* Handle input path with regards to custom css
Before, when someone like the Reference set their extra css as
"theme/reference.css" in their book.toml, this path would be treated as
relative to the invocation of mdbook, and not respect the input path. This
PR modifies these relative paths to do so.
Fixes the build of https://github.com/rust-lang/rust/pull/47753 which
blocks updating rustc to mdbook 0.1
* don't use file-name
the style name is theme/reference.css, this results in a Err(StripPrefixError(())), which means that we push only the file_name, losing the theme bit
* Added documentation to the `config` module
* Added an example to the `config` module
* Updated the docs in lib.rs regarding implementing backends
* Started writing an alternate backends walkthrough
* Mentioned the output.foo.command key
* Added example output
* Added a config section to the backends tutorial
* Finished off the backends tutorial
* Made sure travis checks mdbook-wordcount
* Fixed the broken link at in the user guide
* Changed how travis builds the project
* Added a conclusion
* Went through and documented a lot of stuff
* Added a preprocessors chapter and updated For Developers
* fix(theme/index): Use nav element for Table of Content
* fix(renderer/html_handlebars/helpers/toc): Use ol instead of ul
Chapters and sections are ordered, so we should use the appropriate HTML tag.
* fix(renderer/html_handlebars/helpers/toc): Hide section number from screen readers
Screen readers have this functionality build-in, no need to present this. Ideally, this should not even be in the DOM tree, since the numbers can be shown by using CSS.
* fix(theme/index): Remove tabIndex="-1" from .page
Divs are not focusable by default
* fix(theme): Make sidebar accessible
Using aria-hidden (together with tabIndex) takes the links out of the tab order.
http://heydonworks.com/practical_aria_examples/#progressive-collapsibles
* fix(theme/index): Wrap content inside main tag
The main tag helps users skip additional content on the page.
* fix(theme/book): Don't focus .page on page load
The main content is identified by the main tag, not by auto-focusing it on page load.
* fix(theme/index): Make page controls accessible
* fix: Make theme selector accessible
- Use ul and li (since it is a list)
- Add aria-expanded and aria-haspopup to the toggle button
- Use button instead of div (buttons are accessible by default)
- Handle Esc key (close popup)
- Adjust CSS to keep same visual style
* fix(theme/stylus/sidebar): Make link clickable area wider
Links now expand to fill the entire row.
* fix(theme): Wrap header buttons and improve animation performance
Previously, the header had a fixed height, which meant that sometimes the print button was not visible. Animating the left property is expensive, which lead to laggy animations - transform is much cheaper and has the same effect.
* fix(theme/stylus/theme-popup): Theme button inherits color
Bug introduced while making the popup accessible
* fix(theme/book): Handle edge case when toggling sidebar
Bug introduced when switching from animating left to using transform.
* Added a mechanism for creating alternate backends
* Added a CmdRenderer and the ability to have multiple renderers
* Made MDBook::load() autodetect renderers
* Added a couple methods to RenderContext
* Converted RenderContext.version to a String
* Made sure all alternate renderers are invoked as `mdbook-*`
* Factored out the logic for determining which renderer to use
* Added tests for renderer detection
* Made it so `mdbook test` works on the book-example again
* Updated the "For Developers" docs
* Removed `[output.epub]` from the example book's book.toml
* Added a bit more info on how backends should work
* Added a `destination` key to the RenderContext
* Altered how we wait for an alternate backend to finish
* Refactored the Renderer trait to not use MDBook and moved livereload to the template
* Moved info for developers out of the book.toml format chapter
* MOAR docs
* MDBook::build() no longer takes &mut self
* Replaced a bunch of println!()'s with proper log macros
* Cleaned up the build() method and backend discovery
* Added a couple notes and doc-comments
* Found a race condition when backends exit really quickly
* Added support for backends with arguments
* Fixed a funny doc-comment
* Prohibit 'print.md' files
Fix#258 by emmiting an error whenever an mdBook contains a
"print.md" file in its root.
* Apply suggested changes
* Add tests
* Refactor navigation helpers
* Target::find: take previous_item by reference
This makes more sense for find as an interface, though it causes a
second clone in some cases. Maybe rustc is smart here?
* Test next and previous navigation helpers
* Add more next/previous tests
On the web, the normalized path separator is forward-slash (`/`), so we
use the built-in `is_separator()` method to replace any path separator
with the forward-slash, to ensure consistent output on unix and windows
machines.
- Replaced link parser with a Regex
- Implemented {{#include}} links
- Will display relatively nice error when cannot open {{#}} linked file
- Escaped links no longer render with escape char
- utils::fs::file_to_path no takes AsRef<Path>
- sorted export/mod in lib.rs
`HtmlConfig` was both guaranteed to exist within `BookConfig`
and `expect`ed in few places.
This simplifies the API a little by representing the fact that
`HtmlConfig` is currently mandatory for proper mdBook binary operation.
- removing need to explicitly use `Path::new` all over the place
- removed warnings from doctests (normally invisible unless `cargo test -- --nocapture`)
- no doctests are norun/ignore now
- updated docs both in book-example and in docs not to refer to nonexisting API's
Fixes a lot of browser incompatibilities in localStorage/cookie handling
Including but not limited to:
- loss of styling and functionality on chromium private mode
- loss of styling and functionality on safari and safari private mode
- awaiting verification if problems in mobile safari are solved.
Header fragment links now use "id" attribute instead of the depreciated
"name" attribute.
Similar headers are given numbered ids to avoid id collisions.
For instance, if there are three headers named "Example", their ids
would be "#example", "#example-1", and "#example-2" respectively.
This project already had a transitive dependency on regex; let's use it.
This isn't the most efficient solution, but it should be fine. It ends
up doing five full scans of the text. There's probably an easier way but
I'm mostly just trying to get this to work for now.
This also implements the same algorithm that rustdoc does for generating
the name for the link.
Fixes#204
Path_Ext has been stabilized in the Standard Library, the temporary copy I had can go.
I found a fs::create_dir_all method that does exactly what create_path was doing, but better... create_path is thus replaced with that.
