Bugzilla – Full Text Bug Listing
| Summary: | false positive for 'no return in function returning non void' with switch statement | ||
|---|---|---|---|
| Product: | [Tools] CDT | Reporter: | Andrew Eidsness <eclipse> |
| Component: | cdt-codan | Assignee: | CDT Codan Inbox <cdt-codan-inbox> |
| Status: | RESOLVED FIXED | QA Contact: | Elena Laskavaia <elaskavaia.cdt> |
| Severity: | normal | ||
| Priority: | P3 | CC: | cdtdoug, matt, yevshif, zeratul976 |
| Version: | 8.0 | ||
| Target Milestone: | 8.5 | ||
| Hardware: | PC | ||
| OS: | Windows XP | ||
| Whiteboard: | |||
Yes, dead code confuses the checker- looks like the same issue as in bug 348386 I am seeing something similar in latest trunk (Build id: 20111215-0110), but I don't think it's necessarily related to dead code:
inline const char *ICAP_Scan_status_to_string(ICAP_Scan_status iss)
{
switch (iss)
{
case ISS_NOT_SCANNED:
return "ICAP_COMMUNICATION_ERROR";
case ISS_NO_MODIFICATION:
return "ICAP_NO_MODIFICATION";
case ISS_REPLACEMENT:
return "ICAP_REPLACED";
case ISS_SCAN_IN_PROGRESS:
return "ISS_SCAN_IN_PROGRESS";
case ISS_CORRUPT_UNSCANNED_OBJECT:
return "ISS_CORRUPT_UNSCANNED_OBJECT";
default:
return "unknown";
}
}
Here, this switch doesn't cover all of the cases in the enum -- the default case isn't dead code (not that I think CODAN is thinking *that* deeply about it ;)). All paths return a value, so this is a false positive.
Can you send another bug about this - this seems to be a different issue This appears to be fixed in http://git.eclipse.org/c/cdt/org.eclipse.cdt.git/commit/?id=16037a5f38845acb7194d12d70ba0ed6b5d04b90. fixed |
Build Identifier: Build id: 20110615-0604 "No return in function returning non-void" reported for the following: bool func( int i ) { switch( i ) { case 0: return true; default: return false; break; } } If I remove the final break statement then the problem goes away. The break statement is unreachable and therefore totally useless; but it seems to be confusing the checker. Reproducible: Always