K4 NT, K3, Kobo Touch
|
@roger64 - is there anything that didn't work or wasn't clear about the build instructions?
Code:
mkdir $HOME/sigil-run mkdir $HOME/sigil-build cd $HOME/sigil-build export LD_LIBRARY_PATH=/opt/Qt5.0.1/5.0.1/gcc/lib cmake -G "Unix Makefiles" -DCMAKE_INSTALL_PREFIX=$HOME/tmp/sigil-run -DCMAKE_BUILD_TYPE=Release -DFORCE_BUNDLED_COPIES=1 $HOME/sigil-source $HOME/sigil-run/bin/sigil As far as the instructions go, I think I'll split it into build and run so its clear you always need to include the LD_LIBRARY_PATH, which you can easily do in a script. 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. |
02-19-2013, 01:10 PM | #30 |
Well trained by Cats
Posts: 29,800
Karma: 54830978
Join Date: Aug 2009
Location: The Central Coast of California
Device: Kobo Libra2,Kobo Aura2v1, K4NT(Fixed: New Bat.), Galaxy Tab A
|
@Meme
I will use only one 'minimal match' from now on (BTW, I was trying all combinations, my norm is (.+?) and it was being too greedy..the all of a sudden it was behaving.) |
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Sigil 0.6.2 Released | user_none | Sigil | 89 | 05-04-2014 06:42 AM |
Sigil 0.6.0 Released | user_none | Sigil | 93 | 11-24-2012 06:50 PM |
Sigil 0.5.3 Released | user_none | Sigil | 85 | 05-13-2012 05:29 AM |
Sigil 0.4.2 Released | user_none | Sigil | 41 | 10-26-2011 06:03 AM |
Sigil 0.4.1 Released | user_none | Sigil | 50 | 09-01-2011 04:58 PM |