So after writing he previous post on this matter I came to two conclsions
- This website, and this induvidual project, requires its own build log - This was changed. The commit hashes are listed below
site commit: 1503286ff0b39d3f502a70373e803ec0b381940d
sigs commit: 507310741ac6370d93f8b2e72f343456c81f7de1
- I am overcomplicating this
Hugo, unsuprisingly, has a features within it which solve every issue which I am trying to fix myself.
Custom Output Formats
- From digging around stack overflow and several carefully worded messages to our favourite dimwitted genius (chat gpt) i came across this page. If you read the page at some point you will encouter a string stating
“A page can be output in as many output formats as your want, and you can have an infinate amount of output formats desfined as long as they resolve a unique path on the file system”
This essentially does what I planned to do with my previous plan. The old plan was to have an entire sperate directory inside the content folder, which contained the signed posts and instructions in the front matter to use a different theme. This essentially doees the same but in a tidier way.
To do this i basically need to edit my hugo.toml file with the [outputFormats] section. I need to define two formats, one which uses my current theme and one which outputs it in plain text (which there is a option for). It will look something like this
[outputFormats]
[outputFormats.standardFormat]
name = "standard-format"
baseName = "index"
isPlainText = false
mediaType = "text/html"
path = ""
permalinkable = true
weight = 1
[outputFormats.sigFormat]
name = "sig-format"
baseName = "sig"
isPlainText = true
mediaType = "text/html"
path = "sigs"
permalinkable = true
weight = 2
To take advantage of these two outputs I will need to call to these in the post-front matter. An example of this is shown below
---
title: 'test-page'
date: 2023-11-11T12:09:10Z
draft: false
outputs:
- standard-format
- sig-format
---
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
this is a message signed by me - eddiequinn
-----BEGIN PGP SIGNATURE-----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=hJ13
-----END PGP SIGNATURE-----
What this will do, assuming a post called “test-page”, will generate two pages, where the first one comes out regular and the second comes out plain text - links shown below:
Am I under the impression this will work first time? lol no?
Custom single.html
One of the biggest issues I had with implimenting this before was it wound up making the pages look horrific. This is allot of the reason why I want the sigs on their own plaintext page, which is seperate from the actual post. There is a tiny problem wit this however, that being that I am storig the pages in the content directory sig appended. What this means is that I will need to define a custom layout for pages, where it strips all PGP artifacts from the page.
Originally I was going to use ReplaceRE to do this. What this would do is run the .content variable throug a regex filter when the site is built. This would work if I didnt have pgp sigs in posts, like just above this. Im reluctant to say regex has limitation that would prevent me from doing this, but it definatley doesnt cope well with this kind of strings inside of strings enviroment. What I am instead going to do, is combine regex with python to do this.
As with all other snippets posted here, please do not run this as go “ThAt DoEsNt WoRk” - The below snippet is just a rough draft. I havent debugged it, I havent even run it yet. I just wanted to get something typed out.
import re
def strip_outer_pgp(text):
# Define regular expressions for matching PGP artifacts
pgp_signature_pattern = re.compile(r'(?s)^-----BEGIN PGP SIGNATURE-----.+?-----END PGP SIGNATURE-----\n', re.MULTILINE)
pgp_signed_message_pattern = re.compile(r'(?s)^-----BEGIN PGP SIGNED MESSAGE-----.+?-----BEGIN PGP SIGNATURE-----\n', re.MULTILINE)
# Find the outermost PGP signature and signed message
pgp_signature_match = pgp_signature_pattern.search(text)
pgp_signed_message_match = pgp_signed_message_pattern.search(text)
if pgp_signature_match and pgp_signed_message_match and pgp_signature_match.start() < pgp_signed_message_match.start():
# Remove the outermost PGP signature and signed message
stripped_text = text[:pgp_signature_match.end()] + text[pgp_signed_message_match.end():]
return stripped_text.strip()
# If no matches are found, return the original text
return text
if __name__ == "__main__":
input string = input("Please input string")
result = strip_outer_pgp(input_string)
print(result)
Ill then have to put something like this in the layout file so this filter is used
{{ $content := .Content }}
{{ $filteredContent := printf "%s" $content | exec "python filter.py" }}
<div class = "content"> {{ $filteredContent | safeHTML }} </div>
Conclusion
I will have to debug the things ive discovered and coded today, and im definatley going to have to make a new branch on my hugo files for this while it is worked on. Atleast I have an idea on how I can achieve this now.