omit unnecessary spaces around <b>, <em>, and <strike>

fixes #324

ChangeLog.rst

@@ -5,6 +5,7 @@ UNRELEASED
 * Feature #318: Make padded tables more similar to pandoc's pipe_tables.
 * Add support for Python 3.9.
 * Fix extra line breaks inside html link text (between '[' and ']')
+* Fix #324: unnecessary spaces around ``<b>``, ``<em>``, and ``strike`` tags.
 
 2020.1.16
 =========

html2text/__init__.py

@@ -3,6 +3,7 @@
 import html.entities
 import html.parser
 import re
+import string
 import urllib.parse as urlparse
 from textwrap import wrap
 from typing import Dict, List, Optional, Tuple, Union
@@ -404,13 +405,13 @@ def handle_tag(
                 self.blockquote -= 1
                 self.p()
 
-        def no_preceding_space(self: HTML2Text) -> bool:
-            return bool(
-                self.preceding_data and re.match(r"[^\s]", self.preceding_data[-1])
-            )
-
         if tag in ["em", "i", "u"] and not self.ignore_emphasis:
-            if start and no_preceding_space(self):
+            if (
+                start
+                and self.preceding_data
+                and self.preceding_data[-1] not in string.whitespace
+                and self.preceding_data[-1] not in string.punctuation
+            ):
                 emphasis = " " + self.emphasis_mark
             else:
                 emphasis = self.emphasis_mark
@@ -420,7 +421,11 @@ def no_preceding_space(self: HTML2Text) -> bool:
                 self.stressed = True
 
         if tag in ["strong", "b"] and not self.ignore_emphasis:
-            if start and no_preceding_space(self):
+            if (
+                start
+                and self.preceding_data
+                and self.preceding_data[-1] == self.strong_mark[0]
+            ):
                 strong = " " + self.strong_mark
             else:
                 strong = self.strong_mark
@@ -430,7 +435,7 @@ def no_preceding_space(self: HTML2Text) -> bool:
                 self.stressed = True
 
         if tag in ["del", "strike", "s"]:
-            if start and no_preceding_space(self):
+            if start and self.preceding_data and self.preceding_data[-1] == "~":
                 strike = " ~~"
             else:
                 strike = "~~"
@@ -826,7 +831,7 @@ def handle_data(self, data: str, entity_char: bool = False) -> None:
             self.preceding_stressed = True
         elif self.preceding_stressed:
             if (
-                re.match(r"[^\s.!?]", data[0])
+                re.match(r"[^][(){}\s.!?]", data[0])
                 and not hn(self.current_tag)
                 and self.current_tag not in ["a", "code", "pre"]
             ):

test/emphasis_preserved_whitespace.html

@@ -16,5 +16,12 @@
 <p><em>emphasis</em>.</p>
 <p><em>emphasis</em>?</p>
 <p><em>emphasis</em>!</p>
+<p>(<em>emphasis</em>)</p>
+<p>[<b>bold</b>}</p>
+<p>(<strike>strike</strike>]</p>
+
+<!-- Strong and strike characters have space  -->
+<p>*<b>bold</b></p>
+<p>~<strike>strike</strike></p>
 
 <p><em>em1</em><em>em2</em></p>

test/emphasis_preserved_whitespace.md

@@ -24,5 +24,15 @@ _emphasis_?
 
 _emphasis_!
 
+(_emphasis_)
+
+[**bold**}
+
+(~~strike~~]
+
+* **bold**
+
+~ ~~strike~~
+
 _em1_ _em2_