Spaces:
Sleeping
Sleeping
abnerguzman
commited on
Commit
•
7b11578
1
Parent(s):
1f71a40
Update demo.py
Browse files
demo.py
CHANGED
@@ -480,18 +480,20 @@ def format_chunk_texts_for_display2(url, cid_ctext_tuples):
|
|
480 |
print(f"cid={cid_ctext_tuple[0]}:")
|
481 |
print(f"```{ctext}```")
|
482 |
|
483 |
-
needs_ellipsis = False
|
484 |
if j < n_chunks-1 and ids_l[j] != ids_l[j+1]:
|
485 |
needs_ellipsis = True
|
486 |
if len(ctext) > 512:
|
487 |
-
|
488 |
-
|
489 |
for pidx, quote_pattern in enumerate(quote_pattern_l):
|
490 |
match = re.search(quote_pattern, ctext)
|
491 |
if match:
|
492 |
print(f"pidx={pidx} found match: {match}")
|
493 |
first_quote_idx = match.span()[0]
|
494 |
break
|
|
|
|
|
495 |
|
496 |
if first_quote_idx >= 0:
|
497 |
ctext = ctext[first_quote_idx:]
|
|
|
480 |
print(f"cid={cid_ctext_tuple[0]}:")
|
481 |
print(f"```{ctext}```")
|
482 |
|
483 |
+
first_quote_idx, needs_ellipsis = -1, False
|
484 |
if j < n_chunks-1 and ids_l[j] != ids_l[j+1]:
|
485 |
needs_ellipsis = True
|
486 |
if len(ctext) > 512:
|
487 |
+
_first_quote_idx = ctext.find(quote_start) # TODO: remove with better set of regex
|
488 |
+
print(f"_first_quote_idx={_first_quote_idx}")
|
489 |
for pidx, quote_pattern in enumerate(quote_pattern_l):
|
490 |
match = re.search(quote_pattern, ctext)
|
491 |
if match:
|
492 |
print(f"pidx={pidx} found match: {match}")
|
493 |
first_quote_idx = match.span()[0]
|
494 |
break
|
495 |
+
first_quote_idx = min(first_quote_idx, _first_quote_idx)
|
496 |
+
print(f"first_quote_idx={first_quote_idx}")
|
497 |
|
498 |
if first_quote_idx >= 0:
|
499 |
ctext = ctext[first_quote_idx:]
|