TS-5009 CID 1022011 Logically dead code removal

[ngara]

Please review - I changed this code to ink_release_assert instead of the following code, which seems to either return '\0' or *b->start(); but will ink_assert in debug mode.

ink_assert(!"out of range");
const: At condition !!b, the value of b must be equal to 0.
null: At condition !!b, the value of b must be NULL.
dead_error_condition: The condition !!b cannot be true.
700 if (unlikely(b)) {

CID 1022011 (#1 of 1): Logically dead code (DEADCODE)
dead_error_line: Execution cannot reach this statement: return b->start();.
701 return *b->start();
702 }

[zwoop]

[approve ci]

[atsci]

Linux build failed! See https://ci.trafficserver.apache.org/job/Github-Linux/1004/ for details.

[atsci]

FreeBSD build successful! See https://ci.trafficserver.apache.org/job/Github-FreeBSD/1111/ for details.

[zwoop]

@ngara I think this is failing because of clang-format. I recommend you copy the pre-commit githook into your .git/hooks tree.

[ngara]

I put the plugin hook in and got it working. Thanks for the suggestion.

Nathan

On Oct 26, 2016, at 9:01 PM, Leif Hedstrom notifications@github.com wrote:

@ngara https://github.com/ngara I think this is failing because of clang-format. I recommend you copy the pre-commit githook into your .git/hooks tree.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub #1152 (comment), or mute the thread https://github.com/notifications/unsubscribe-auth/AA51eE0tGOe6QyG0UPgODLOK41L_-Xutks5q4CILgaJpZM4KhylU.

[zwoop]

[approve ci]

[ngara]

I took a close look at this and I didn't make the change that caused clang-format to fail here

[atsci]

Linux build failed! See https://ci.trafficserver.apache.org/job/Github-Linux/1008/ for details.

[ngara]

OK, figured it out. removing a variable made the equals signs want to line up differently on a line i didnt modify. Fixed!

[atsci]

FreeBSD build successful! See https://ci.trafficserver.apache.org/job/Github-FreeBSD/1114/ for details.

[ngara]

Maybe this should be an ink_abort instead of ink_release_assert?

[zwoop]

This seems reasonable to me, I will commit unless someone else disagrees.