diff options
| author | yipstanley <stanley_yip@brown.edu> | 2019-07-27 20:12:10 -0400 |
|---|---|---|
| committer | yipstanley <stanley_yip@brown.edu> | 2019-07-27 20:12:10 -0400 |
| commit | f62c275090a0b227a5f77ed14e0a6c7d19d20052 (patch) | |
| tree | 6e72aacb0ca2c2527c6dfc6e19900030eed57111 /src/scraping/buxton/scraper.py | |
| parent | d53ad748d90ca1c863a7ef52d0835573ce967a54 (diff) | |
| parent | 5cbbd71ceb98a554109ac2a4469a02b28a836e61 (diff) | |
Merge branch 'master' of https://github.com/browngraphicslab/Dash-Web into pdf_paste_backlink
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...") |