Re: [Expectperl-discuss] $ in regex has duplicate meaning/ambiguity

[Jo R. <jr...@sv...>]

Obviously I will test, but the description is

  \Z - Matches only at the end of a string or before a newline
  \z - Matches only at the end of a string 

What is "a string" in this context?  In my coding brain, the entire buffer
is a string...  if it means the entire buffer, then this is exactly the
opposite of what I want.  I want a newline but NOT the end of the buffer...

On Thu, Mar 09, 2006 at 10:08:45AM -0500, Chris Snyder wrote:
> Have you looked at \z and \Z as to whether they satisfy your needs?
> 
> Page 179 of the camel book, 3rd edition.
> 
> On Thu, 2006-03-09 at 01:25 -0800, Jo Rhett wrote:
> > I'm having some difficulty with some expect sequences that I finally
> > figured out, but others may run into this.  The documentations says that 
> > ^ will match the beginning of a line, and $ match the end... but $ also
> > matches the end of the buffer, so unless the far side is outputting a \r in
> > addition to a linefeed, it can be pretty tough trying to match a specific
> > line.  (watch in debug, and often a buffer ends mid-line...)
> > 
> > I'm also seeing scenaries where /^\s+(Text)$/ will match across multiple 
> > linefeeds prior to finding "Text".  Like "\n\n\nText\n".
> >  
> > Are there some delimiters that can be useful to match end of line but not
> > end of buffer?  And any way to guarantee that /^...$/ stays on a single
> > line?
> > 
> > I'm using /^[ \t]+Text\r$/ and it meets my needs for the current situation,
> > but some of the things I'm working with won't output carriage returns so...
> > 

-- 
Jo Rhett
senior geek
SVcolo : Silicon Valley Colocation