gogs
/
dnd-pdf-to-txt
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

gather pages

master
bel 2023-02-21 12:52:57 -07:00
parent 689998d71f
commit 006a66941f
1 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
main.py
View File

@ -14,6 +14,7 @@ def log(*args):
def de_column_ify(): def de_column_ify():
q = queue.Queue(maxsize=4) q = queue.Queue(maxsize=4)
with pdfplumber.open(config.INPUT) as pdf: with pdfplumber.open(config.INPUT) as pdf:
cropped_pages = []
with ThreadPool(4) as pool: with ThreadPool(4) as pool:
for i in range(len(pdf.pages)): for i in range(len(pdf.pages)):
pool.apply_async(de_columnify_page, (q, config.INPUT, pdf.pages[i], )) pool.apply_async(de_columnify_page, (q, config.INPUT, pdf.pages[i], ))
@ -21,8 +22,12 @@ def de_column_ify():
log("getting", i, "of", len(pdf.pages)) log("getting", i, "of", len(pdf.pages))
got = q.get() got = q.get()
for got_i in got[1]: for got_i in got[1]:
if got_i.chars: cropped_pages.append((got[0], got_i.path))
debug.debug_show(debug.debug_im(got_i.page)) #if got_i.chars:
# debug.debug_show(debug.debug_im(got_i.page))
cropped_pages = sorted(cropped_pages)
for cropped_page in cropped_pages:
print(cropped_page)
def de_columnify_page(q, path, page): def de_columnify_page(q, path, page):
result = cluster.Chars(path, page.chars, page).divide_into_columns() result = cluster.Chars(path, page.chars, page).divide_into_columns()