best way to generate TOCs: 4 scenarios?

[hapax legomenon]

I am still not getting this ebook format conversion thing.

I know how to use Mobicreator, less familiar with Calibre, vaguely familiar with Book Creator.

Basically I need to take html files and make TOCs for them. Some sort of gui tool would be nice, or at least some tool that doesn't involve a lot of typing on command line.

here are some scenarios:

1)i have written static html chapters, and I need merely to create a main TOC. Preferably more than one levels and with illustration. these are my own writings, so my goal is clean code and ability to maintain over time.

2)I have used a tool to download recursively some html pages (such as pages from a website). like downloadthemall. The html is messy, has a lot of random code, but the links basically work. I need to create an index page.

3) I have chapters which are actually a bunch of text files generated by random people. would be nice to generate a toc.

4)i have a single html file with internal links/toc, but lots of graphics in a separate directory.

What is the best method for generating TOCs for each scenario?

On another note, is there a utility that gives you the ability to strip out certain kinds of code (javascript, etc) from html files?

[AZed]

I'm considering writing a TOC generator for EBook::Tools v0.3, and the ability to strip <script> and <noscript> blocks already exists in v.02. Can you give me a few more details about exactly what you would want the TOC to link to for each case?

I'm going to make a few guesses about how I'd approach things, and you can tell me if it's what you're looking for.

1) Add IDs to every <h1> and <h2> element and link to those in the TOC? I have no idea what you mean by 'with illustration'.

2) I'm not sure what you would want the index page to point to. You want a way to extract all of the links from inside the messy HTML, and present a simple list of links on a clean page?

3) The TOC file would be one-level list of links to each file?

4) Again, not sure what you want your TOC to look like, or why the "internal toc" isn't working for you. How did you want the graphics to show up in the TOC (and why did you want graphics to show up in a TOC -- I don't think I've ever seen a TOC done that way).

[hapax legomenon]

the issue mainly is, what do you do with "found html" over which you have no control?

I'm talking about static html sites where there may not exist a clean TOC.

Are there any ways to autogenerate this kind of TOC? I come across hundreds of static html sites which I'd like to translate into a portable ebook format. In most cases, I just cut and paste the text, but I lose out on that.

The workflows described here assume the formatter has some control over what kind of html pages he has to deal with.

I guess HTML Tidy can clean up the code, maybe you can use XSLT to remove javascript and then autogenerate a TOC based on the title or H1 tag.

Are there any automated tools for doing this?

for one thing, I have a Sony PRS-505, and I don't know how to make ebooks out of a dozen html files (using Calibre for instance).

[kovidgoyal]

html2epub --help

you can autogenerate a multilevel TOC by select elements witht he full expressivity of XPath

[RWood]

With BookDesigner it can be a hit-or-miss thing. Sometimes it works fine and other times it will (on its own) decide that certain text lines are TOC elements. In most cases it is easy enough to correct these problems using 'Element Browser' under the Tools menu item. There is also a facility in BookDesigner that I use for generating the TOC -- under the Insert menu is 'Insert TOC (all)' that places a TOC at the start of the file just after the title and author. (It can be moved.).

[=X=]

It sounds like you just want to read Static HTML pages on your Reader. Have you looked into BookIt. This is a firefox plug in that turns a web page into a LRF eBook?

BookCreator is not really designed for creating ebook from different sources. Neither is BookDesigner.

If it is one book you are trying to create building it from multiple web sites. I think the best took is MobiCreate. Here you can load many different files into one OPF project. MobiCreate then turns them into HTML.

Once you have your eBook you can use Calibre to create the LRF book.

=X=

[hapax legomenon]

Wow, thanks for your comments! I totally had not heard of BookIt