[prev in list] [next in list] [prev in thread] [next in thread] 

List:       osdl-lsb-discuss
Subject:    [lsb-discuss] Adding Python to the LSB
From:       mats.d.wichmann () intel ! com (Wichmann, Mats D)
Date:       2006-07-26 13:05:59
Message-ID: 3F62CBEE02D6404E98C65934617EB5824819FD () fmsmsx414 ! amr ! corp ! intel ! com
[Download RAW message or body]


>Here are a few things to consider about adding Python to the LSB. I am
>in favor of Python being added but just wanted to point out 
>these issues
>as needing resolution.
>
>Python itself is not LSB compliant as the following interfaces 
>which are
>not part of LSB today are used:
>
>mremap   => I requested addition of this interface to the LSB in a 
>            separate e-mail
>tmpnam or tmpnam_r  depending on the configuration => can probably be 
>            replaced with mkstemp but will have to be pushed upstream
>hstrerror  => marked as obsolete in the man page but I do not 
>know of a 
>              suitable replacement
>gethostbyaddr_r, gethostbyname_r  => requested these for re-inclusion, 
>           or this couild be pushed upstream for use of getaddrinfo()
>inet_aton  => I requested addition of this interface to the LSB in a 
>              separate e-mail
>
>Comments?

I don't think any of these are a big issue.
We'll look at the interfaces, but meanwhile, a Python built
in the LSB build environment *does* come out LSB conforming,
due to the configuration checks doing the right thing,
excepting a couple of spots where the configure script
is not quite correct.

But the main point is it's not required for the Python
binary to itself be LSB-conforming (although as just noted,
it's possible to make it so with little trouble), the LSB
requirement on commands is that they provide conforming
behavior - only.  So /bin/sh needs to behave like a
conforming shell, but does not itself have to be an
LSB-conforming binary.

-- mats


[prev in list] [next in list] [prev in thread] [next in thread]