How to make Sigil live (in calibre)

[DrChiper]

Originally Posted by Divingduck

No, I did this only with mouse action.

You can try it with Sigil as the delete option there is also at the top. If it's a mouse bump, then it should also occur there. If not, its indeed curious.

[DrChiper]

Kovid, using the file rename option, I noticed that unlike Sigil, you cannot rename all selected files at once with auto increment numbering. As this was a handy feature, could it also be on your to-do list for the near future?

[samy2]

I got this error casually buy selecting a file. But if ignore this message, it still works.

calibre, version 1.14.0
FEHLER: Nicht abgefangener Fehler: <b>IOError</b>:[Errno 2] Datei oder Verzeichnis nicht gefunden: u'/tmp/calibre_1.14.0_tmp_TY6Y32/uKlYU_/00000-TWk6d2/OEBPS/text/content002_1.xhtml'

calibre 1.14  isfrozen: True is64bit: True
Linux-3.10.9-200.fc19.x86_64-x86_64-with-fedora-19-Schrödinger’s_Cat Linux ('64bit', 'ELF')
('Linux', '3.10.9-200.fc19.x86_64', '#1 SMP Wed Aug 21 19:27:58 UTC 2013')
Python 2.7.5
Linux: ('Fedora', '19', 'Schr\xc3\xb6dinger\xe2\x80\x99s Cat')
Traceback (most recent call last):
  File "site-packages/calibre/gui2/tweak_book/preview.py", line 521, in refresh
  File "site-packages/calibre/gui2/tweak_book/preview.py", line 135, in add_request
  File "site-packages/calibre/gui2/tweak_book/preview.py", line 43, in get_data
  File "site-packages/calibre/ebooks/oeb/polish/container.py", line 422, in raw_data
  File "site-packages/calibre/ebooks/oeb/polish/container.py", line 757, in open
IOError: [Errno 2] Datei oder Verzeichnis nicht gefunden: u'/tmp/calibre_1.14.0_tmp_TY6Y32/uKlYU_/00000-TWk6d2/OEBPS/text/content002_1.xhtml'

[Katsunami]

In case Kovid doesn't read German:

FEHLER: Nicht abgefangener Fehler: <b>IOError</b>:[Errno 2] Datei oder Verzeichnis nicht gefunden: u'/tmp/calibre_1.14.0_tmp_TY6Y32/uKlYU_/00000-TWk6d2/OEBPS/text/content002_1.xhtml'

ERROR: Unhandled Error: <b>IOError</b>:[Errno 2] File or folder not found: u'/tmp/calibre_1.14.0_tmp_TY6Y32/uKlYU_/00000-TWk6d2/OEBPS/text/content002_1.xhtml'

[DrChiper]

Using a regex expression like <p>[a-z] (trying to find paragraphs which start with lowercase), the search also finds uppercase items: thus "<p>and ...", "<p>Become...", etc. I thus must explicitly check the "case sensitive" option to get only the lowercases.

However, using regex I explicitly must be able to define just the regex lowercase via [a-z] for matches and the "case sensitive" should not overrule this. In fact I think that the "case sensitive" setting must be disabled for regex fully.

Can this behavior be confirmed?

[arspr]

Originally Posted by kovidgoyal

Finally, there will be an error check tool that will tell you about unparseable XML in the book, along with other useful (as opposed to the junk produced by epubcheck) errors.

That's all what is really needed. With a "Check Book/CSS/HTML" or whatever, you can always look for errors and fixing them manually. And this "advanced" feature would stay out of the way of "basic" users.

(I've got the feeling that all of us were actually thinking in the same functionality but with a different shape. Something about not noticing the forest because the first trees in view.)

[kovidgoyal]

@DrChiper: I dont see much need for auto-renaming files, since the file browser displays the files in spine order anyway, regardless of their names. I do plan on creating a tool to bulk rename all file, putting them into a "standard" file structure.

The treatment of character ranges depends on the case sensitive setting. This is particularly useful with unicode ranges, since the regex engine supports full unicode case folding. In any case, this is not something in calibre's control, it comes from the regex engine https://pypi.python.org/pypi/regex

@samy2: Is that for just one particular file/book or all of them?

[kovidgoyal]

@DrChiper: You should be happy to learn that I decided to implement direct clicking on links in the preview panel, instead of requiring a right click to follow the link. https://github.com/kovidgoyal/calibr...9f5aff918342cb

[DrChiper]

Originally Posted by kovidgoyal

@DrChiper: You should be happy to learn that I decided to implement direct clicking on links in the preview panel, instead of requiring a right click to follow the link. https://github.com/kovidgoyal/calibr...9f5aff918342cb

Indeed, I am pleased! Thanks Kovid!

[samy2]

Originally Posted by kovidgoyal

@samy2: Is that for just one particular file/book or all of them?

Seem to be just one particular book. I can't reproduce it with other books. And now i tried it again with the same book, but this error message does not appear.

[DrChiper]

Originally Posted by kovidgoyal

@DrChiper: I dont see much need for auto-renaming files, since the file browser displays the files in spine order anyway, regardless of their names. I do plan on creating a tool to bulk rename all file, putting them into a "standard" file structure.

It is not that I try to sort files this way, but want to give them a more meaningful name during touch-up as sometimes the original filenames are quite bizzare. Using Sigil I can select a file range and rename them for instance as Picture0000 - Picture0099.

[kovidgoyal]

@samy2: Well if you can reproduce it again, send the book to me.

[DrChiper]

Originally Posted by kovidgoyal

The treatment of character ranges depends on the case sensitive setting. This is particularly useful with unicode ranges, since the regex engine supports full unicode case folding. In any case, this is not something in calibre's control, it comes from the regex engine [url]https://pypi.python.org/pypi/regex[/ur]

I see. Well we are now aware of this so I can compensate for this behavior.

[Katsunami]

I've just noticed that the latest Calibre installed E-Book Tweak 64-bit. Looks great I'll start to fiddle with it, as I'm rearranging some Gutenberg texts I've started one in Sigil a few days ago, so I'll finish it there, but I'll start the next in E-Book Tweak and see how far I'll get.

There is one thing that seems to be incorrect (cosmetic problem): The window titles have &-signs in them, presumably to denote a hotkey. Examples:

&Files Browser
&Inspector
File &Preview

If it makes any difference, the operating system is Windows 7 x64 Professional.

[Perkin]

One thing I've found is that when 'beautifying' & nbsp; entities get changed to spaces, which is what I don't want.
Edit: Also does it for the 'Fix html' function.

Quite often there's

<p>&nbsp;</p>

which will then get ignored rather than having a 'scene-break' empty paragraph.

Also a few instances inside paragraphs where you have deliberate extra spaces or other such occurrences which then get 'reduced' to normal spaces which because of the way they're rendered by readers is now wrong.