[Ada] Fix completion for multiple function matches

Message ID 55E55526.2000303@adacore.com
State New, archived
Headers

Commit Message

Pierre-Marie de Rodat Sept. 1, 2015, 7:35 a.m. UTC
  On 08/31/2015 07:25 PM, Joel Brobecker wrote:
> Overall, the patch looks like definite progress as is, and I'd like
> to see it pushed to avoid both the unsollicited MCQ and the ensuing
> SEGV. However, I'm wondering if it might still be leaving the hole
> in the completion.
>
> In particular, what happens to the completion of "p ambig" if
> you add a second function called, say "Aux_Pck.Ambiguous_Prog"?
> In other words, I am wondering if the fact that we're returning
> zero when parsing for completion might not unexpectedly filter out
> some matches. Regardless of the answer, that's probably a worthwhile
> adjustment to the tests you propose.

I adjusted the testcase in the way I think you meant: I added the 
Ambiguous_Prog procedure to Aux_Pck and added a test to ask the 
completion of "p ambig": GDB correctly reports two possible completions: 
"p ambiguous_func" and "p ambiguous_proc". This looks independent of 
function resolution in ada-lang, and that is fortunate in our case. :-)

Is this what you had in mind?

> So, here's the plan I propose: We adjust the very minor suggestions
> I made below, adjust the testcase as suggested above, do a quick
> second round of review and then push the patch. If if turns out
> there is a hole (I am not sure), then we can put that on the list
> of things to improve and decide when we want to look at it.

Sure: here's the updated patch, re-regtested on x86_64-linux.

>> +  /* If we got multiple matches, ask the user which one to use.  Don't do this
>> +     interactive thing during completion, though: it is irritating to get a
>> +     prompt in this context.  */
>
> I wouldn't say "irritating" as, to me, this is a bug, not an
> inconvenience. So, I would change the comment to say something like:
>
>    /* If we got multiple matches, ask the user which one to use.  Don't do this
>       interactive thing during completion, though, as the purpose of the
>       completion is providing a list of all possible matches.  Prompting the
>       user to filter it down would be completely unexpected in this case.  */

Done.

>> +# Usually, parsing a function name that is ambiguous yields a menu through
>> +# which users can select a specific function. This should not happen during
>> +# completion, though.
>
> Missing second space after a period.

Done.

>> +test_gdb_complete "ambig" \
>> +                  "p ambiguous_func"
>> +test_gdb_complete "ambiguous_func" \
>> +                  "p ambiguous_func"
>> diff --git a/gdb/testsuite/gdb.ada/complete/aux_pck.adb b/gdb/testsuite/gdb.ada/complete/aux_pck.adb
>> new file mode 100644
>> index 0000000..18e060e
>> --- /dev/null
>> +++ b/gdb/testsuite/gdb.ada/complete/aux_pck.adb
>> @@ -0,0 +1,23 @@
>> +--  Copyright 2008-2015 Free Software Foundation, Inc.
>
> I *think* you want to only list 2015 in this case. Unless I missed
> the fact that you copied this file from elsewhere...

Copy-paste error, indeed. ;-) Done.

>> diff --git a/gdb/testsuite/gdb.ada/complete/aux_pck.ads b/gdb/testsuite/gdb.ada/complete/aux_pck.ads
>> new file mode 100644
>> index 0000000..36fdb6c
>> --- /dev/null
>> +++ b/gdb/testsuite/gdb.ada/complete/aux_pck.ads
>> @@ -0,0 +1,20 @@
>> +--  Copyright 2008-2015 Free Software Foundation, Inc.
>
> Same here.

Likewise.

Thank you for reviewing!
  

Comments

Joel Brobecker Sept. 1, 2015, 12:15 p.m. UTC | #1
> I adjusted the testcase in the way I think you meant: I added the
> Ambiguous_Prog procedure to Aux_Pck and added a test to ask the completion
> of "p ambig": GDB correctly reports two possible completions: "p
> ambiguous_func" and "p ambiguous_proc". This looks independent of function
> resolution in ada-lang, and that is fortunate in our case. :-)
> 
> Is this what you had in mind?

Yep :)

> gdb/ChangeLog:
> 
> 	* ada-lang.c (ada_resolve_function): Do not ask the user what
> 	match to use when in completion mode.
> 
> gdb/testsuite/ChangeLog:
> 
> 	* gdb.ada/complete.exp: Add "pck.ambiguous_func" to the relevant
> 	expected outputs.  Add two testcases for completing ambiguous
> 	functions.
> 	* gdb.ada/complete/aux_pck.adb: New file.
> 	* gdb.ada/complete/aux_pck.ads: New file.
> 	* gdb.ada/complete/foo.adb: Pull Aux_Pck and call the two
> 	Ambiguous_Func functions.
> 	* gdb.ada/complete/pck.ads: Add an Ambiguous_Func function.
> 	* gdb.ada/complete/pck.adb: Likewise.

Looks good. Go ahead and push :).

Thank you, Pierre-Marie.
  
Pierre-Marie de Rodat Sept. 1, 2015, 12:54 p.m. UTC | #2
On 09/01/2015 02:15 PM, Joel Brobecker wrote:
>> Is this what you had in mind?
>
> Yep :)

Cool!

> Looks good. Go ahead and push :).
>
> Thank you, Pierre-Marie.

This is pushed. Thank you very much!
  

Patch

From 1ce5e145bd6147507fcae4fb614b6cb4fdefd2bc Mon Sep 17 00:00:00 2001
From: Pierre-Marie de Rodat <derodat@adacore.com>
Date: Mon, 31 Aug 2015 16:04:07 +0200
Subject: [PATCH] [Ada] Fix completion for multiple function matches

Before this change, trying to complete an expression ending with an
ambiguous function name (i.e. for which there are multiple matches)
would display a menu with a prompt for the user to pick one. For
instance:

    (gdb) p func<tab>Multiple matches for func
    [0] cancel
    [1] pack2.func at pack2.adb:5
    [2] pack.func at pack.adb:5
    >

This is not user friendly and actually triggered a segmentation fault
after the user did pick one. It is not clear whether the segmentation
fault needs a separate fix, but this is the only known case which
exhibits it at the moment, and this case must be fixed itself.

The problem lies in ada-lang.c (ada_resolve_function): when we got
multiple matches, we should not display the menu if we are in completion
mode. This patch adjusts the corresponding condition accordingly.

gdb/ChangeLog:

	* ada-lang.c (ada_resolve_function): Do not ask the user what
	match to use when in completion mode.

gdb/testsuite/ChangeLog:

	* gdb.ada/complete.exp: Add "pck.ambiguous_func" to the relevant
	expected outputs.  Add two testcases for completing ambiguous
	functions.
	* gdb.ada/complete/aux_pck.adb: New file.
	* gdb.ada/complete/aux_pck.ads: New file.
	* gdb.ada/complete/foo.adb: Pull Aux_Pck and call the two
	Ambiguous_Func functions.
	* gdb.ada/complete/pck.ads: Add an Ambiguous_Func function.
	* gdb.ada/complete/pck.adb: Likewise.

Tested on x86_64-linux, no regression.
---
 gdb/ada-lang.c                             |  6 +++++-
 gdb/testsuite/gdb.ada/complete.exp         | 13 +++++++++++++
 gdb/testsuite/gdb.ada/complete/aux_pck.adb | 28 ++++++++++++++++++++++++++++
 gdb/testsuite/gdb.ada/complete/aux_pck.ads | 21 +++++++++++++++++++++
 gdb/testsuite/gdb.ada/complete/foo.adb     |  6 +++++-
 gdb/testsuite/gdb.ada/complete/pck.adb     |  5 +++++
 gdb/testsuite/gdb.ada/complete/pck.ads     |  2 ++
 7 files changed, 79 insertions(+), 2 deletions(-)
 create mode 100644 gdb/testsuite/gdb.ada/complete/aux_pck.adb
 create mode 100644 gdb/testsuite/gdb.ada/complete/aux_pck.ads

diff --git a/gdb/ada-lang.c b/gdb/ada-lang.c
index 7e6b6dc..a7809ff 100644
--- a/gdb/ada-lang.c
+++ b/gdb/ada-lang.c
@@ -3655,9 +3655,13 @@  ada_resolve_function (struct block_symbol syms[],
         }
     }
 
+  /* If we got multiple matches, ask the user which one to use.  Don't do this
+     interactive thing during completion, though, as the purpose of the
+     completion is providing a list of all possible matches.  Prompting the
+     user to filter it down would be completely unexpected in this case.  */
   if (m == 0)
     return -1;
-  else if (m > 1)
+  else if (m > 1 && !parse_completion)
     {
       printf_filtered (_("Multiple matches for %s\n"), name);
       user_select_syms (syms, m, 1);
diff --git a/gdb/testsuite/gdb.ada/complete.exp b/gdb/testsuite/gdb.ada/complete.exp
index 9919bdf..0c4f03f 100644
--- a/gdb/testsuite/gdb.ada/complete.exp
+++ b/gdb/testsuite/gdb.ada/complete.exp
@@ -140,6 +140,7 @@  test_gdb_complete "external_ident" \
 test_gdb_complete "pck" \
                   [multi_line "(p pck\\.ad\[sb\])?" \
                               "(p pck\\.ad\[sb\])?" \
+                              "p pck.ambiguous_func" \
                               "p pck.external_identical_one" \
                               "p pck.inner.inside_variable" \
                               "p pck.local_identical_one" \
@@ -151,6 +152,7 @@  test_gdb_complete "pck" \
 test_gdb_complete "pck." \
                   [multi_line "(p pck\\.ad\[sb\])?" \
                               "(p pck\\.ad\[sb\])?" \
+                              "p pck.ambiguous_func" \
                               "p pck.external_identical_one" \
                               "p pck.inner.inside_variable" \
                               "p pck.local_identical_one" \
@@ -181,3 +183,14 @@  if { [readline_is_used] } {
 	}
     }
 }
+
+# Usually, parsing a function name that is ambiguous yields a menu through
+# which users can select a specific function.  This should not happen during
+# completion, though.
+test_gdb_complete "ambig" \
+                  [multi_line "p ambiguous_func" \
+                              "p ambiguous_proc" ]
+test_gdb_complete "ambiguous_f" \
+                  "p ambiguous_func"
+test_gdb_complete "ambiguous_func" \
+                  "p ambiguous_func"
diff --git a/gdb/testsuite/gdb.ada/complete/aux_pck.adb b/gdb/testsuite/gdb.ada/complete/aux_pck.adb
new file mode 100644
index 0000000..b15912a
--- /dev/null
+++ b/gdb/testsuite/gdb.ada/complete/aux_pck.adb
@@ -0,0 +1,28 @@ 
+--  Copyright 2015 Free Software Foundation, Inc.
+--
+--  This program is free software; you can redistribute it and/or modify
+--  it under the terms of the GNU General Public License as published by
+--  the Free Software Foundation; either version 3 of the License, or
+--  (at your option) any later version.
+--
+--  This program is distributed in the hope that it will be useful,
+--  but WITHOUT ANY WARRANTY; without even the implied warranty of
+--  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+--  GNU General Public License for more details.
+--
+--  You should have received a copy of the GNU General Public License
+--  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package body Aux_Pck is
+
+   procedure Ambiguous_Func is
+   begin
+      null;
+   end Ambiguous_Func;
+
+   procedure Ambiguous_Proc is
+   begin
+      null;
+   end Ambiguous_Proc;
+
+end Aux_Pck;
diff --git a/gdb/testsuite/gdb.ada/complete/aux_pck.ads b/gdb/testsuite/gdb.ada/complete/aux_pck.ads
new file mode 100644
index 0000000..f3476c6
--- /dev/null
+++ b/gdb/testsuite/gdb.ada/complete/aux_pck.ads
@@ -0,0 +1,21 @@ 
+--  Copyright 2015 Free Software Foundation, Inc.
+--
+--  This program is free software; you can redistribute it and/or modify
+--  it under the terms of the GNU General Public License as published by
+--  the Free Software Foundation; either version 3 of the License, or
+--  (at your option) any later version.
+--
+--  This program is distributed in the hope that it will be useful,
+--  but WITHOUT ANY WARRANTY; without even the implied warranty of
+--  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+--  GNU General Public License for more details.
+--
+--  You should have received a copy of the GNU General Public License
+--  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+package Aux_Pck is
+
+   procedure Ambiguous_Func;
+   procedure Ambiguous_Proc;
+
+end Aux_Pck;
diff --git a/gdb/testsuite/gdb.ada/complete/foo.adb b/gdb/testsuite/gdb.ada/complete/foo.adb
index 58d0ee3..98f1c1e 100644
--- a/gdb/testsuite/gdb.ada/complete/foo.adb
+++ b/gdb/testsuite/gdb.ada/complete/foo.adb
@@ -13,7 +13,8 @@ 
 --  You should have received a copy of the GNU General Public License
 --  along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
-with Pck; use Pck;
+with Aux_Pck; use Aux_Pck;
+with Pck;     use Pck;
 
 procedure Foo is
    Some_Local_Variable : Integer := 1;
@@ -21,5 +22,8 @@  procedure Foo is
 begin
    My_Global_Variable := Some_Local_Variable + 1; -- START
    Proc (External_Identical_Two);
+   Aux_Pck.Ambiguous_Func;
+   Aux_Pck.Ambiguous_Proc;
+   Pck.Ambiguous_Func;
 end Foo;
 
diff --git a/gdb/testsuite/gdb.ada/complete/pck.adb b/gdb/testsuite/gdb.ada/complete/pck.adb
index 51dd725..10680dc 100644
--- a/gdb/testsuite/gdb.ada/complete/pck.adb
+++ b/gdb/testsuite/gdb.ada/complete/pck.adb
@@ -21,4 +21,9 @@  package body Pck is
       Inner.Inside_Variable := Not_In_Scope + I;
    end Proc;
 
+   procedure Ambiguous_Func is
+   begin
+      null;
+   end Ambiguous_Func;
+
 end Pck;
diff --git a/gdb/testsuite/gdb.ada/complete/pck.ads b/gdb/testsuite/gdb.ada/complete/pck.ads
index ab2c47b..e85f566 100644
--- a/gdb/testsuite/gdb.ada/complete/pck.ads
+++ b/gdb/testsuite/gdb.ada/complete/pck.ads
@@ -31,4 +31,6 @@  package Pck is
 
    procedure Proc (I : Integer);
 
+   procedure Ambiguous_Func;
+
 end Pck;
-- 
2.5.1