Can we add a generated JSON list of FEPs to the repo?

ActivityPub
1

Hi! It’d be fantastic if we had a machine readable index of FEPs and their frontmatter available in the repository in codeberg, which would allow software to integrate that data.

Would this be possible?

2

cc @silverpill or someone

3

I they don’t want to add it to the repo, it wouldn’t be difficult to write a small script or web app to generate it.

4

We can do this. There’s a script that facilitators run every time a new proposal is registered: create_readme.py. It adds new FEPs to the table in README.md file, and it can generate a JSON file as well.

5

Awesome! I thought that might be possible. I don’t mind if it’s JSON or RSS come to think of it, as long as it’s machine readable

1 Like
6

It’s good to know that the scripts are still in use and working. This means that my coding can survive 6 months of me being MIA.

My general opinion of having something as the script below as part of the process: It creates another failure point. As maintaining this type of script creates another burden, I would like to avoid it.

Anyway, the correct place to look would be fep/scripts/tools/__init__.py at main - fediverse/fep - Codeberg.org

Or the script

import json

from tools import FepFile, get_fep_ids, build_url_link

fep_files = [FepFile(fep) for fep in get_fep_ids()]
fep_files = sorted(
    fep_files,
    key=lambda x: (x.parsed_frontmatter["dateReceived"], x.parsed_frontmatter["slug"]),
)

result = []

for fep in fep_files:
    link = f"[FEP-{fep.fep}: {fep.title}](./{fep.filename})"
    parsed = fep.parsed_frontmatter

    tracking_issue = build_url_link(parsed["trackingIssue"])

    if "dateFinalized" in parsed:
        date_final = parsed["dateFinalized"]
    elif "dateWithdrawn" in parsed:
        date_final = parsed["dateWithdrawn"]
    else:
        date_final = "-"
    result.append(
        {
            "slug": f"FEP-{fep.fep}",
            "title": fep.title,
            "trackingIssue": tracking_issue,
            "link": f"https://w3id.org/fep/{fep.filename}",
            "dateFinalized": date_final
        }
    )

print(json.dumps(result, indent=2))

saved as scripts/make_json.py works. You don’t even need to install any dependencies for it.

2 Likes
7

I also wouldn’t mind if this was done as a CI job to push data somewhere

1 Like
8

See:

Unless there is a volunteer to maintain the FEP CI or funding, I feel the burden is too great.

9

I opened a pull request (WIP): #426 - WIP: Create FEP index in JSON format - fediverse/fep - Codeberg.org
We can generate JSON index at the same time as readme table

10

@thisismissem did you want something like this JSON file? fep/index.json at 9562596336256502412ccc540150f2a34c26e2d1 - fediverse/fep - Codeberg.org

(from #426 - WIP: Create FEP index in JSON format - fediverse/fep - Codeberg.org)

11

yep, but maybe unwrap the quotes on the fep slug?

- "slug": "\"ae0c\"",
+ "slug": "ae0c",
12

perhaps this is workable? Working with Codeberg's CI | Codeberg Documentation

Maybe SWF could pay to host a woodpecker CI instance for ActivityPub spec/process stuff?

cc @eprodrom

1 Like
13

Sure. Updated the PR and marked it as ready for review

We already use Codeberg CI (although it stopped working several days ago, I haven’t looked into that yet).

1 Like
14

@thisismissem My pull request has been merged and FEP index in JSON format is now available at fep/index.json at main - fediverse/fep - Codeberg.org