Openembedded Bitbake Development
 help / color / mirror / Atom feed
From: Richard Purdie <richard.purdie@linuxfoundation.org>
To: Robert Yang <liezhi.yang@windriver.com>
Cc: bitbake-devel@lists.openembedded.org
Subject: Re: [PATCH 2/2] taskdata.py: set PREFERRED_VERSION when it is in the item
Date: Mon, 09 Sep 2013 16:32:53 +0100	[thread overview]
Message-ID: <1378740773.3484.126.camel@ted> (raw)
In-Reply-To: <27659d52912f96cb30a1a2d44638b262b50c1acc.1378121138.git.liezhi.yang@windriver.com>

On Mon, 2013-09-02 at 09:41 -0400, Robert Yang wrote:
> There is an error if we:
> 
> $ bitbake make (the make-3.82 will be built)
> // Edit make.inc
> $ bitbake make-3.81
> [snip]
>  *** 0004:    mfile = open(manifest)
>      0005:    entries = mfile.readlines()
>      0006:    mfile.close()
>      0007:
>      0008:    for entry in entries:
> Exception: IOError: [Errno 2] No such file or directory: xxx
> [snip]
> 
> Set the PREFERRED_VERSION if we find the version info in the item would fix the
> problem, I think that we need iterate in all the packages since it seems that
> we can't know wether the item is just a PN or PN-PV unless we compare them,
> what can I think is check len(all_p) == 1, but this doesn't work when the same
> recipe is multiple layers.
> 
> [YOCTO #5067]
> 
> Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
> ---
>  bitbake/lib/bb/taskdata.py |   15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/bitbake/lib/bb/taskdata.py b/bitbake/lib/bb/taskdata.py
> index 58fe199..38a3191 100644
> --- a/bitbake/lib/bb/taskdata.py
> +++ b/bitbake/lib/bb/taskdata.py
> @@ -430,6 +430,21 @@ class TaskData:
>  
>          all_p = dataCache.providers[item]
>  
> +        # Check whether item has the one of following formats and set the
> +        # PREFERRED_VERSION if it does:
> +        # - <pn>-<pv>
> +        # - <pn>-<pv>-<pr> (when no PE)
> +        # - <pn>-<pe>_<pv>-<pr> (when PE)
> +        for fn in all_p:
> +            pn = dataCache.pkg_fn[fn]
> +            pe = dataCache.pkg_pepvpr[fn][0]
> +            pv = dataCache.pkg_pepvpr[fn][1]
> +            pr = dataCache.pkg_pepvpr[fn][2]
> +            if not pe and (item == "%s-%s" % (pn, pv) or item == "%s-%s-%s" % (pn, pv, pr)):
> +                cfgData.setVar("PREFERRED_VERSION_" + pn, pv)
> +            elif pe and item == "%s-%s_%s-%s" % (pn, pe, pv, pr):
> +                cfgData.setVar("PREFERRED_VERSION_" + pn, pv)
> +
>          eligible, foundUnique = bb.providers.filterProviders(all_p, item, cfgData, dataCache)
>          eligible = [p for p in eligible if not self.getfn_id(p) in self.failed_fnids]
>  


This is a neat solution and I'm torn over it. On the one hand it does
address the problem. On the other hand:

a) Its implementation specific knowledge since bitbake.conf in OE-Core
is responsible for the extra PROVIDES
b) It would confuse things if some recipe does DEPENDS = "make-3.81"
c) The performance of this code isn't going to be good

b) can be fixed by adding a bb.warn(), c) might be either not much of an
issue or could be fixed, a) is a real problem though.

I've put a counter proposal on the OE-Core mailing list suggesting we
simplify PROVIDES in bitbake.conf. Depending on the feedback, we can
either take that patch or this one with a bb.warn() added. Need some
user feedback...

Cheers,

Richard



      reply	other threads:[~2013-09-09 15:33 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-02 13:41 [PATCH 0/2] taskdata.py: set PREFERRED_VERSION when it is in the item Robert Yang
2013-09-02 13:41 ` [PATCH 1/2] providers.py: enhance the runtime debug degbug messgae Robert Yang
2013-09-02 13:41 ` [PATCH 2/2] taskdata.py: set PREFERRED_VERSION when it is in the item Robert Yang
2013-09-09 15:32   ` Richard Purdie [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1378740773.3484.126.camel@ted \
    --to=richard.purdie@linuxfoundation.org \
    --cc=bitbake-devel@lists.openembedded.org \
    --cc=liezhi.yang@windriver.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox