esbi-cli 0.2.1__py3-none-any.whl

esbi_cli/capture/legacy.py

@@ -0,0 +1,49 @@
+"""One-time import of an older notes folder: Links/DD-MM-YYYY.md files (one [url] per line) and PDFs/."""
+
+import re
+from dataclasses import dataclass
+from datetime import date, datetime
+from pathlib import Path
+
+from esbi_cli.queue import Queue
+
+_URL = re.compile(r"https?://[^\s\]\)>]+")
+
+
+@dataclass
+class ImportResult:
+    added: int = 0
+    duplicates: int = 0
+
+
+def _file_date(path: Path) -> date | None:
+    """The day a legacy links file stands for, from its DD-MM-YYYY name."""
+    try:
+        return datetime.strptime(path.stem, "%d-%m-%Y").date()
+    except ValueError:
+        return None
+
+
+def _link_file_order(path: Path) -> tuple[int, date | str]:
+    """Chronological by file name; oddly named files go last."""
+    day = _file_date(path)
+    return (0, day) if day else (1, path.stem)
+
+
+def import_legacy(legacy_vault: Path, queue: Queue) -> ImportResult:
+    """Queue every URL and PDF of the legacy vault. Read-only: never touches the old files."""
+    result = ImportResult()
+
+    def add(target: str, **details) -> None:
+        if queue.add(target, origin="legacy", **details):
+            result.added += 1
+        else:
+            result.duplicates += 1
+
+    for links_file in sorted((legacy_vault / "Links").glob("*.md"), key=_link_file_order):
+        saved_on = _file_date(links_file)
+        for url in _URL.findall(links_file.read_text(encoding="utf-8")):
+            add(url, captured=saved_on)  # oldest file first, so the first save date wins
+    for pdf in sorted((legacy_vault / "PDFs").glob("*.pdf")):
+        add(str(pdf.resolve()), label=pdf.stem)
+    return result