Added tool to generate old-new-URLs list

dump_s9y_urls_to_csv.py

@@ -0,0 +1,37 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+import csv
+import sqlite3
+from os.path import basename
+
+JEKYLL_BASE_URL = "https://old.domain.com/"
+JEKYLL_PATH_CUT = 7    # number of characters to strip from beginning of file path, e.g. "jekyll/" = 7
+S9Y_SQLITE_FILE = "serendipity.db"
+S9Y_BASE_URL = "https://new.domain.com/"
+
+print("Reading permalinks from {}".format(S9Y_SQLITE_FILE))
+db = sqlite3.connect(S9Y_SQLITE_FILE)
+req = db.execute("SELECT permalink, entry_id FROM permalinks WHERE type='entry'")
+response = req.fetchall()
+s9y_entries = {}
+for r in response:
+    (url, entry_id) = r
+    s9y_entries[int(entry_id)] = str(url)
+
+print("Reading Jekyll paths, matching and writing jekyll2s9y_urls.csv.")
+with open("jekyll2s9y_urls.csv", "wt") as output:
+    output.write("jekyll_path,jekyll_url,entry_id,s9y_url\n")
+    with open("jekyll_paths.csv", "rt") as f:
+        for row in csv.DictReader(f):
+            jekyll_path = row["jekyll_path"]
+            jekyll_url = JEKYLL_BASE_URL + jekyll_path[JEKYLL_PATH_CUT:].replace("/_posts/", "/").replace(".md", ".html")
+            # Strip date from file path
+            jekyll_basename = basename(jekyll_url)
+            jekyll_date = jekyll_basename[:11]   # "2020-02-20-" = 11 characters (incl. last minus)
+            jekyll_url = jekyll_url.replace(jekyll_date, "")
+            entry_id = int(row["s9y_entry_id"])
+            s9y_url = S9Y_BASE_URL + s9y_entries[entry_id]
+            output.write("\"{}\",\"{}\",{},\"{}\"\n".format(jekyll_path, jekyll_url, entry_id, s9y_url))
+
+print("jekyll2s9y_urls.csv written. All done.")