Lady’s Gitweb - x_status_git/blob - post-receive

   1 #!/usr/bin/env python3
   2 from glob import iglob
   3 from itertools import starmap
   4 import json
   5 from os import mkdir
   6 from os.path import exists
   7 from pathlib import Path
   8 import re
   9 from shutil import copy2, rmtree
  10 from subprocess import run
  11 from sys import stdin
  12 from warnings import warn
  13 from xml.dom import XHTML_NAMESPACE
  14 from xml.dom.minidom import getDOMImplementation
  15
  16 GIT_DIRECTORY = "/home/USERNAME/Status.git"
  17 BUILD_DIRECTORY = "/home/USERNAME/status.site.example/.build"
  18 PUBLIC_DIRECTORY = "/home/USERNAME/status.site.example/public"
  19 PUBLIC_URL = "https://status.site.example"
  20 LANG = "en"
  21 LIVE_BRANCH = "live"
  22
  23 if stdin.read().split()[-1] == f"refs/heads/{LIVE_BRANCH}":
  24
  25         print(f"This is an update to the '{LIVE_BRANCH}' branch; regenerating site…")
  26
  27         # Set up the build directory.
  28         if exists(BUILD_DIRECTORY):
  29                 rmtree(BUILD_DIRECTORY)
  30         run(["git", "clone", "--local", "--branch", "live", GIT_DIRECTORY, BUILD_DIRECTORY], capture_output=True, encoding="utf-8")
  31
  32         # Set up various containers.
  33         months = {}
  34         topics = {}
  35
  36         # Create an XML representation of the provided status text.
  37         def statusxml (text, version="1.0"):
  38                 doc = getDOMImplementation().createDocument(None, "article", None)
  39                 articleElt = doc.documentElement
  40                 articleElt.setAttribute("xmlns", XHTML_NAMESPACE)
  41                 articleElt.setAttribute("lang", LANG)
  42                 for para in text.split("\n\n"):
  43                         paraElt = articleElt.appendChild(doc.createElement("p"))
  44                         for component in re.findall(r'<[a-z]+:[^\s]*>(?:="[^\n"]+")?|\n|[^<\n]+|<(?![a-z]+:[^\s]*>)', para):
  45                                 if component == "\n":
  46                                         paraElt.appendChild(doc.createElement("br"))
  47                                 elif re.fullmatch(r'<[a-z]+:[^\s]*>(?:="[^\n"]+")?', component):
  48                                         href = component.split(">", maxsplit=1)[0][1:]
  49                                         anchorElt = paraElt.appendChild(doc.createElement("a"))
  50                                         anchorElt.setAttribute("href", href)
  51                                         anchorElt.setAttribute("rel", "noreferrer")
  52                                         anchorElt.appendChild(doc.createTextNode(component if len(href) == len(component) - 2 else component[len(href)+4:-1]))
  53                                 else:
  54                                         paraElt.appendChild(doc.createTextNode(component))
  55                 return articleElt.toxml()
  56
  57         # Map status paths to status objects, or None if there is an error.
  58         #
  59         # The provided path must be to a `text` object.
  60         def statusmap (topic, path):
  61                 status = { "@type": "MicroblogPost" }
  62                 version_path = next(path.parent.glob("0=*"))
  63                 if version_path and version_path.name != "0=x_status_git_1.0":
  64                         warn(f"Unrecognized version for {path}; skipping.")
  65                         return None
  66                 if topic:
  67                         status["subject"] = topic
  68                 author_path = next(path.parent.glob("1=*"))
  69                 if author_path:
  70                         status["author"] = { "name": author_path.name[2:] }
  71                         with author_path.open("r", encoding="utf-8") as text:
  72                                 status["author"]["@id"] = text.read().strip()
  73                 date_path = next(path.parent.glob("3=*"))
  74                 datetime = ""
  75                 if date_path:
  76                         with date_path.open("r", encoding="utf-8") as text:
  77                                 datetime = text.read().strip()
  78                                 status["created"] = datetime
  79                 else:
  80                         warn(f"Missing date for {path}; skipping.")
  81                         return None
  82                 identifier_path = next(path.parent.glob("4=*"))
  83                 identifier = ""
  84                 if identifier_path:
  85                         identifier = identifier_path.name[2:]
  86                         status["@id"] = f"{PUBLIC_URL}/topics/{topic}/{identifier}" if topic else f"{PUBLIC_URL}/{datetime[0:7]}/{identifier}"
  87                         with identifier_path.open("r", encoding="utf-8") as text:
  88                                 status["identifier"] = text.read().strip()
  89                 else:
  90                         warn(f"Missing identifier for {path}; skipping.")
  91                         return None
  92                 with path.open("r", encoding="utf-8") as text:
  93                         status["content"] = statusxml(text.read().strip())
  94                 return (datetime, identifier, status)
  95
  96         # Get status paths.
  97         status_paths = []
  98         for yearpath in Path(f"{BUILD_DIRECTORY}/").glob("[0-9][0-9][0-9][0-9]"):
  99                 for monthpath in yearpath.glob("[0-9][0-9]"):
 100                         for daypath in monthpath.glob("[0-9][0-9]"):
 101                                 for statuspath in daypath.glob("*/text"):
 102                                         status_paths.append((None, statuspath))
 103         for topicpath in Path(f"{BUILD_DIRECTORY}/").glob("topic/*"):
 104                 for hash0path in topicpath.glob("[0-9a-f]"):
 105                         for hash1path in hash0path.glob("[0-9a-f]"):
 106                                 for hash2path in hash1path.glob("[0-9a-f]"):
 107                                         for hash3path in hash2path.glob("[0-9a-f]"):
 108                                                 for statuspath in hash3path.glob("*/text"):
 109                                                         status_paths.append((topicpath.name, statuspath))
 110
 111         # Build status objects and listings.
 112         for (datetime, identifier, status) in sorted(filter(None, starmap(statusmap, status_paths))):
 113                 if "subject" in status:
 114                         topic = status["subject"]
 115                         if topic not in topics:
 116                                 topics[topic] = { "@context": { "@language": LANG, "activity": "https://www.w3.org/ns/activitystreams#", "dct": "http://purl.org/dc/terms/", "foaf": "http://xmlns.com/foaf/0.1/", "sioc": "http://rdfs.org/sioc/ns#", "sioct": "http://rdfs.org/sioc/types#", "OrderedCollection": "activity:OrderedCollection", "Thread": "sioc:Thread", "MicroblogPost": "sioct:MicroblogPost", "items": { "@id": "activity:items", "@type": "@id", "@container": "@list" }, "created": { "@id": "dct:created", "@type": "http://www.w3.org/2001/XMLSchema#dateTime" }, "creator": { "@id": "dct:creator", "@type": "@id" }, "identifier": { "@id":  "dct:identifier", "@type": "http://www.w3.org/2001/XMLSchema#anyURI" }, "subject": "dct:subject", "name": "foaf:name", "content": { "@id": "sioc:content", "@type": "http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral" } }, "@id": f"{PUBLIC_URL}/topics/{topic}", "@type": ["OrderedCollection", "Thread"], "items": [], "subject": topic }
 117                         topics[topic]["items"].append(status)
 118                 else:
 119                         yyyy_mm = datetime[0:7]
 120                         if yyyy_mm not in months:
 121                                 months[yyyy_mm] = { "@context": { "@language": LANG, "activity": "https://www.w3.org/ns/activitystreams#", "dct": "http://purl.org/dc/terms/", "foaf": "http://xmlns.com/foaf/0.1/", "sioc": "http://rdfs.org/sioc/ns#", "sioct": "http://rdfs.org/sioc/types#", "OrderedCollectionPage": "activity:OrderedCollectionPage", "Thread": "sioc:Thread", "MicroblogPost": "sioct:MicroblogPost", "current": { "@id": "activity:current", "@type": "@id" }, "first": { "@id": "activity:first", "@type": "@id" }, "items": { "@id": "activity:items", "@type": "@id", "@container": "@list" }, "partOf": { "@id": "activity:partOf", "@type": "@id" }, "prev": { "@id": "activity:prev", "@type": "@id" }, "next": { "@id": "activity:next", "@type": "@id" }, "created": { "@id": "dct:created", "@type": "http://www.w3.org/2001/XMLSchema#dateTime" }, "creator": { "@id": "dct:creator", "@type": "@id" }, "identifier": { "@id":  "dct:identifier", "@type": "http://www.w3.org/2001/XMLSchema#anyURI" }, "name": "foaf:name", "content": { "@id": "sioc:content", "@type": "http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral" } }, "@id": f"{PUBLIC_URL}/{yyyy_mm}", "@type": ["OrderedCollectionPage", "Thread"], "items": [], "partOf": f"{PUBLIC_URL}/statuses" }
 122                         months[yyyy_mm]["items"].append(status)
 123
 124         # Set up the public directory.
 125         if exists(PUBLIC_DIRECTORY):
 126                 rmtree(PUBLIC_DIRECTORY)
 127         mkdir(PUBLIC_DIRECTORY)
 128
 129         # Copy H·T·M·L files to their expected locations.
 130         copy2(f"{BUILD_DIRECTORY}/index.html", f"{PUBLIC_DIRECTORY}/index.html")
 131         copy2(f"{BUILD_DIRECTORY}/about.html", f"{PUBLIC_DIRECTORY}/.about.html")
 132         copy2(f"{BUILD_DIRECTORY}/status.html", f"{PUBLIC_DIRECTORY}/.status.html")
 133         copy2(f"{BUILD_DIRECTORY}/statuses.html", f"{PUBLIC_DIRECTORY}/.statuses.html")
 134         copy2(f"{BUILD_DIRECTORY}/topic.html", f"{PUBLIC_DIRECTORY}/.topic.html")
 135         copy2(f"{BUILD_DIRECTORY}/topics.html", f"{PUBLIC_DIRECTORY}/.topics.html")
 136
 137         # Output “about” metadata
 138         if not exists(f"{PUBLIC_DIRECTORY}/about"):
 139                 mkdir(f"{PUBLIC_DIRECTORY}/about")
 140         with open(f"{PUBLIC_DIRECTORY}/about/index.jsonld", "w", encoding="utf-8") as f:
 141                 json.dump({ "@context": { "@language": LANG, "activity": "https://www.w3.org/ns/activitystreams#", "sioc": "http://rdfs.org/sioc/ns#", "sioct": "http://rdfs.org/sioc/types#", "Forum": "sioc:Forum", "Thread": "sioc:Thread", "Microblog": "sioct:Microblog", "streams": { "@id": "activity:streams", "@type": "@id" } }, "@id": f"{PUBLIC_URL}", "@type": "Microblog", "streams": [{ "@id": f"{PUBLIC_URL}/statuses", "@type": "Thread" }, { "@id": f"{PUBLIC_URL}/topics", "@type": "Forum" }] }, f, ensure_ascii=False, allow_nan=False)
 142
 143         # Output month‐based listings and the non‐topic index
 144         statuspairs = list(enumerate(months.items()))
 145         for (index, (yyyy_mm, ld)) in statuspairs:
 146                 if not exists(f"{PUBLIC_DIRECTORY}/{yyyy_mm}"):
 147                         mkdir(f"{PUBLIC_DIRECTORY}/{yyyy_mm}")
 148                 ld["first"] = f"{PUBLIC_URL}/{statuspairs[0][1][0]}"
 149                 ld["current"] = f"{PUBLIC_URL}/{statuspairs[-1][1][0]}"
 150                 if index > 0:
 151                         ld["prev"] = f"{PUBLIC_URL}/{statuspairs[index - 1][1][0]}"
 152                 if index < len(statuspairs) - 1:
 153                         ld["next"] = f"{PUBLIC_URL}/{statuspairs[index + 1][1][0]}"
 154                 with open(f"{PUBLIC_DIRECTORY}/{yyyy_mm}/index.jsonld", "w", encoding="utf-8") as f:
 155                         json.dump(ld, f, ensure_ascii=False, allow_nan=False)
 156         if not exists(f"{PUBLIC_DIRECTORY}/statuses"):
 157                 mkdir(f"{PUBLIC_DIRECTORY}/statuses")
 158         with open(f"{PUBLIC_DIRECTORY}/statuses/index.jsonld", "w", encoding="utf-8") as f:
 159                 json.dump({ "@context": { "@language": LANG, "activity": "https://www.w3.org/ns/activitystreams#", "sioc": "http://rdfs.org/sioc/ns#", "OrderedCollection": "activity:OrderedCollection", "Thread": "sioc:Thread", "current": { "@id": "activity:current", "@type": "@id" }, "first": { "@id": "activity:first", "@type": "@id" }, "has_parent": { "@id": "sioc:has_parent", "@type": "id" } }, "@id": f"{PUBLIC_URL}/statuses", "@type": ["OrderedCollection", "Thread"], "first": f"{PUBLIC_URL}/{statuspairs[0][1][0]}", "current": f"{PUBLIC_URL}/{statuspairs[-1][1][0]}", "has_parent": f"{PUBLIC_URL}" }, f, ensure_ascii=False, allow_nan=False)
 160
 161         # Output topic‐based listings and the topic index
 162         if not exists(f"{PUBLIC_DIRECTORY}/topics"):
 163                 mkdir(f"{PUBLIC_DIRECTORY}/topics")
 164         for (topic, ld) in topics.items():
 165                 if not exists(f"{PUBLIC_DIRECTORY}/topics/{topic}"):
 166                         mkdir(f"{PUBLIC_DIRECTORY}/topics/{topic}")
 167                 with open(f"{PUBLIC_DIRECTORY}/topics/{topic}/index.jsonld", "w", encoding="utf-8") as f:
 168                         json.dump(ld, f, ensure_ascii=False, allow_nan=False)
 169         with open(f"{PUBLIC_DIRECTORY}/topics/index.jsonld", "w", encoding="utf-8") as f:
 170                 json.dump({ "@context": { "@language": LANG, "activity": "https://www.w3.org/ns/activitystreams#", "dct": "http://purl.org/dc/terms/", "sioc": "http://rdfs.org/sioc/ns#", "Collection": "activity:Collection", "Forum": "sioc:Forum", "items": { "@id": "activity:items", "@type": "@id" }, "has_parent": { "@id": "sioc:has_parent", "@type": "id" }, "subject": "dct:subject" }, "@id": f"{PUBLIC_URL}/topics", "@type": ["Collection", "Forum"], "items": list(map(lambda a: { "@id": a["@id"], "subject": a["subject"] }, topics.values())), "has_parent": f"{PUBLIC_URL}" }, f, ensure_ascii=False, allow_nan=False)
 171
 172         # Remove the build directory.
 173         rmtree(BUILD_DIRECTORY)