This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [RFA 1/3] Linespec rewrite: Parsing


On Sat, Mar 24, 2012 at 1:23 PM, Keith Seitz <keiths@redhat.com> wrote:
>
> On 03/24/2012 06:04 AM, asmwarrior wrote:
>
>> [debug]> break
>> "E:/code/cb/cb_trunk/src/plugins/codecompletion/parser/token.cpp:64"
>> [debug]unmatched quote
>
>
> This is now an illegal linespec. You do not need to quote this at all. This is one of the motivating factors for doing this rewrite: the quoting is _out of control_. [If you do want to quote it, ONLY quote the filename portion.]
>
> It appears you may be using the patches I submitted to the list. I've committed one or two additional patches for problems I've found to the archer branch, one of which deal specifically with quoting and embedded colons.
>
> I encourage you to try that branch:
>
> (gdb) break E:/code/cb/cb_trunk/src/plugins/codecompletion/parser/token.cpp:64
>
> No source file named E:/code/cb/cb_trunk/src/plugins/codecompletion/parser/token.cpp.
>
> Leaving the quoting in place will force the lookup of the function "E:/.../token.cpp:64", which I'm guessing is not what you really want. :-)

This is the change of behavior I think most likely to blow up.
-break-insert proxies straight through to the CLI break command; have
you checked what various other IDEs send today?

Is a special case (yeah, I know) for a quoted function ending in
:[0-9]* practical?

--
Thanks,
Daniel


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]