Hack Hugo Post Metadata With Python

A while back, I rejigged the sections on my site to better reflect how I think and write.
Which meant, all the urls, on all my posts changed, since they now used the new category as a slug, instead of ye old /blog.

For e.g. https://janusworx.com/blog/using-hugo-variables-to-help-with-mailto-links-in-hugo/ …
was now at, https://janusworx.com/work/using-hugo-variables-to-help-with-mailto-links-in-hugo/

After searching a bit, I found that Hugo supported aliases. For me, it would redirect the original /blog path url to its new location
Ass I had to do, was add an aliases: ["/blog/old-post-slug"] line to each post’s metadata.¹
Line 4 in the snippet below shows, what I added to fix the post above.

1
2
3
4
5
6
7
8

---
title: "Using Hugo Variables to Help With Mailto Links in Hugo"
date: 2024-05-30T18:17:35+05:30
aliases: ["/blog/using-hugo-variables-to-help-with-mailto-links-in-hugo"]
categories: ["work"]
tags: [100WordHabit, Dgplug, Hugo]
summary: Shortcodes! Hugo Variables in Shortcodes!
---

I did not want to do this by hand for 800+ posts.
One stroke of luck for me, was that I let Hugo use its default behaviour of generating url slugs from the file names. So even if the category slugs had changed (from /blog to /work or from /blog to /personal), the url slugs would stay the same. Which meant, I could whip up a script to run through all my markdown posts and add the alias line.
So I did.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12

from pathlib import Path

INPUT_FOLDER = Path("old-posts-folder")
OUTPUT_FOLDER = Path("modified-posts-folder")

for each_file in INPUT_FOLDER.iterdir():
    with open(each_file, 'r') as file_to_read:
        alias_derived_from_file = each_file.stem
        contents_as_list = file_to_read.readlines()
        contents_as_list.insert(3, f"aliases: [\"/blog/{alias_derived_from_file}\"]\n")
        with open(Path(OUTPUT_FOLDER, each_file.name), 'w+') as file_to_write:
            file_to_write.writelines(contents_as_list)

It takes all the posts from my old folder, inserts the alias line and puts them into a new folder.² In essence, take each file, figure out the url slug from the file name, read in the contents as a list, insert my alias at position 3 (fourth actually. zero based indexing) of the list (below the title and date) and then write it all out to a new file.

I ran it, published the site and then went to check on the old urls with bated breath.
Hurrah, it all worked :)