Invalid escape sequence in part_query.py

[devbisme]

Invalid escape sequence in part_query.py

[xesscorp]

Even a blind squirrel finds a nut once and a while!

[xesscorp]

I'm confused by this. I want to match the literal ( character at the start of the string, not start a group. All the man pages for the re module (from 2.7 up to 3.8) state that I can use \( or [(] to do that. So why is this flagged as deprecated? Is this a spurious error message, or am I overlooking something?

[johnthagen]

@xesscorp I think the correct way to create regular expression strings to avoid any of these issues is to use a raw string: r"(\s*pad\s+([^\s)]+)", that avoids the issue of having to distinguish between escaping in the string (which at least in Python 3 can mean unicode escapes) and using \ in the regular expression string.

See https://stackoverflow.com/a/50504635/1398841

[xesscorp]

OK, I see now. I made the change to a raw string and checked it in. Thanks!

[johnthagen]

@xesscorp I think this was fixed in error: 27cb168

I believe you want to remove the \ before the first ( since a raw string will keep the \ now.

r"\(\s*pad\s+([^\s)]+)
r"(\s*pad\s+([^\s)]+)

[xesscorp]

The regex is trying to detect a starting paren of a pad definition. If I removed the \, then the paren would be looked at as the start of a group. That's not what I want. There would also be unbalanced grouping parens: two ( and only one ).

[johnthagen]

@xesscorp Ah, I see what you're saying. You're right, you did it correctly. 👍