1 files changed, 128 insertions, 0 deletions
diff --git a/extract.py b/extract.py
new file mode 100755
index 0000000..368211c
--- /dev/null
+++ b/extract.py
@@ -0,0 +1,128 @@
+#!/usr/bin/env python3
+
+import fnmatch
+import hashlib
+import json
+import os
+import shutil
+import sqlite3
+import sys
+import time
+import urllib.parse
+import argparse
+
+import trac2md
+
+wiki_query = '''
+  SELECT
+    name, 
+    author,
+    version, 
+    time / 1000000 AS time, 
+    text 
+  FROM wiki
+  ORDER BY
+    name, version
+'''
+
+attachment_query = '''
+  SELECT
+    id,
+    filename,
+    size,
+    author,
+    description,
+    ipnr,
+    time / 1000000 AS createdtime
+  FROM
+    attachment
+  WHERE
+    type = 'wiki'
+  ORDER BY
+    filename, time
+'''
+
+def attachment_link(row):
+    h   = lambda whatever: hashlib.sha1(whatever.encode()).hexdigest()
+    h1  = h(row.id)
+    h2  = h(row.filename)
+    fn2 = os.path.splitext(row["filename"])[1]
+    return \
+        os.path.join("attachments", "wiki", h1[:3], h1, h2 + fn2), \
+        os.path.join("pelican", "content", urllib.parse.quote(row.id, ""), row.filename)
+
+class Filter:
+
+    def __init__(self, filename = "filter.json"):
+        with open(filename) as f:
+            filter = json.load(f)
+        if not all(action in "-+" for action, pattern in filter):
+            sys.exit("Bad action \"{}\" in filter".format(action))
+        self.filter = tuple((action == "+", pattern) for action, pattern in filter)
+
+    def __call__(self, name):
+        for action, pattern in self.filter:
+            if fnmatch.fnmatch(name, pattern):
+                return action
+        return True
+
+class Row(sqlite3.Row):
+
+    def __getattr__(self, name):
+        return self[name]
+
+    @property
+    def isotime(self):
+        return time.strftime("%Y-%m-%d %H:%M", time.gmtime(self.time))
+
+def markdown_header(row, first_published):
+    if row.name in first_published:
+        modtime = "Modified: {}\n".format(row.isotime)
+    else:
+        modtime = ""
+        first_published[row.name] = row.isotime
+    return "Title: {}\nAuthor: {}\nDate: {}\n{}\n".format(row.name, row.author, first_published[row.name], modtime)
+
+def main():
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--source-url")
+    args = ap.parse_args()
+
+    for dn in ("wiki", "pelican"):
+        shutil.rmtree(dn)
+
+    for dn in ("wiki", "pelican/content"):
+        os.makedirs(dn)
+
+    os.link("pelicanconf.py", "pelican/pelicanconf.py")
+
+    wiki_to_markdown = trac2md.Trac2Markdown(args.source_url)
+
+    keep = Filter()
+
+    first_published = {}
+
+    db = sqlite3.connect("trac.db")
+    db.row_factory = Row
+
+    for row in db.execute(wiki_query):
+        if keep(row.name):
+            slug = urllib.parse.quote(row.name, "")
+            #print(slug, row.version)
+            with open("wiki/{}.trac".format(slug), "w") as f:
+                f.write(row.text)
+            md = markdown_header(row, first_published) + wiki_to_markdown(row.text, slug)
+            with open("pelican/content/{}.md".format(slug), "w") as f:
+                f.write(md)
+
+    for row in db.execute(attachment_query):
+        src, dst = attachment_link(row)
+        #print("{} => {}".format(dst, src))
+        if not os.path.isdir(os.path.dirname(dst)):
+            os.makedirs(os.path.dirname(dst))
+        os.link(src, dst)
+
+    db.close()
+
+if __name__ == "__main__":
+    main()