Fixed os linebreaks

README.md

@@ -2,12 +2,37 @@
 
 [![Build Status](https://travis-ci.org/openlegaldata/legal-md.svg?branch=master)](https://travis-ci.org/openlegaldata/legal-md) [![Coverage Status](https://coveralls.io/repos/github/openlegaldata/legal-md/badge.svg?branch=master)](https://coveralls.io/github/openlegaldata/legal-md?branch=master)
 
+*!!! THIS EXTENSION IS STILL WORK-IN-PROGRESS !!!*
 
-WIP
+Extension for [Python-Markdown](https://python-markdown.github.io/). Supported features:
 
-Install locally:
+- Line numbers: Convert paragraph numbers into HTML tables
+- Special links: Semantic annotation and citation markers
+
+
+## Install
 
 ```
+pip install git+https://github.com/openlegaldata/legal-md.git#egg=legal-md
+
+# Install locally (dev purpose)
 pip install -e /var/www/apps/oldp/app/
 ```
 
+## Usage
+
+```python
+import markdown
+
+md_str = '# Title\n'
+md_str += '1| Paragraph with line number\n'
+
+html = markdown.markdown(md_str, extensions=[
+    'legal_md.extensions.line_numbers',
+])
+
+```
+
+## License
+
+not licensed yet

legal_md/extensions/line_numbers.py

@@ -1,3 +1,4 @@
+import logging
 import re
 
 from bs4 import BeautifulSoup
@@ -13,6 +14,8 @@
 PLACEHOLDER_BEGIN = '{{{{{'
 PLACEHOLDER_END = '}}}}}'
 
+logger = logging.getLogger(__name__)
+
 
 class LineNumberExtension(Extension):
     """Line number extension.
@@ -85,7 +88,10 @@ def run(self, text):
         out_lines = []
         last_with_line_no = False
         last_line_no = 0
-        lines = text.split('\n')
+        lines = text.splitlines()
+
+        # print('Post lines: %s' % lines)
+
         for i, line in enumerate(lines):
             is_last_line = i == len(lines) - 1
 
@@ -152,7 +158,11 @@ def run(self, text):
 
             out_lines.append(line)
 
-        return '\n'.join(out_lines)
+        out_str = '\n'.join(out_lines)
+
+        # print('Output lines: %s' % out_str)
+
+        return out_str
 
 
 class LineNumberPreprocessor(Preprocessor):
@@ -168,74 +178,25 @@ def run(self, lines):
         ln_marker_close = PLACEHOLDER_BEGIN + '/' + PLACEHOLDER_END
         ln_with_marker = []  # store index of lines with number
 
+        # print('Input lines: %s' % lines)
+
         for i, line in enumerate(lines):
             match = re.search(pattern, line)
 
             if match:  # If line marker found
                 line_content = match.group(2)
+
+                # Replace line breaks regardless of the OS
+                line_content = line_content.replace('\r', '').replace('\n', '')
+
                 ln_marker = PLACEHOLDER_BEGIN + match.group(1) + PLACEHOLDER_END
 
                 # Write marker at the end of line
                 lines[i] = line_content + ln_marker + ln_marker_close
                 ln_with_marker.append(i)
 
-        # print('\n'.join(lines) + '\n-------------------')
-
         return lines
 
-    def _run(self, lines):
-        """Process line number marks."""
-
-        # Find and process critic marks
-        # text = '\n'.join(lines)
-        out_lines = []
-        pattern = r'^([0-9]+|#|:)\|\s(.*)$'
-        found_ln_marker = False
-        ln_marker_close = None
-
-        for i, l in enumerate(lines):
-            is_last_line = i == len(lines) - 1
-            # print('PRE: %s' % l)
-            match = re.search(pattern, l)
-            if match:
-                # print('found LN')
-                line_content = match.group(2)
-                ln_marker = PLACEHOLDER_BEGIN + match.group(1) + PLACEHOLDER_END
-                ln_marker_close = PLACEHOLDER_BEGIN + '/' + PLACEHOLDER_END
-
-                if is_last_line:
-                    # last line, set closing marker
-                    l = line_content + ln_marker + ln_marker_close
-                    ln_marker_close = None
-                else:
-                    # not last line
-                    if lines[i + 1].strip() == '':
-                        # next line is empty
-                        l = line_content + ln_marker + ln_marker_close
-                        ln_marker_close = None
-                    else:
-                        # next line is not empty, write marker to end of the line which next lines is empty
-                        l = line_content + ln_marker
-                        pass
-                    pass
-            else:
-                if ln_marker_close is not None:
-                    if is_last_line:
-                        # Set close marker if is last line
-                        l = l + ln_marker_close
-                        ln_marker_close = None
-                    else:
-                        # Set close marker if next line is empty
-                        if lines[i + 1].strip() == '':
-                            l = l + ln_marker_close
-                            ln_marker_close = None
-                        else:
-                            pass
-
-            out_lines.append(l)
-
-        return out_lines
-
 
 def makeExtension(**kwargs):  # pragma: no cover
     return LineNumberExtension(**kwargs)

legal_md/tests/__init__.py

@@ -11,12 +11,15 @@ class MarkdownExtensionTest(TestCase):
         ]
     RESOURCE_DIR = os.path.join(os.path.dirname(os.path.realpath(__file__)), 'resources')
 
+    def convert_md(self, md_content):
+        return markdown.markdown(md_content, extensions=self.extensions)
+
     def assert_md_file(self, resource_name, msg=None):
         """Compare resource markdown file with corresponding html file"""
         with open(os.path.join(self.RESOURCE_DIR, resource_name + '.md')) as md_file:
             with open(os.path.join(self.RESOURCE_DIR, resource_name + '.html')) as html_file:
                 md_content = md_file.read()
                 expected = html_file.read().strip()
-                actual = markdown.markdown(md_content, extensions=self.extensions)
+                actual = self.convert_md(md_content)
 
                 self.assertEqual(expected, actual, msg)

legal_md/tests/resources/os_line_breaks.html

@@ -0,0 +1,8 @@
+<table class="table table-lines">
+<tr class="line" data-line="1"><td class="line-number"><a href="#L1">1</a></td><td class="line-content"><p>
+ ab
+</p></td></tr>
+<tr class="line" data-line="2"><td class="line-number"><a href="#L2">2</a></td><td class="line-content"><p>
+ cd
+</p></td></tr>
+</table>

legal_md/tests/test_line_numbers.py

@@ -1,3 +1,5 @@
+import os
+
 from legal_md.tests import MarkdownExtensionTest
 
 
@@ -16,6 +18,11 @@ def test_mixed(self):
     def test_case_text(self):
         self.assert_md_file('case_text')
 
+    def test_os_line_breaks(self):
+        """Test on line breaks (e.g. \r\n, ...) """
+        with open(os.path.join(self.RESOURCE_DIR, 'os_line_breaks.html')) as html_file:
+            self.assertEqual(self.convert_md('1| ab\r\n\r\n2| cd'), html_file.read().strip())
+
     def test_lines_with_single_line_breaks(self):
         self.assert_md_file('lines_with_single_line_breaks')