FU Logo
  • Startseite
  • Kontakt
  • Impressum
  • Home
  • Listenauswahl
  • Anleitungen

Re: [Seqan-dev] C++11 features

<-- thread -->
<-- date -->
  • From: "Holtgrewe, Manuel" <manuel.holtgrewe@fu-berlin.de>
  • To: SeqAn Development <seqan-dev@lists.fu-berlin.de>
  • Date: Fri, 08 Mar 2013 18:03:29 +0100
  • Reply-to: SeqAn Development <seqan-dev@lists.fu-berlin.de>
  • Subject: Re: [Seqan-dev] C++11 features

Hi Hannes,

at the moment, the policy is that the library should use C++98 only and C++11 is OK for applications.

The reasoning is that there is currently no schedule for switching to C++11 and we do not have terribly much experience with it. Also, it is unclear which features are available in which compilers and since VS 2010 is still an important target and does not have variadic templates IIRC, it would be better to have code that does not rely on these features.

Also, the GCC STL team reported that they had to go over each line and possibly adjust it because of rvalue references.

We plan to head for C++11 support in SeqAn 2.0 with a planned schedule, informed decisions which compilers support which features and also which compilers are used by our user base.

HTH,
Manuel
________________________________________
From: Hannes Hauswedell [hauswedell@mi.fu-berlin.de]
Sent: Friday, March 08, 2013 12:32 PM
To: seqan-dev@lists.fu-berlin.de
Subject: [Seqan-dev] C++11 features

Hi everyone,

I wanted to ask what the state of C++11 features in SeqAn is. The
homepage mentions C++98, but in the medium term something like SeqAn
would obviously benefit significantly from the new features.
And considiering that the Debian stable computers at the institute
already support many features relevant to SeqAn, wide-spread support
doesn't seem to be too far off.

Is there some sort of timeline or migration plan? Is there e.g. a macro
that I could enclose code with that requires C++11 features?
This would permit adding new features and new code that already works
with C++11 without having to bump the general requirement of the
library. Ultimately the transition would be smoother, because not all
existing code would have to migrated at once and new library features
could already build on new C++ features.

The resaon I am asking is that I am currently working on
Blast-statistics and Blast-IO for the library. All basic features are of
course C++98, but some stuff simply isn't viable without C++11 -- or at
least it isn't viable to do, when C++11 support will come sooner or
later anyway. One such example is the feature in Blast-IO to have the
programmer specifiy the amount (and type) of output columns in tabuler
format. I have implemented this through Variadic Templates, which can be
done with easily readable and reusable code -- without Variadic
Templates on the other hand one would have to define n functions for 1
to n parameters or use some obscure form of linked or nested templates...

Thanks,
Hannes

_______________________________________________
seqan-dev mailing list
seqan-dev@lists.fu-berlin.de
https://lists.fu-berlin.de/listinfo/seqan-dev



<-- thread -->
<-- date -->
  • Follow-Ups:
    • Re: [Seqan-dev] C++11 features
      • From: Hannes Hauswedell <hauswedell@mi.fu-berlin.de>
  • References:
    • [Seqan-dev] C++11 features
      • From: Hannes Hauswedell <hauswedell@mi.fu-berlin.de>
  • seqan-dev - March 2013 - Archives indexes sorted by:
    [ thread ] [ subject ] [ author ] [ date ]
  • Complete archive of the seqan-dev mailing list
  • More info on this list...

Hilfe

  • FAQ
  • Dienstbeschreibung
  • ZEDAT Beratung
  • postmaster@lists.fu-berlin.de

Service-Navigation

  • Startseite
  • Listenauswahl

Einrichtung Mailingliste

  • ZEDAT-Portal
  • Mailinglisten Portal