Merge pull request #151 from mailgun/sergey/reshape

reshape data as suggested by sklearn

setup.py

@@ -29,7 +29,7 @@ class InstallCommand(install):
 
 
 setup(name='talon',
-      version='1.4.0',
+      version='1.4.3',
       description=("Mailgun library "
                    "to extract message quotations and signatures."),
       long_description=open("README.rst").read(),

talon/html_quotations.py

@@ -94,6 +94,12 @@ def cut_microsoft_quote(html_message):
         #outlook 2007, 2010 (american)
         "//div[@style='border:none;border-top:solid #B5C4DF 1.0pt;"
         "padding:3.0pt 0in 0in 0in']|"
+        #outlook 2013 (international)
+        "//div[@style='border:none;border-top:solid #E1E1E1 1.0pt;"
+        "padding:3.0pt 0cm 0cm 0cm']|"
+        #outlook 2013 (american)
+        "//div[@style='border:none;border-top:solid #E1E1E1 1.0pt;"
+        "padding:3.0pt 0in 0in 0in']|"
         #windows mail
         "//div[@style='padding-top: 5px; "
         "border-top-color: rgb(229, 229, 229); "

talon/quotations.py

@@ -42,6 +42,8 @@ RE_ON_DATE_SMB_WROTE = re.compile(
             u'På',
             # Swedish, Danish
             'Den',
+            # Vietnamese
+            u'Vào',
         )),
         # Date and sender separator
         u'|'.join((
@@ -64,6 +66,8 @@ RE_ON_DATE_SMB_WROTE = re.compile(
             'schrieb',
             # Norwegian, Swedish
             'skrev',
+            # Vietnamese
+            u'đã viết',
         ))
     ))
 # Special case for languages where text is translated like this: 'on {date} wrote {somebody}:'

talon/signature/extraction.py

@@ -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
 
 

tests/text_quotations_test.py

@@ -401,6 +401,14 @@ Op 17-feb.-2015, om 13:18 heeft Julius Caesar <pantheon@rome.com> het volgende g
 Small batch beard laboris tempor, non listicle hella Tumblr heirloom.
 """))
 
+def test_vietnamese_from_block():
+    eq_('Hello', quotations.extract_from_plain(
+    u"""Hello
+
+Vào 14:24 8 tháng 6, 2017, Hùng Nguyễn <hungnguyen@xxx.com> đã viết:
+
+> Xin chào
+"""))
 
 def test_quotation_marker_false_positive():
     msg_body = """Visit us now for assistance...