Kepubify v3

[pgaskin]

I'm planning to work on kepubify v3 sometime soon. The goals will be:

• More predictable and intuitive output locations and options (this will be the main feature).
• Improved CLI and logging.
• More unit tests and code cleanup.
• Automatically detect and sync with a kobo.
• Improved XHTML parsing and generation.
• Possible performance improvements.
• Possible pre-importing to the kobo.
• More comprehensive help text.
• Possibly a GUI or installer with explorer integration for Windows.
• The Calibre extension may be ready.
• Possibly caching for performance improvements.
• A better README.
• More later.

If anybody has any suggestions/comments, I'd be happy to consider them.

[pgaskin]

Here is the tentative help text so far.

kepubify - Convert EPUBs to Kobo EPUBs.

Usage: kepubify [options] <input>...

Options:
    -o, --output <path>       The output file/directory for converted files (default: the orignal filename with .kepub.epub).
                              For directories, the default is the directory name with the suffix _converted. If it exists, it
                              will be replaced (unless --update is specified), and if not, it will be created. In the case
                              that multiple input folders are provided, they will be put in a subdirectory, unless
                              --no-preserve-dirs is specified. You can also specify KOBO as the path to automatically detect
                              a kobo and convert to a directory in it named kepubify.
    -n, --no-preserve-dirs    Do not preserve the directory structure of the source.
    -i, --inplace             Convert files in-place. Cannot be used with --output or --no-preserve-dirs.
    -u, --update              Only converts unconverted files (automatically enabled if output is KOBO).
        --kobo-import         EXPERIMENTAL: Pre-import books when KOBO is the output (compatible with firmware 4.12.*****).
                              This adds the book and chapter information to Content, generates the cover thumbnails, and 
                              adds the chapters to volume_shortcovers.
    -h, --hyphenate           Force enables hyphenation.
    -H, --no-hyphenate        Force disables hyphenation.
    -f, --fullscreen-fixes    Adds css to fix fullscreen issues on older firmware.
    -c, --css <css>           Adds custom css styles to the converted file.
        --inline-stylesheets  Inlines all stylesheets to fix issues on some firmware versions.
        --calibre             Uses .kepub as the extension. This option may also do additional things in the future.
    -v, --verbose             Shows more detailed output.
    -d, --debug               Shows even more detailed output (implies verbose) (use this when submitting bug reports).
    -h, --help                Show this help text.
        --version             Show the current version (kepubify v3.0.0-alpha).

Arguments:
    <input>: The path to the input files/folders. Kepubify will refuse to convert files which are already .kepub.epubs. Errors
	in each file will be ignored until the end, where a list of errors, if any, will be shown. If a path does not exist, kepubify
	will exit with an error and will not convert any of the files. If a file is specified more than once, only the last occurence
	will be counted. TODO: more info.

Links:
    Website      - https://pgaskin.net/kepubify
    Source Code  - https://github.com/geek1011/kepubify
    Bugs/Support - https://github.com/geek1011/kepubify/issues
    MobileRead   - http://mr.gd/forums/showthread.php?t=295287

And some conversion examples:

== Examples of conversion inputs/outputs ==
These examples demonstrate the behaviour of
different options. The outputs list the
files which are written to by kepubify.

input:
./test1/book0.epub
./test1/book1.epub
./test1/dir1/book2.epub
./test1/dir1/book3.epub
./test1/dir2/book4.epub
./book5.epub
./test2/dir3/book6.epub
./test2/dir4/book7.epub
./test3/book0.kepub.epub
./test3/book1.kepub
./test3/dir1/book2.kepub.epub

kepubify test1:
./test1_converted/book0.kepub.epub
./test1_converted/book1.kepub.epub
./test1_converted/dir1/book2.kepub.epub
./test1_converted/dir1/book3.kepub.epub
./test1_converted/dir2/book4.kepub.epub

kepubify test1/dir1:
./dir1_converted/book2.kepub.epub
./dir1_converted/book3.kepub.epub

kepubify --output test3 test1:
./test3/dir1/book2.kepub.epub
./test3/dir1/book3.kepub.epub
./test3/dir2/book4.kepub.epub

kepubify --output test3 --no-preserve-dirs test1:
./test3/book2.kepub.epub
./test3/book3.kepub.epub
./test3/book4.kepub.epub

kepubify --no-preserve-dirs test1:
./book2.kepub.epub
./book3.kepub.epub
./book4.kepub.epub

kepubify test1 test2 book5.epub:
./test1_converted/book0.kepub.epub
./test1_converted/book1.kepub.epub
./test1_converted/dir1/book2.kepub.epub
./test1_converted/dir1/book3.kepub.epub
./test1_converted/dir2/book4.kepub.epub
./test2_converted/dir3/book6.kepub.epub
./test2_converted/dir4/book7.kepub.epub
./book5.kepub.epub

kepubify --inplace test1:
./test1/book0.kepub.epub
./test1/book1.kepub.epub
./test1/dir1/book2.kepub.epub
./test1/dir1/book3.kepub.epub
./test1/dir2/book4.kepub.epub

kepubify --inplace --calibre test1:
./test1/book0.kepub
./test1/book1.kepub
./test1/dir1/book2.kepub
./test1/dir1/book3.kepub
./test1/dir2/book4.kepub

kepubify --output test4 test1 test2 book5.epub:
./test4/test1/book0.kepub.epub
./test4/test1/book1.kepub.epub
./test4/test1/dir1/book2.kepub.epub
./test4/test1/dir1/book3.kepub.epub
./test4/test1/dir2/book4.kepub.epub
./test4/test2/dir3/book6.kepub.epub
./test4/test2/dir4/book7.kepub.epub
./test4/book5.kepub.epub

kepubify --output whatever.kepub.epub book5.epub:
./whatever.kepub.epub

kepubify test1/book0.epub book5.epub:
./book0.kepub.epub
./book5.kepub.epub

[jgneff]

Have you considered packaging Kepubify v3 as a snap for Linux? There are instructions for Go at the bottom of the Snapcraft page.

[pgaskin]

OK, I'll consider making a snap for this version. I've heard of snaps before, but I don't use Ubuntu very much (I'm mainly a Debian/Arch person, and prefer their native packaging systems).

[jgneff]

I'm mainly a Debian/Arch person, and prefer their native packaging systems.

Even better, then, have you thought about getting Kepubify into Debian? That would make it available as part of any Debian-based distribution, including Ubuntu, and allow us to install it from a trusted third-party source without hesitation. That's the best goal for important software like Kepubify, but it takes a bit more patience than just packaging a snap. For details, see the links under the section Through Debian in the first answer to this question.

[pgaskin]

Yep. I already have a package generated with NFPM and hosted on my repo (https://deb.geek1011.net). I'm also familiar with proper deb packaging (e.g. using dh_make and debuild) (more than snap). I'm planning to work on getting it into debian along with some other packages I currently build locally as soon as I have enough spare time to go through the process.

[pgaskin]

I'm nearly finished v3. All I have left to do is updating the docs, possibly adding Explorer integration for Windows (edit: put off until later, it needs more thought put into the design), improving the build scripts, writing a few more tests, writing the release notes, and manually testing a few things.

[pgaskin]

Note to self (about future explorer extension):

	MsgBox = func(hwnd uintptr, caption, title string, flags uint) int {
		ret, _, _ := syscall.NewLazyDLL("user32.dll").NewProc("MessageBoxW").Call(
			uintptr(hwnd),
			uintptr(unsafe.Pointer(syscall.StringToUTF16Ptr(caption))),
			uintptr(unsafe.Pointer(syscall.StringToUTF16Ptr(title))),
			uintptr(flags))
		return int(ret)
	}