Add misc scripts to repoitory
This commit is contained in:
@@ -0,0 +1,40 @@
|
||||
import csv
|
||||
from pathlib import Path
|
||||
|
||||
|
||||
def collect_first_tokens_from_full_table_files(directory: Path) -> list[str]:
|
||||
first_tokens: list[str] = []
|
||||
|
||||
for file_path in directory.iterdir():
|
||||
# Match file names like "Full_Table_List_..." regardless of case.
|
||||
if file_path.is_file() and file_path.name.lower().startswith("full_table_list"):
|
||||
with file_path.open("r", encoding="utf-8", errors="ignore") as f:
|
||||
for line in f:
|
||||
parts = line.strip().split()
|
||||
if parts:
|
||||
first_tokens.append(parts[0])
|
||||
|
||||
return first_tokens
|
||||
|
||||
|
||||
def main() -> None:
|
||||
current_directory = Path.cwd()
|
||||
tokens = collect_first_tokens_from_full_table_files(current_directory)
|
||||
tokens.sort()
|
||||
|
||||
for token in tokens:
|
||||
print(token)
|
||||
|
||||
output_file = current_directory / "full_table_list_tokens.csv"
|
||||
with output_file.open("w", newline="", encoding="utf-8") as csv_file:
|
||||
writer = csv.writer(csv_file)
|
||||
writer.writerow(["value"])
|
||||
for token in tokens:
|
||||
writer.writerow([token])
|
||||
|
||||
print(f"\nTotal extracted values: {len(tokens)}")
|
||||
print(f"Saved CSV file: {output_file}")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
Reference in New Issue
Block a user