[RFA] gdb.base/callfuncs.exp: make all test names unique

[RFA] gdb.base/callfuncs.exp: make all test names unique

[Michael Elizabeth Chastain]

This patch changes gdb.base/callfuncs.exp to make all test names unique.

Testing: I tested this on native Red Hat Linux 7 and native Solaris 2.8.

OK to apply?

Michael

===

2001-05-19  Michael Chastain  <chastain@redhat.com>

	* gdb.base/callfuncs.exp: Make all test names unique.

Index: gdb/testsuite/gdb.base/callfuncs.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.base/callfuncs.exp,v
retrieving revision 1.4
diff -c -3 -p -r1.4 callfuncs.exp
*** gdb/testsuite/gdb.base/callfuncs.exp	2001/03/06 08:21:50	1.4
--- gdb/testsuite/gdb.base/callfuncs.exp	2001/05/20 05:39:03
*************** if { $hp_aCC_compiler } {
*** 292,299 ****
  
  # Make sure that malloc gets called and that the floating point unit
  # is initialized via a call to t_double_values.
! gdb_test "next" "t_double_values\\(double_val1, double_val2\\);.*"
! gdb_test "next" "t_structs_c\\(struct_val1\\);.*"
  
  # Save all register contents.
  do_get_all_registers
--- 292,299 ----
  
  # Make sure that malloc gets called and that the floating point unit
  # is initialized via a call to t_double_values.
! gdb_test "next" "t_double_values\\(double_val1, double_val2\\);.*" "next 1"
! gdb_test "next" "t_structs_c\\(struct_val1\\);.*" "next 2"
  
  # Save all register contents.
  do_get_all_registers

Re: [RFA] gdb.base/callfuncs.exp: make all test names unique

[Michael Snyder]

Michael Elizabeth Chastain wrote:
> 
> This patch changes gdb.base/callfuncs.exp to make all test names unique.
> 
> Testing: I tested this on native Red Hat Linux 7 and native Solaris 2.8.
> 
> OK to apply?
> 
> Michael

Michael, this is great, but see below:

> 
> ===
> 
> 2001-05-19  Michael Chastain  <chastain@redhat.com>
> 
>         * gdb.base/callfuncs.exp: Make all test names unique.
> 
> Index: gdb/testsuite/gdb.base/callfuncs.exp
> ===================================================================
> RCS file: /cvs/src/src/gdb/testsuite/gdb.base/callfuncs.exp,v
> retrieving revision 1.4
> diff -c -3 -p -r1.4 callfuncs.exp
> *** gdb/testsuite/gdb.base/callfuncs.exp        2001/03/06 08:21:50     1.4
> --- gdb/testsuite/gdb.base/callfuncs.exp        2001/05/20 05:39:03
> *************** if { $hp_aCC_compiler } {
> *** 292,299 ****
> 
>   # Make sure that malloc gets called and that the floating point unit
>   # is initialized via a call to t_double_values.
> ! gdb_test "next" "t_double_values\\(double_val1, double_val2\\);.*"
> ! gdb_test "next" "t_structs_c\\(struct_val1\\);.*"
> 
>   # Save all register contents.
>   do_get_all_registers
> --- 292,299 ----
> 
>   # Make sure that malloc gets called and that the floating point unit
>   # is initialized via a call to t_double_values.
> ! gdb_test "next" "t_double_values\\(double_val1, double_val2\\);.*" "next 1"
> ! gdb_test "next" "t_structs_c\\(struct_val1\\);.*" "next 2"
> 
>   # Save all register contents.
>   do_get_all_registers

I wonder if it would not be preferable to use "next over t_double_values"
and "next over t_structs_c".  Since you are proposing to do a lot of these
mods, I have some concern about the testsuite becoming filled with names
such as "next 1" and "next 2" which, although practically speaking just as
useful, are somewhat opaque.

Re: [RFA] gdb.base/callfuncs.exp: make all test names unique

[Fernando Nasser]

Michael Snyder wrote:
> 
> Michael Elizabeth Chastain wrote:
> >
> > This patch changes gdb.base/callfuncs.exp to make all test names unique.
> >
> > Testing: I tested this on native Red Hat Linux 7 and native Solaris 2.8.
> >
> > OK to apply?
> >
> > Michael
> 
> Michael, this is great, but see below:
> 
> >
> > ===
> >
> > 2001-05-19  Michael Chastain  <chastain@redhat.com>
> >
> >         * gdb.base/callfuncs.exp: Make all test names unique.
> >
> > Index: gdb/testsuite/gdb.base/callfuncs.exp
> > ===================================================================
> > RCS file: /cvs/src/src/gdb/testsuite/gdb.base/callfuncs.exp,v
> > retrieving revision 1.4
> > diff -c -3 -p -r1.4 callfuncs.exp
> > *** gdb/testsuite/gdb.base/callfuncs.exp        2001/03/06 08:21:50     1.4
> > --- gdb/testsuite/gdb.base/callfuncs.exp        2001/05/20 05:39:03
> > *************** if { $hp_aCC_compiler } {
> > *** 292,299 ****
> >
> >   # Make sure that malloc gets called and that the floating point unit
> >   # is initialized via a call to t_double_values.
> > ! gdb_test "next" "t_double_values\\(double_val1, double_val2\\);.*"
> > ! gdb_test "next" "t_structs_c\\(struct_val1\\);.*"
> >
> >   # Save all register contents.
> >   do_get_all_registers
> > --- 292,299 ----
> >
> >   # Make sure that malloc gets called and that the floating point unit
> >   # is initialized via a call to t_double_values.
> > ! gdb_test "next" "t_double_values\\(double_val1, double_val2\\);.*" "next 1"
> > ! gdb_test "next" "t_structs_c\\(struct_val1\\);.*" "next 2"
> >
> >   # Save all register contents.
> >   do_get_all_registers
> 
> I wonder if it would not be preferable to use "next over t_double_values"
> and "next over t_structs_c".  Since you are proposing to do a lot of these
> mods, I have some concern about the testsuite becoming filled with names
> such as "next 1" and "next 2" which, although practically speaking just as
> useful, are somewhat opaque.

Good catch.  This will all become irrelevant if the changes I am
thinking of happen, but for now it would be nice to have the more
descriptive name.

Michael (Chastain), please check it in your patch with these changes. 
Thank you very much to help eliminating these ambiguities.

-- 
Fernando Nasser
Red Hat Canada Ltd.                     E-Mail:  fnasser@redhat.com
2323 Yonge Street, Suite #300
Toronto, Ontario   M4P 2C9

Re: [RFA] gdb.base/callfuncs.exp: make all test names unique

[Michael Elizabeth Chastain]

Okay, I'll drop callfuncs.exp from this round.

I have seven other patches.  Do you think any of them are good enough
to merge?

Michael

Re: [RFA] gdb.base/callfuncs.exp: make all test names unique

[Fernando Nasser]

Michael Elizabeth Chastain wrote:
> 
> chastain> ! gdb_test "next" "t_double_values\\(double_val1, double_val2\\);.*" "next 1"
> chastain> ! gdb_test "next" "t_structs_c\\(struct_val1\\);.*" "next 2"
> 
> snyder> I wonder if it would not be preferable to use "next over t_double_values"
> snyder> and "next over t_structs_c".
> 
> nasser> Michael (Chastain), please check it in your patch with these changes.
> nasser> Thank you very much to help eliminating these ambiguities.
> 
> Oops, actually that should be "next to t_double_values" rather than
> "next over t_double_values" and so on.  It still looks better than
> "next" or "next 1".
> 
> Works for me, I will use "next to t_double_values" and check it in.
> 

Thanks again.

Cheers,
Fernando

-- 
Fernando Nasser
Red Hat Canada Ltd.                     E-Mail:  fnasser@redhat.com
2323 Yonge Street, Suite #300
Toronto, Ontario   M4P 2C9

Re: [RFA] gdb.base/callfuncs.exp: make all test names unique

[Michael Elizabeth Chastain]

Committed.

I have four more files in progress:

  arithmet.exp
  assign.exp
  ena-dis-br.exp
  logical.exp

These are much bigger changes.  These files use a lot of
send_gdb/gdb_expect calls that can be calls to gdb_test instead, which
makes the files a lot shorter.  Plus I vaguely recall that people prefer
gdb_test over send_gdb/gdb_expect because it standardizes all the error
and timeout handling.

I'm planning to put up the first one (arithmet.exp) for comments this
weekend.  Any guidance here would be appreciated.

If anyone else wants to join the fun, here are more files with duplicate
test names:

  gdb.base/maint.exp
  gdb.base/miscexprs.exp
  gdb.base/opaque.exp
  gdb.base/page.exp
  gdb.base/pointers.exp
  gdb.base/printcmds.exp
  gdb.base/radix.exp
  gdb.base/relational.exp
  gdb.base/restore.exp
  gdb.base/return2.exp
  gdb.base/scope.exp
  gdb.base/shlib-call.exp
  gdb.base/signals.exp
  gdb.c++/annota2.exp
  gdb.c++/anon-union.exp
  gdb.c++/demangle.exp
  gdb.c++/inherit.exp
  gdb.c++/ref-types.exp
  gdb.chill/chexp.exp
  gdb.fortran/types.exp
  gdb.threads/pthreads.exp
  gdb.trace/passcount.exp

Michael

Re: [RFA] gdb.base/callfuncs.exp: make all test names unique

[Fernando Nasser]

Michael Elizabeth Chastain wrote:
> 
> Committed.
> 

Thanks.


> I have four more files in progress:
> 
>   arithmet.exp
>   assign.exp
>   ena-dis-br.exp
>   logical.exp
> 
> These are much bigger changes.  These files use a lot of
> send_gdb/gdb_expect calls that can be calls to gdb_test instead, which
> makes the files a lot shorter.  Plus I vaguely recall that people prefer
> gdb_test over send_gdb/gdb_expect because it standardizes all the error
> and timeout handling.
> 

Yes, that is it.  There are a few occasions where we still cannot use gdb_test, unfortunately -- I plan to do something about that.  But but getting rid of as many send_gdb/gdb_expect pairs as we currently can is a nice move.


> I'm planning to put up the first one (arithmet.exp) for comments this
> weekend.  Any guidance here would be appreciated.
> 

Great.  Thanks again.




-- 
Fernando Nasser
Red Hat - Toronto                       E-Mail:  fnasser@redhat.com
2323 Yonge Street, Suite #300
Toronto, Ontario   M4P 2C9