chore: init

sew_csvs.py

@@ -0,0 +1,85 @@
+#!/usr/bin/env python3
+"""Sew all frame CSVs in order; deduplicate overlap at boundaries (scroll down)."""
+import csv
+import os
+import re
+import sys
+
+from row_eq import row_equal
+
+
+def frame_number_from_name(name):
+    m = re.match(r"^(\d+)\.csv$", name)
+    return int(m.group(1)) if m else None
+
+
+def list_frame_csvs(frames_dir):
+    names = []
+    for name in os.listdir(frames_dir):
+        n = frame_number_from_name(name)
+        if n is not None:
+            names.append((n, name))
+    return [name for _, name in sorted(names)]
+
+
+def read_csv(path):
+    with open(path, newline="", encoding="utf-8") as f:
+        return list(csv.reader(f))
+
+
+def write_csv(path, rows):
+    with open(path, "w", newline="", encoding="utf-8") as f:
+        csv.writer(f).writerows(rows)
+
+
+def overlap_length(acc_rows, next_rows, num_cols=6):
+    max_k = min(len(acc_rows), len(next_rows))
+    for k in range(max_k, 0, -1):
+        if k > len(acc_rows) or k > len(next_rows):
+            continue
+        match = True
+        for i in range(k):
+            if not row_equal(acc_rows[-k + i], next_rows[i], num_cols):
+                match = False
+                break
+        if match:
+            return k
+    return 0
+
+
+def dedupe_rows(rows, num_cols=6):
+    seen = set()
+    out = []
+    for row in rows:
+        key = tuple((row[i] if i < len(row) else "").strip() for i in range(num_cols))
+        if key in seen:
+            continue
+        seen.add(key)
+        out.append(row)
+    return out
+
+
+def sew(frames_dir, out_path, num_cols=6, dedupe=True):
+    csv_names = list_frame_csvs(frames_dir)
+    if not csv_names:
+        return 0
+    acc = read_csv(os.path.join(frames_dir, csv_names[0]))
+    for name in csv_names[1:]:
+        next_rows = read_csv(os.path.join(frames_dir, name))
+        k = overlap_length(acc, next_rows, num_cols)
+        acc.extend(next_rows[k:])
+    if dedupe:
+        acc = dedupe_rows(acc, num_cols)
+    write_csv(out_path, acc)
+    return len(acc)
+
+
+def main():
+    frames_dir = sys.argv[1] if len(sys.argv) > 1 else "frames"
+    out_path = sys.argv[2] if len(sys.argv) > 2 else "result.csv"
+    n = sew(frames_dir, out_path)
+    print(f"Wrote {n} rows to {out_path}")
+
+
+if __name__ == "__main__":
+    main()