diff mbox

PR driver/67613 - spell suggestions for misspelled command line options

Message ID 1446648825-40027-1-git-send-email-dmalcolm@redhat.com
State New
Headers show

Commit Message

David Malcolm Nov. 4, 2015, 2:53 p.m. UTC
This patch adds hints to the option-not-found error in the driver,
using the Levenshtein distance implementation posted here:
"[PATCH 0/2] Levenshtein-based suggestions (v3)"
  https://gcc.gnu.org/ml/gcc-patches/2015-10/msg03379.html

It splits out the identifier-based implementation into a new
spellcheck-tree.c, keeping the core in spellcheck.c, since the tree
checking code needed for IDENTIFIER_POINTER etc isn't available in
the driver.

Bootstrapped&regrtested on top of the other patch kit (with nit fixes)
on x86_64-pc-linux-gnu; adds 5 PASS results to gcc.sum.

OK for trunk as a followup to that patch kit?

gcc/ChangeLog:
	PR driver/67613
	* Makefile.in (GCC_OBJS): Add spellcheck.o.
	(OBJS): Add spellcheck-tree.o.
	* gcc.c: Include "spellcheck.h".
	(suggest_option): New function.
	(driver::handle_unrecognized_options): Call suggest_option to
	provide a hint about misspelled options.
	* spellcheck.c: Update file comment.
	(levenshtein_distance): Convert 4-param implementation from static
	to extern scope.  Remove note about unit tests from leading
	comment for const char * implementation.  Move tree
	implementation to...
	* spellcheck-tree.c: New file.
	* spellcheck.h (levenshtein_distance):  Add 4-param decl.

gcc/testsuite/ChangeLog:
	PR driver/67613
	* gcc/testsuite/gcc.dg/spellcheck-options-1.c: New file.
	* gcc/testsuite/gcc.dg/spellcheck-options-2.c: New file.
---
 gcc/Makefile.in                             |  3 +-
 gcc/gcc.c                                   | 51 ++++++++++++++++++++++++++++-
 gcc/spellcheck-tree.c                       | 39 ++++++++++++++++++++++
 gcc/spellcheck.c                            | 21 ++----------
 gcc/spellcheck.h                            |  4 +++
 gcc/testsuite/gcc.dg/spellcheck-options-1.c |  4 +++
 gcc/testsuite/gcc.dg/spellcheck-options-2.c |  5 +++
 7 files changed, 107 insertions(+), 20 deletions(-)
 create mode 100644 gcc/spellcheck-tree.c
 create mode 100644 gcc/testsuite/gcc.dg/spellcheck-options-1.c
 create mode 100644 gcc/testsuite/gcc.dg/spellcheck-options-2.c

Comments

Bernd Schmidt Nov. 6, 2015, 10:19 a.m. UTC | #1
On 11/04/2015 03:53 PM, David Malcolm wrote:
> This patch adds hints to the option-not-found error in the driver,
> using the Levenshtein distance implementation posted here:
> "[PATCH 0/2] Levenshtein-based suggestions (v3)"
>    https://gcc.gnu.org/ml/gcc-patches/2015-10/msg03379.html
>
> It splits out the identifier-based implementation into a new
> spellcheck-tree.c, keeping the core in spellcheck.c, since the tree
> checking code needed for IDENTIFIER_POINTER etc isn't available in
> the driver.

Nice. Ok.


Bernd
diff mbox

Patch

diff --git a/gcc/Makefile.in b/gcc/Makefile.in
index 9fb643e..a57bd40 100644
--- a/gcc/Makefile.in
+++ b/gcc/Makefile.in
@@ -1151,7 +1151,7 @@  CXX_TARGET_OBJS=@cxx_target_objs@
 FORTRAN_TARGET_OBJS=@fortran_target_objs@
 
 # Object files for gcc many-languages driver.
-GCC_OBJS = gcc.o gcc-main.o ggc-none.o
+GCC_OBJS = gcc.o gcc-main.o ggc-none.o spellcheck.o
 
 c-family-warn = $(STRICT_WARN)
 
@@ -1395,6 +1395,7 @@  OBJS = \
 	simplify-rtx.o \
 	sparseset.o \
 	spellcheck.o \
+	spellcheck-tree.o \
 	sreal.o \
 	stack-ptr-mod.o \
 	statistics.o \
diff --git a/gcc/gcc.c b/gcc/gcc.c
index bbc9b23..ef0fb4e 100644
--- a/gcc/gcc.c
+++ b/gcc/gcc.c
@@ -42,6 +42,7 @@  compilation is specified by a string called a "spec".  */
 #include "opts.h"
 #include "params.h"
 #include "filenames.h"
+#include "spellcheck.h"
 
 
 
@@ -7598,6 +7599,45 @@  driver::maybe_putenv_OFFLOAD_TARGETS () const
   offload_targets = NULL;
 }
 
+/* Helper function for driver::handle_unrecognized_options.
+
+   Given an unrecognized option BAD_OPT (without the leading dash),
+   locate the closest reasonable matching option (again, without the
+   leading dash), or NULL.  */
+
+static const char *
+suggest_option (const char *bad_opt)
+{
+  const cl_option *best_option = NULL;
+  edit_distance_t best_distance = MAX_EDIT_DISTANCE;
+
+  for (unsigned int i = 0; i < cl_options_count; i++)
+    {
+      edit_distance_t dist = levenshtein_distance (bad_opt,
+						   cl_options[i].opt_text + 1);
+      if (dist < best_distance)
+	{
+	  best_distance = dist;
+	  best_option = &cl_options[i];
+	}
+    }
+
+  if (!best_option)
+    return NULL;
+
+  /* If more than half of the letters were misspelled, the suggestion is
+     likely to be meaningless.  */
+  if (best_option)
+    {
+      unsigned int cutoff = MAX (strlen (bad_opt),
+				 strlen (best_option->opt_text + 1)) / 2;
+      if (best_distance > cutoff)
+	return NULL;
+    }
+
+  return best_option->opt_text + 1;
+}
+
 /* Reject switches that no pass was interested in.  */
 
 void
@@ -7605,7 +7645,16 @@  driver::handle_unrecognized_options () const
 {
   for (size_t i = 0; (int) i < n_switches; i++)
     if (! switches[i].validated)
-      error ("unrecognized command line option %<-%s%>", switches[i].part1);
+      {
+	const char *hint = suggest_option (switches[i].part1);
+	if (hint)
+	  error ("unrecognized command line option %<-%s%>;"
+		 " did you mean %<-%s%>?",
+		 switches[i].part1, hint);
+	else
+	  error ("unrecognized command line option %<-%s%>",
+		 switches[i].part1);
+      }
 }
 
 /* Handle the various -print-* options, returning 0 if the driver
diff --git a/gcc/spellcheck-tree.c b/gcc/spellcheck-tree.c
new file mode 100644
index 0000000..d203776
--- /dev/null
+++ b/gcc/spellcheck-tree.c
@@ -0,0 +1,39 @@ 
+/* Find near-matches for identifiers.
+   Copyright (C) 2015 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC 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, or (at your option) any later
+version.
+
+GCC 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 GCC; see the file COPYING3.  If not see
+<http://www.gnu.org/licenses/>.  */
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "tm.h"
+#include "tree.h"
+#include "spellcheck.h"
+
+/* Calculate Levenshtein distance between two identifiers.  */
+
+edit_distance_t
+levenshtein_distance (tree ident_s, tree ident_t)
+{
+  gcc_assert (TREE_CODE (ident_s) == IDENTIFIER_NODE);
+  gcc_assert (TREE_CODE (ident_t) == IDENTIFIER_NODE);
+
+  return levenshtein_distance (IDENTIFIER_POINTER (ident_s),
+			       IDENTIFIER_LENGTH (ident_s),
+			       IDENTIFIER_POINTER (ident_t),
+			       IDENTIFIER_LENGTH (ident_t));
+}
diff --git a/gcc/spellcheck.c b/gcc/spellcheck.c
index 4d4c176..31fd22b 100644
--- a/gcc/spellcheck.c
+++ b/gcc/spellcheck.c
@@ -1,4 +1,4 @@ 
-/* Find near-matches for strings and identifiers.
+/* Find near-matches for strings.
    Copyright (C) 2015 Free Software Foundation, Inc.
 
 This file is part of GCC.
@@ -30,7 +30,7 @@  along with GCC; see the file COPYING3.  If not see
 
    This implementation uses the Wagner-Fischer algorithm.  */
 
-static edit_distance_t
+edit_distance_t
 levenshtein_distance (const char *s, int len_s,
 		      const char *t, int len_t)
 {
@@ -112,25 +112,10 @@  levenshtein_distance (const char *s, int len_s,
   return result;
 }
 
-/* Calculate Levenshtein distance between two nil-terminated strings.
-   This exists purely for the unit tests.  */
+/* Calculate Levenshtein distance between two nil-terminated strings.  */
 
 edit_distance_t
 levenshtein_distance (const char *s, const char *t)
 {
   return levenshtein_distance (s, strlen (s), t, strlen (t));
 }
-
-/* Calculate Levenshtein distance between two identifiers.  */
-
-edit_distance_t
-levenshtein_distance (tree ident_s, tree ident_t)
-{
-  gcc_assert (TREE_CODE (ident_s) == IDENTIFIER_NODE);
-  gcc_assert (TREE_CODE (ident_t) == IDENTIFIER_NODE);
-
-  return levenshtein_distance (IDENTIFIER_POINTER (ident_s),
-			       IDENTIFIER_LENGTH (ident_s),
-			       IDENTIFIER_POINTER (ident_t),
-			       IDENTIFIER_LENGTH (ident_t));
-}
diff --git a/gcc/spellcheck.h b/gcc/spellcheck.h
index 58355d6..673a756 100644
--- a/gcc/spellcheck.h
+++ b/gcc/spellcheck.h
@@ -24,6 +24,10 @@  typedef unsigned int edit_distance_t;
 const edit_distance_t MAX_EDIT_DISTANCE = UINT_MAX;
 
 extern edit_distance_t
+levenshtein_distance (const char *s, int len_s,
+		      const char *t, int len_t);
+
+extern edit_distance_t
 levenshtein_distance (const char *s, const char *t);
 
 extern edit_distance_t
diff --git a/gcc/testsuite/gcc.dg/spellcheck-options-1.c b/gcc/testsuite/gcc.dg/spellcheck-options-1.c
new file mode 100644
index 0000000..cd5fdca
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/spellcheck-options-1.c
@@ -0,0 +1,4 @@ 
+/* { dg-do compile } */
+/* { dg-options "-Wcoercion" } */
+/* { dg-error "unrecognized command line option '-Wcoercion'; did you mean '-Wconversion'?"  "" { target *-*-* } 0 } */
+
diff --git a/gcc/testsuite/gcc.dg/spellcheck-options-2.c b/gcc/testsuite/gcc.dg/spellcheck-options-2.c
new file mode 100644
index 0000000..786266d
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/spellcheck-options-2.c
@@ -0,0 +1,5 @@ 
+/* { dg-do compile } */
+/* { dg-options "-Wthis-should-not-get-a-hint" } */
+/* { dg-bogus "did you mean" "" { target *-*-* } 0 } */
+/* { dg-error "unrecognized command line option '-Wthis-should-not-get-a-hint'"  "" { target *-*-* } 0 } */
+