[erlang-questions] dialyzer and re(3)

[Paul Guyot]

Le 11 nov. 08 à 12:00, erlang-questions-request@REDACTED a écrit :

> Date: Tue, 11 Nov 2008 12:33:44 +0200
> From: Kostis Sagonas <kostis@REDACTED>
> Subject: Re: [erlang-questions] dialyzer and re(3)
> To: erlang-questions@REDACTED
> Message-ID: <49195F88.3020302@REDACTED>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>> dialyzer does not know native functions of re(3) module and this  
>> makes
>> it impossible to check our code that now uses re:compile/2, unless we
>> remove re.beam from the plt.
>
> I do not think you should worry about the above or take any action.   
> The
> fact that dialyzer does not know anything about some function of the  
> re
> module in no way affects the correctness of the analysis.  It just  
> means
> that dialyzer might not be able to detect all discrepancies  
> involving re
>
> Nothing to worry about...

I am not worrying about dialyzer's error message. Our continuous  
integration server runs dialyzer and fails if dialyzer reports  
anything of this level. I don't want to ignore dialyzer errors, so the  
workaround was to remove re.beam from the PLT -- therefore calls to  
re:compile/2 just becomes calls to an unknown function in an unknwon  
module, and this is not fatal. I am just wondering if there is a  
better workaround.

While we appreciate dialyzer very much and we rely on it, we are  
somewhat disappointed by its current limitations and horrible  
slowness, especially when compared to ocaml compiler which does type  
inference and checking faster by a magnitude...

Paul