A common bug in published code • Hacker News Board

Re: A common bug in published code

#51

Post by mbenjaminsmith » Fri, Apr 22, 2011, 2:38 AM UTC

bnewbold wrote: post #24

samps wrote: post #13

Someone should write a tool that scrapes these results and automatically files bugs at the associated bug trackers for open-source projects.

I've been thinking of doing something and i'm looking for people to work with! Notes here: http://memeschemes.com/code_search/

A bug in that program would certainly be wonderful. Have fun when it files a few million unnecessary bug reports while you're sleeping.

Re: A common bug in published code

#52

Post by dgallagher » Fri, Apr 22, 2011, 2:48 AM UTC

ghewgill wrote: post #16

Last year I found several instances of the printf format "200%d" when dealing with years. Some still exist: http://hewgill.com/journal/entries/548-y2010-bugs-found-via-...

I think that's called job security in some circles. ;)

Re: A common bug in published code

#53

Post by rmccue » Fri, Apr 22, 2011, 2:54 AM UTC

For those who haven't picked up the error: Math.random() returns a value between 0 and 1, and the typecasting floors the value.

Re: A common bug in published code

#54

Post by mail2345 » Fri, Apr 22, 2011, 3:00 AM UTC

mbenjaminsmith wrote: post #51

bnewbold wrote: post #24

Earlier quoted context omitted.

I've been thinking of doing something and i'm looking for people to work with! Notes here: http://memeschemes.com/code_search/

A bug in that program would certainly be wonderful. Have fun when it files a few million unnecessary bug reports while you're sleeping.

Have it require approval for each bug report. While it isn't as fast as a fully automatic bug finding system, it is faster than a manual find and file. Plus it takes care of those situations where it was intended(poor coding, but it works).

Re: A common bug in published code

#55

Post by ghempton » Fri, Apr 22, 2011, 3:29 AM UTC

Majority of these results are examples and test cases which is probably why the bugs have not manifested themselves.

Re: A common bug in published code

#56

Post by Splines » Fri, Apr 22, 2011, 3:39 AM UTC

benburleson wrote: post #28

ignifero wrote: post #12

A-mazing Also, notice the difference in number of results between C and C++ below: http://www.google.com/codesearch?hl=en&lr=&q=if%5Cs*... vs http://www.google.com/codesearch?hl=en&lr=&q=if%5Cs*...

That's why I write comparisons like if (CONSTANT == variable)

I know it's safer, but personally I find it harder to read. Especially if the constant is a #define or enum and is similar in format to the variable in question.

Re: A common bug in published code

#57

Post by Carlfish » Fri, Apr 22, 2011, 4:33 AM UTC

ChuckMcM wrote: post #32

One would hope at some point a tool would warn you about this since clearly it could optimize (int) Math.random() to just 0. [queue debate about tools that hold your hands vs understanding what you are actually writing] I wonder if they did the search for if (x = y) bug pre-gcc-4.x-warn what sort of numbers they would get.

I believe this is the tool you are looking for.

http://findbugs.sourceforge.net/bugDescriptions.html#RV_01_T...

Re: A common bug in published code

#58

Post by bobds » Fri, Apr 22, 2011, 4:54 AM UTC

bnewbold wrote: post #24

samps wrote: post #13

Someone should write a tool that scrapes these results and automatically files bugs at the associated bug trackers for open-source projects.

I've been thinking of doing something and i'm looking for people to work with! Notes here: http://memeschemes.com/code_search/

I've had some similar ideas myself, consider me interested. I've written some (very little) code to search projects among the big open source hosts (SF, Google, Github etc). Also have a nice domain for it.

Re: A common bug in published code

#59

Post by polynomial » Fri, Apr 22, 2011, 5:14 AM UTC

klodolph wrote: post #19

ilcavero wrote: post #11

the main cause here is that randomness is hard to test so that code was just assumed to be correct. With randomness you can never be sure, http://www.random.org/analysis/

Yes, but... (int)Math.random() always returns zero. A little suspicious.

You can never be sure the next one won't be a 1!

Re: A common bug in published code

#60

Post by mirkules » Fri, Apr 22, 2011, 6:06 AM UTC

baltcode wrote: post #17

Apparently, python only has 5 instances of the corresponding error: http://www.google.com/codesearch?hl=en&lr=&q=\s%2Bin... Python-Java flame-war, anyone?

Clearly, PHP is a better language than both: int rand($min, $max)

ducks

All joking aside, I'm just pointing out the absurdity of comparing one language to another based on how easy it is to shoot oneself in the foot.