git-auto | Simple Shell Script To Do Git Commit

I have a small script which makes some changes to certain files and push those changes. However after the script runs I get the PR url and I have to manually submit it from the browser. How to automate it so that even the pull request is created automatically.

I had a few questions about automatic git pulls on a remote server. I am aware there are several questions like this, but I wasn't sure what steps to take exactly, and I don't want to mess up my current setup with a mistake :/

To wit, the environment is on a Google Cloud VM. I am running a flask-based website that renders each page with the render_template() function.

The website resides inside its git folder, i.e. I never set up a bare repo and copied stuff. When I set it up a couple years ago, I just did git clone repo-url, then inside the repo directory, did flask run. Then I set up nginx to connect to the site's socket created with uwsgi inside the repo directory.

--

It has been working fine. I make changes locally to the content, push to github, then log in to the VM, and perform a git pull.

I want to do this automatically. I tried adding a cron job to do this, where the job basically ran a script, and the script did the git pull. Script content was:

I am trying to run a script for git auto push when we shut down the PC. But my script and service file only execute the commit command and after that, the network goes down. So, it does not execute the push command.

How can I complete the execution of the script before the network goes down?

My service file:

I don't know the main reason but once in a while git completely removes the content in a buffer/file that I am working on if the commit is not completed successfully.

After file's content is cleaned, file's undo-tree is not able to fetch the latest changes using emacs as well, instead its tree is completely empty. I am not able to find the latest changes I have and they do not show up in stash.

In the background: zsh's git auto-fetch is running, I am not sure would that have any affect.

=> Is there anyway to prevent for git to completely delete the content of a file?

• The file (scripts/lib.py) is not located in the .gitignore or ~/.gitignore_global.

• git check-ignore -v -- scripts/lib.py returns empty

• Please note that the file remains as it is with 0 bytes

I work with postman collections in git. Postman does a lot of things well, but the id regeneration that occur when you import are not ideal.

Essentially importing a postman collection, and exporting it again results in a change for every id

e.g. output from git diff-index -p HEAD --

First of all on zsh I am using git-auto-fetch plugin. This may messes up if it is automatically take place right after squash is done and led me end up in the rebase.

I just want to update my previous commit with the new one. I believe squashing is solution for this, where commit and squash it with the previous commit; hence latest commit is updated with the new one.

If the latest commit is: