[RFA] Fix "set" handling of Python parameters

[RFA] Fix "set" handling of Python parameters

[Tom Tromey]

It's long bothered me that setting a Python parameter from the CLI
will print the "set" help text by default.  I think usually "set"
commands should be silent.  And, while you can modify this behavior a
bit by providing a "get_set_string" method, if this method returns an
empty string, a blank line will be printed.

This patch removes the "help" behavior and changes the get_set_string
behavior to avoid printing a blank line.  The code has a comment about
preserving API behavior, but I don't think this is truly important;
and in any case the workaround -- implementing get_set_string -- is
trivial.

Regression tested on x86-64 Fedora 26.

gdb/ChangeLog
2018-04-26  Tom Tromey  <tom@tromey.com>

	* NEWS: Mention new "set" behavior.
	* python/py-param.c (get_set_value): Don't print an empty string.
	Don't call get_doc_string.

gdb/doc/ChangeLog
2018-04-26  Tom Tromey  <tom@tromey.com>

	* python.texi (Parameters In Python): Update get_set_string
	documentation.
---
 gdb/ChangeLog                             |  6 ++++++
 gdb/NEWS                                  |  6 ++++++
 gdb/doc/ChangeLog                         |  5 +++++
 gdb/doc/python.texi                       | 14 ++++++++------
 gdb/python/py-param.c                     | 11 +++--------
 gdb/testsuite/gdb.python/py-parameter.exp |  4 ++--
 6 files changed, 30 insertions(+), 16 deletions(-)

diff --git a/gdb/NEWS b/gdb/NEWS
index 63fe30d175..c992048634 100644
--- a/gdb/NEWS
+++ b/gdb/NEWS
@@ -24,6 +24,12 @@ set|show record btrace cpu
   Controls the processor to be used for enabling errata workarounds for
   branch trace decode.
 
+* Python API
+
+  ** A gdb.Parameter will no longer print the "set" help text on an
+     ordinary "set"; instead by default a "set" will be silent unless
+     the get_set_string method returns a non-empty string.
+
 * New targets
 
 RiscV ELF			riscv*-*-elf
diff --git a/gdb/doc/python.texi b/gdb/doc/python.texi
index ebd48fffe7..7bbfb09d04 100644
--- a/gdb/doc/python.texi
+++ b/gdb/doc/python.texi
@@ -3722,14 +3722,16 @@ parameter.  It can be read and assigned to just as any other
 attribute.  @value{GDBN} does validation when assignments are made.
 @end defvar
 
-There are two methods that should be implemented in any
-@code{Parameter} class.  These are:
+There are two methods that may be implemented in any @code{Parameter}
+class.  These are:
 
 @defun Parameter.get_set_string (self)
-@value{GDBN} will call this method when a @var{parameter}'s value has
-been changed via the @code{set} API (for example, @kbd{set foo off}).
-The @code{value} attribute has already been populated with the new
-value and may be used in output.  This method must return a string.
+If this method exists, @value{GDBN} will call it when a
+@var{parameter}'s value has been changed via the @code{set} API (for
+example, @kbd{set foo off}).  The @code{value} attribute has already
+been populated with the new value and may be used in output.  This
+method must return a string.  If the returned string is not empty,
+@value{GDBN} will present it to the user.
 @end defun
 
 @defun Parameter.get_show_string (self, svalue)
diff --git a/gdb/python/py-param.c b/gdb/python/py-param.c
index 0f8d9b6b42..77311ad1bd 100644
--- a/gdb/python/py-param.c
+++ b/gdb/python/py-param.c
@@ -381,15 +381,10 @@ get_set_value (const char *args, int from_tty,
 	  return;
 	}
     }
-  else
-    {
-      /* We have to preserve the existing < GDB 7.3 API.  If a
-	 callback function does not exist, then attempt to read the
-	 set_doc attribute.  */
-      set_doc_string  = get_doc_string (obj, set_doc_cst);
-    }
 
-  fprintf_filtered (gdb_stdout, "%s\n", set_doc_string.get ());
+  const char *str = set_doc_string.get ();
+  if (str != nullptr && str[0] != '\0')
+    fprintf_filtered (gdb_stdout, "%s\n", str);
 }
 
 /* A callback function that is registered against the respective
diff --git a/gdb/testsuite/gdb.python/py-parameter.exp b/gdb/testsuite/gdb.python/py-parameter.exp
index 3cd1198462..f52d737dbc 100644
--- a/gdb/testsuite/gdb.python/py-parameter.exp
+++ b/gdb/testsuite/gdb.python/py-parameter.exp
@@ -151,7 +151,7 @@ gdb_py_test_multiple "Simple gdb booleanparameter" \
    "end"
 
 gdb_test "show print test-nodoc-param" "This command is not documented.*" "show parameter on"
-gdb_test "set print test-nodoc-param off" "This command is not documented.*" "turn off parameter"
+gdb_test_no_output "set print test-nodoc-param off" "turn off parameter"
 gdb_test "show print test-nodoc-param" "This command is not documented.*.*" "show parameter off"
 gdb_test "python print (test_nodoc_param.value)" "False" "test parameter value"
 gdb_test "help show print test-nodoc-param" "This command is not documented.*" "test show help"
@@ -173,7 +173,7 @@ gdb_py_test_multiple "Simple gdb booleanparameter" \
 
 gdb_test "python print (test_param.value)" "True" "test parameter value"
 gdb_test "show print test-param" "State of the Test Parameter on.*" "show parameter on"
-gdb_test "set print test-param off" "Set the state of the Test Parameter.*" "turn off parameter"
+gdb_test_no_output "set print test-param off" "turn off parameter"
 gdb_test "show print test-param" "State of the Test Parameter off.*" "show parameter off"
 gdb_test "python print (test_param.value)" "False" "test parameter value"
 gdb_test "help show print test-param" "State of the Test Parameter.*" "test show help"
-- 
2.13.6

Re: [RFA] Fix "set" handling of Python parameters

[Eli Zaretskii]

> From: Tom Tromey <tom@tromey.com>
> Cc: Tom Tromey <tom@tromey.com>
> Date: Thu, 26 Apr 2018 16:51:40 -0600
> 
> +* Python API
> +
> +  ** A gdb.Parameter will no longer print the "set" help text on an
> +     ordinary "set"; instead by default a "set" will be silent unless
> +     the get_set_string method returns a non-empty string.

The code mentions compatibility with GDB < 7.3, so shouldn't NEWS at
least indicate that this breaks that compatibility?

> +If this method exists, @value{GDBN} will call it when a
> +@var{parameter}'s value has been changed via the @code{set} API (for
> +example, @kbd{set foo off}).  The @code{value} attribute has already
> +been populated with the new value and may be used in output.  This
> +method must return a string.  If the returned string is not empty,
> +@value{GDBN} will present it to the user.

So there's no way for a Python extension to display an empty string,
even if it wants to do that?

Anyway, the documentation parts are approved.  Thanks.

Re: [RFA] Fix "set" handling of Python parameters

[Phil Muldoon]

On 26/04/18 23:51, Tom Tromey wrote:

> diff --git a/gdb/python/py-param.c b/gdb/python/py-param.c
> index 0f8d9b6b42..77311ad1bd 100644
> --- a/gdb/python/py-param.c
> +++ b/gdb/python/py-param.c
> @@ -381,15 +381,10 @@ get_set_value (const char *args, int from_tty,
>  	  return;
>  	}
>      }
> -  else
> -    {
> -      /* We have to preserve the existing < GDB 7.3 API.  If a
> -	 callback function does not exist, then attempt to read the
> -	 set_doc attribute.  */
> -      set_doc_string  = get_doc_string (obj, set_doc_cst);
> -    }
>  
> -  fprintf_filtered (gdb_stdout, "%s\n", set_doc_string.get ());
> +  const char *str = set_doc_string.get ();
> +  if (str != nullptr && str[0] != '\0')
> +    fprintf_filtered (gdb_stdout, "%s\n", str);
>  }
>  
>  /* A callback function that is registered against the respective
> diff --git a/gdb/testsuite/gdb.python/py-parameter.exp b/gdb/testsuite/gdb.python/py-parameter.exp
> index 3cd1198462..f52d737dbc 100644
> --- a/gdb/testsuite/gdb.python/py-parameter.exp
> +++ b/gdb/testsuite/gdb.python/py-parameter.exp
> @@ -151,7 +151,7 @@ gdb_py_test_multiple "Simple gdb booleanparameter" \
>     "end"
>  
>  gdb_test "show print test-nodoc-param" "This command is not documented.*" "show parameter on"
> -gdb_test "set print test-nodoc-param off" "This command is not documented.*" "turn off parameter"
> +gdb_test_no_output "set print test-nodoc-param off" "turn off parameter"
>  gdb_test "show print test-nodoc-param" "This command is not documented.*.*" "show parameter off"
>  gdb_test "python print (test_nodoc_param.value)" "False" "test parameter value"
>  gdb_test "help show print test-nodoc-param" "This command is not documented.*" "test show help"
> @@ -173,7 +173,7 @@ gdb_py_test_multiple "Simple gdb booleanparameter" \
>  
>  gdb_test "python print (test_param.value)" "True" "test parameter value"
>  gdb_test "show print test-param" "State of the Test Parameter on.*" "show parameter on"
> -gdb_test "set print test-param off" "Set the state of the Test Parameter.*" "turn off parameter"
> +gdb_test_no_output "set print test-param off" "turn off parameter"
>  gdb_test "show print test-param" "State of the Test Parameter off.*" "show parameter off"
>  gdb_test "python print (test_param.value)" "False" "test parameter value"
>  gdb_test "help show print test-param" "State of the Test Parameter.*" "test show help"

I have no comments other than I read the patch and it LGTM.

Cheers
Phil

Re: [RFA] Fix "set" handling of Python parameters

[Tom Tromey]

>>>>> "Tom" == Tom Tromey <tom@tromey.com> writes:

Tom> It's long bothered me that setting a Python parameter from the CLI
Tom> will print the "set" help text by default.  I think usually "set"
Tom> commands should be silent.  And, while you can modify this behavior a
Tom> bit by providing a "get_set_string" method, if this method returns an
Tom> empty string, a blank line will be printed.

Tom> This patch removes the "help" behavior and changes the get_set_string
Tom> behavior to avoid printing a blank line.  The code has a comment about
Tom> preserving API behavior, but I don't think this is truly important;
Tom> and in any case the workaround -- implementing get_set_string -- is
Tom> trivial.

Tom> Regression tested on x86-64 Fedora 26.

Ping.

Tom

Re: [RFA] Fix "set" handling of Python parameters

[Joel Brobecker]

Hi Tom,

> gdb/ChangeLog
> 2018-04-26  Tom Tromey  <tom@tromey.com>
> 
> 	* NEWS: Mention new "set" behavior.
> 	* python/py-param.c (get_set_value): Don't print an empty string.
> 	Don't call get_doc_string.
> 
> gdb/doc/ChangeLog
> 2018-04-26  Tom Tromey  <tom@tromey.com>
> 
> 	* python.texi (Parameters In Python): Update get_set_string
> 	documentation.

So sorry again for the long delay in getting this patch reviewed.
Looks good!

-- 
Joel