This is the mail archive of the
mailing list for the GDB project.
Re: [PATCH] An implementation of pipe to make I/O communication between gdb and shell.
- From: Pedro Alves <pedro at codesourcery dot com>
- To: Abhijit Halder <abhijit dot k dot halder at gmail dot com>
- Cc: Jan Kratochvil <jan dot kratochvil at redhat dot com>, gdb-patches at sourceware dot org, Sergio Durigan Junior <sergiodj at redhat dot com>, Tom Tromey <tromey at redhat dot com>
- Date: Fri, 5 Aug 2011 11:36:14 +0100
- Subject: Re: [PATCH] An implementation of pipe to make I/O communication between gdb and shell.
- References: <CAOhZP9zU2K01zFDkChtutGT6yJjUeWYeGRNAAgPgM5rKwrgysg@mail.gmail.com> <firstname.lastname@example.org> <CAOhZP9zNY7YkFbRk5=_RXMoaXcijyY7Q674iSxKym8DvYkOWsQ@mail.gmail.com>
On Friday 05 August 2011 11:24:54, Abhijit Halder wrote:
> > the delimiter would be `dthread'. I see no reason to
> > require it to be a single character.
> Just a thought. Can't using a single character make the life simple?
To who? I imagine myself stumbling on some expression in some
language that conflicts with |, and without much thinking,
pipe XXX randomexpr XXX randomcommand
instead of having to think, or possibly try/fail more than one
time, which characters are allowed and wouldn't conflict with
both randomexpr and randomcommand.
> That will surely reduce the options for delimiter. But as you said,
> like sed this can be used.
Well, I said like a grandchild of sed and cat EOF, the latter
refering to "here documents", which accept strings as
delimiters. :-) Sorry for not being clear.