# Github webhook scripts

These are scripts to setup the basic webhook folder structure,
the clone base for one campaign and a simple crontab script to pull data from the repository

## Scripts

- base_setup.sh: setup for the folder structure, users, etc
- new_clone.sh: Basic clone script
- git_sync.sh: The script to run in crontab, to sync the changes
- switch_branch.sh: Switch from one branch to the other
- create_ssh_config.sh: setup ssh key and ssh config entry

## Run commands

### Settings

in `config/webhook.cfg` the following settings have to be entered

```ini
GIT_WEBHOOK_BASE_FOLDER="<BASE FOLDER>"
WWW_GROUP="<APACHE GROUP NAME>"
SUDO_USER="<SUDO USER NAME TO USE>"
USE_SUDO=<1: use sudo, 0: no sudo used>
```

The script will create the sudo user if needed automatically

### base_setup.sh

as is, if the folder exists it will only copy the scripts, will not alter or change anything.

Planned to get auto fixes for wrong ACL, etc or missing config settings

### SSH Key generation

A SSH Key has to be generated for each git respository that will be synced and the configuration has to be added to the ssh config file

```sh
create_ssh_config.sh [repo url full] ([jump proxy])
```

> [!notice]
> This will currently output the command to create the SSH key and the host config to add to the ssh config file

### new_clone.sh

Create a new clone

```sh
new_clone.sh -r <Repository URL> -b <Branch Name> [-H <Host name>] [-f <Folder name>] [-n <Remote name>]
```

The [host (-H)] is the SSH Host name entry, this will replace any "[host]:" in the [repository]. If the [host (-H)] is not set the host set in the [repository] will be used. If nothing found the script will abort. Note that if [host (-H)] is not set the host name will be the Repository name which has to match the SSH config setting.

A [branch (-B)] name must be set all the time.

An override folder naem can be set with `-f`

An optional [remote name] can be set, if not set "origin" will be used.

### git_sync.sh

Syncs changes, this is the script that should be placed into the crontab

```sh
git_sync.sh [repository] [branch] ([remote name])
```

The [repository] the the folder name inside the clone folder, on new clone it is output in the info line "... into (GIT REPOSITORY FOLDER)". All repository folders are in the "clone-base/" folder. The base folder is defined in the config/webhook.crf file.

A [branch] name must be set all the time.

An optional [remote name] can be set, if not set "origin" will be used.

All sync progress will be written to the log folder inside the base folder, the log file has the name of [repository]

Sample

```log
[2025-07-04 16:06:31] [<uniq id>] [START] git merge <repository folder> <remote name>/<branch>
[2025-07-04 16:06:31] [<uniq id>] Updating <sha>..<sha>
<git inforation>
[2025-07-04 16:06:31] [<uniq id>] [END]
```

#### What happens to local changes

Any local changes are stored in a backup branch and then reset and cleaned up. No local changes should be commited or synced up, this would break the pull only flow.

If there are changes and they have been reset it will show up in the log like this

```log
...
[2026-01-22 14:31:02] [F4222EC4] [!] Local or cached changes detected, creating backup branch and resetting changes
[2026-01-22 14:31:02] [F4222EC4] Reset log: HEAD is now at 633f6de Uppdate 14
...
```

If a clean up was run too it will show up as "Clean log" after the Reset log.

The reset will only reset local or cached changs (added) but not changed that have been commited. If changes have been commited a manual reset has to be done.

## TODO

Future versions will hold an incoming webhook handler and a polling scripts (systemd based)