1 files changed, 52 insertions, 0 deletions
diff --git a/bjoern/videoanalyse/combine_ocr-logs.py b/bjoern/videoanalyse/combine_ocr-logs.py
new file mode 100644
index 0000000..76c59ed
--- /dev/null
+++ b/bjoern/videoanalyse/combine_ocr-logs.py
@@ -0,0 +1,52 @@
+#!/usr/bin/env python3
+
+import argparse
+from pathlib import Path
+from datetime import datetime, timedelta
+import pandas as pd
+import csv
+
+argparser = argparse.ArgumentParser(
+    description="Combines results of OCR analysis with log files"
+)
+argparser.add_argument(
+    "vp_dir", help="Directory containing analysis_results.csv and VPCODE.csv"
+)
+
+args = argparser.parse_args()
+
+vp_path = Path(args.vp_dir)
+
+video_path = next(vp_path.glob("*.mkv"))
+date_format = "%Y-%m-%d %H-%M-%S"
+video_date = datetime.strptime(video_path.stem, date_format)
+print(video_date)
+# video_delta = timedelta(hours=video_date.hour, minutes=video_date.minute, seconds=video_date.second)
+
+def add_video_time_to_start(x, video_date):
+    start = timedelta(seconds=int(x))
+    return (start + video_date).time().isoformat()
+
+analysis = pd.read_csv(vp_path / "analysis_results.csv")
+analysis["Starttime"] = analysis["start_time"].apply(add_video_time_to_start, args=(video_date,))
+
+logs = pd.read_csv(vp_path / f"{vp_path.name}.csv")
+
+def get_log_url(start_time):
+    start_time = datetime.strptime(start_time, "%H:%M:%S")
+
+    for _, row in logs.iterrows():
+        log_start = datetime.strptime(row[0], "%H:%M:%S")
+        log_end = datetime.strptime(row[1], "%H:%M:%S")
+        if start_time >= log_start and start_time <= log_end:
+            return row[3]
+    return 0
+
+
+
+analysis["log_url"] = analysis.apply(
+        lambda row: get_log_url(row.Starttime), axis=1
+    )
+
+
+analysis.to_csv(vp_path / "merged.csv", quoting=csv.QUOTE_NONNUMERIC)