There was a bug that caused a double "html/html" in front of the link
URLs for each file listed in the "Files" view.
This PR fixes this. It's an ugly fix, because I just replaced "html"
with "" as the function argument, but it works? haha. Feel free to do
this the proper way, if such a way exists!
- Updated documentation, so it's very nice for pho4cexa <3
- Moved helper/example files to an assets directory (feel free to
rearrange all of these again if you have any better ideas :D
- Renamed and moved tutorial file to make it more descriptive
- Removed old environment variables from post-receive hook example file
CLONE-URL, TITLE, and H1 are no longer processed. instead, the user
should modify the templates/default.html to their liking.
the output used to be structured like this:
target-html-directory/
+-- index.html
+-- files.html
+-- README.md.html
+-- (all the other files)
the output is now structured like this:
target-html-directory/
+-- index.html
+-- html/
+-- files.html
+-- README.md.html
+-- (all the other files...)
this makes it so people who are hosting bare repos for cloning on static
webservers will have only two new directory entries (index.html and
html) in the bare repo directory, reducing the risk of clobbering
something.
finally, i tried something hacky with gitattributes to get an automatic
version idenifier to show up. but it turns out this is not the git hash
of the commit, but instead the hash of the blob for main.scm, which
could remain the same across releases. doesn't hurt, so i'll look for a
better approach in the future.
well this was a journey :)
behavior when running `repo2html <output directory>` remains the same;
we apply our internal sxml template as usual.
to use ersatz templates, run like this:
`repo2html <output directory> <template directory>`
the program will look for a "default.html" file in that directory.
this code feels like it could be way shorter and prettier
and it's broken for a few edge cases, here's some i can think of:
- it makes no attempt to ensure the ids that it assigns are unique on
the page
- there might yet be weird characters inappropriate for an id that it
uses anyway
- it doesn't make an attempt to limit the length of the id
but other than that it pretty much works
