Result #
#! /run/current-system/sw/bin/bash
cd ..
# Get the total number of commits
commit_count=$(git rev-list --count HEAD)
# Loop through each commit in reverse order (last to first)
for (( i=1; i<=$commit_count; i++ )); do
# Calculate commit hash for the current iteration
commit_hash=$(git rev-list --max-count=1 --skip=$((commit_count - i)) HEAD)
# Get the hash of the previous commit (or an empty string if it's the first commit)
if [ $i -eq 1 ]; then
previous_commit=""
else
previous_commit=$(git rev-list --max-count=1 --skip=$((commit_count - i + 1)) HEAD)
fi
# Generate the diff, using the empty tree for the initial commit if necessary
if [ -z "$previous_commit" ]; then
git diff 4b825dc642cb6eb9a060e54bf8d69288fbee4904 $commit_hash > "commit_$(printf "%03d" $i)_$(git rev-parse --short $commit_hash).diff"
else
git diff $previous_commit $commit_hash > "commit_$(printf "%03d" $i)_$(git rev-parse --short $commit_hash).diff"
fi
done
echo "Diff files for all commits have been created."
Current Date: December 29, 2024
With documenting my experiences to blog and documentation; I’ve been starting the process to understand how I could get the Git information presented clearly into the appropriate resources, or as clear as possible.
Get Commits #
As a start, I began to conceptualize how to approach this from top, bottom approach. I began my experience with asking AI this:
Question #
I have a Git Repository and a Hugo Site.
I want to build a blog on this Hugo Site. The purpose of this blog is share my experiences with different systems I’ve been learning about, based on using commits
Is there a way I could easily capture the Commits and share on a post?
Answer #
To integrate Git commits into your Hugo blog posts in a way that allows you to share your experiences with different systems, you can follow these steps:
- Setup Your Environment
Ensure you have Hugo and Git installed on your system. Your Hugo site should already be set up with a basic structure.
- Capture Git Commits
Scripting: You can write a script (in bash, Python, or another language) to extract commit information from your Git repository. Here’s how you might do this with bash:
bash
git log --pretty=format:"%h %ad %s" --date=short -n 5 > commits.txtThis command lists the last 5 commits with their hash, date, and message. You might customize this to fit your needs, like changing the number of commits, or formatting.
Automation: Automate this script to run each time you want to update your blog post with new commit information. You could include this in a CI/CD pipeline or run it manually before generating your site.
…
Resources for Reference:
- For scripting to capture Git commits, consider looking into Git documentation for more options on formatting log output.
- For integrating with Hugo, you might find it useful to understand how Hugo handles data files and template syntax.
- For automation through GitHub Actions or similar CI/CD, explore how others have set up Hugo with GitHub Pages.
This method allows you to keep your blog posts updated with your Git commit history, providing a narrative of your development journey with various systems.
Commits, got it! #
So, I have a way to get the commits. Helpful overall. I want to present the exact changes to another person that would read the content of my side which led to these followup questions:
Question #
Ok, it’s helpful to know I could share in that manner with a script. How would I present the exact code changes from a commit?
Answer #
To present the exact code changes from a specific Git commit in your Hugo blog post, you can use git diff along with Hugo’s Markdown capabilities. Here’s how you can do it:
- Generate the Diff
Use git diff to see the changes made in a specific commit. Here’s how you can do it:
bash
git diff <commit_hash>~1..<commit_hash>Replace <commit_hash> with the actual hash of the commit you’re interested in. The ~1 refers to the commit before the one you specified, showing the changes introduced by that commit.
Or, if you just want to see the changes from the previous commit:
bash
git show <commit_hash>To save this diff for use in your blog post:
bash
git diff <commit_hash>~1..<commit_hash> > commit_diff.txt
- Format the Diff for Markdown
Git diff output can be directly used in Markdown with some formatting. Here’s how you can present it:
markdown
Here's the exact change from commit `3d7546c`: ```diff <paste the output from commit_diff.txt here>The `diff` language in Markdown code blocks will highlight additions (with `+`) and deletions (with `-`).
- Example in Hugo Post
In your Hugo post, your Markdown might look like this:
markdown
--- title: "Detailed Changes in System XYZ" date: 2024-12-22 --- Here's the exact change from commit `3d7546c`: ```diff diff --git a/path/to/file b/path/to/file index 1234567..89abcde 100644 --- a/path/to/file +++ b/path/to/file @@ -1,3 +1,4 @@ -Old line +New line added +Another new line Unchanged lineThis commit introduced a fix for an authentication issue by adding necessary checks.
…
This approach allows you to showcase specific code changes directly within your blog posts, giving a clear view of what was modified in each commit. Remember to use syntax highlighting for better readability, which Hugo supports through its integration with libraries like Highlight.js or Prism.js.
Question #
Write a loop script to pull the diff from all commits in repo, each commit would have it’s own file with a prefix of first commit to last commit.
Answer #
Here’s a bash script that will loop through all commits in a Git repository, generate diffs for each commit, and save them in individual files named with a prefix indicating their order from the first to the last commit:
#!/bin/bash
# Get the total number of commits
commit_count=$(git rev-list --count HEAD)
# Loop through each commit in reverse order (last to first)
for (( i=1; i<=$commit_count; i++ )); do
# Calculate commit hash for the current iteration
commit_hash=$(git rev-list --max-count=1 --skip=$((commit_count - i)) HEAD)
# Get the hash of the previous commit (or an empty string if it's the first commit)
if [ $i -eq 1 ]; then
previous_commit=""
else
previous_commit=$(git rev-list --max-count=1 --skip=$((commit_count - i + 1)) HEAD)
fi
# Generate the diff, using the empty tree for the initial commit if necessary
if [ -z "$previous_commit" ]; then
git diff 4b825dc642cb6eb9a060e54bf8d69288fbee4904 $commit_hash > "commit_$(printf "%03d" $i)_$(git rev-parse --short $commit_hash).diff"
else
git diff $previous_commit $commit_hash > "commit_$(printf "%03d" $i)_$(git rev-parse --short $commit_hash).diff"
fi
done
echo "Diff files for all commits have been created."
Explanation:
- Commit Count: We first determine the total number of commits in the repository.
- Loop: We loop through commits from the last to the first, which helps in maintaining the order in file names.
- Commit Hash: For each iteration, we get the hash of the commit we’re currently processing.
- Previous Commit: We determine the previous commit’s hash, except for the first commit where there isn’t one, so we use the empty tree’s hash as a placeholder.
- Diff Generation:
- For the first commit, we compare against the empty tree (4b825dc642cb6eb9a060e54bf8d69288fbee4904), which is a constant known hash in Git for an empty repository state.
- For subsequent commits, we compare against the previous commit’s hash.
- File Naming: Files are named with a three-digit prefix for ordering (001, 002, etc.), followed by the short hash of the commit for clarity and uniqueness.
Note:
- This script assumes you’re running it in the root directory of your Git repository.
- The script uses printf “%03d” to ensure the commit number is always three digits, aiding in correct sorting.
- git rev-parse –short gives a shortened version of the commit hash for readability.
Make sure to make the script executable with chmod +x script_name.sh before running it.