[CWB] "Segmentation fault (core dumped)" on various versions

[Scott Sadowsky]

Hi Stefan,

That worked just fine. Here's the output:

$ cwb-lexdecode -f -p '.ai' PERS-DIVER-USENET
    165 Tai
     57 dai
    357 tai
      7 Mai
      3 Kai
    357 vai
     23 Vai
      6 rai
     81 cai
     13 sai
      4 Dai
      1 uai
     32 hai
      1 Jai
     23 mai
      7 fai
      2 lai
      2 Wai
      2 wai
      6 nai
      2 pai
      5 Sai
      4 bai
      4 Cai
      5 kai
      4 gai
      1 Bai
      2 Rai
      1 Lai
      1 Fai
      1 jai
$

Cheers,
Scott

On Tue, Jul 23, 2013 at 5:47 AM, Stefan Evert <stefanML at collocations.de>wrote:

>
> On 23 Jul 2013, at 05:12, Scott Sadowsky <ssadowsky at gmail.com> wrote:
>
> > [no corpus]> PERS-DIVER-USENET
> > PERS-DIVER-USENET> [word = ".+ai"]
> > Segmentation fault (core dumped)
> >
> > [no corpus]> PERS-DIVER-USENET
> > PERS-DIVER-USENET> [word = ".*ai"];
> > Segmentation fault (core dumped)
> >
> > [no corpus]> PERS-DIVER-USENET
> > PERS-DIVER-USENET> [word = ".ai"];
> > Segmentation fault (core dumped)
> >
> > However, for what it's worth, this does NOT happen in cases such as the
> following:
> >
> > [no corpus]> PERS-DIVER-USENET
> > PERS-DIVER-USENET> [word = "cas(a|o)(s|)"];
>
> This is rather strange.
>
> IIRC, I've recently had some segmentation faults with damaged index files,
> though it would have to be some word matched by all three patterns (i.e.
> /.ai/).
>
> Can you try the following on the command-line:
>
>         cwb-lexdecode -f -p '.ai' PERS-DIVER-USENET
>
> Best,
> Stefan



scott.sadowsky ufrontera cl
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://devel.sslmit.unibo.it/pipermail/cwb/attachments/20130723/83b1fdcf/attachment.html>