question on mi-console.exp kfail case

question on mi-console.exp kfail case

[Bob Rossi]

Hi,

Andrew, I've mailed you personally cause I think you originally wrote
the comment below,

I've got the entire testsuite passing with the new syntax checker for
the MI output. However, I noticed the testcase mi?-console.exp has this,

   gdb_expect {
       -re "@\"H\"\r\n.*@\"e\"\r\n.*@\"l\"\r\n.*@\"l\"\r\n.*@\"o\"\r\n.*@\" \"\r\n.*@\"\\\\\\\\\"\r\n.*@\"\\\\\"\"\r\n.*@\"!\"\r\n.*@\"\\\\r\"\r\n.*@\"\\\\n\"\r\n" {
       pass "Hello message"
       }
       -re "Hello" {

       # Probably a native system where GDB doesn't have direct
       # control over the inferior console.
       # For this to work, GDB would need to run the inferior process
       # under a PTY and then use the even-loops ability to wait on
       # multiple event sources to channel the output back through the
       # MI.

       kfail "gdb/623" "Hello message"
       }
       timeout {
       fail "Hello message (timeout)"
       }
   }

In particular, I don't know what the first regular expression is, and
it's the one that allows for a "pass". What is it trying to match?

When I run this testcase, I get,
   KFAIL: gdb.mi/mi-console.exp: Hello message (PRMS: gdb/623)

The output of the inferior is simply "Hello \"!\r\n". Since I have the
testsuite set up so the inferior output is redirected to it's own pty,
would it be fine for me to match the expected output with a PASS?

Or is this a kfail for some reason I just don't understand?

Thanks,
Bob Rossi

Re: question on mi-console.exp kfail case

[Andrew Cagney]

Bob Rossi wrote:
> Hi,
> 
> Andrew, I've mailed you personally cause I think you originally wrote
> the comment below,
> 
> I've got the entire testsuite passing with the new syntax checker for
> the MI output. However, I noticed the testcase mi?-console.exp has this,

See my reply to your patch.  I suspect we ``violently agree with each 
other''.   The comment, written in ~'98, reflects the position taken tat 
that time.  It's been superseeded by other events - MI clients end up 
doing what I thought was "hard".

Andrew

>    gdb_expect {

This is now wrong.

GDB would only produce this when connected to a remote target.  That 
code has been superseeded by remote file I/O.

>        -re "@\"H\"\r\n.*@\"e\"\r\n.*@\"l\"\r\n.*@\"l\"\r\n.*@\"o\"\r\n.*@\" \"\r\n.*@\"\\\\\\\\\"\r\n.*@\"\\\\\"\"\r\n.*@\"!\"\r\n.*@\"\\\\r\"\r\n.*@\"\\\\n\"\r\n" {
>        pass "Hello message"
>        }

This is now right (but must come from the PTY).

>        -re "Hello" {
> 
>        # Probably a native system where GDB doesn't have direct
>        # control over the inferior console.
>        # For this to work, GDB would need to run the inferior process
>        # under a PTY and then use the even-loops ability to wait on
>        # multiple event sources to channel the output back through the
>        # MI.
> 
>        kfail "gdb/623" "Hello message"
>        }
>        timeout {
>        fail "Hello message (timeout)"
>        }
>    }
> 
> In particular, I don't know what the first regular expression is, and
> it's the one that allows for a "pass". What is it trying to match?
> 
> When I run this testcase, I get,
>    KFAIL: gdb.mi/mi-console.exp: Hello message (PRMS: gdb/623)
> 
> The output of the inferior is simply "Hello \"!\r\n". Since I have the
> testsuite set up so the inferior output is redirected to it's own pty,
> would it be fine for me to match the expected output with a PASS?
> 
> Or is this a kfail for some reason I just don't understand?

Andrew