better prettify implementation + new tests + docs

docs/_build/html/genindex.html

@@ -112,6 +112,7 @@ <h1 id="index">Index</h1>
 <div class="genindex-jumpbox">
  <a href="#C"><strong>C</strong></a>
  | <a href="#I"><strong>I</strong></a>
+ | <a href="#P"><strong>P</strong></a>
  | <a href="#R"><strong>R</strong></a>
  | <a href="#S"><strong>S</strong></a>
  | <a href="#U"><strong>U</strong></a>
@@ -178,6 +179,16 @@ <h2 id="I">I</h2>
   </dl></td>
 </tr></table>
 
+<h2 id="P">P</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%" valign="top"><dl>
+
+  <dt><a href="index.html#string_utils.prettify">prettify() (in module string_utils)</a>
+  </dt>
+
+  </dl></td>
+</tr></table>
+
 <h2 id="R">R</h2>
 <table style="width: 100%" class="indextable genindextable"><tr>
   <td style="width: 33%" valign="top"><dl>

docs/_build/html/index.html

@@ -498,6 +498,35 @@ <h1>Welcome to Python String Utils&#8217;s documentation!<a class="headerlink" h
 </table>
 </dd></dl>
 
+<dl class="function">
+<dt id="string_utils.prettify">
+<code class="descclassname">string_utils.</code><code class="descname">prettify</code><span class="sig-paren">(</span><em>string</em><span class="sig-paren">)</span><a class="headerlink" href="#string_utils.prettify" title="Permalink to this definition">¶</a></dt>
+<dd><p>Turns an ugly text string into a beautiful one by applying a regex pipeline which ensures the following:</p>
+<ul class="simple">
+<li>String cannot start or end with spaces</li>
+<li>String cannot have multiple sequential spaces, empty lines or punctuation (except for &#8221;?&#8221;, &#8221;!&#8221; and &#8221;.&#8221;)</li>
+<li>Arithmetic operators (&#8220;+&#8221;, &#8220;-&#8221;, &#8220;/&#8221;, &#8220;*&#8221;, &#8220;=&#8221;) must have one, and only one space before and after themselves</li>
+<li>The first letter after a dot, an exclamation or a question mark must be uppercase</li>
+<li>One, and only one space should follow a dot, an exclamation or a question mark</li>
+<li>Text inside double quotes cannot start or end with spaces, but one, and only one space must come first and     after quotes (foo&#8221; bar&#8221;baz -&gt; foo &#8220;bar&#8221; baz)</li>
+<li>Text inside round brackets cannot start or end with spaces, but one, and only one space must come first and     after brackets (&#8220;foo(bar )baz&#8221; -&gt; &#8220;foo (bar) baz&#8221;)</li>
+<li>Percentage sign (&#8220;%&#8221;) cannot be preceded by a space if there is a number before (&#8220;100 %&#8221; -&gt; &#8220;100%&#8221;)</li>
+<li>Saxon genitive is correct (&#8220;Dave&#8217; s dog&#8221; -&gt; &#8220;Dave&#8217;s dog&#8221;)</li>
+</ul>
+<table class="docutils field-list" frame="void" rules="none">
+<col class="field-name" />
+<col class="field-body" />
+<tbody valign="top">
+<tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>string</strong> &#8211; String to manipulate</td>
+</tr>
+<tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body">Prettified string.</td>
+</tr>
+<tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body">str</td>
+</tr>
+</tbody>
+</table>
+</dd></dl>
+
 </div>
 <div class="section" id="indices-and-tables">
 <h1>Indices and tables<a class="headerlink" href="#indices-and-tables" title="Permalink to this headline">¶</a></h1>

setup.py

@@ -5,7 +5,7 @@
 
 setup(
     name='python-string-utils',
-    version='0.2.0',
+    version='0.3.0',
     description='Utility functions for strings checking and manipulation.',
     long_description=long_description,
     author='Davide Zanotti',

string_utils.py

@@ -6,7 +6,7 @@
 import random
 
 # module settings
-__version__ = '0.2.0'
+__version__ = '0.3.0'
 __all__ = [
     'is_string',
     'is_url',
@@ -81,11 +81,11 @@
     'RIGHT_SPACE': re.compile(
         r'('
         r'(?<=[^\s\d]),(?=[^\s\d])|\s,\s|\s,(?=[^\s\d])|\s,(?!.)|'  # comma (,)
-        r'(?<=[^\s\d])\.(?=[^\s\d])|\s\.\s|\s\.(?=[^\s\d])|\s\.(?!.)|'  # dot (.)
+        r'(?<=[^\s\d\.])\.+(?=[^\s\d\.])|\s\.+\s|\s\.+(?=[^\s\d])|\s\.+(?!\.)|'  # dot (.)
         r'(?<=\S);(?=\S)|\s;\s|\s;(?=\S)|\s;(?!.)|'  # semicolon (;)
         r'(?<=\S):(?=\S)|\s:\s|\s:(?=\S)|\s:(?!.)|'  # colon (:)
-        r'(?<=\S)!(?=\S)|\s!\s|\s!(?=\S)|\s!(?!.)|'  # exclamation (!)
-        r'(?<=\S)\?(?=\S)|\s\?\s|\s\?(?=\S)|\s\?(?!.)|'  # question (?)
+        r'(?<=[^\s!])!+(?=[^\s!])|\s!+\s|\s!+(?=[^\s!])|\s!+(?!!)|'  # exclamation (!)
+        r'(?<=[^\s\?])\?+(?=[^\s\?])|\s\?+\s|\s\?+(?=[^\s\?])|\s\?+(?!\?)|'  # question (?)
         r'\d%(?=\S)|(?<=\d)\s%\s|(?<=\d)\s%(?=\S)|(?<=\d)\s%(?!.)'  # percentage (%)
         r')',
         re.MULTILINE | re.DOTALL
@@ -471,6 +471,27 @@ def strip_html(string, keep_tag_content=False):
 
 
 def prettify(string):
+    """
+    Turns an ugly text string into a beautiful one by applying a regex pipeline which ensures the following:
+
+    - String cannot start or end with spaces
+    - String cannot have multiple sequential spaces, empty lines or punctuation (except for "?", "!" and ".")
+    - Arithmetic operators ("+", "-", "/", "*", "=") must have one, and only one space before and after themselves
+    - The first letter after a dot, an exclamation or a question mark must be uppercase
+    - One, and only one space should follow a dot, an exclamation or a question mark
+    - Text inside double quotes cannot start or end with spaces, but one, and only one space must come first and \
+    after quotes (foo" bar"baz -> foo "bar" baz)
+    - Text inside round brackets cannot start or end with spaces, but one, and only one space must come first and \
+    after brackets ("foo(bar )baz" -> "foo (bar) baz")
+    - Percentage sign ("%") cannot be preceded by a space if there is a number before ("100 %" -> "100%")
+    - Saxon genitive is correct ("Dave' s dog" -> "Dave's dog")
+
+
+    :param string: String to manipulate
+    :return: Prettified string.
+    :rtype: str
+    """
+
     def remove_duplicates(regex_match):
         return regex_match.group(1)[0]
 

tests.py

@@ -1080,29 +1080,35 @@ def test_should_replace_multiple_percentage_with_single_ones(self):
         self.assertEqual('%', prettify('%%%'))
         self.assertEqual('A % b % c', prettify('a %% b %%%%%% c'))
 
-    def test_should_put_space_after_comma_if_missing(self):
+    def test_should_add_space_after_comma_if_missing(self):
         self.assertEqual('One, two, three', prettify('one,two,three'))
 
+    def test_should_not_add_right_space_after_dot_for_numbers(self):
+        self.assertEqual('12,55', prettify('12,55'))
+
     def test_should_remove_space_before_comma(self):
         self.assertEqual('One, two, three', prettify('one , two , three'))
 
     def test_should_uppercase_first_letter_after_period(self):
         self.assertEqual('Foo. Bar', prettify('Foo. bar'))
 
-    def test_should_put_space_after_period_if_missing(self):
+    def test_should_add_space_after_period_if_missing(self):
         self.assertEqual('One. Two. Three', prettify('one.two.three'))
 
+    def test_should_not_add_right_space_after_comma_for_numbers(self):
+        self.assertEqual('12.55', prettify('12.55'))
+
     def test_should_remove_space_before_period(self):
         self.assertEqual('One. Two. Three', prettify('one . two . three'))
 
-    def test_should_put_space_after_colon_if_missing(self):
+    def test_should_add_space_after_colon_if_missing(self):
         self.assertEqual('Test: this', prettify('Test:this'))
 
     def test_should_remove_space_before_colon(self):
         self.assertEqual('Test: this', prettify('Test :this'))
         self.assertEqual('Test:', prettify('Test :'))
 
-    def test_should_put_space_after_semicolon_if_missing(self):
+    def test_should_add_space_after_semicolon_if_missing(self):
         self.assertEqual('Test; this', prettify('Test;this'))
 
     def test_should_remove_space_before_semicolon(self):
@@ -1112,7 +1118,7 @@ def test_should_remove_space_before_semicolon(self):
     def test_should_uppercase_first_letter_after_exclamation(self):
         self.assertEqual('Foo! Bar', prettify('Foo! bar'))
 
-    def test_should_put_space_after_exclamation_if_missing(self):
+    def test_should_add_space_after_exclamation_if_missing(self):
         self.assertEqual('Test! This', prettify('Test!this'))
 
     def test_should_remove_space_before_exclamation(self):
@@ -1122,7 +1128,7 @@ def test_should_remove_space_before_exclamation(self):
     def test_should_uppercase_first_letter_after_question(self):
         self.assertEqual('Foo? Bar', prettify('Foo? bar'))
 
-    def test_should_put_space_after_question_if_missing(self):
+    def test_should_add_space_after_question_if_missing(self):
         self.assertEqual('Test? This', prettify('Test?this'))
 
     def test_should_remove_space_before_question(self):
@@ -1165,6 +1171,18 @@ def test_should_add_spaces_around_multiplication_if_missing(self):
         self.assertEqual('5 * 2 = 10', prettify('5 *2 = 10'))
         self.assertEqual('5 * 2 = 10', prettify('5*2 = 10'))
 
+    def test_triple_dot_preserved(self):
+        self.assertEqual('Test...', prettify('Test...'))
+        self.assertEqual('Test... This', prettify('Test...This'))
+
+    def test_triple_exclamation_preserved(self):
+        self.assertEqual('Test!!!', prettify('Test!!!'))
+        self.assertEqual('Test!!! This', prettify('Test!!!This'))
+
+    def test_triple_question_preserved(self):
+        self.assertEqual('Test???', prettify('Test???'))
+        self.assertEqual('Test??? This', prettify('Test???This'))
+
     def test_should_prettify_string_as_expected(self):
         original = ' unprettified string ,, like this one,will be"prettified" .it\' s awesome!( like python)) '
         pretty = 'Unprettified string, like this one, will be "prettified". It\'s awesome! (like python)'