#!/usr/bin/env python3 import json, sys from os import path from datetime import datetime from urllib import request def render_time(when): return datetime.strptime(when, "%Y-%m-%dT%H:%M:%S.%f%z").astimezone().strftime("%F %T %Z") def render_who_when(who, when, how): return f"""**{who["displayName"]} | @{who["handle"]}** {how} at {render_time(when)}:""" # the markdown is vaguely based around Zulip's, but massaged to look more like cohost's html formatting def render(post): rendered = [] for block in post["content"]: if "markdown" in block: rendered.append(block["markdown"]) elif "ask" in block: ask = block["ask"] content, sent, who = ask["content"], ask["sentAt"], ask["who"] if who.get("anon"): who_when = f"anon asked at {render_time(sent)}" else: who_when = render_who_when(ask["who"], sent, "asked") rendered.append(f"""{who_when}: ```quote {content} ``` """) else: img = block["img"] alt, url = img["altText"], img["fileURL"] if url is None: print("Warning: found image with missing url...") continue alt = alt or "" basename = url.split("/")[-2] + path.splitext(url.split("/")[-1])[1] dst = f"../img/{basename}" if not path.exists(dst): with request.urlopen(url) as download: img = download.read() with open(dst, 'wb') as f: f.write(img) dst = "../" + dst # markdown doesn't allow newlines in image alt text # TODO: figure out how to actually keep these: https://tech.lgbt/@jyn/112117398042554191 alt = alt.replace("\n", " ") rendered.append(f"![{alt}]({dst})") rendered = "\n\n".join(rendered) tags = ", ".join("#" + tag for tag in post["tags"]) who_when = render_who_when(post["poster"], post["publishedAt"], "said") return f"""{who_when} {rendered} {tags}""" for post in json.load(sys.stdin): rendered = "\n\n".join([render(share) for share in post["shareTree"]] + [render(post)]) with open(f"""rendered/{post["filename"]}.md""", "w") as f: f.write(rendered)