]> Lady’s Gitweb - x_status_git/blob - post-receive
Update README.markdown
[x_status_git] / post-receive
1 #!/usr/bin/env python3
2 from datetime import datetime as dt, timezone
3 from glob import iglob
4 from itertools import starmap
5 import json
6 from os import mkdir
7 from os.path import exists
8 from pathlib import Path
9 import re
10 from shutil import copy2, rmtree
11 from subprocess import run
12 from sys import stdin
13 from warnings import warn
14 from xml.dom import XHTML_NAMESPACE
15 from xml.dom.minidom import getDOMImplementation, parseString
16
17 GIT_DIRECTORY = "/home/USERNAME/Status.git"
18 BUILD_DIRECTORY = "/home/USERNAME/status.site.example/.build"
19 PUBLIC_DIRECTORY = "/home/USERNAME/status.site.example/public"
20 PUBLIC_URL = "https://status.site.example"
21 LANG = "en"
22 LIVE_BRANCH = "live"
23
24 UTC = timezone.utc
25 CURRENT_DATETIME = f"{dt.now(UTC).replace(tzinfo=None).isoformat(timespec='seconds')}Z"
26 ATOM_NAMESPACE = "http://www.w3.org/2005/Atom"
27
28 if stdin.read().split()[-1] == f"refs/heads/{LIVE_BRANCH}":
29
30 print(f"This is an update to the '{LIVE_BRANCH}' branch; regenerating site…")
31
32 # Set up the build directory.
33 if exists(BUILD_DIRECTORY):
34 rmtree(BUILD_DIRECTORY)
35 cloneresult = run(["git", "clone", "--local", "--branch", LIVE_BRANCH, GIT_DIRECTORY, BUILD_DIRECTORY], capture_output=True, encoding="utf-8")
36 # if cloneresult.stderr:
37 # print(cloneresult.stderr)
38 cloneresult.check_returncode()
39
40 # Set up various containers.
41 months = {}
42 topics = {}
43
44 # Create an XML representation of the provided status text.
45 def statusxml (text, version="1.0"):
46 doc = getDOMImplementation().createDocument(None, "article", None)
47 articleElt = doc.documentElement
48 articleElt.setAttribute("xmlns", XHTML_NAMESPACE)
49 articleElt.setAttribute("lang", LANG)
50 for para in text.split("\n\n"):
51 paraElt = articleElt.appendChild(doc.createElement("p"))
52 for component in re.findall(r'<[a-z]+:[^\s]*>(?:="[^\n"]+")?|\n|[^<\n]+|<(?![a-z]+:[^\s]*>)', para):
53 if component == "\n":
54 paraElt.appendChild(doc.createElement("br"))
55 elif re.fullmatch(r'<[a-z]+:[^\s]*>(?:="[^\n"]+")?', component):
56 href = component.split(">", maxsplit=1)[0][1:]
57 anchorElt = paraElt.appendChild(doc.createElement("a"))
58 anchorElt.setAttribute("href", href)
59 anchorElt.setAttribute("rel", "noreferrer")
60 anchorElt.appendChild(doc.createTextNode(component if len(href) == len(component) - 2 else component[len(href)+4:-1]))
61 else:
62 paraElt.appendChild(doc.createTextNode(component))
63 return articleElt.toxml()
64
65 # Map status paths to status objects, or None if there is an error.
66 #
67 # The provided path must be to a `text` object.
68 def statusmap (topic, path):
69 status = { "@type": "MicroblogPost" }
70 version_path = next(path.parent.glob("0=*"))
71 if version_path and version_path.name != "0=x_status_git_1.0":
72 warn(f"Unrecognized version for {path}; skipping.")
73 return None
74 if topic:
75 status["subject"] = topic
76 author_path = next(path.parent.glob("1=*"))
77 if author_path:
78 status["author"] = { "name": author_path.name[2:] }
79 with author_path.open("r", encoding="utf-8") as text:
80 status["author"]["@id"] = text.read().strip()
81 title_path = next(path.parent.glob("2=*"))
82 if title_path:
83 with title_path.open("r", encoding="utf-8") as text:
84 title = text.read().strip()
85 status["title"] = title
86 date_path = next(path.parent.glob("3=*"))
87 datetime = ""
88 if date_path:
89 with date_path.open("r", encoding="utf-8") as text:
90 datetime = text.read().strip()
91 status["created"] = datetime
92 else:
93 warn(f"Missing date for {path}; skipping.")
94 return None
95 identifier_path = next(path.parent.glob("4=*"))
96 identifier = ""
97 if identifier_path:
98 identifier = identifier_path.name[2:]
99 status["@id"] = f"{PUBLIC_URL}/topics/{topic}/{identifier}" if topic else f"{PUBLIC_URL}/statuses/{datetime[0:7]}/{identifier}"
100 with identifier_path.open("r", encoding="utf-8") as text:
101 status["identifier"] = text.read().strip()
102 else:
103 warn(f"Missing identifier for {path}; skipping.")
104 return None
105 with path.open("r", encoding="utf-8") as text:
106 status["content"] = statusxml(text.read().strip())
107 return (datetime, identifier, status)
108
109 def atomForLD (ld):
110 doc = getDOMImplementation().createDocument(None, "feed", None)
111 atomElt = doc.documentElement
112 atomElt.setAttribute("xmlns", ATOM_NAMESPACE)
113 atomElt.setAttribute("xml:lang", LANG)
114 subject = ld["subject"] if "subject" in ld else "Statuses"
115 titleElt = atomElt.appendChild(doc.createElement("title"))
116 titleElt.appendChild(doc.createTextNode(f"{subject} @ {PUBLIC_URL}"))
117 updatedElt = atomElt.appendChild(doc.createElement("updated"))
118 updatedElt.appendChild(doc.createTextNode(CURRENT_DATETIME))
119 generatorElt = atomElt.appendChild(doc.createElement("generator"))
120 generatorElt.appendChild(doc.createTextNode("x_status_git"))
121 generatorElt.setAttribute("uri", "https://git.ladys.computer/x_status_git")
122 atomLinks = {}
123 if "OrderedCollectionPage" in ld["@type"]:
124 idElt = atomElt.appendChild(doc.createElement("id"))
125 idElt.appendChild(doc.createTextNode(f"{PUBLIC_URL}/statuses"))
126 atomLinks["alternate"] = f"{PUBLIC_URL}/statuses"
127 atomLinks["current"] = f"{PUBLIC_URL}/statuses.atom"
128 atomLinks["self"] = atomLinks["current"] if ld["@id"] == ld["current"] else f"{ld['@id']}.atom"
129 if "prev" in ld:
130 atomLinks["prev-archive"] = f"{ld['prev']}.atom"
131 if "next" in ld and ld["next"] != ld["current"]:
132 atomLinks["next-archive"] = f"{ld['next']}.atom"
133 else:
134 idElt = atomElt.appendChild(doc.createElement("id"))
135 idElt.appendChild(doc.createTextNode(ld["@id"]))
136 atomLinks["alternate"] = ld["@id"]
137 atomLinks["self"] = f"{ld['@id']}.atom"
138 for (rel, href) in atomLinks.items():
139 linkElt = atomElt.appendChild(doc.createElement("link"))
140 linkElt.setAttribute("rel", rel)
141 linkElt.setAttribute("href", href)
142 for item in ld["items"]:
143 entryElt = atomElt.appendChild(doc.createElement("entry"))
144 title = item["title"] if "title" in item else item["content"][0:27] + "…"
145 titleElt = entryElt.appendChild(doc.createElement("title"))
146 titleElt.appendChild(doc.createTextNode(title))
147 idElt = entryElt.appendChild(doc.createElement("id"))
148 idElt.appendChild(doc.createTextNode(item["@id"]))
149 updatedElt = entryElt.appendChild(doc.createElement("updated"))
150 updatedElt.appendChild(doc.createTextNode(CURRENT_DATETIME))
151 if "created" in item:
152 publishedElt = entryElt.appendChild(doc.createElement("published"))
153 publishedElt.appendChild(doc.createTextNode(item["created"]))
154 authorElt = entryElt.appendChild(doc.createElement("author"))
155 if "author" in item:
156 nameElt = authorElt.appendChild(doc.createElement("name"))
157 nameElt.appendChild(doc.createTextNode(item["author"]["name"]))
158 uriElt = authorElt.appendChild(doc.createElement("uri"))
159 uriElt.appendChild(doc.createTextNode(item["author"]["@id"]))
160 else:
161 nameElt = authorElt.appendChild(doc.createElement("name"))
162 nameElt.appendChild(doc.createTextNode("Anonymous"))
163 contentElt = entryElt.appendChild(doc.createElement("content"))
164 contentElt.setAttribute("type", "xhtml")
165 contentDiv = contentElt.appendChild(doc.createElement("div"))
166 contentDiv.setAttribute("xmlns", XHTML_NAMESPACE)
167 contentDiv.setAttribute("lang", LANG)
168 for child in list(parseString(item["content"]).documentElement.childNodes):
169 contentDiv.appendChild(child)
170 return (atomLinks["self"], atomElt.toxml())
171
172 # Get status paths.
173 status_paths = []
174 for yearpath in Path(f"{BUILD_DIRECTORY}/").glob("[0-9][0-9][0-9][0-9]"):
175 for monthpath in yearpath.glob("[0-9][0-9]"):
176 for daypath in monthpath.glob("[0-9][0-9]"):
177 for statuspath in daypath.glob("*/text"):
178 status_paths.append((None, statuspath))
179 for topicpath in Path(f"{BUILD_DIRECTORY}/").glob("topic/*"):
180 for hash0path in topicpath.glob("[0-9a-f]"):
181 for hash1path in hash0path.glob("[0-9a-f]"):
182 for hash2path in hash1path.glob("[0-9a-f]"):
183 for hash3path in hash2path.glob("[0-9a-f]"):
184 for statuspath in hash3path.glob("*/text"):
185 status_paths.append((topicpath.name, statuspath))
186
187 # Build status objects and listings.
188 for (datetime, identifier, status) in sorted(filter(None, starmap(statusmap, status_paths))):
189 if "subject" in status:
190 topic = status["subject"]
191 if topic not in topics:
192 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 }
193 topics[topic]["items"].append(status)
194 else:
195 yyyy_mm = datetime[0:7]
196 if yyyy_mm not in months:
197 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" }
198 months[yyyy_mm]["items"].append(status)
199
200 # Set up the public directory.
201 if exists(PUBLIC_DIRECTORY):
202 rmtree(PUBLIC_DIRECTORY)
203 mkdir(PUBLIC_DIRECTORY)
204
205 # Copy H·T·M·L files to their expected locations.
206 copy2(f"{BUILD_DIRECTORY}/index.html", f"{PUBLIC_DIRECTORY}/index.html")
207 copy2(f"{BUILD_DIRECTORY}/about.html", f"{PUBLIC_DIRECTORY}/.about.html")
208 copy2(f"{BUILD_DIRECTORY}/status.html", f"{PUBLIC_DIRECTORY}/.status.html")
209 copy2(f"{BUILD_DIRECTORY}/statuses.html", f"{PUBLIC_DIRECTORY}/.statuses.html")
210 copy2(f"{BUILD_DIRECTORY}/topic.html", f"{PUBLIC_DIRECTORY}/.topic.html")
211 copy2(f"{BUILD_DIRECTORY}/topics.html", f"{PUBLIC_DIRECTORY}/.topics.html")
212
213 # Output “about” metadata
214 if not exists(f"{PUBLIC_DIRECTORY}/about"):
215 mkdir(f"{PUBLIC_DIRECTORY}/about")
216 with open(f"{PUBLIC_DIRECTORY}/about/index.jsonld", "w", encoding="utf-8") as f:
217 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)
218
219 # Output month‐based listings and the non‐topic index
220 if not exists(f"{PUBLIC_DIRECTORY}/statuses"):
221 mkdir(f"{PUBLIC_DIRECTORY}/statuses")
222 statuspairs = list(enumerate(months.items()))
223 for (index, (yyyy_mm, ld)) in statuspairs:
224 if not exists(f"{PUBLIC_DIRECTORY}/statuses/{yyyy_mm}"):
225 mkdir(f"{PUBLIC_DIRECTORY}/statuses/{yyyy_mm}")
226 ld["first"] = f"{PUBLIC_URL}/statuses/{statuspairs[0][1][0]}"
227 ld["current"] = f"{PUBLIC_URL}/statuses/{statuspairs[-1][1][0]}"
228 if index > 0:
229 ld["prev"] = f"{PUBLIC_URL}/statuses/{statuspairs[index - 1][1][0]}"
230 if index < len(statuspairs) - 1:
231 ld["next"] = f"{PUBLIC_URL}/statuses/{statuspairs[index + 1][1][0]}"
232 with open(f"{PUBLIC_DIRECTORY}/statuses/{yyyy_mm}/index.jsonld", "w", encoding="utf-8") as f:
233 json.dump(ld, f, ensure_ascii=False, allow_nan=False)
234 atomlink, atomxml = atomForLD(ld)
235 with open(f"{PUBLIC_DIRECTORY}/{atomlink[len(PUBLIC_URL):-5]}/index.atom", "w", encoding="utf-8") as f:
236 f.write(atomxml)
237 with open(f"{PUBLIC_DIRECTORY}/statuses/index.jsonld", "w", encoding="utf-8") as f:
238 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)
239
240 # Output topic‐based listings and the topic index
241 if not exists(f"{PUBLIC_DIRECTORY}/topics"):
242 mkdir(f"{PUBLIC_DIRECTORY}/topics")
243 for (topic, ld) in topics.items():
244 if not exists(f"{PUBLIC_DIRECTORY}/topics/{topic}"):
245 mkdir(f"{PUBLIC_DIRECTORY}/topics/{topic}")
246 with open(f"{PUBLIC_DIRECTORY}/topics/{topic}/index.jsonld", "w", encoding="utf-8") as f:
247 json.dump(ld, f, ensure_ascii=False, allow_nan=False)
248 atomlink, atomxml = atomForLD(ld)
249 with open(f"{PUBLIC_DIRECTORY}/{atomlink[len(PUBLIC_URL):-5]}/index.atom", "w", encoding="utf-8") as f:
250 f.write(atomxml)
251 with open(f"{PUBLIC_DIRECTORY}/topics/index.jsonld", "w", encoding="utf-8") as f:
252 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)
253
254 # Remove the build directory.
255 rmtree(BUILD_DIRECTORY)
This page took 0.084833 seconds and 5 git commands to generate.