From 8782d3018e5cbc4ef04e4b8e74b90025de3004b3 Mon Sep 17 00:00:00 2001 From: MarkKoz Date: Wed, 6 May 2020 14:18:40 -0700 Subject: Code block: fix find_code_blocks iteration and missing return * Add named capture groups to the regex --- bot/cogs/codeblock/parsing.py | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-) diff --git a/bot/cogs/codeblock/parsing.py b/bot/cogs/codeblock/parsing.py index 88a5c7b7a..9adb4e0ab 100644 --- a/bot/cogs/codeblock/parsing.py +++ b/bot/cogs/codeblock/parsing.py @@ -21,13 +21,13 @@ TICKS = { } RE_CODE_BLOCK = re.compile( fr""" - ( - ([{''.join(TICKS)}]) # Put all ticks into a character class within a group. - \2{{2}} # Match the previous group 2 more times to ensure it's the same char. + (?P + (?P[{''.join(TICKS)}]) # Put all ticks into a character class within a group. + \2{{2}} # Match previous group 2 more times to ensure the same char. ) - ([^\W_]+\n)? # Optionally match a language specifier followed by a newline. - (.+?) # Match the actual code within the block. - \1 # Match the same 3 ticks used at the start of the block. + (?P[^\W_]+\n)? # Optionally match a language specifier followed by a newline. + (?P.+?) # Match the actual code within the block. + \1 # Match the same 3 ticks used at the start of the block. """, re.DOTALL | re.VERBOSE ) @@ -52,14 +52,19 @@ def find_code_blocks(message: str) -> Sequence[CodeBlock]: one code block right, they already know how to fix the rest themselves. """ code_blocks = [] - for _, tick, language, content in RE_CODE_BLOCK.finditer(message): - language = language.strip() - if tick == BACKTICK and language: + for match in RE_CODE_BLOCK.finditer(message): + # Used to ensure non-matched groups have an empty string as the default value. + groups = match.groupdict("") + language = groups["lang"].strip() # Strip the newline cause it's included in the group. + + if groups["tick"] == BACKTICK and language: return () - elif len(content.split("\n", 3)) > 3: - code_block = CodeBlock(content, language, tick) + elif len(groups["code"].split("\n", 3)) > 3: + code_block = CodeBlock(groups["code"], language, groups["tick"]) code_blocks.append(code_block) + return code_blocks + def is_python_code(content: str) -> bool: """Return True if `content` is valid Python consisting of more than just expressions.""" -- cgit v1.2.3