Restrict accidental-ordered-list fix to post start

Previously this was also trying to catch ones at the beginning of new
paragraphs, but that seems to mostly just be causing unexpected issues
when people create ordered lists with a blank line between items. This
can probably be done properly in the future, but just restricting it to
the start of posts is probably better for now.

tildes/tests/test_markdown.py

@@ -70,11 +70,7 @@ def test_deliberate_ordered_list():
 
 def test_accidental_ordered_list():
     """Ensure a common "accidental" ordered list gets escaped."""
-    markdown = (
-        "What year did this happen?\n\n"
-        "1975. It was a long time ago.\n\n"
-        "But I remember it like it was yesterday."
-    )
+    markdown = "1975. It was a long time ago."
     html = convert_markdown_to_safe_html(markdown)
 
     assert "<ol" not in html

tildes/tildes/lib/markdown.py

@@ -80,9 +80,9 @@ HTML_ATTRIBUTE_WHITELIST = {
 PROTOCOL_WHITELIST = ("http", "https")
 
 # Regex that finds ordered list markdown that was probably accidental - ones being
-# initiated by anything except "1."
+# initiated by anything except "1." at the start of a post
 BAD_ORDERED_LIST_REGEX = re.compile(
-    r"((?:\A|\n\n)"  # Either the start of the entire text, or a new paragraph
+    r"((?:\A)"  # The start of the entire text
     r"(?!1\.)\d+)"  # A number that isn't "1"
     r"\.\s"  # Followed by a period and a space
 )
@@ -156,13 +156,12 @@ def escape_accidental_ordered_lists(markdown: str) -> str:
     """Escape markdown that's probably an accidental ordered list.
 
     It's a common markdown mistake to accidentally start a numbered list, by beginning a
-    post or paragraph with a number followed by a period. For example, someone might try
-    to write "1975. It was a long time ago.", and the result will be a comment that says
-    "1. It was a long time ago." since that gets parsed into a numbered list.
+    post with a number followed by a period. For example, someone might try to write
+    "1975. It was a long time ago.", and the result will be a comment that says "1. It
+    was a long time ago." since that gets parsed into a numbered list.
 
     This fixes that quirk of markdown by escaping anything that would start a numbered
-    list except for "1. ". This will cause a few other edge cases, but I believe they're
-    less common/important than fixing this common error.
+    list at the beginning of a post, except for "1. ".
     """
     return BAD_ORDERED_LIST_REGEX.sub(r"\1\\. ", markdown)