×

Caspershire Meta (csmeta) has been running for almost 10 years now since July 2014. It went through several static content generators (SCGs), the latest one being Hugo. Now, it is time for another migration. I decided to retire csmeta, replacing it with Caspershire Aurora (auroracs, located at aurora.caspershire.net). Eleventy has been the default option for new simple projects because I found out that I spent less time setting it up compared to Hugo. For a slightly complicated project, I would choose AstroJS^[1] as it gives a much more superior control at the expense of longer dev time.

Initially, I planned to migrate contents from the hugo-based SCGs to eleventy, but the plan changed and I wished to start anew. When the migration plan was formulated, I thought of amending the frontmatter in all of the *.md markdown files. It used to look like this:

---
date: 2023-08-19T17:40:00-04:00
title: "Plotting for Fragment Length from Nanopore FASTQ Data"
url: checking-for-fragment-length
description: "a comparison of speed between programs written in Python vs. Golang"
---

... and now I want it to look like this:

---
date: 2023-08-19T17:40:00-04:00
title: "Plotting for Fragment Length from Nanopore FASTQ Data"
tags: "post"
permalink: "checking-for-fragment-length/"
---

Three fields needed modification:

• url is now permalink, with a compulsory trailing slash / and quotes pair "..".
• description field is removed.
• tags field is added.

I could use tools like sed and yq to automate this process on 144 markdown *.md files, but I decided to write a python script. At first I thought it would be simple. It was not, and the process took me roughly an hour to get the script working. Why? RegEx and the behavior or Python's fileinput. Inside the src/modify.py:

import glob
import fileinput
import re


def modify() -> None:
    for f in glob.glob("*.md"):
        with fileinput.input(f, inplace=True, backup=".bak") as f:
            for line in f:
                modify_permalink(line)
    
    for f in glob.glob("*.md"):
        with fileinput.input(f, inplace=True, backup=".bak") as f:
            for line in f:
                add_tag(line)

    for f in glob.glob("*.md"):
        with fileinput.input(f, inplace=True, backup=".bak") as f:
            for line in f:
                remove_description(line)


def modify_permalink(l: str):
    """
    Modify from
      url: post-slug
    To
      permalink: 'post-slug/'  
    """
    url = re.search(r"^url: (.*)", l.strip())
    if url:
        change_link = "permalink: " + "\"" + url.group(1) + "/\""
        print(change_link)
    else:
        print(l, end="")


def add_tag(l: str):
    """
    Adding the 'tags' field underneath permalink
    """
    tag_location = re.search(r"^permalink: (.*)", l.strip())
    if tag_location:
        tag = tag_location.group(0) + "\ntags: \"posts\""
        print(tag)
    else:
        print(l, end="")


def remove_description(l: str):
    l = re.sub(r"^description: (.*)", "", l.strip())
    print(l)

I stopped it halfway because there would be cases where special handling(s) might be required. It would not be fun checking on them one-by-one, especially when there are more than 100 posts to go over.