release_notes: Handle reference links without names like [1]

Some links in WordPress have text like [1], [2], etc.

When they get converted into text the "[1]" is kept and a
reference is added so we have something like "[[1]][2]" in the
final text.

The fix is to completely remove the [[1]] part as it is not
necessary anymore.

Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@cyberdimension.org>

1 files changed, 41 insertions, 0 deletions

diff --git a/release_notes.py b/release_notes.py
index 95ae737..85ab84c 100755
--- a/release_notes.py
+++ b/release_notes.py
@@ -34,12 +34,51 @@ try:
 except:
     pass
 
+import os
+import re
 import sys
 
 def usage(progname):
     print("{} path/to/file.html".format(progname))
     sys.exit(1)
 
+# A "[1]" in the html becomes "[[1]][6]" in text.
+# As we already uses references at the end a [6] would
+# be enough.
+def fix_wordpress_references_link(string):
+    open_square_bracket = re.escape('[')
+    close_square_bracket = re.escape(']')
+    whitespaces = '\s*'
+    numbers = '\d+'
+
+    # [ [ 1 ] ] [ 6 ]
+    wordpress_link_regex = \
+        \
+          open_square_bracket + whitespaces \
+        + open_square_bracket + whitespaces \
+        + numbers + whitespaces \
+        + close_square_bracket + whitespaces \
+        + close_square_bracket + whitespaces \
+        \
+        + open_square_bracket + whitespaces \
+        + numbers + whitespaces \
+        + close_square_bracket + whitespaces \
+
+    results = re.findall(wordpress_link_regex, string)
+
+    part_to_remove = '^' \
+        + open_square_bracket + whitespaces \
+        + open_square_bracket + whitespaces \
+        + numbers + whitespaces \
+        + close_square_bracket + whitespaces \
+        + close_square_bracket + whitespaces \
+
+    for result in results:
+        replacement = re.sub(part_to_remove, '', result)
+        string = string.replace(result, replacement)
+
+    return string
+
 def convert(html_file_path):
     with open(html_file_path) as html_file:
         try:
@@ -67,6 +106,8 @@ def convert(html_file_path):
     parser = HTML2Text()
     text = parser.handle(soup.decode())
 
+    text = fix_wordpress_references_link(text)
+
     return text
 
 def main():