Initial commit: FCCS extraction tool

fccs_reorganize.py

@@ -0,0 +1,142 @@
+"""
+Step 3: Reorganize flat exported files into a proper folder structure.
+
+Parses FCCS export filenames using known folder names and date anchors
+to reconstruct the original drawer/folder hierarchy.
+
+Filename format:
+  {drawer_id}_{client_name}_{folder_name}_{MM-DD-YYYY}_{document_name}.ext
+  Example: 01069_SMITH, BOB_2007 Tax Documents_02-10-2009_2007 Form 1040A Page 1.pdf
+
+Output structure:
+  output/{drawer_id}_{client_name}/{folder_name}/{original_filename}
+
+Files that cannot be parsed go to output/_unparsed/ for manual review.
+"""
+
+import os
+import re
+import shutil
+import sys
+
+from fccs_config import parse_args, load_config, make_logger, load_folder_list
+
+
+def parse_filename(filename, known_folders):
+    """
+    Parse an FCCS export filename into (drawer_id, client_name, folder_name,
+    date, doc_name_with_ext) or return None if it cannot be parsed.
+
+    Strategy: anchor on drawer_id (left), date MM-DD-YYYY (middle), and match
+    a known folder name between client_name and date.
+    """
+    stem, ext = os.path.splitext(filename)
+
+    # Drawer ID is always the first underscore-delimited token
+    sep = stem.find("_")
+    if sep == -1:
+        return None
+    drawer_id = stem[:sep]
+    rest = stem[sep + 1:]
+
+    # Find all date-pattern occurrences (MM-DD-YYYY) in rest
+    # We search with a leading underscore context to ensure proper boundaries
+    date_pattern = re.compile(r"_(\d{2}-\d{2}-\d{4})_")
+    # Prepend underscore so the first potential date at position 0 of rest is caught
+    search_str = "_" + rest
+    matches = list(date_pattern.finditer(search_str))
+
+    if not matches:
+        return None
+
+    # Try each date match; known_folders is already sorted longest-first
+    for m in matches:
+        date_str = m.group(1)
+        # Position in `rest` where date starts (adjust for prepended _)
+        date_start_in_rest = m.start() - 1  # -1 for the prepended _
+        # But the match includes the leading _, so the actual content before date:
+        before_date = rest[:date_start_in_rest]
+        after_date = rest[date_start_in_rest + len(date_str) + 1:]  # +1 for trailing _
+
+        # Try to match a known folder at the end of before_date
+        for folder in known_folders:
+            if before_date.endswith(folder):
+                # Check there's an underscore separator before the folder name
+                prefix_end = len(before_date) - len(folder)
+                if prefix_end > 0 and before_date[prefix_end - 1] == "_":
+                    client_name = before_date[: prefix_end - 1]
+                    doc_name = after_date + ext
+                    return (drawer_id, client_name, folder, date_str, doc_name)
+
+    return None
+
+
+def main():
+    args = parse_args()
+    cfg = load_config(args.config)
+    log = make_logger(cfg.get("paths", "log_file"))
+
+    export_dir = cfg.get("paths", "export_dir")
+    output_dir = cfg.get("paths", "output_dir")
+    folder_list_path = cfg.get("paths", "folder_list")
+
+    if not os.path.isdir(export_dir):
+        log(f"ERROR: export directory not found: {export_dir}")
+        sys.exit(1)
+
+    known_folders = load_folder_list(folder_list_path)
+    if not known_folders:
+        log("WARNING: folder list is empty. All files will go to _unparsed/.")
+    # Sort longest-first to prevent partial matches
+    known_folders.sort(key=len, reverse=True)
+
+    files = [
+        f for f in os.listdir(export_dir)
+        if os.path.isfile(os.path.join(export_dir, f))
+    ]
+
+    log(f"Found {len(files)} files to reorganize")
+
+    success = 0
+    failed = 0
+
+    for filename in files:
+        result = parse_filename(filename, known_folders)
+
+        if result is None:
+            log(f"  UNPARSED: {filename}")
+            dest_dir = os.path.join(output_dir, "_unparsed")
+            failed += 1
+        else:
+            drawer_id, client_name, folder_name, date, doc_name = result
+            dest_dir = os.path.join(
+                output_dir,
+                f"{drawer_id}_{client_name}",
+                folder_name,
+            )
+            success += 1
+
+        os.makedirs(dest_dir, exist_ok=True)
+        src = os.path.join(export_dir, filename)
+        dst = os.path.join(dest_dir, filename)
+
+        # Handle duplicate filenames
+        if os.path.exists(dst):
+            base, fext = os.path.splitext(filename)
+            counter = 1
+            while os.path.exists(dst):
+                dst = os.path.join(dest_dir, f"{base}_{counter}{fext}")
+                counter += 1
+            log(f"  DUPLICATE renamed: {filename} -> {os.path.basename(dst)}")
+
+        shutil.copy2(src, dst)
+
+    log("=" * 60)
+    log(f"Reorganization complete: {success} organized, {failed} unparsed")
+    if failed:
+        log(f"Review unparsed files in: {os.path.join(output_dir, '_unparsed')}")
+    log("=" * 60)
+
+
+if __name__ == "__main__":
+    main()