Hacker News Reader

Comment by blueblob | Hacker News Reader

Can't you make the argument that even in C/C++ they could have enforced different "Language rules" like forcing no warnings? When compiling with gcc you could use -Wunreachable-code. Then it is part of the process.

EDIT: with -Werror it will make this bug an error but not necessarily the class of bugs.

View on HN

an unreachable code warning would have caught this particular defect, but it wouldn't have helped if the duplicated line were something other than a goto. I think the policy of always using curly braces for conditionals (even if only one line) in c like languages is a good one.

> an unreachable code warning would have caught this particular defect, but it wouldn't have helped if the duplicated line were something other than a goto.

In that case always using curly braces or using a language that requires an "end" statement after the conditionally executed code may not have helped either. Imagine the incorrectly repeated statement was "a = a + 1" or "error_mask ^= error_x" etc. Putting the erroneous line inside the conditional doesn't erase the error, it just modifies the conditions under which it executes. That's about as likely to hang you as save you.

Is there any language rule that can save you from incorrectly repeating an operation that is not idempotent?

gcc has not warned about unreachable code for years[1] now :(

clang -Wunreachable-code does, though.

[1]: http://gcc.gnu.org/ml/gcc-help/2011-05/msg00360.html

Wow, that's interesting! Thanks for that. I just saw it in a manpage, didn't realize it doesn't currently work. That's what I get for R-ingTFM :-D

Of course. I didn't want to muddie the issue, but I would consider compilation flags, transpilation (e.g., coffeescript) and linting (e.g., jshint, etc...) to reside between "Language rules" and "Unit tests."

Reply on news.ycombinator.com