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         cloneresult = run(["git", "clone", "--local", "--branch", LIVE_BRANCH, GIT_DIRECTORY, BUILD_DIRECTORY], capture_output=True, encoding="utf-8")
  31         # if cloneresult.stderr:
  32                 # print(cloneresult.stderr)
  33         cloneresult.check_returncode()
  34
  35         # Set up various containers.
  36         months = {}
  37         topics = {}
  38
  39         # Create an XML representation of the provided status text.
  40         def statusxml (text, version="1.0"):
  41                 doc = getDOMImplementation().createDocument(None, "article", None)
  42                 articleElt = doc.documentElement
  43                 articleElt.setAttribute("xmlns", XHTML_NAMESPACE)
  44                 articleElt.setAttribute("lang", LANG)
  45                 for para in text.split("\n\n"):
  46                         paraElt = articleElt.appendChild(doc.createElement("p"))
  47                         for component in re.findall(r'<[a-z]+:[^\s]*>(?:="[^\n"]+")?|\n|[^<\n]+|<(?![a-z]+:[^\s]*>)', para):
  48                                 if component == "\n":
  49                                         paraElt.appendChild(doc.createElement("br"))
  50                                 elif re.fullmatch(r'<[a-z]+:[^\s]*>(?:="[^\n"]+")?', component):
  51                                         href = component.split(">", maxsplit=1)[0][1:]
  52                                         anchorElt = paraElt.appendChild(doc.createElement("a"))
  53                                         anchorElt.setAttribute("href", href)
  54                                         anchorElt.setAttribute("rel", "noreferrer")
  55                                         anchorElt.appendChild(doc.createTextNode(component if len(href) == len(component) - 2 else component[len(href)+4:-1]))
  56                                 else:
  57                                         paraElt.appendChild(doc.createTextNode(component))
  58                 return articleElt.toxml()
  59
  60         # Map status paths to status objects, or None if there is an error.
  61         #
  62         # The provided path must be to a `text` object.
  63         def statusmap (topic, path):
  64                 status = { "@type": "MicroblogPost" }
  65                 version_path = next(path.parent.glob("0=*"))
  66                 if version_path and version_path.name != "0=x_status_git_1.0":
  67                         warn(f"Unrecognized version for {path}; skipping.")
  68                         return None
  69                 if topic:
  70                         status["subject"] = topic
  71                 author_path = next(path.parent.glob("1=*"))
  72                 if author_path:
  73                         status["author"] = { "name": author_path.name[2:] }
  74                         with author_path.open("r", encoding="utf-8") as text:
  75                                 status["author"]["@id"] = text.read().strip()
  76                 title_path = next(path.parent.glob("2=*"))
  77                 if title_path:
  78                         with title_path.open("r", encoding="utf-8") as text:
  79                                 title = text.read().strip()
  80                                 status["title"] = title
  81                 date_path = next(path.parent.glob("3=*"))
  82                 datetime = ""
  83                 if date_path:
  84                         with date_path.open("r", encoding="utf-8") as text:
  85                                 datetime = text.read().strip()
  86                                 status["created"] = datetime
  87                 else:
  88                         warn(f"Missing date for {path}; skipping.")
  89                         return None
  90                 identifier_path = next(path.parent.glob("4=*"))
  91                 identifier = ""
  92                 if identifier_path:
  93                         identifier = identifier_path.name[2:]
  94                         status["@id"] = f"{PUBLIC_URL}/topics/{topic}/{identifier}" if topic else f"{PUBLIC_URL}/statuses/{datetime[0:7]}/{identifier}"
  95                         with identifier_path.open("r", encoding="utf-8") as text:
  96                                 status["identifier"] = text.read().strip()
  97                 else:
  98                         warn(f"Missing identifier for {path}; skipping.")
  99                         return None
 100                 with path.open("r", encoding="utf-8") as text:
 101                         status["content"] = statusxml(text.read().strip())
 102                 return (datetime, identifier, status)
 103
 104         # Get status paths.
 105         status_paths = []
 106         for yearpath in Path(f"{BUILD_DIRECTORY}/").glob("[0-9][0-9][0-9][0-9]"):
 107                 for monthpath in yearpath.glob("[0-9][0-9]"):
 108                         for daypath in monthpath.glob("[0-9][0-9]"):
 109                                 for statuspath in daypath.glob("*/text"):
 110                                         status_paths.append((None, statuspath))
 111         for topicpath in Path(f"{BUILD_DIRECTORY}/").glob("topic/*"):
 112                 for hash0path in topicpath.glob("[0-9a-f]"):
 113                         for hash1path in hash0path.glob("[0-9a-f]"):
 114                                 for hash2path in hash1path.glob("[0-9a-f]"):
 115                                         for hash3path in hash2path.glob("[0-9a-f]"):
 116                                                 for statuspath in hash3path.glob("*/text"):
 117                                                         status_paths.append((topicpath.name, statuspath))
 118
 119         # Build status objects and listings.
 120         for (datetime, identifier, status) in sorted(filter(None, starmap(statusmap, status_paths))):
 121                 if "subject" in status:
 122                         topic = status["subject"]
 123                         if topic not in topics:
 124                                 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", "title": "dct:title", "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 }
 125                         topics[topic]["items"].append(status)
 126                 else:
 127                         yyyy_mm = datetime[0:7]
 128                         if yyyy_mm not in months:
 129                                 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", "title": "dct:title", "content": { "@id": "sioc:content", "@type": "http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral" } }, "@id": f"{PUBLIC_URL}/statuses/{yyyy_mm}", "@type": ["OrderedCollectionPage", "Thread"], "items": [], "partOf": f"{PUBLIC_URL}/statuses" }
 130                         months[yyyy_mm]["items"].append(status)
 131
 132         # Set up the public directory.
 133         if exists(PUBLIC_DIRECTORY):
 134                 rmtree(PUBLIC_DIRECTORY)
 135         mkdir(PUBLIC_DIRECTORY)
 136
 137         # Copy H·T·M·L files to their expected locations.
 138         copy2(f"{BUILD_DIRECTORY}/index.html", f"{PUBLIC_DIRECTORY}/index.html")
 139         copy2(f"{BUILD_DIRECTORY}/about.html", f"{PUBLIC_DIRECTORY}/.about.html")
 140         copy2(f"{BUILD_DIRECTORY}/status.html", f"{PUBLIC_DIRECTORY}/.status.html")
 141         copy2(f"{BUILD_DIRECTORY}/statuses.html", f"{PUBLIC_DIRECTORY}/.statuses.html")
 142         copy2(f"{BUILD_DIRECTORY}/topic.html", f"{PUBLIC_DIRECTORY}/.topic.html")
 143         copy2(f"{BUILD_DIRECTORY}/topics.html", f"{PUBLIC_DIRECTORY}/.topics.html")
 144
 145         # Output “about” metadata
 146         if not exists(f"{PUBLIC_DIRECTORY}/about"):
 147                 mkdir(f"{PUBLIC_DIRECTORY}/about")
 148         with open(f"{PUBLIC_DIRECTORY}/about/index.jsonld", "w", encoding="utf-8") as f:
 149                 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)
 150
 151         # Output month‐based listings and the non‐topic index
 152         if not exists(f"{PUBLIC_DIRECTORY}/statuses"):
 153                 mkdir(f"{PUBLIC_DIRECTORY}/statuses")
 154         statuspairs = list(enumerate(months.items()))
 155         for (index, (yyyy_mm, ld)) in statuspairs:
 156                 if not exists(f"{PUBLIC_DIRECTORY}/statuses/{yyyy_mm}"):
 157                         mkdir(f"{PUBLIC_DIRECTORY}/statuses/{yyyy_mm}")
 158                 ld["first"] = f"{PUBLIC_URL}/statuses/{statuspairs[0][1][0]}"
 159                 ld["current"] = f"{PUBLIC_URL}/statuses/{statuspairs[-1][1][0]}"
 160                 if index > 0:
 161                         ld["prev"] = f"{PUBLIC_URL}/statuses/{statuspairs[index - 1][1][0]}"
 162                 if index < len(statuspairs) - 1:
 163                         ld["next"] = f"{PUBLIC_URL}/statuses/{statuspairs[index + 1][1][0]}"
 164                 with open(f"{PUBLIC_DIRECTORY}/statuses/{yyyy_mm}/index.jsonld", "w", encoding="utf-8") as f:
 165                         json.dump(ld, f, ensure_ascii=False, allow_nan=False)
 166         with open(f"{PUBLIC_DIRECTORY}/statuses/index.jsonld", "w", encoding="utf-8") as f:
 167                 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}/statuses/{statuspairs[0][1][0]}", "current": f"{PUBLIC_URL}/statuses/{statuspairs[-1][1][0]}", "has_parent": f"{PUBLIC_URL}" }, f, ensure_ascii=False, allow_nan=False)
 168
 169         # Output topic‐based listings and the topic index
 170         if not exists(f"{PUBLIC_DIRECTORY}/topics"):
 171                 mkdir(f"{PUBLIC_DIRECTORY}/topics")
 172         for (topic, ld) in topics.items():
 173                 if not exists(f"{PUBLIC_DIRECTORY}/topics/{topic}"):
 174                         mkdir(f"{PUBLIC_DIRECTORY}/topics/{topic}")
 175                 with open(f"{PUBLIC_DIRECTORY}/topics/{topic}/index.jsonld", "w", encoding="utf-8") as f:
 176                         json.dump(ld, f, ensure_ascii=False, allow_nan=False)
 177         with open(f"{PUBLIC_DIRECTORY}/topics/index.jsonld", "w", encoding="utf-8") as f:
 178                 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)
 179
 180         # Remove the build directory.
 181         rmtree(BUILD_DIRECTORY)