diff options
| author | Fawn <fangrui_tong@brown.edu> | 2019-07-27 17:19:00 -0400 |
|---|---|---|
| committer | Fawn <fangrui_tong@brown.edu> | 2019-07-27 17:19:00 -0400 |
| commit | f02e8416239b612f40dd2c65d6ca9173e81fb1a9 (patch) | |
| tree | 6541914e8d71f7dd57e66ed7fbc7679820d7594b /src/scraping/buxton/scraper.py | |
| parent | 905d77804bf9a810822e5e43f0cd06019aef7620 (diff) | |
| parent | 606e87ccd9c06c83114d5962382cd76def86f103 (diff) | |
Merge branch 'master' of https://github.com/browngraphicslab/Dash-Web into schema_view_improvements_2
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...") |