eqin comments differ from implementation

[arno formella]

Hi,

  using indent 2.2.10 I observed that the break-function-decl-args-end
long option was not recognized well.

  digging into the source I found that the comment of function "eqin"
says that "true" is returned when the two argument strings are equal,
however, its implementation returns "true" whenever the first argument
is a prefix of the second one, which explains the observed behavior.

  suggestion:  insert  the line

if(*s2) return false;

  just before asigning the value to achieve what I guess was intented.
(every pair of options but the two
"break-function-decl-args"-"break-function-decl-args-end" and
"dont-"break-function-decl-args"-"dont-break-function-decl-args-end"
are prefix-free, which give a hint why the bug has been hidden for years :-).

best regards

Arno Formella

PS. remember the saying from Norm Schryer:
"If the code and comments disagree, both are probably wrong".

_______________________________________________
bug-indent mailing list
bug-indent@...
https://lists.gnu.org/mailman/listinfo/bug-indent

[Petr Pisar-3]

On Thu, Dec 15, 2011 at 10:00:18AM +0100, arno formella wrote:

>   using indent 2.2.10 I observed that the break-function-decl-args-end
> long option was not recognized well.
>
>   digging into the source I found that the comment of function "eqin"
> says that "true" is returned when the two argument strings are equal,
> however, its implementation returns "true" whenever the first argument
> is a prefix of the second one, which explains the observed behavior.
>
>   suggestion:  insert  the line
>
> if(*s2) return false;
>

This is not proper fix. It breaks parsing other options. Run "make -C

regression" after compilation to see all the failures.

-- Petr


_______________________________________________
bug-indent mailing list
bug-indent@...
https://lists.gnu.org/mailman/listinfo/bug-indent