Skip to main content

Sed: Normalize markdown file with Regex

· One min read
papa31

Once upon a time I found this page:

http://sethrobertson.github.io/GitBestPractices/ from this repo.

Initial file​

I downloaded it from the repo's gh-pages branch and saved as git-best-initial.md.

After grabbing via Web Clipper (Chrome browser plugin), this file needed to be normalized. The Table of Content (TOC) links in this file do not work correctly, and some code blocks look ugly.

Normalize with sed & Regular Expressions​

Initially I was processing the file along with the TOC:

cat git-best-initial.md | sed -e 's/ \/>/><\/a>/g' -e 's/\* \(\w.*$\)/\* **\1**/g' -e 's/^    //g' -e 's/^### /### \&emsp; /g' | less

But then I deleted the TOC because it is automatically generated by Docusaurus (from the h2 & h3 headers) and simplified the code a bit:

cat git-best-initial.md | sed -e 's/\*   \(.*$\)/\* **\1**/g' -e 's/^### /### \&emsp; /g' > result.md

Resulting file​

Finally I manually edited the resulting file and got this: Git Best Practices