Git2sC

git2sc is a program to sync a git documentation repository to Confluence.

Install

git clone https://git.paradigmadigital.com/seguridad/git2sc
cd git2sc
virtualenv -p python3 git2sc
source git2sc/bin/activate
pip3 install -r requirements.txt
python3 setup.py install

If you don't have pandoc and asciidoctor installed, do so:

sudo apt-get install pandoc asciidoctor

Set up the credentials of a user that can read and write the pages in the environment variables:

• GIT2SC_API_URL: https://company.atlassian.net/wiki/rest/api
• GIT2SC_AUTH: username:password

Usage

All commands require that you specify the space id of your project. To obtain this value you have to go to your confluence page which may look like this

https://company.atlassian.net/wiki/spaces/SPACE/overview

Where SPACE is the space id.

Update an article

This command will update the confluence article with id {{ article_id }} with the content of the file in the path {{ content }}.

git2sc {{ space }} article update {{ article_id }} {{ content }}

Create an article

This command will create an confluence article under the {{ space }} space, with title {{ title }} and with the content of the file in the path {{ content }}.

git2sc {{ space }} article create {{ title }} {{ content }}

If you want to make the article a children of another article use the -p {{ parent_id }} flag

git2sc {{ space }} article create -p {{ parent_id }} {{ title }} {{ content }}

Delete an article

This command will delete the confluence article with id {{ article_id }}.

git2sc {{ space }} article delete {{ article_id }}

Upload a directory

This command will upload all the contents of a directory to the main page of confluence creating a hierarchy of articles equally to the directory tree.

For each directory it will try to load the README.adoc or README.md to the directory confluence page.

Even if confluence uses page_ids there can't be two articles with the same title, so if you have two files with the same name it will create it with name_1, name_2, and so on. Check this issue to view the improvements. Try to avoid having files with the same name on your repo for flawless results.

git2sc {{ space }} upload {{ directory_path }}

Optionally you can exclude some files and directories (by default .git, .gitignore, and .gitmodules)

git2sc {{ space }} upload {{ directory_path }} --exclude file1 directory1 file2

If you don't want to upload the directory to the main page but starting from an article you can specify it with the -p flag. Beware in this case, the confluence page name is the basename of the directory_path therefore avoid using . as directory_path.

git2sc {{ space }} upload {{ directory_path }} -p {{ parent_id }}

Sync a directory

This command will sync all the contents of a directory to the main page of confluence creating a hierarchy of articles equally to the directory tree.

For each directory it will try to load the README.adoc or README.md to the directory confluence page.

Even if confluence uses page_ids there can't be two articles with the same title, so if you have two files with the same name, the upload directory command will create it with name_1, name_2, and so on. Check this issue to view the improvements. But the sync command won't update the files with *_1, *_2, ... So try to avoid having files with the same names on your directory.

git2sc {{ space }} sync {{ directory_path }}

Optionally you can exclude some files and directories (by default .git, .gitignore, and .gitmodules)

git2sc {{ space }} sync {{ directory_path }} --exclude file1 directory1 file2

Test

To run the tests first install tox

pip3 install tox

And then run the tests

tox

Authors

jamatute@paradigmadigital.com