[Patch] Fix meaningless (char (*)[13]) to (char *) cast
 

Not a bug, but to fix a code readability problem.

A pointer to array is rarely useful in C. Here, passing m_start_time instead of &m_start_time as function arguments is what the code intended.

(In prime95 / mprime "menu.c")
[CODE]diff -r -u a/linux/menu.c b/linux/menu.c
--- a/linux/menu.c 2016-09-15 10:15:10.000000000 +0800
+++ b/linux/menu.c 2017-02-22 23:30:39.806257928 +0800
@@ -866,8 +866,8 @@
askNum ("Daytime P-1/ECM stage 2 memory in MB", &m_day_memory, 8, max_mem);
askNum ("Nighttime P-1/ECM stage 2 memory in MB", &m_night_memory, 8, max_mem);
if (m_day_memory != m_night_memory) {
- askStr ("Daytime begins at", (char *) &m_start_time, 12);
- askStr ("Daytime ends at", (char *) &m_end_time, 12);
+ askStr ("Daytime begins at", m_start_time, 12);
+ askStr ("Daytime ends at", m_end_time, 12);
}
}

@@ -887,8 +887,8 @@
delete_timed_event (TE_COMM_SERVER);
UpdateEndDates ();
}
- new_day_start_time = strToMinutes ((char *) &m_start_time);
- new_day_end_time = strToMinutes ((char *) &m_end_time);
+ new_day_start_time = strToMinutes (m_start_time);
+ new_day_end_time = strToMinutes (m_end_time);
if (m_memory_editable &&
(day_memory != m_day_memory ||
night_memory != m_night_memory ||[/CODE]

My personal style, not rigidly adhered to over the years, is to use

[CODE]
some_proc_not_modifying_int_var (int_variable)
some_proc_not_modifying_char_var (char_variable)
some_proc_modifying_int_var (&int_variable)
some_proc_modifying_char_var (&char_variable)
[/CODE]

so that the & is a quick clue to the reader that the proc being called may be modifying the variable.

Unfortunately, somewhere over the last 20 years, one or more compilers started whining about "some_proc_modifying_char_var (&char_variable)" saying that &char[13] is not compatible with proc expecting char *. Thus, I tossed in the (char *) cast.

Since I'm no fan of needless casts, I've made the change as you suggested.

BTW, you'll probably find a lot more needless (char *) typecasts if you look real hard.

[QUOTE=Prime95;454015]My personal style, not rigidly adhered to over the years, is to use

[CODE]
some_proc_not_modifying_int_var (int_variable)
some_proc_not_modifying_char_var (char_variable)
some_proc_modifying_int_var (&int_variable)
some_proc_modifying_char_var (&char_variable)
[/CODE]

so that the & is a quick clue to the reader that the proc being called may be modifying the variable.

Unfortunately, somewhere over the last 20 years, one or more compilers started whining about "some_proc_modifying_char_var (&char_variable)" saying that &char[13] is not compatible with proc expecting char *. Thus, I tossed in the (char *) cast.

Since I'm no fan of needless casts, I've made the change as you suggested.[/QUOTE]

I have already got used to the C convention that for some_proc_modifying_C_string (pointer_to_char) the string argument does not need an "&" operator.

It might be just me, but if I'm going to indicate a string argument will never be modified, I'll just make the parameter type (const char *) on the prototype.