From 0e6d5f993c5407f3f6722908c23b3dc2e7eb78e3 Mon Sep 17 00:00:00 2001 From: Sergey Obukhov Date: Mon, 23 Oct 2017 16:24:20 -0700 Subject: [PATCH] fix appointments in text --- setup.py | 2 +- talon/quotations.py | 6 +++--- tests/text_quotations_test.py | 32 ++++++++++++++++++++++++++++++++ 3 files changed, 36 insertions(+), 4 deletions(-) diff --git a/setup.py b/setup.py index 13f83d61..f9305ff6 100755 --- a/setup.py +++ b/setup.py @@ -29,7 +29,7 @@ def finalize_options(self): setup(name='talon', - version='1.4.4', + version='1.4.5', description=("Mailgun library " "to extract message quotations and signatures."), long_description=open("README.rst").read(), diff --git a/talon/quotations.py b/talon/quotations.py index 40e66d4a..996f79dd 100644 --- a/talon/quotations.py +++ b/talon/quotations.py @@ -165,15 +165,15 @@ RE_FROM_COLON_OR_DATE_COLON, # 02.04.2012 14:20 пользователь "bob@example.com" < # bob@xxx.mailgun.org> написал: - re.compile("(\d+/\d+/\d+|\d+\.\d+\.\d+).*@", re.S), + re.compile("(\d+/\d+/\d+|\d+\.\d+\.\d+).*\s\S+@\S+", re.S), # 2014-10-17 11:28 GMT+03:00 Bob < # bob@example.com>: - re.compile("\d{4}-\d{2}-\d{2}\s+\d{2}:\d{2}\s+GMT.*@", re.S), + re.compile("\d{4}-\d{2}-\d{2}\s+\d{2}:\d{2}\s+GMT.*\s\S+@\S+", re.S), # Thu, 26 Jun 2014 14:00:51 +0400 Bob : re.compile('\S{3,10}, \d\d? \S{3,10} 20\d\d,? \d\d?:\d\d(:\d\d)?' '( \S+){3,6}@\S+:'), # Sent from Samsung MobileName wrote: - re.compile('Sent from Samsung .*@.*> wrote'), + re.compile('Sent from Samsung.* \S+@\S+> wrote'), RE_ANDROID_WROTE, RE_POLYMAIL ] diff --git a/tests/text_quotations_test.py b/tests/text_quotations_test.py index 77ef08ca..84be09dd 100644 --- a/tests/text_quotations_test.py +++ b/tests/text_quotations_test.py @@ -119,6 +119,38 @@ def test_pattern_on_date_somebody_sent(): eq_("Test reply", quotations.extract_from_plain(msg_body)) +def test_appointment(): + msg_body = """Response + +10/19/2017 @ 9:30 am for physical therapy +Bla +1517 4th Avenue Ste 300 +London CA 19129, 555-421-6780 + +John Doe, FCLS +Mailgun Inc +555-941-0697 + +From: from@example.com [mailto:from@example.com] +Sent: Wednesday, October 18, 2017 2:05 PM +To: John Doer - SIU +Subject: RE: Claim # 5551188-1 + +Text""" + + expected = """Response + +10/19/2017 @ 9:30 am for physical therapy +Bla +1517 4th Avenue Ste 300 +London CA 19129, 555-421-6780 + +John Doe, FCLS +Mailgun Inc +555-941-0697""" + eq_(expected, quotations.extract_from_plain(msg_body)) + + def test_line_starts_with_on(): msg_body = """Blah-blah-blah On blah-blah-blah"""