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: [PATCH] [Ada] Fix completion for multiple function matches


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!

--
Pierre-Marie de Rodat
>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


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