pdf-annotation-merge/app/annotate.py

118 lines
3.0 KiB
Python

#!/usr/bin/env python3
from pathlib import Path
import argparse
import pandas as pd
import fitz # PyMuPDF
BASE_DIR = Path(__file__).resolve().parents[1]
INPUT_DIR = BASE_DIR / "input"
OUTPUT_DIR = BASE_DIR / "output"
DATA_DIR = BASE_DIR / "data"
def add_textbox(page, row):
"""
Adds a Bluebeam-friendly note:
- Square annotation: white fill, blue border
- FreeText annotation: red text
"""
x = float(row["x"])
y = float(row["y"])
w = float(row["w"])
h = float(row["h"])
rect = fitz.Rect(x, y, x + w, y + h)
text = str(row["note_text"]).replace("\\n", "\n")
border_width = float(row.get("border_width", 1.5) or 1.5)
box = page.add_rect_annot(rect)
box.set_colors(
stroke=(0, 0, 1), # blue border
fill=(1, 1, 1), # white fill
)
box.set_border(width=border_width)
box.set_info({
"title": str(row.get("author", "PDF Annotation Merge")),
"subject": str(row.get("category", "Drafter Note")) + " Box",
"content": text,
})
box.update(opacity=1)
inset = 6
text_rect = fitz.Rect(x + inset, y + inset, x + w - inset, y + h - inset)
note = page.add_freetext_annot(
text_rect,
text,
fontsize=float(row.get("font_size", 10) or 10),
fontname="helv",
text_color=(1, 0, 0), # red text
fill_color=None,
align=fitz.TEXT_ALIGN_LEFT,
)
note.set_info({
"title": str(row.get("author", "PDF Annotation Merge")),
"subject": str(row.get("category", "Drafter Note")),
"content": text,
})
note.update()
def generate_annotations(csv_path):
df = pd.read_csv(csv_path)
OUTPUT_DIR.mkdir(exist_ok=True)
required = {"file", "page", "x", "y", "w", "h", "note_text"}
missing = required - set(df.columns)
if missing:
raise ValueError(f"CSV is missing required columns: {sorted(missing)}")
for filename, group in df.groupby("file"):
input_pdf = INPUT_DIR / filename
output_pdf = OUTPUT_DIR / filename.replace(".pdf", "_annotated.pdf")
if not input_pdf.exists():
print(f"SKIP: missing input file: {input_pdf}")
continue
doc = fitz.open(input_pdf)
for _, row in group.iterrows():
page_number = int(row["page"]) - 1
if page_number < 0 or page_number >= len(doc):
print(f"SKIP: invalid page {row['page']} for {filename}")
continue
page = doc[page_number]
add_textbox(page, row)
doc.save(output_pdf, garbage=4, deflate=True)
doc.close()
print(f"WROTE: {output_pdf}")
def main():
parser = argparse.ArgumentParser(description="Generate PDF annotations from CSV data.")
parser.add_argument(
"-c",
"--csv",
default=str(DATA_DIR / "annotations.csv"),
help="Path to annotations CSV file. Default: data/annotations.csv",
)
args = parser.parse_args()
generate_annotations(Path(args.csv))
if __name__ == "__main__":
main()