HomeHarvest/homeharvest/cli.py

74 lines
2.0 KiB
Python
Raw Normal View History

2023-09-19 13:01:39 -07:00
import argparse
import datetime
from homeharvest import scrape_property
def main():
parser = argparse.ArgumentParser(description="Home Harvest Property Scraper")
2023-09-19 19:13:20 -07:00
parser.add_argument("location", type=str, help="Location to scrape (e.g., San Francisco, CA)")
2023-09-19 13:43:24 -07:00
2023-09-19 13:01:39 -07:00
parser.add_argument(
2023-09-19 13:43:24 -07:00
"-s",
2023-09-19 13:01:39 -07:00
"--site_name",
type=str,
nargs="*",
default=None,
2023-09-19 13:43:24 -07:00
help="Site name(s) to scrape from (e.g., realtor, zillow)",
2023-09-19 13:01:39 -07:00
)
2023-09-19 13:43:24 -07:00
2023-09-19 13:01:39 -07:00
parser.add_argument(
2023-09-19 13:43:24 -07:00
"-l",
2023-09-19 13:01:39 -07:00
"--listing_type",
type=str,
default="for_sale",
choices=["for_sale", "for_rent", "sold"],
help="Listing type to scrape",
)
2023-09-19 13:43:24 -07:00
2023-09-19 13:01:39 -07:00
parser.add_argument(
2023-09-19 13:43:24 -07:00
"-o",
2023-09-19 13:01:39 -07:00
"--output",
type=str,
default="excel",
choices=["excel", "csv"],
help="Output format",
)
2023-09-19 13:43:24 -07:00
2023-09-19 13:01:39 -07:00
parser.add_argument(
2023-09-19 13:43:24 -07:00
"-f",
2023-09-19 13:01:39 -07:00
"--filename",
type=str,
default=None,
help="Name of the output file (without extension)",
)
2023-09-20 18:24:18 -07:00
parser.add_argument(
"-k",
"--keep_duplicates",
action="store_true",
help="Keep duplicate properties based on address"
)
2023-09-19 19:13:20 -07:00
parser.add_argument("-p", "--proxy", type=str, default=None, help="Proxy to use for scraping")
2023-09-19 13:43:24 -07:00
2023-09-19 13:01:39 -07:00
args = parser.parse_args()
2023-09-19 13:43:24 -07:00
2023-09-20 18:24:18 -07:00
result = scrape_property(args.location, args.site_name, args.listing_type, proxy=args.proxy, keep_duplicates=args.keep_duplicates)
2023-09-19 13:01:39 -07:00
if not args.filename:
timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
args.filename = f"HomeHarvest_{timestamp}"
if args.output == "excel":
output_filename = f"{args.filename}.xlsx"
result.to_excel(output_filename, index=False)
print(f"Excel file saved as {output_filename}")
elif args.output == "csv":
output_filename = f"{args.filename}.csv"
result.to_csv(output_filename, index=False)
print(f"CSV file saved as {output_filename}")
if __name__ == "__main__":
main()