7 7
from urllib.parse import parse_qsl, quote, unquote, urlencode, urlsplit, urlunsplit
8 8 9 9
from django.utils.encoding import punycode
10 -
from django.utils.functional import Promise, keep_lazy, keep_lazy_text
10 +
from django.utils.functional import Promise, cached_property, keep_lazy, keep_lazy_text
11 11
from django.utils.http import RFC3986_GENDELIMS, RFC3986_SUBDELIMS
12 12
from django.utils.regex_helper import _lazy_re_compile
13 13
from django.utils.safestring import SafeData, SafeString, mark_safe
@@ -225,6 +225,16 @@ def unquote_quote(segment):
225 225
return urlunsplit((scheme, netloc, path, query, fragment))
226 226 227 227 228 +
class CountsDict(dict):
229 +
def __init__(self, *args, word, **kwargs):
230 +
super().__init__(*args, *kwargs)
231 +
self.word = word
232 + 233 +
def __missing__(self, key):
234 +
self[key] = self.word.count(key)
235 +
return self[key]
236 + 237 + 228 238
class Urlizer:
229 239
"""
230 240
Convert any URLs in text into clickable links.
@@ -330,40 +340,72 @@ def trim_url(self, x, *, limit):
330 340
return x
331 341
return "%s…" % x[: max(0, limit - 1)]
332 342 343 +
@cached_property
344 +
def wrapping_punctuation_openings(self):
345 +
return "".join(dict(self.wrapping_punctuation).keys())
346 + 347 +
@cached_property
348 +
def trailing_punctuation_chars_no_semicolon(self):
349 +
return self.trailing_punctuation_chars.replace(";", "")
350 + 351 +
@cached_property
352 +
def trailing_punctuation_chars_has_semicolon(self):
353 +
return ";" in self.trailing_punctuation_chars
354 + 333 355
def trim_punctuation(self, word):
334 356
"""
335 357
Trim trailing and wrapping punctuation from `word`. Return the items of
336 358
the new state.
337 359
"""
338 -
lead, middle, trail = "", word, ""
360 +
# Strip all opening wrapping punctuation.
361 +
middle = word.lstrip(self.wrapping_punctuation_openings)
362 +
lead = word[: len(word) - len(middle)]
363 +
trail = ""
364 + 339 365
# Continue trimming until middle remains unchanged.
340 366
trimmed_something = True
341 -
while trimmed_something:
367 +
counts = CountsDict(word=middle)
368 +
while trimmed_something and middle:
342 369
trimmed_something = False
343 370
# Trim wrapping punctuation.
344 371
for opening, closing in self.wrapping_punctuation:
345 -
if middle.startswith(opening):
346 -
middle = middle[len(opening) :]
347 -
lead += opening
348 -
trimmed_something = True
349 -
# Keep parentheses at the end only if they're balanced.
350 -
if (
351 -
middle.endswith(closing)
352 -
and middle.count(closing) == middle.count(opening) + 1
353 -
):
354 -
middle = middle[: -len(closing)]
355 -
trail = closing + trail
356 -
trimmed_something = True
357 -
# Trim trailing punctuation (after trimming wrapping punctuation,
358 -
# as encoded entities contain ';'). Unescape entities to avoid
359 -
# breaking them by removing ';'.
360 -
middle_unescaped = html.unescape(middle)
361 -
stripped = middle_unescaped.rstrip(self.trailing_punctuation_chars)
362 -
if middle_unescaped != stripped:
363 -
punctuation_count = len(middle_unescaped) - len(stripped)
364 -
trail = middle[-punctuation_count:] + trail
365 -
middle = middle[:-punctuation_count]
372 +
if counts[opening] < counts[closing]:
373 +
rstripped = middle.rstrip(closing)
374 +
if rstripped != middle:
375 +
strip = counts[closing] - counts[opening]
376 +
trail = middle[-strip:]
377 +
middle = middle[:-strip]
378 +
trimmed_something = True
379 +
counts[closing] -= strip
380 + 381 +
rstripped = middle.rstrip(self.trailing_punctuation_chars_no_semicolon)
382 +
if rstripped != middle:
383 +
trail = middle[len(rstripped) :] + trail
384 +
middle = rstripped
366 385
trimmed_something = True
386 + 387 +
if self.trailing_punctuation_chars_has_semicolon and middle.endswith(";"):
388 +
# Only strip if not part of an HTML entity.
389 +
amp = middle.rfind("&")
390 +
if amp == -1:
391 +
can_strip = True
392 +
else:
393 +
potential_entity = middle[amp:]
394 +
escaped = html.unescape(potential_entity)
395 +
can_strip = (escaped == potential_entity) or escaped.endswith(";")
396 + 397 +
if can_strip:
398 +
rstripped = middle.rstrip(";")
399 +
amount_stripped = len(middle) - len(rstripped)
400 +
if amp > -1 and amount_stripped > 1:
401 +
# Leave a trailing semicolon as might be an entity.
402 +
trail = middle[len(rstripped) + 1 :] + trail
403 +
middle = rstripped + ";"
404 +
else:
405 +
trail = middle[len(rstripped) :] + trail
406 +
middle = rstripped
407 +
trimmed_something = True
408 + 367 409
return lead, middle, trail
368 410 369 411
@staticmethod
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4