Remove pandas dependency from Bluebeam BCI generator

app/bluebeam_bci.py

@@ -2,14 +2,14 @@
 
 from pathlib import Path
 import argparse
-import pandas as pd
+import csv
 
 BASE_DIR = Path(__file__).resolve().parents[1]
 DATA_DIR = BASE_DIR / "data"
 OUTPUT_DIR = BASE_DIR / "output"
 
 def clean_value(v):
-    if pd.isna(v):
+    if v is None:
         return ""
     return str(v).strip()
 
@@ -20,89 +20,80 @@ def bci_escape(v):
     v = v.replace("\n", "\\n")
     return v
 
-def build_column_dict(row, exclude_cols):
-    data = {}
-    for col in row.index:
-        if col in exclude_cols:
-            continue
-        val = clean_value(row[col])
-        if val != "":
-            data[col] = val
-    return data
+def read_csv(path):
+    with open(path, newline="", encoding="utf-8-sig") as f:
+        return list(csv.DictReader(f))
 
 def dict_to_bci_payload(d):
     parts = []
     for k, v in d.items():
+        if clean_value(v) != "":
             parts.append(f"'{bci_escape(k)}':'{bci_escape(v)}'")
     return "{" + ",".join(parts) + "}"
 
 def main():
     parser = argparse.ArgumentParser(description="Generate Bluebeam BCI script from Markups CSV + update CSV.")
-    parser.add_argument("--bb-csv", default=str(DATA_DIR / "bluebeam_markups.csv"), help="Bluebeam Markups List CSV export")
-    parser.add_argument("--updates-csv", default=str(DATA_DIR / "bluebeam_updates.csv"), help="Your update CSV")
-    parser.add_argument("--out", default=str(OUTPUT_DIR / "update_columns.bci"), help="Output BCI file")
-    parser.add_argument("--pdf-path", default=r"C:\PATH\TO\TARGET.pdf", help="Windows path to target PDF for the BCI Open() command")
-    parser.add_argument("--match-column", default="Comment", help="Column in Bluebeam CSV to match against match_value")
-    parser.add_argument("--page-column", default="Page Index", help="Bluebeam page index column")
-    parser.add_argument("--id-column", default="ID", help="Bluebeam markup ID column")
-    parser.add_argument("--contains", action="store_true", help="Use contains match instead of exact match")
+    parser.add_argument("--bb-csv", default=str(DATA_DIR / "bluebeam_markups.csv"))
+    parser.add_argument("--updates-csv", default=str(DATA_DIR / "bluebeam_updates.csv"))
+    parser.add_argument("--out", default=str(OUTPUT_DIR / "update_columns.bci"))
+    parser.add_argument("--pdf-path", default=r"C:\PATH\TO\TARGET.pdf")
+    parser.add_argument("--match-column", default="Comment")
+    parser.add_argument("--page-column", default="Page Index")
+    parser.add_argument("--id-column", default="ID")
+    parser.add_argument("--contains", action="store_true")
     args = parser.parse_args()
 
-    bb_csv = Path(args.bb_csv)
-    updates_csv = Path(args.updates_csv)
-    out_path = Path(args.out)
-
     OUTPUT_DIR.mkdir(exist_ok=True)
 
-    bb = pd.read_csv(bb_csv, dtype=str).fillna("")
-    upd = pd.read_csv(updates_csv, dtype=str).fillna("")
+    bb_rows = read_csv(args.bb_csv)
+    update_rows = read_csv(args.updates_csv)
+
+    if not bb_rows:
+        raise ValueError("Bluebeam CSV is empty.")
+    if not update_rows:
+        raise ValueError("Updates CSV is empty.")
 
     required_bb = {args.page_column, args.id_column, args.match_column}
-    missing_bb = required_bb - set(bb.columns)
+    missing_bb = required_bb - set(bb_rows[0].keys())
     if missing_bb:
         raise ValueError(f"Bluebeam CSV missing required columns: {sorted(missing_bb)}")
 
-    required_upd = {"match_value"}
-    missing_upd = required_upd - set(upd.columns)
-    if missing_upd:
-        raise ValueError(f"Updates CSV missing required columns: {sorted(missing_upd)}")
-
-    lines = []
-    lines.append(f"Open('{args.pdf_path}', '')")
+    if "match_value" not in update_rows[0]:
+        raise ValueError("Updates CSV missing required column: match_value")
 
+    lines = [f"Open('{args.pdf_path}', '')"]
     total_matches = 0
 
-    for _, urow in upd.iterrows():
-        match_value = clean_value(urow["match_value"])
+    for urow in update_rows:
+        match_value = clean_value(urow.get("match_value"))
         if not match_value:
             continue
 
-        if args.contains:
-            mask = bb[args.match_column].astype(str).str.contains(match_value, na=False, regex=False)
-        else:
-            mask = bb[args.match_column].astype(str).str.strip() == match_value
-
-        matches = bb[mask]
-
-        col_data = build_column_dict(urow, exclude_cols={"match_value"})
+        col_data = {k: v for k, v in urow.items() if k != "match_value" and clean_value(v) != ""}
         if not col_data:
             continue
 
         payload = dict_to_bci_payload(col_data)
 
-        for _, brow in matches.iterrows():
-            page_index = clean_value(brow[args.page_column])
-            markup_id = clean_value(brow[args.id_column])
+        for brow in bb_rows:
+            target = clean_value(brow.get(args.match_column))
 
-            if page_index == "" or markup_id == "":
+            matched = match_value in target if args.contains else target == match_value
+            if not matched:
+                continue
+
+            page_index = clean_value(brow.get(args.page_column))
+            markup_id = clean_value(brow.get(args.id_column))
+
+            if not page_index or not markup_id:
                 continue
 
             lines.append(f'ColumnDataSet({page_index},"{markup_id}","{payload}")')
             total_matches += 1
 
-    lines.append("Save()")
-    lines.append("Close()")
+    lines += ["Save()", "Close()"]
 
+    out_path = Path(args.out)
     out_path.write_text("\n".join(lines) + "\n", encoding="utf-8")
 
     print(f"WROTE: {out_path}")