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 2/2] namespace support


Tom Tromey wrote:
"Sami" == Sami Wagiaalla <swagiaal@redhat.com> writes:

Sami> Import data structure ( struct using_direct) is limited to Sami> representing import statements where the destination is a parent Sami> of the source (true for anonymous namespaces). This patch removes Sami> this limitation, allowing each import to represent independent Sami> import source and destination.

Thanks.

Aside from some nits this looks pretty good.

Is this just a "technical" patch?  Or does it enable some new
user-visible functionality?  The former is fine, but if it is the
latter, how about a test case?


I have added a test case.


Sami> +	* cp-support.h (cp_add_using, cp_add_using_directive): Now take char*
Sami> +	inner, char* outer arguments. Updated callers.

I suppose these arguments ought to be renamed, now that you have
renamed the struct fields...


Agreed. Is it cool if I add this to the struct field rename patch.



Space before open paren.


Sami> -	      cp_add_using_directive (name,
Sami> -				      previous_component == 0
Sami> -				      ? 0 : previous_component - 2,
Sami> -				      next_component);
Sami> +	      cp_add_using_directive (outer, name);

Do we know for sure that NAME is \0-terminated here?
It needs to be due to the change to cp_add_using_directive.


I am sure it is. However, after looking further into it I realized that the desired name does not always end at the null terminator, but it can also end at the first occurrence of '(' or '<'. So I changed the patch to be more compatible with the old code. Please see patch.






2009-06-24 Sami Wagiaalla <swagiaal@redhat.com>


     	* dwarf2read.c (read_import_statement): Properly set import location
     	and destination.
     	* cp-support.h (cp_add_using, cp_add_using_directive): Now take char*
     	inner, char* outer arguments. Updated callers.

2009-06-24 Sami Wagiaalla <swagiaal@redhat.com>

     	* gdb.cp/namespace-nested-import.cc: New test.
     	* gdb.cp/namespace-nested-import.exp: New test.



diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index 95d41e5..46d0ef5 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,10 @@
+2009-06-24  Sami Wagiaalla  <swagiaal@redhat.com>
+
+ 	* dwarf2read.c (read_import_statement): Properly set import location
+ 	and destination.
+ 	* cp-support.h (cp_add_using, cp_add_using_directive): Now take char*
+ 	inner, char* outer arguments. Updated callers.
+
 2009-06-23  Ulrich Weigand  <uweigand@de.ibm.com>

 	* valops.c (value_one): Reimplement broken decimal-float case.
diff --git a/gdb/cp-namespace.c b/gdb/cp-namespace.c
index f3b2194..a4ad62a 100644
--- a/gdb/cp-namespace.c
+++ b/gdb/cp-namespace.c
@@ -103,14 +103,22 @@ cp_scan_for_anonymous_namespaces (const struct symbol *symbol)
 			  "(anonymous namespace)",
 			  ANONYMOUS_NAMESPACE_LEN) == 0)
 	    {
+	      int outer_len = (previous_component == 0 ? 0 : previous_component - 2);
+	      int inner_len = next_component;
+	      char *outer = alloca (outer_len + 1);
+	      char *inner = alloca (inner_len + 1);
+
+	      strncpy(outer, name, outer_len);
+	      strncpy(inner, name, inner_len);
+
+	      outer[outer_len] = '\0';
+	      inner[inner_len] = '\0';
+
 	      /* We've found a component of the name that's an
 		 anonymous namespace.  So add symbols in it to the
 		 namespace given by the previous component if there is
 		 one, or to the global namespace if there isn't.  */
-	      cp_add_using_directive (name,
-				      previous_component == 0
-				      ? 0 : previous_component - 2,
-				      next_component);
+	      cp_add_using_directive (outer, inner);
 	    }
 	  /* The "+ 2" is for the "::".  */
 	  previous_component = next_component + 2;
@@ -121,16 +129,11 @@ cp_scan_for_anonymous_namespaces (const struct symbol *symbol)
     }
 }

-/* Add a using directive to using_list.  NAME is the start of a string
-   that should contain the namespaces we want to add as initial
-   substrings, OUTER_LENGTH is the end of the outer namespace, and
-   INNER_LENGTH is the end of the inner namespace.  If the using
-   directive in question has already been added, don't add it
-   twice.  */
+/* Add a using directive to using_list. If the using directive in question
+   has already been added, don't add it twice.  */

 void
-cp_add_using_directive (const char *name, unsigned int outer_length,
-			unsigned int inner_length)
+cp_add_using_directive (const char *outer, const char *inner)
 {
   struct using_direct *current;
   struct using_direct *new;
@@ -139,14 +142,13 @@ cp_add_using_directive (const char *name, unsigned int outer_length,

   for (current = using_directives; current != NULL; current = current->next)
     {
-      if ((strncmp (current->inner, name, inner_length) == 0)
-	  && (strlen (current->inner) == inner_length)
-	  && (strlen (current->outer) == outer_length))
+      if (strcmp (current->inner, inner) == 0
+          && strcmp (current->outer, outer) == 0)
 	return;
     }

-  using_directives = cp_add_using (name, inner_length, outer_length,
-                                   using_directives);
+  using_directives = cp_add_using (outer, inner, using_directives);
+
 }

 /* Record the namespace that the function defined by SYMBOL was
@@ -197,26 +199,22 @@ cp_is_anonymous (const char *namespace)
 	  != NULL);
 }

-/* Create a new struct using direct whose inner namespace is the
-   initial substring of NAME of leng INNER_LEN and whose outer
-   namespace is the initial substring of NAME of length OUTER_LENGTH.
+/* Create a new struct using direct whose inner namespace is INNER
+   and whose outer namespace is OUTER.
    Set its next member in the linked list to NEXT; allocate all memory
    using xmalloc.  It copies the strings, so NAME can be a temporary
    string.  */

 struct using_direct *
-cp_add_using (const char *name,
-	      unsigned int inner_len,
-	      unsigned int outer_len,
+cp_add_using (const char *outer,
+              const char *inner,
 	      struct using_direct *next)
 {
   struct using_direct *retval;

-  gdb_assert (outer_len < inner_len);
-
   retval = xmalloc (sizeof (struct using_direct));
-  retval->inner = savestring (name, inner_len);
-  retval->outer = savestring (name, outer_len);
+  retval->inner = savestring (inner, strlen(inner));
+  retval->outer = savestring (outer, strlen(outer));
   retval->next = next;

   return retval;
diff --git a/gdb/cp-support.h b/gdb/cp-support.h
index e577f7d..ea36608 100644
--- a/gdb/cp-support.h
+++ b/gdb/cp-support.h
@@ -76,13 +76,11 @@ extern struct type *cp_lookup_rtti_type (const char *name,

extern int cp_is_anonymous (const char *namespace);

-extern void cp_add_using_directive (const char *name,
-				    unsigned int outer_length,
-				    unsigned int inner_length);
+extern void cp_add_using_directive (const char *outer,
+                                    const char *inner);

-extern struct using_direct *cp_add_using (const char *name,
-					  unsigned int inner_len,
-					  unsigned int outer_len,
+extern struct using_direct *cp_add_using (const char *outer,
+                                          const char *inner,
 					  struct using_direct *next);

 extern void cp_initialize_namespace (void);
diff --git a/gdb/dwarf2read.c b/gdb/dwarf2read.c
index f5901c7..a2f2e81 100644
--- a/gdb/dwarf2read.c
+++ b/gdb/dwarf2read.c
@@ -3046,6 +3046,9 @@ read_import_statement (struct die_info *die, struct dwarf2_cu *cu)
   struct attribute *import_attr;
   struct die_info *imported_die;
   const char *imported_name;
+  const char *imported_name_prefix;
+  const char *import_prefix;
+  char *canonical_name;

   import_attr = dwarf2_attr (die, DW_AT_import, cu);
   if (import_attr == NULL)
@@ -3096,8 +3099,27 @@ read_import_statement (struct die_info *die, struct dwarf2_cu *cu)

/* FIXME: dwarf2_name (die); for the local name after import. */

-  using_directives = cp_add_using (imported_name, strlen (imported_name), 0,
-                                   using_directives);
+  /* Figure out where the statement is being imported to.  */
+  import_prefix = determine_prefix (die, cu);
+
+  /* Figure out what the scope of the imported die is and prepend it
+     to the name of the imported die.  */
+  imported_name_prefix = determine_prefix (imported_die, cu);
+
+  if (strlen (imported_name_prefix) > 0)
+    {
+      canonical_name = alloca (strlen (imported_name_prefix) + 2 + strlen (imported_name) + 1);
+      strcpy (canonical_name, imported_name_prefix);
+      strcat (canonical_name, "::");
+      strcat (canonical_name, imported_name);
+    }
+  else
+    {
+      canonical_name = alloca (strlen (imported_name) + 1);
+      strcpy (canonical_name, imported_name);
+    }
+
+  using_directives = cp_add_using (import_prefix,canonical_name, using_directives);
 }

 static void
@@ -5037,9 +5059,7 @@ read_namespace (struct die_info *die, struct dwarf2_cu *cu)
       if (is_anonymous)
 	{
 	  const char *previous_prefix = determine_prefix (die, cu);
-	  cp_add_using_directive (TYPE_NAME (type),
-				  strlen (previous_prefix),
-				  strlen (TYPE_NAME (type)));
+	  cp_add_using_directive (previous_prefix, TYPE_NAME (type));
 	}
     }

diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog
index 8732519..a8de198 100644
--- a/gdb/testsuite/ChangeLog
+++ b/gdb/testsuite/ChangeLog
@@ -1,3 +1,8 @@
+2009-06-24  Sami Wagiaalla  <swagiaal@redhat.com>
+
+ 	* gdb.cp/namespace-nested-import.cc: New test.
+ 	* gdb.cp/namespace-nested-import.exp: New test.
+
 2009-06-23  Sami Wagiaalla  <swagiaal@redhat.com>

 	* gdb.cp/namespace-using.exp: New test.
diff --git a/gdb/testsuite/gdb.cp/namespace-nested-import.cc b/gdb/testsuite/gdb.cp/namespace-nested-import.cc
new file mode 100644
index 0000000..be07311
--- /dev/null
+++ b/gdb/testsuite/gdb.cp/namespace-nested-import.cc
@@ -0,0 +1,12 @@
+namespace A{
+  namespace B{
+    namespace C{
+      int       x = 5;
+    }
+  }
+}
+
+int main(){
+  using namespace A::B;
+  return C::x;
+}
\ No newline at end of file
diff --git a/gdb/testsuite/gdb.cp/namespace-nested-import.exp b/gdb/testsuite/gdb.cp/namespace-nested-import.exp
new file mode 100644
index 0000000..986c109
--- /dev/null
+++ b/gdb/testsuite/gdb.cp/namespace-nested-import.exp
@@ -0,0 +1,48 @@
+# Copyright 2008 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/>.
+
+if $tracelevel then {
+    strace $tracelevel
+}
+
+set prms_id 0
+set bug_id 0
+
+set testfile namespace-nested-import
+set srcfile ${testfile}.cc
+set binfile ${objdir}/${subdir}/${testfile}
+if  { [gdb_compile "${srcdir}/${subdir}/${srcfile}" "${binfile}" executable {debug c++}] != "" } {
+    untested "Couldn't compile test program"
+    return -1
+}
+
+# Get things started.
+
+gdb_exit
+gdb_start
+gdb_reinitialize_dir $srcdir/$subdir
+gdb_load ${binfile}
+
+############################################
+# Test printing of a variable from a nested
+# in a namespace inner to the one which has
+# been imported.
+
+if ![runto_main] then {
+    perror "couldn't run to breakpoint main"
+    continue
+}
+
+gdb_test "print C::x" "= 5"
\ No newline at end of file



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