Merge pull request #286 from byexamples/Issue-275-Ignore-empty-space-at-begin

Issue 275 ignore empty space at begin

Author: eldipa

.gitignore

@@ -18,3 +18,4 @@ prof-traces
 test/ds/good.args
 test/autocomplete_byexample.sh
 .check-secrets-impl.sh
+build/

byexample/example.py

@@ -82,6 +82,7 @@ class Example(object):
 
     >>> example.options
     {'capture': True,
+     'ignore_first_empty_lines': True,
      'input_prefix_range': (6, 12),
      'norm_ws': False,
      'rm': [],

byexample/expected.py

@@ -32,7 +32,7 @@ class _LinearExpected(Expected):
         >>> from byexample.options import Options
         >>> from byexample.finder import _build_fake_example as build_example
 
-        >>> opts = {'norm_ws': False, 'tags': True, 'capture': True, 'rm': [], 'type': False, 'input_prefix_range': (6,12)}
+        >>> opts = {'norm_ws': False, 'tags': True, 'capture': True, 'rm': [], 'type': False, 'input_prefix_range': (6,12), 'ignore_first_empty_lines': True}
 
         Consider the following example with a named capture in the expected:
 
@@ -145,7 +145,7 @@ class _LinearExpected(Expected):
 
         (See byexample.parser docs)
 
-        >>> opts = {'norm_ws': True, 'tags': True, 'capture': True, 'rm': [], 'type': False, 'input_prefix_range': (6, 12)}
+        >>> opts = {'norm_ws': True, 'tags': True, 'capture': True, 'rm': [], 'type': False, 'input_prefix_range': (6, 12), 'ignore_first_empty_lines': True}
         >>> ex = build_example('f()', '\n  <a>A \n\nB <bc> C\n<c>', opts=opts)
         >>> exp = ex.expected
 

byexample/finder.py

@@ -51,7 +51,8 @@ class F:
                 'capture': True,
                 'rm': [],
                 'type': False,
-                'input_prefix_range': (6, 12)
+                'input_prefix_range': (6, 12),
+                'ignore_first_empty_lines': True,
             }
         )
     parser.extract_options = lambda x: opts

byexample/init.py

@@ -386,6 +386,12 @@ def get_default_options_parser(cmdline_args):
     options_parser.add_flag(
         "norm-ws", default=False, help="ignore the amount of whitespaces."
     )
+    options_parser.add_flag(
+        "ignore-first-empty-lines",
+        default=True,
+        help=
+        "ignore any empty or whitespace-only lines at the begin of the got string."
+    )
     options_parser.add_flag(
         "pass",
         default=False,

byexample/parser.py

@@ -193,7 +193,8 @@ def parse(self, example, concerns):
         input_prefix_len_range = options['input_prefix_range']
         expected_regexs, charnos, rcounts, tags_by_idx, input_list = self.expected_as_regexs(
             example.expected_str, options['tags'], options['capture'],
-            options['type'], options['norm_ws'], input_prefix_len_range
+            options['type'], options['norm_ws'], input_prefix_len_range,
+            options['ignore_first_empty_lines']
         )
 
         ExpectedClass = _LinearExpected
@@ -230,8 +231,14 @@ def parse(self, example, concerns):
 
     @profile
     def expected_as_regexs(
-        self, expected, tags_enabled, capture_enabled, input_enabled,
-        normalize_whitespace, input_prefix_len_range
+        self,
+        expected,
+        tags_enabled,
+        capture_enabled,
+        input_enabled,
+        normalize_whitespace,
+        input_prefix_len_range,
+        ignore_first_empty_lines=True
     ):
         r'''
         From the expected string create a list of regular expressions that
@@ -262,8 +269,8 @@ def expected_as_regexs(
 
         We return the regexs
 
-            >>> regexs
-            ('\\A', 'a', '(?P<foo>.*?)', 'b', '(?P<bar>.*?)', 'c', '\\n*\\Z')
+            >>> regexs          # byexample: +norm-ws
+            ('\\A(?:[ \\t]*\\n)*?', 'a', '(?P<foo>.*?)', 'b', '(?P<bar>.*?)', 'c', '\\n*\\Z')
 
             >>> m = re.compile(''.join(regexs), re.MULTILINE | re.DOTALL)
             >>> m.match('axxbyyyc').groups()
@@ -300,7 +307,7 @@ def expected_as_regexs(
             >>> regexs, _, _, tags_by_idx, _ = _as_regexs(expected, normalize_whitespace=True)
 
             >>> regexs          # byexample: +norm-ws
-            ('\\A', 'a', '(?:.*?)(?<!\\s)', '\\s+(?!\\s)', '(?P<foo_bar>.*?)', 'c', '\\s*\\Z')
+            ('\\A\\s*?', 'a', '(?:.*?)(?<!\\s)', '\\s+(?!\\s)', '(?P<foo_bar>.*?)', 'c', '\\s*\\Z')
 
             >>> tags_by_idx
             {2: None, 4: 'foo-bar'}
@@ -315,7 +322,7 @@ def expected_as_regexs(
             >>> regexs, _, _, tags_by_idx, _ = _as_regexs(expected)
 
             >>> regexs
-            ('\\A', 'a<foo>b<bar>c', '\\n*\\Z')
+            ('\\A(?:[ \\t]*\\n)*?', 'a<foo>b<bar>c', '\\n*\\Z')
 
             >>> tags_by_idx
             {}
@@ -324,11 +331,24 @@ def expected_as_regexs(
             >>> regexs, _, _, tags_by_idx, _ = _as_regexs(expected)
 
             >>> regexs
-            ('\\A', 'a', '(?:.*?)', 'b<bar>c', '\\n*\\Z')
+            ('\\A(?:[ \\t]*\\n)*?', 'a', '(?:.*?)', 'b<bar>c', '\\n*\\Z')
 
             >>> tags_by_idx
             {2: None}
 
+        When ignore_first_empty_lines is False the begin anchor is a plain \\A,
+        so the got string must start exactly at the first expected character.
+
+            >>> regexs, _, _, _, _ = _as_regexs('foo', ignore_first_empty_lines=False)
+
+            >>> regexs
+            ('\\A', 'foo', '\\n*\\Z')
+
+            >>> regexs, _, _, _, _ = _as_regexs('foo', normalize_whitespace=True, ignore_first_empty_lines=False)
+
+            >>> regexs
+            ('\\A', 'foo', '\\s*\\Z')
+
         '''
         if capture_enabled:
             tag_regexs = self.tag_regexs()
@@ -338,12 +358,12 @@ def expected_as_regexs(
         if normalize_whitespace:
             sm = SM_NormWS(
                 tag_regexs, self.input_regexs(), self.ellipsis_marker(),
-                input_prefix_len_range
+                input_prefix_len_range, ignore_first_empty_lines
             )
         else:
             sm = SM_NotNormWS(
                 tag_regexs, self.input_regexs(), self.ellipsis_marker(),
-                input_prefix_len_range
+                input_prefix_len_range, ignore_first_empty_lines
             )
 
         return sm.parse(expected, tags_enabled, input_enabled)
@@ -433,7 +453,7 @@ def _extend_parser_and_parse_options_strictly_and_cache(self, optlist):
 >>> regexs, _, _, _, _ = _as_regexs(expected, normalize_whitespace=True)
 
 >>> regexs
-('\\A',
+('\\A\\s*?',
  'ex',
  '\\s',
  '(?:\\s*(?!\\s)(?:.+)(?<!\\s))?',
@@ -450,7 +470,7 @@ def _extend_parser_and_parse_options_strictly_and_cache(self, optlist):
 >>> regexs, _, _, _, _ = _as_regexs(expected, normalize_whitespace=True)
 
 >>> regexs
-('\\A',
+('\\A\\s*?',
  'ex',
  '\\s',
  '(?:\\s*(?!\\s)(?P<foo>.+?)(?<!\\s))?',