Sigil 0.7 Beta (0.6.901) Available

[DiapDealer]

Originally Posted by roger64

Compiling on Linux

I installed the full pre-compiled Qt binary at the standard place: ~/Qt5.0.1

cmake reports that it does not find its path, asks me to tell him but I do not know how to write up properly this command line on the terminal.

.../...
 Could not find a package configuration file provided by "Qt5Core" with any
  of the following names:

    Qt5CoreConfig.cmake
    qt5core-config.cmake

  Add the installation prefix of "Qt5Core" to CMAKE_PREFIX_PATH or set
  "Qt5Core_DIR" to a directory containing one of the above files.  If
  "Qt5Core" provides a separate development package or SDK, be sure it has
  been installed.
.../...

See if anything here helps.

http://www.mobileread.com/forums/sho...d.php?t=203562

[roger64]

Originally Posted by DiapDealer

See if anything here helps.

http://www.mobileread.com/forums/sho...d.php?t=203562

YES. Thank you very much.

The path I had to use was (YMMV):

cmake -G "Unix Makefiles" -DCMAKE_PREFIX_PATH=~/Qt5.0.1/5.0.1/gcc_64/lib/cmake/ -DCMAKE_BUILD_TYPE=Release ~/Raccourcis/Sigil-0.6.901-Code

During the make process, It nearly stopped for this reason

Spoiler:

[ 99%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/MainUI/OPFModel.cpp.o
In file included from /home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/smart_ptr/shared_ptr.hpp:30:0,
from /home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/shared_ptr.hpp:17,
from /home/roger/Raccourcis/Sigil-0.6.901-Code/src/Sigil/BookManipulation/XhtmlDoc.h:26,
from /home/roger/Raccourcis/Sigil-0.6.901-Code/src/Sigil/BookManipulation/Book.h:33,
from /home/roger/Raccourcis/Sigil-0.6.901-Code/src/Sigil/MainUI/OPFModel.cpp:27:
/home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/checked_delete.hpp: In instantiation of ‘void boost::checked_delete(T*) [with T = boost::error_info<boost::tag_original_exception_ty pe, const std::type_info*>]’:
/home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/smart_ptr/detail/shared_count.hpp:95:13: required from ‘boost::detail::shared_count::shared_count(Y* ) [with Y = boost::error_info<boost::tag_original_exception_ty pe, const std::type_info*>]’
/home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/smart_ptr/shared_ptr.hpp:183:50: required from ‘boost::shared_ptr<T>::shared_ptr(Y*) [with Y = boost::error_info<boost::tag_original_exception_ty pe, const std::type_info*>; T = boost::error_info<boost::tag_original_exception_ty pe, const std::type_info*>]’
/home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/exception/info.hpp:171:69: required from ‘const E& boost::exception_detail::set_info(const E&, const boost::error_info<Tag, T>&) [with E = boost::unknown_exception; Tag = boost::tag_original_exception_type; T = const std::type_info*]’
/home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/exception/info.hpp:192:46: required from ‘typename boost::enable_if<boost::exception_detail::derives_ boost_exception<E>, const E&>::type boost:perator<<(const E&, const boost::error_info<Tag, T>&) [with E = boost::unknown_exception; Tag = boost::tag_original_exception_type; T = const std::type_info*; typename boost::enable_if<boost::exception_detail::derives_ boost_exception<E>, const E&>::type = const boost::unknown_exception&]’
/home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/exception/detail/exception_ptr.hpp:182:13: required from ‘void boost::unknown_exception::add_original_type(const E&) [with E = std::exception]’
/home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/exception/detail/exception_ptr.hpp:161:32: required from here
/home/roger/Raccourcis/Sigil-0.6.901-Code/src/BoostParts/boost/checked_delete.hpp:34:5: warning: deleting object of polymorphic class type ‘boost::error_info<boost::tag_original_exception_t ype, const std::type_info*>’ which has non-virtual destructor might cause undefined behaviour [-Wdelete-non-virtual-dtor]
[ 99%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/MainUI/OPFModelItem.cpp.o
[ 99%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/MainUI/NCXModel.cpp.o
[ 99%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/MainUI/ValidationResultsView.cpp.o
[ 99%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/qrc_main.cpp.o
[100%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/qrc_About.cpp.o
[100%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/qrc_javascript.cpp.o
[100%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/qrc_icon.cpp.o
[100%] Building CXX object src/Sigil/CMakeFiles/sigil.dir/sigil_automoc.cpp.o
Linking CXX executable ../../bin/sigil
[100%] Built target sigil
roger@lmde64:~/Raccourcis/Sigil-07$

Despite of these quirks, build finished successfully.

But when I tried to launch Sigil in /usr/local/bin however, it reported again a probable new path problem:

error while loading shared libraries libQt5Widgets.so.5: no such file .../...

This concludes my homework for today.

[DiapDealer]

Originally Posted by roger64

This concludes my homework for today.

But will it launch from the build directory (before make install)? If it does (dictionaries and other resources might be unavailable--just looking for a successful program launch here), then you're still in business. Look at that same thread for hints on adding the QT5/lib directory to the LD_LIBRARY_PATH environment variable.

I launch Sigil with a custom script that updates the LD_LIBRARY_PATH environment variable beforehand. I use this simple bash script:

#!/bin/sh

# Adds the directory the Qt5 libs are located
# in to the LD_LIBRARY_PATH before launching sigil.

QTLIB_DIR="/location/of/your/Qt5/lib"

if [ -z "$LD_LIBRARY_PATH" ]; then
  LD_LIBRARY_PATH="$QTLIB_DIR"
else
  LD_LIBRARY_PATH="$QTLIB_DIR:$LD_LIBRARY_PATH"
fi

export LD_LIBRARY_PATH

exec sigil "$@"

Obviously change QTLIB_DIR to the location where your QT5 libs are located (and potentially the name/location of the "real" Sigil executable--the one "make install" creates--might need to be modified).

If it all works, I change the real executable (/usr/local/bin) to something like sigil-real and change the custom launch script (also in /usr/local/bin) to sigil so all shortcuts and *.desktop file associations still function as expected.

[roger64]

Thank you for the script. I will try tomorrow to adapt the paths.

[Notjohn]

Originally Posted by meme

The beta User Guide should be up soon.

Funnily enough, I Googled a Sigil question yesterday, and Google took me to the appropriate page in the 0.7 User Guide.

(And yes, the question was answered there. Sorry, I can't remember what it was.)

[theducks]

Could I slip in a Feature change request?
In the HTML TOC (Sigil Generated), the styles are inline.

I would like to see them moved to their own stylesheet: SGC-TOC.CSS (reserved name)

the second part is a rule: that if the style sheet Exists, it will not be overwritten, which allows the user to custom style that sheet without fear that regenerating a TOC will stomp on their tweaks. The burden is on the user to ONLY use those SGC selectors in stylesheet edits.

(BTW delete unused styles tool will clean any unused. that may want to be blocked)

[meme]

Why is it with a beta release, after all the features are done, there's always a request for more features

It seems reasonable - certainly using a stylesheet is better than inline styles as everyone always says, so I guess we should follow that preference.

The Index HTML file will also need the same approach since it uses inline styles as well.

[theducks]

Originally Posted by meme

Why is it with a beta release, after all the features are done, there's always a request for more features

It seems reasonable - certainly using a stylesheet is better than inline styles as everyone always says, so I guess we should follow that preference.

The Index HTML file will also need the same approach since it uses inline styles as well.

I just had to make your day.

Well, really, I had just made a HTML TOC and was looking at it and going I wonder if I could <center> those chapter items.

Then I thought up the case where I needed the custom styling to survive regeneration.

Hey! It proves someone uses those features

[Toxaris]

It will also make it easier for the users to create and reuse their own stylesheets specifically for the TOC and/or index. It makes perfect sense to utilize a seperate stylesheet for that.

[Moonraker]

Thank you for this marvelous release of Sigil. I just adore the Preview pane.

I am sending a donation to you.

[kovidgoyal]

I'm rather interested to see how Qt 5 performs. I am tempted to build calibre with Qt 5 (mainly for the webkit update) but, I dont think its mature enough as yet.

[theducks]

Originally Posted by kovidgoyal

I'm rather interested to see how Qt 5 performs. I am tempted to build calibre with Qt 5 (mainly for the webkit update) but, I dont think its mature enough as yet.

One thing I noticed: Drop caps in Sigil, now look almost the same on my PEz Before, there were noticeable differences in placement.

[meme]

Originally Posted by kovidgoyal

I'm rather interested to see how Qt 5 performs. I am tempted to build calibre with Qt 5 (mainly for the webkit update) but, I dont think its mature enough as yet.

I'm sure user_none could give you more feedback on this. It seems to work mostly ok, but there are rough issues - context menu on MAC, slow selection of highlighted text (I think this was mentioned for 5.0.2).

[Jabby]

Running Windows 7.
With entries in either the Find or the Find & Replace fields, there is no match count. The words "Matches Found" is there but no count (inclduing "No Matches).

Did anyone consider enhancing the readability of the F&R fields for this release. I guess I'm asking if you have decided not to do it.

[meme]

Originally Posted by Jabby

Running Windows 7.
With entries in either the Find or the Find & Replace fields, there is no match count. The words "Matches Found" is there but no count (inclduing "No Matches).

I don't see that. When I enter ">" in the Find box and press Count All, it shows the count.

What language are you using for the ui? What settings in Find (though that shouldn't matter). It's possible that the translation syntax used for that string is not working right. What about Replace All? Or try Autofill in Index or Clips - that uses the same string and should give a count of how many entries were added.

Did anyone consider enhancing the readability of the F&R fields for this release. I guess I'm asking if you have decided not to do it.

Probably got lost as there wasn't an issue. A quick test shows it's possible to configure the font and size of the text in the Find and Replace boxes, so I'll look at an option to set it. It may be limited in size due to the text box, etc.