Change type of struct complaints::series

Message ID 1438811079-25809-1-git-send-email-simon.marchi@ericsson.com
State New, archived
Headers

Commit Message

Simon Marchi Aug. 5, 2015, 9:44 p.m. UTC
  Found while processing the C++ enum changes.  It seems like series
should be of type enum complaint_series, instead of adding a cast.

If somebody could guide me a little bit, I think I could clean up the
comments around this.  The comments on enum complaint_series and the
series field seem redundant and maybe out of date.  Do the "case N" in
the comments refer to the enum values?  If so they don't seem to match
the actual values.

gdb/ChangeLog:

	* complaints.c (enum complaint_series): Add newlines.
	(struct complaints) <series>: Change type to enum
	complaint_series.
	(symfile_complaint_hook): Use equivalent enum value
	ISOLATED_MESSAGE instead of 0.
---
 gdb/complaints.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
  

Comments

Pedro Alves Aug. 6, 2015, 8:22 a.m. UTC | #1
On 08/05/2015 10:44 PM, Simon Marchi wrote:
> Found while processing the C++ enum changes.  It seems like series
> should be of type enum complaint_series, instead of adding a cast.
> 
> If somebody could guide me a little bit, I think I could clean up the
> comments around this.  The comments on enum complaint_series and the
> series field seem redundant and maybe out of date.  Do the "case N" in
> the comments refer to the enum values?  If so they don't seem to match
> the actual values.

'git blame' is your friend.  :-)  Seems to me they do refer to
the enum values -- see b9caf505.  That commit added the enum, and moved around
the "case N" comments.  But, it also added a fourth state, and missed updating
the N numbers to account for it.  E.g., "case 2: Subsequent message" is a typo
that should say "case 3".  Definitely agreed that all these comments
are redundant.

> 
> gdb/ChangeLog:
> 
> 	* complaints.c (enum complaint_series): Add newlines.
> 	(struct complaints) <series>: Change type to enum
> 	complaint_series.
> 	(symfile_complaint_hook): Use equivalent enum value
> 	ISOLATED_MESSAGE instead of 0.

This is OK.

Thanks,
Pedro Alves
  

Patch

diff --git a/gdb/complaints.c b/gdb/complaints.c
index dbacb2a..3566904 100644
--- a/gdb/complaints.c
+++ b/gdb/complaints.c
@@ -34,11 +34,14 @@  extern void _initialize_complaints (void);
 enum complaint_series {
   /* Isolated self explanatory message.  */
   ISOLATED_MESSAGE,
+
   /* First message of a series, includes an explanation.  */
   FIRST_MESSAGE,
+
   /* First message of a series, but does not need to include any sort
      of explanation.  */
   SHORT_FIRST_MESSAGE,
+
   /* Subsequent message of a series that needs no explanation (the
      user already knows we have a problem so we can just state our
      piece).  */
@@ -75,7 +78,7 @@  struct complaints
      must start off with explanation.  case 2: Subsequent message of a
      series that needs no explanation (the user already knows we have
      a problem so we can just state our piece).  */
-  int series;
+  enum complaint_series series;
 
   /* The explanatory messages that should accompany the complaint.
      NOTE: cagney/2002-08-14: In a desperate attempt at being vaguely
@@ -99,7 +102,7 @@  static struct explanation symfile_explanations[] = {
 
 static struct complaints symfile_complaint_book = {
   &complaint_sentinel,
-  0,
+  ISOLATED_MESSAGE,
   symfile_explanations
 };
 struct complaints *symfile_complaints = &symfile_complaint_book;