diff options
| author | Bob Zeleznik <zzzman@gmail.com> | 2019-07-27 16:18:26 -0400 |
|---|---|---|
| committer | Bob Zeleznik <zzzman@gmail.com> | 2019-07-27 16:18:26 -0400 |
| commit | 7f8281ca3b6fdbda7dae624bcad307d3ccdcac7b (patch) | |
| tree | 6bd96340abb1a4a0e2f66352f3de504572a625be /src/scraping/buxton/scraper.py | |
| parent | 60aa214501f7932e5efee68fd878ef367fc2a31b (diff) | |
| parent | 2daee90fae5c93082434bdf34fc046e0ea871800 (diff) | |
Merge branch 'master' of https://github.com/browngraphicslab/Dash-Web
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...") |