Sigil, the ePub editor

[delphidb96]

Originally Posted by Valloric

Images are the only exception. The last folder from which you loaded an image is not currently stored. It will be in the next release.

You end up having not just one working directory, but three: one for loading files, one for saving files and one for importing images from. Of course, you can make all three the same.

I believe this implements your desired functionality and more.

Sigh....

Okay, *I* tend to write on three different machines. (And when I'm not writing on my main machine, I email the additions/changes to my main machine.) *I* then put any add/change files into my 'current working' folders. (Yeeees, I *do* work on multiple novels concurrently.) I've *seen* that Sigil doesn't automatically go to the 'current working folder' when adding a file. It goes to the Sigil program folder.

What I would like to have happen is that information is stored in the Sigil .sgf file that points Sigil to the 'current working folder' for *every* file search - by default. It would be 'nice' if one could specify 'tables' (excel spreadsheets and the like) and 'images' (cover art, paragraph-break clip-art, etc.) sub-folders, but just having a 'default' working folder for any generic file search would be a minimum.

Derek

[Valloric]

Originally Posted by delphidb96

What I would like to have happen is that information is stored in the Sigil .sgf file that points Sigil to the 'current working folder' for *every* file search - by default. It would be 'nice' if one could specify 'tables' (excel spreadsheets and the like) and 'images' (cover art, paragraph-break clip-art, etc.) sub-folders, but just having a 'default' working folder for any generic file search would be a minimum.

The functionality described in my previous post is the one every other application that remembers folders takes. It is also what is specified in the Windows User Experience guide (and Apple's HIG too, I think), which I tend to follow.

Your idea of storing the locations of folders you are loading content from directly into a file format does not appeal to me. Sometimes I have to say "No" to feature requests, and this seems to be one of those times.

Please don't be offended by that.

[am_mread]

Valloric...

Thanks for your work on this app! I know it's in it's infancy, but it looks to be a great tool for the workflow of my eBook conversion process!

Will there be a way to create TOC entries at some time?

Thanks again...

[Valloric]

Originally Posted by am_mread

Will there be a way to create TOC entries at some time?

You can create TOC entries by using headings in your book. Look through this thread for more information.

[am_mread]

Originally Posted by Valloric

You can create TOC entries by using headings in your book. Look through this thread for more information.

Yep, I realized it just after I hit submit, and posted the question.. Thanks!

[EowynCarter]

Please do.

Got it.
If you have two existing headers, adding one in between mess up the ids. (Maybe in addition to page breaks)
Nope, actually sigil ignore precious header, and starts counting to one.

Ticket on it's way (if i can figure out how google code works)
Edit : it's issue 36.

[crutledge]

Originally Posted by Valloric

Oh. I get what you're saying. Try to make a test case that can invoke this bug, then create an issue with the test case. This would be a major bug and would get looked at ASAP, but I can't do it if I can't replicate the behaviour.

I now think that this was my error. If I reproduce it I'll post an issue.

Although I managed to create an Epub with minor problems, I think that all were already covered. Howvere, after examining the code, I don't see any reason that a save of the html file could not be easily converted t0 MOBI. I'll wait for the html save.

Great job. Bless you my son.
Charlie

[Valloric]

Originally Posted by crutledge

I now think that this was my error. If I reproduce it I'll post an issue.

I created an issue for this and fixed it already . At least I think I did. I'm pretty sure. You're using the Mac or Linux version, right? Do comment on the issue report if you see it again in a later version.

Originally Posted by crutledge

Although I managed to create an Epub with minor problems, I think that all were already covered. Howvere, after examining the code, I don't see any reason that a save of the html file could not be easily converted t0 MOBI. I'll wait for the html save.

HTML export could possibly end up in the next release.

Originally Posted by crutledge

Great job. Bless you my son.

[ldolse]

Thanks for this software, this will be a big help, especially in conjunction with Calibre.

I don't recall seeing it mentioned so far in this thread, I'm having a bit of trouble with and HTML source file and linked images. The HTML file itself loads in fine, but the images don't load. The source of the HTML file is from Acrobat Pro, it converted a PDF book to HTML with extracted images. All the images wound up in a subfolder called 'images', which is at the same level as the HTML file in the directory hierarchy. Is this to be expected? (note they load fine in a browser)

I haven't played with it much yet, but seems pretty cool so far. Was glad to see the split view - One thing that would be sweet for this view would be to have the book pane and code pane linked, or at least have some way to click on one side and have the other side show you the same detail. A collapsible XML tree might also be nice. This input is based off my experiences with Firebug, which I use pretty often when working with html content. Not sure whether trying to leverage firebug's code is an option, it's open source with a bsd style license.

Edit - I see reading back through the thread that syncing the two views was already discussed. Still, Firebug might be worth a look for some ideas as to how they handle synchronization, navigating the XML tree, etc.

[AprilHare]

Originally Posted by FrancisT

on a clean(ish) *buntu 9.04 you need to add (using synaptic or apt-get) support for libqt4-svg and libqt4-webkit which are not included by default (and their various dependencies). It looks like calibre makes you get these (at least the source code version does) but if you have an otherwise clean system you don't get them.

Maybe this could be added to the sigil home page or to the announcement at the top of this thread?

Francis

Those packages are already installed here.
Still no dice.

[crutledge]

Valloric,
One addition that I would greatly appreciate. Provide the means to adjust the editor font. My old eyeballs are not what they used to be.

Thanks

[Valloric]

Originally Posted by ldolse

I don't recall seeing it mentioned so far in this thread, I'm having a bit of trouble with and HTML source file and linked images. The HTML file itself loads in fine, but the images don't load. The source of the HTML file is from Acrobat Pro, it converted a PDF book to HTML with extracted images. All the images wound up in a subfolder called 'images', which is at the same level as the HTML file in the directory hierarchy. Is this to be expected? (note they load fine in a browser)

Please create an issue and attach the HTML and the folder where the images are located (before import).

And yes, Sigil will import your images and transfer them to a "images" folder in the temporary folder where all your book's resources are being stored.

Originally Posted by AprilHare

Those packages are already installed here.
Still no dice.

Hm. Open and issue. This warrants more attention. I'll be working on unifying the build process under CMake today, and if I get that done, you should be able to build it from source with ease.

Originally Posted by crutledge

Valloric,
One addition that I would greatly appreciate. Provide the means to adjust the editor font. My old eyeballs are not what they used to be.

I was thinking of creating a zoom slider like Word 2007 has. Open an issue so I don't forget.

[Valloric]

For those who want more up-to-date information on Sigil's development, you can always see the live ChangeLog here. That file is linked directly from the repository, and it is updated as bugs are fixed and features are added. You can also subscribe to RSS and Atom feeds here.

[DaleDe]

Originally Posted by Valloric

For those who want more up-to-date information on Sigil's development, you can always see the live ChangeLog here. That file is linked directly from the repository, and it is updated as bugs are fixed and features are added. You can also subscribe to RSS and Atom feeds here.

Great, I made a wiki article Sigil to promote the site a bit.

[ldolse]

Regarding the images that weren't imported:

Originally Posted by Valloric

Please create an issue and attach the HTML and the folder where the images are located (before import).

And yes, Sigil will import your images and transfer them to a "images" folder in the temporary folder where all your book's resources are being stored.

Opened, bug #39

A note regarding issue #21 TOC H1/Span issue, just wanted to note I'm seeing this too. Unfortunately the workaround of "Don't do it" isn't really a workaround. Many of us are format shifting and this weird kind of formatting happens with a lot of tools which convert to html.

Below is some other feedback/wishlist type stuff. I can open some bugs for anything that sounds amenable.

CSS handling:

• Would be awesome if all the css styles showed up in the drop down list of styles. Right now everything not in a heading tag just shows up as 'normal', regardless of the number of styles that are actually in the epub.
• Related to that, if that were supported it would be great if a user could create human readable names for the styles, but this could be mitigated if the styles in the drop-down just listed the attributes set in the style
• Equally great would be to be able to search or highlight document text by style. Often in format shifted content you get a large number of styles and it's difficult to tell the good from the bad so that one can homogenize them

Would be great to see the following types of things be inputs into the TOC editor:

• Support listing all items with specific css style(s) applied
• Support listing all items matching a regex
• Possibly support more generic things like bolded text, or text above a certain size, etc

Other formatting niceties

• Would be great to be able to insert page breaks as an alternative to Chapter markers
• Margin settings would be nice, though this could be delayed if css elements were automatically added to the drop-down list as requested above
• Being able to set Blockquotes with a button would be a nice alternative to margins as well
• Raised Caps and Drop Caps support would be great, one way to handle this would be for Sigil to create but not use a css style that does each. If the all the css styles showed up in the drop down list the user need only apply the style. If they prefer slightly different settings they need only edit the css from the default