Re: [PATCH] Remove redundant assignments
From: Junio C Hamano <hidden>
Date: 2016-06-15 22:52:32
Jan Stępień [off-list ref] writes:
There were two redundant variable assignments in transport.c and wt-status.c. Removing them hasn't introduced any compiler warnings or regressions.
It may not have added any warnings for _you_. These "type a = a;" initializations (they are not even assignments and would not result in any code in the output for sane compilers) are idioms to squelch "unused" warnings from versions of compiler whose data flow analysis is less than perfect by telling them that the author of the section of the code knows what s/he is doing. At least, that is why these pseudo initializations were added in these codepaths. I personally feel that they should be made into useless but real assignments (which may result in a useless assignment emited in the resulting object code) rather than removing like your patch did which unfortunately forces people with imperfect compilers to live with the warnings, if one finds them ugly.
quoted hunk
Signed-off-by: Jan Stępień <redacted> --- transport.c | 2 +- wt-status.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)diff --git a/transport.c b/transport.c index 51814b5..5143718 100644 --- a/transport.c +++ b/transport.c@@ -105,7 +105,7 @@ static void insert_packed_refs(const char *packed_refs, struct ref **list) return; for (;;) { - int cmp = cmp, len; + int cmp, len; if (!fgets(buffer, sizeof(buffer), f)) { fclose(f);diff --git a/wt-status.c b/wt-status.c index 70fdb76..35f61f4 100644 --- a/wt-status.c +++ b/wt-status.c@@ -229,7 +229,7 @@ static void wt_status_print_change_data(struct wt_status *s, { struct wt_status_change_data *d = it->util; const char *c = color(change_type, s); - int status = status; + int status; char *one_name; char *two_name; const char *one, *two;