Re: [isabelle] Isabelle2016-RC0 - A fast implementation of concat for strings?



Hi,

I refer to my mail from 29.09.2015, "A fast implementation of concat
for strings?".

In Isabelle 2016, concat is still slow (which makes printing large
datatypes with value[code] quite slow). Same behavior as in
Isabelle2015.

value[code] "replicate 1000
''xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx''"
(*5.895s*)

value[code] "concat (replicate 1000
''xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx'')"
(*67.556s*)

Urgency: none

Best Regards,
  Cornelius

2016-01-01 20:17 GMT+01:00 Makarius <makarius at sketis.net>:
> Dear Isabelle users,
>
> the coming Isabelle2016 release is scheduled for February 2016, after the
> next big Java 8 update by Oracle in January and some weeks before the
> deadline of ITP 2016.
>
> To get started with systematic testing there is now the relatively early
> http://isabelle.in.tum.de/website-Isabelle2016-RC0 (corresponding to
> Isabelle/e18444532fce and AFP/c62777f3e932).
>
> The website, NEWS, ANNOUNCE etc. are already mostly up-to-date.  Some
> documentation is still lagging behind, notably the Isabelle/jEdit manual.
> There are further fine points still to be sorted out.
>
>
> When discussing problems, observations, suggustions, etc. the mail subject
> line should be changed to something meaningful (but the release candidate
> number still given in the message body).
>
> As usual it is important to keep general laws of causality in mind: release
> candidates may still change, but the final release is final. Although this
> is tautological, in the past few releases we've often had complaints right
> after final lift-off, when it was too late.
>
> So the best time to start testing is now.
>
>
>         Makarius
>




This archive was generated by a fusion of Pipermail (Mailman edition) and MHonArc.