recipes that don't list where their source is
not sure what the policy is but, occasionally, i run across a recipe
that doesn't identify the location of the source code (just in case i want to peruse it). case in point: kexec-tools on master branch:
http://cgit.openembedded.org/openembedded-core/tree/meta/recipes-kernel/kexec/kexec-tools_2.0.20.bb
combining HOMEPAGE and SRC_URI, you get little more than the directory to download versioned tarballs, but nothing about the location of the source code itself, if it's available.
is there a protocol for identifying the (accessible) location of the
source code, or is that simply up to the maintainer to provide it if he or she wants?
rday
Signature · 7 lines
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================Quoted 6 lines · On Sat, 2020-03-21 at 08:26 -0400, [email protected] wrote:
not sure what the policy is but, occasionally, i run across a
recipe
that doesn't identify the location of the source code (just in case i want to peruse it). case in point: kexec-tools on master branch:http://cgit.openembedded.org/openembedded-core/tree/meta/recipes-kernel/kexec/kexec-tools_2.0.20.bb
Surely the recipe doesn't work if you can't access the source code? The source code is clearly in the tarball.
Quoted 8 lines
combining HOMEPAGE and
SRC_URI, you get little more than the directory
to download versioned tarballs, but nothing about the location of the source code itself, if it's available.is there a protocol for identifying the (accessible) location of
the source code, or is that simply up to the maintainer to provide it if he or she wants?
Do you actually mean the upstream source control system supporting the source code?
There is no requirement to have that. I agree its useful at times and its up to the maintainer. Some source code doesn't have it publicly available.
I'm actually a big supporter of devupstream being enabled in recipes which would add more of this information to recipes. We don't have enough maintainers to sustain regular testing of that though.
Cheers,
Richard
Quoted 10 lines · Richard Purdie
Quoted 6 lines · On Sat, 2020-03-21 at 08:26 -0400, [email protected] wrote:
not sure what the policy is but, occasionally, i run across a
recipe
that doesn't identify the location of the source code (just in case i want to peruse it). case in point: kexec-tools on master branch:http://cgit.openembedded.org/openembedded-core/tree/meta/recipes-kernel/kexec/kexec-tools_2.0.20.bb
Surely the recipe doesn't work if you can't access the source code? The source code is clearly in the tarball.
i realize that ... in the recipe i mentioned, that information is
already in the SRC_URI.
Quoted 8 lines
Quoted 7 lines
combining HOMEPAGE and
SRC_URI, you get little more than the directory to download versioned tarballs, but nothing about the location of the source code itself, if it's available.is there a protocol for identifying the (accessible) location of
the source code, or is that simply up to the maintainer to provide it if he or she wants?
Quoted 6 lines
Do you actually mean the upstream source control system supporting the source code?
There is no requirement to have that. I agree its useful at times and its up to the maintainer. Some source code doesn't have it publicly available.
i understand that as well, just wanted to clarify as to whether it's
considered a best practice to do that, and whether it's a personal call on part of the maintainer.
rday