7 Commits

Author SHA1 Message Date
Sergey Obukhov
60637ff13a Merge pull request #152 from mailgun/sergey/v1.4.4
bump version
2017-08-24 16:00:05 -07:00
Sergey Obukhov
df8259e3fe bump version 2017-08-24 15:58:53 -07:00
Sergey Obukhov
aab3b1cc75 Merge pull request #150 from ezrapagel/fix_greedy_dash_regex
android_wrote regex incorrectly matching
2017-08-24 15:52:29 -07:00
Sergey Obukhov
9492b39f2d Merge branch 'master' into fix_greedy_dash_regex 2017-08-24 15:39:28 -07:00
Sergey Obukhov
b9ac866ea7 Merge pull request #151 from mailgun/sergey/reshape
reshape data as suggested by sklearn
2017-08-24 12:04:58 -07:00
Sergey Obukhov
678517dd89 reshape data as suggested by sklearn 2017-08-24 12:03:47 -07:00
Ezra Pagel
221774c6f8 android_wrote regex was incorrectly iterating characters in 'wrote', resulting in greedy regex that
matched many strings with dashes
2017-08-21 12:47:06 -05:00
4 changed files with 16 additions and 3 deletions

View File

@@ -29,7 +29,7 @@ class InstallCommand(install):
setup(name='talon',
version='1.4.2',
version='1.4.4',
description=("Mailgun library "
"to extract message quotations and signatures."),
long_description=open("README.rst").read(),

View File

@@ -147,7 +147,7 @@ RE_FROM_COLON_OR_DATE_COLON = re.compile(u'(_+\r?\n)?[\s]*(:?[*]?{})[\s]?:[*]?.*
RE_ANDROID_WROTE = re.compile(u'[\s]*[-]+.*({})[ ]*[-]+'.format(
u'|'.join((
# English
'wrote'
'wrote',
))), re.I)
# Support polymail.io reply format

View File

@@ -32,7 +32,7 @@ RE_REVERSE_SIGNATURE = re.compile(r'''
def is_signature_line(line, sender, classifier):
'''Checks if the line belongs to signature. Returns True or False.'''
data = numpy.array(build_pattern(line, features(sender)))
data = numpy.array(build_pattern(line, features(sender))).reshape(1, -1)
return classifier.predict(data) > 0

View File

@@ -778,3 +778,16 @@ def test_split_email():
expected_markers = "stttttsttttetesetesmmmmmmssmmmmmmsmmmmmmmm"
markers = quotations.split_emails(msg)
eq_(markers, expected_markers)
def test_feedback_below_left_unparsed():
msg_body = """Please enter your feedback below. Thank you.
------------------------------------- Enter Feedback Below -------------------------------------
The user experience was unparallelled. Please continue production. I'm sending payment to ensure
that this line is intact."""
parsed = quotations.extract_from_plain(msg_body)
eq_(msg_body, parsed.decode('utf8'))