help-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Differences with perld amd cmdline perl -d


From: Harry Putnam
Subject: Differences with perld amd cmdline perl -d
Date: Tue, 18 Feb 2003 16:59:17 GMT
User-agent: Gnus/5.090008 (Oort Gnus v0.08) Emacs/21.3.50 (i686-pc-linux-gnu)

I notice a number of instances where running perld (perldebugger) from
emacs, give very different results from running perl -d at the
command line, on the same material.

Apparently something happens between emacs and terminal or somewhere
that changes things enough to get very different and somewhat useless
results. 

Rather than post long winded perl code, as example I'll just say the
command looks like:

  perl -d ./pdb.pl -l 6 7 -c <test.g 
at the perld prompt.

Running the command from the command line, the perl debugger steps
thru the code and shows no problems.

Running it in emacs, it enters a `while' loop and never leaves that
loop.

  while (<STDIN>) {
    chomp;
    print FILE_BK "
  }

Pressing `p "<$_>\n"  always reports a blank line
  <>

As you see, the loop is confined to STDIN which is why the file is
fed with `< test.g' syntax.  The script itself is intended to work only
in a pipe or with  `cmd < infile' syntax.

Apparently emacs is confused about what is STDIN and never sees the
end of it. I suspect something technical about how emacs sees the
command to be the problem.

But have also noticed many other instances where at the cmdline the
debugger works as expected, like setting a break point then pressing
`r' to get to it.

On some scripts (not all or even most).  The same proceedure in emacs
will not stop at the break point and just run right thru and finish,
maybe reporting problems.  So there must be certain things that are
seen differently when viewed thru the emacs layer compared to from the
command line.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]