diff options
| author | Sam Wilkins <samwilkins333@gmail.com> | 2019-07-27 16:40:26 -0400 |
|---|---|---|
| committer | Sam Wilkins <samwilkins333@gmail.com> | 2019-07-27 16:40:26 -0400 |
| commit | a75494ad339b656242dab033f86cbeb13dc882b8 (patch) | |
| tree | 060976e1ed80f97f90ad83a4b1821b79046b786d /src/scraping/buxton/scraper.py | |
| parent | d009df0c7142c04947193fb5670974017919bd05 (diff) | |
| parent | 7f8281ca3b6fdbda7dae624bcad307d3ccdcac7b (diff) | |
Merged with master
Diffstat (limited to 'src/scraping/buxton/scraper.py')
| -rw-r--r-- | src/scraping/buxton/scraper.py | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/src/scraping/buxton/scraper.py b/src/scraping/buxton/scraper.py index 48b8fe3fa..182b22a1a 100644 --- a/src/scraping/buxton/scraper.py +++ b/src/scraping/buxton/scraper.py @@ -1,4 +1,5 @@ import os +from shutil import copyfile import docx2txt from docx import Document from docx.opc.constants import RELATIONSHIP_TYPE as RT @@ -233,6 +234,8 @@ def parse_document(file_name: str): for image in os.listdir(dir_path): count += 1 view_guids.append(write_image(pure_name, image)) + copyfile(dir_path + "/" + image, dir_path + + "/" + image.replace(".", "_o.", 1)) os.rename(dir_path + "/" + image, dir_path + "/" + image.replace(".", "_m.", 1)) print(f"extracted {count} images...") |