~ob/stagit

#stagit

Personal fork of static git page generator. It generates static HTML pages for a git repository.

This fork uses md4c to convert the README markdown into HTML and then shows it in an about page for each repository, this adds a new dependency. On top of that, the assets have been changed, creating a personal theme. The scripts have also been changed to fit my needs.

#Usage

Make files per repository:

$ mkdir -p htmldir && cd htmldir
$ stagit path-to-repo

Make index file for repositories:

$ stagit-index repodir1 repodir2 repodir3 > index.html

#Build and install

$ make
# make install

#Dependencies

C compiler (C99).
libc (tested with OpenBSD, FreeBSD, NetBSD, Linux: glibc and musl).
libgit2 (v0.22+).
POSIX make (optional).
md4c (v0.4.4+).

#Documentation

See man pages: stagit(1) and stagit-index(1).

#Building a static binary

It may be useful to build static binaries, for example to run in a chroot.

It can be done like this at the time of writing (v0.24):

cd libgit2-src

# change the options in the CMake file: CMakeLists.txt
BUILD_SHARED_LIBS to OFF (static)
CURL to OFF              (not needed)
USE_SSH OFF              (not needed)
THREADSAFE OFF           (not needed)
USE_OPENSSL OFF          (not needed, use builtin)

mkdir -p build && cd build
cmake ../
make
make install

#Extract owner field from git config

A way to extract the gitweb owner for example in the format:

[gitweb]
	owner = Name here

Script:

#!/bin/sh
awk '/^[ 	]*owner[ 	]=/ {
	sub(/^[^=]*=[ 	]*/, "");
	print $0;
}'

#Set clone url for a directory of repos

#!/bin/sh
cd "$dir"
for i in *; do
	test -d "$i" && echo "git://git.codemadness.org/$i" > "$i/url"
done

#Update files on git push

Using a post-receive hook the static files can be automatically updated. Keep in mind git push -f can change the history and the commits may need to be recreated. This is because stagit checks if a commit file already exists. It also has a cache (-c) option which can conflict with the new history. See stagit(1).

git post-receive hook (repo/.git/hooks/post-receive):

#!/bin/sh
# detect git push -f
force=0
while read -r old new ref; do
	hasrevs=$(git rev-list "$old" "^$new" | sed 1q)
	if test -n "$hasrevs"; then
		force=1
		break
	fi
done

# remove commits and .cache on git push -f
#if test "$force" = "1"; then
# ...
#fi

# see example_create.sh for normal creation of the files.

#Create .tar.gz archives by tag

#!/bin/sh
name="stagit"
mkdir -p archives
git tag -l | while read -r t; do
	f="archives/${name}-$(echo "${t}" | tr '/' '_').tar.gz"
	test -f "${f}" && continue
	git archive \
		--format tar.gz \
		--prefix "${t}/" \
		-o "${f}" \
		-- \
		"${t}"
done

#Features

Log of all commits from HEAD.
Log and diffstat per commit.
Show file tree with linkable line numbers.
Show references: local branches and tags.
Detect README and LICENSE file from HEAD and link it as a webpage.
Detect submodules (.gitmodules file) from HEAD and link it as a webpage.
Atom feed of the commit log (atom.xml).
Atom feed of the tags/refs (tags.xml).
Make index page for multiple repositories with stagit-index.
After generating the pages (relatively slow) serving the files is very fast, simple and requires little resources (because the content is static), only a HTTP file server is required.
Usable with text-browsers such as dillo, links, lynx and w3m.

#Cons

Not suitable for large repositories (2000+ commits), because diffstats are an expensive operation, the cache (-c flag) is a workaround for this in some cases.
Not suitable for large repositories with many files, because all files are written for each execution of stagit. This is because stagit shows the lines of textfiles and there is no "cache" for file metadata (this would add more complexity to the code).
Not suitable for repositories with many branches, a quite linear history is assumed (from HEAD).

In these cases it is better to just use cgit or possibly change stagit to run as a CGI program.
Relatively slow to run the first time (about 3 seconds for sbase, 1500+ commits), incremental updates are faster.
Does not support some of the dynamic features cgit has (this is by design, just use git locally), like:
- Snapshot tarballs per commit.
- File tree per commit.
- History log of branches diverged from HEAD.
- Stats (git shortlog -s).

refs

clone