RE: backslashes and r-strings r"abc"
"Mark Hahn" <[email protected]> Mon, 26 Jul 2004 15:10:54 -0700
| Newsgroups | gmane.comp.lang.prothon.user |
|---|---|
| Message-ID | <000901c4735d$6b9a65d0$0b01a8c0@mark> |
Martin Christensen wrote: >Backslash escaping is widely used, and it'd be gotcha-ish to >not use it as an escape character, if you ask me. I only ever >use raw (r'abc') strings for regular expressions, but for >those, they increase legibility quite a bit, I think. > >As I see it, the best alternative to the usual way of handling >escape characters is for Strings to have a 'process escape >characters' method that would... well, treat escape characters >like escape characters. I'm not certain how clear that was, so >here's an example: > >O>> 'a\tb\nc' >a\tb\nc >O>> 'a\tb\nc'.processEscapeCharacters() >a b >c > >Personally I'm in favour of preserving escape characters by >default, if for no other reason than consistency with just >about every other langauge around. If escape characters are a >Python gotcha, then Python is not the sole perpetrator. as for >r'abc' being a gotcha... I don't know, really. How often is >that notation used outside of regular expressions? For the >longest time, I even thought that it was specific _to_ regular >expressions. :o) I don't mind either, but my opinions on the >subject aren't overly strong. One thing that bothers me about Python's r-strings is that they are supposed to pass backslashes but they throw an exception if you have a backslash as the last character. This makes them useless in one place I've always wanted to use them: r"c:\windows directory\" # throws Python exception So maybe Prothon could at least fix that problem. Does anyone see any reason to preserve Python's restriction on not having a backslash as the last character of a r-string?