vault backup: 2025-12-11 10:42:22

wip/extract-provfråga.py

@@ -20,7 +20,7 @@ def find_date(path: Path):
 def parse_frontmatter(text: str):
     m = FRONTMATTER_RE.match(text)
     if not m:
-        return {"tags": [], "date": ""}, text
+        return {"tags": [], "date": "", "maxpoints": ""}, text
     fm_raw = m.group(1)
     rest = text[m.end():]
     tags = []
@@ -53,7 +53,12 @@ def parse_frontmatter(text: str):
     mdate = re.search(r"^date\s*:\s*(.+)$", fm_raw, re.M)
     if mdate:
         date_val = mdate.group(1).strip().strip('"\'')
-    return {"tags": tags, "date": date_val}, rest
+    # parse maxpoints (accept either maxpoints or max_points)
+    maxpoints_val = ""
+    mmax = re.search(r"^(?:max_points|maxpoints)\s*:\s*(.+)$", fm_raw, re.M)
+    if mmax:
+        maxpoints_val = mmax.group(1).strip().strip('"\'')
+    return {"tags": tags, "date": date_val, "maxpoints": maxpoints_val}, rest
 
 def extract_question_answer(body: str):
     # find first fenced block (prefer spoiler)
@@ -75,7 +80,24 @@ def extract_question_answer(body: str):
         question = body.strip()
     return question, answer
 
-def main(root: Path, out: Path):
+
+def _plain_text_from_md(md_text: str) -> str:
+    """Simple cleanup to produce plain text from markdown for CSV/Excel export.
+    Removes fenced code blocks, basic markdown punctuation, list markers, and collapses whitespace.
+    This is intentionally lightweight (not a full markdown->text renderer) but good enough for spreadsheet import.
+    """
+    # remove fenced code blocks
+    txt = re.sub(r"```.*?```", "", md_text, flags=re.S)
+    # remove inline code markers and emphasis/headers
+    txt = re.sub(r"[#*_`]+", "", txt)
+    # remove list markers at start of lines
+    txt = re.sub(r"^\s*-\s+", "", txt, flags=re.M)
+    # collapse whitespace to single spaces
+    txt = re.sub(r"\s+", " ", txt)
+    return txt.strip()
+
+
+def main(root: Path, out: Path, mode: str = "anki"):
     rows = []
     for md in root.rglob("*.md"):
         # process each markdown file
@@ -86,6 +108,7 @@ def main(root: Path, out: Path):
         date = fm.get("date") or find_date(md.parent)
         qnum = md.stem
         tags = fm.get("tags", [])
+        maxpoints = fm.get("maxpoints", "")
         # choose first tag that's not biokemi or provfråga
         category = ""
         for t in tags:
@@ -101,6 +124,17 @@ def main(root: Path, out: Path):
         question_md = question_md.replace("**Rätt svar**", "")
         question_md = question_md.replace("**Svar**", "")
         question_md = question_md.replace("**Answer**", "")
+
+        if mode == "excel":
+            # produce plain-text (no HTML) rows: date;question_number;question;category;maxpoints;question;answer
+            date_val = date or ""
+            q_plain = _plain_text_from_md(question_md)
+            a_plain = _plain_text_from_md(answer_md)
+            # columns: date;question_number;question;category;maxpoints;question;answer
+            # move category to be the third column as requested: date;question_number;category;question;maxpoints;question;answer
+            rows.append((date_val, qnum, category, q_plain, maxpoints, q_plain, a_plain))
+            continue
+
         # Render question and answer markdown to HTML. Enable common extensions.
         question_html = markdown(question_md, extensions=["fenced_code", "tables"])
         answer_html = markdown(answer_md, extensions=["fenced_code", "tables"])
@@ -116,13 +150,18 @@ def main(root: Path, out: Path):
     out.parent.mkdir(parents=True, exist_ok=True)
     with out.open("w", encoding="utf-8", newline="") as f:
         writer = csv.writer(f, delimiter=";", quoting=csv.QUOTE_ALL)
+        # If exporting for Excel, add a header row matching the columns
+        if mode == "excel":
+            writer.writerow(["date", "question_number", "category", "question", "maxpoints", "question", "answer"])
         #writer.writerow(["fråga", "svar", "kategori"])
         for r in rows:
             writer.writerow(r)
 
+
 if __name__ == "__main__":
     ap = argparse.ArgumentParser(description="Extract questions+answers to CSV")
     ap.add_argument("root", nargs="?", default="content", help="root content folder (default: content)")
     ap.add_argument("-o", "--out", default="output.csv", help="output CSV file (default: output.csv)")
+    ap.add_argument("--mode", choices=["anki", "excel"], default="anki", help="output mode: 'anki' (default) or 'excel' (date;question_number;question;category;maxpoints;question;answer)")
     args = ap.parse_args()
-    main(Path(args.root), Path(args.out))
+    main(Path(args.root), Path(args.out), args.mode)