Spaces:
Runtime error
Runtime error
Commit
·
bee8306
1
Parent(s):
14fefb4
Update app.py
Browse files
app.py
CHANGED
|
@@ -24,6 +24,10 @@ def pdf_to_text(path, start_page=1, end_page=None):
|
|
| 24 |
|
| 25 |
if end_page is None:
|
| 26 |
end_page = total_pages
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
|
| 28 |
text_list = []
|
| 29 |
|
|
@@ -36,6 +40,7 @@ def pdf_to_text(path, start_page=1, end_page=None):
|
|
| 36 |
return text_list
|
| 37 |
|
| 38 |
|
|
|
|
| 39 |
def text_to_chunks(texts, word_length=150, start_page=1):
|
| 40 |
text_toks = [t.split(' ') for t in texts]
|
| 41 |
page_nums = []
|
|
|
|
| 24 |
|
| 25 |
if end_page is None:
|
| 26 |
end_page = total_pages
|
| 27 |
+
else:
|
| 28 |
+
end_page = int(end_page)
|
| 29 |
+
|
| 30 |
+
start_page = int(start_page)
|
| 31 |
|
| 32 |
text_list = []
|
| 33 |
|
|
|
|
| 40 |
return text_list
|
| 41 |
|
| 42 |
|
| 43 |
+
|
| 44 |
def text_to_chunks(texts, word_length=150, start_page=1):
|
| 45 |
text_toks = [t.split(' ') for t in texts]
|
| 46 |
page_nums = []
|