This patch adds in support for automatically detecting and decompressing
both formats gentoo currently ships portage snapshots in. To make the
change as minimal as possible I am using recent gnu/bsd tar ability to
automatically pick the correct format and decompress. This should work
on all gentoo supported platforms including linux, freebsd, and
prefix-macosx.
def set_snapcache_path(self):
if self.settings.has_key("SNAPCACHE"):
@@ -635,7 +643,7 @@ class generic_stage_target(generic_target):
self.settings["source_path"]+"
to "+
self.settings["chroot_path"]+
" (This may take some time) ...
"
- unpack_cmd="tar xjpf "+self.settings["source_path"]+" -C "+
+ unpack_cmd="tar xpf "+self.settings["source_path"]+" -C "+
self.settings["chroot_path"]
error_msg="Tarball extraction of "+
self.settings["source_path"]+" to "+
@@ -646,7 +654,7 @@ class generic_stage_target(generic_target):
self.settings["source_path"]+"
to "+
self.settings["chroot_path"]+
" (This may take some time) ...
"
- unpack_cmd="tar xjpf "+self.settings["source_path"]+" -C "+
+ unpack_cmd="tar xpf "+self.settings["source_path"]+" -C "+
self.settings["chroot_path"]
error_msg="Tarball extraction of "+self.settings["source_path"]+
" to "+self.settings["chroot_path"]+" failed."
@@ -744,7 +752,7 @@ class generic_stage_target(generic_target):
read_from_clst(self.settings["snapshot_cache_path"]+
"catalyst-hash")
destdir=self.settings["snapshot_cache_path"]
- unpack_cmd="tar xjpf "+self.settings["snapshot_path"]+" -C "+destdir
+ unpack_cmd="tar xpf "+self.settings["snapshot_path"]+" -C "+destdir
unpack_errmsg="Error unpacking snapshot"
cleanup_msg="Cleaning up invalid snapshot cache at
"+
self.settings["snapshot_cache_path"]+
@@ -760,7 +768,7 @@ class generic_stage_target(generic_target):
cleanup_errmsg="Error removing existing snapshot directory."
cleanup_msg=
"Cleaning up existing portage tree (This can take a long time)..."
- unpack_cmd="tar xjpf "+self.settings["snapshot_path"]+" -C "+
+ unpack_cmd="tar xpf "+self.settings["snapshot_path"]+" -C "+
self.settings["chroot_path"]+"/usr"
unpack_errmsg="Error unpacking snapshot"
--
1.7.8.6
08-22-2012, 04:13 PM
Richard Farina
Support snapshot.tar.xz format
This version of the patch addresses the concerns from Matt (of me
failing to copy and paste everything) as well as updates the commit
message to show that I have no verified this should work on all gentoo
supported platforms.
Please accept.
thanks,
Zero
On 08/22/2012 12:12 PM, Rick Farina wrote:
> This patch adds in support for automatically detecting and decompressing
> both formats gentoo currently ships portage snapshots in. To make the
> change as minimal as possible I am using recent gnu/bsd tar ability to
> automatically pick the correct format and decompress. This should work
> on all gentoo supported platforms including linux, freebsd, and
> prefix-macosx.
>
> Signed-off-by: Rick Farina <sidhayn@gmail.com>
> ---
> modules/generic_stage_target.py | 18 +++++++++++++-----
> 1 files changed, 13 insertions(+), 5 deletions(-)
>
> diff --git a/modules/generic_stage_target.py b/modules/generic_stage_target.py
> index c4ef239..b971741 100644
> --- a/modules/generic_stage_target.py
> +++ b/modules/generic_stage_target.py
> @@ -404,12 +404,20 @@ class generic_stage_target(generic_target):
>
> def set_snapshot_path(self):
> self.settings["snapshot_path"]=normpath(self.settings["storedir"]+
> - "/snapshots/portage-"+self.settings["snapshot"]+".tar.bz2")
> + "/snapshots/portage-"+self.settings["snapshot"]+".tar.xz")
>
> if os.path.exists(self.settings["snapshot_path"]):
> self.settings["snapshot_path_hash"]=
> generate_hash(self.settings["snapshot_path"],
> hash_function=self.settings["hash_function"],verbose=False)
> + else:
> + self.settings["snapshot_path"]=normpath(self.settings["storedir"]+
> + "/snapshots/portage-"+self.settings["snapshot"]+".tar.bz2")
> +
> + if os.path.exists(self.settings["snapshot_path"]):
> + self.settings["snapshot_path_hash"]=
> + generate_hash(self.settings["snapshot_path"],
> + hash_function=self.settings["hash_function"],verbose=False)
>
> def set_snapcache_path(self):
> if self.settings.has_key("SNAPCACHE"):
> @@ -635,7 +643,7 @@ class generic_stage_target(generic_target):
> self.settings["source_path"]+"
to "+
> self.settings["chroot_path"]+
> " (This may take some time) ...
"
> - unpack_cmd="tar xjpf "+self.settings["source_path"]+" -C "+
> + unpack_cmd="tar xpf "+self.settings["source_path"]+" -C "+
> self.settings["chroot_path"]
> error_msg="Tarball extraction of "+
> self.settings["source_path"]+" to "+
> @@ -646,7 +654,7 @@ class generic_stage_target(generic_target):
> self.settings["source_path"]+"
to "+
> self.settings["chroot_path"]+
> " (This may take some time) ...
"
> - unpack_cmd="tar xjpf "+self.settings["source_path"]+" -C "+
> + unpack_cmd="tar xpf "+self.settings["source_path"]+" -C "+
> self.settings["chroot_path"]
> error_msg="Tarball extraction of "+self.settings["source_path"]+
> " to "+self.settings["chroot_path"]+" failed."
> @@ -744,7 +752,7 @@ class generic_stage_target(generic_target):
> read_from_clst(self.settings["snapshot_cache_path"]+
> "catalyst-hash")
> destdir=self.settings["snapshot_cache_path"]
> - unpack_cmd="tar xjpf "+self.settings["snapshot_path"]+" -C "+destdir
> + unpack_cmd="tar xpf "+self.settings["snapshot_path"]+" -C "+destdir
> unpack_errmsg="Error unpacking snapshot"
> cleanup_msg="Cleaning up invalid snapshot cache at
"+
> self.settings["snapshot_cache_path"]+
> @@ -760,7 +768,7 @@ class generic_stage_target(generic_target):
> cleanup_errmsg="Error removing existing snapshot directory."
> cleanup_msg=
> "Cleaning up existing portage tree (This can take a long time)..."
> - unpack_cmd="tar xjpf "+self.settings["snapshot_path"]+" -C "+
> + unpack_cmd="tar xpf "+self.settings["snapshot_path"]+" -C "+
> self.settings["chroot_path"]+"/usr"
> unpack_errmsg="Error unpacking snapshot"
>
>
08-24-2012, 05:48 PM
Sebastian Pipping
Support snapshot.tar.xz format
On 08/22/2012 06:13 PM, Richard Farina wrote:
> This version of the patch addresses the concerns from Matt (of me
> failing to copy and paste everything) as well as updates the commit
> message to show that I have no verified this should work on all gentoo
> supported platforms.
>
> Please accept.
Quoted patches are no fun. Attach the patch or make the whole mail work
with git-am, please please please.
To be fair, I should mention that I hav no capacities for testing this
myself these days. Hopefully someone else does. Increased chances for
easy patch access ;-)
Best,
Sebastian
08-24-2012, 06:24 PM
Sebastian Pipping
Support snapshot.tar.xz format
On 08/24/2012 07:48 PM, Sebastian Pipping wrote:
> On 08/22/2012 06:13 PM, Richard Farina wrote:
>> This version of the patch addresses the concerns from Matt (of me
>> failing to copy and paste everything) as well as updates the commit
>> message to show that I have no verified this should work on all gentoo
>> supported platforms.
>>
>> Please accept.
>
> Quoted patches are no fun. Attach the patch or make the whole mail work
> with git-am, please please please.
Ignore my conmplaint if the patch was identical to the previous version
and you were just adding another note.
Best,
Sebastian
08-24-2012, 06:38 PM
Richard Farina
Support snapshot.tar.xz format
On 08/24/2012 01:48 PM, Sebastian Pipping wrote:
> On 08/22/2012 06:13 PM, Richard Farina wrote:
>> This version of the patch addresses the concerns from Matt (of me
>> failing to copy and paste everything) as well as updates the commit
>> message to show that I have no verified this should work on all gentoo
>> supported platforms.
>>
>> Please accept.
>
> Quoted patches are no fun. Attach the patch or make the whole mail work
> with git-am, please please please.
I specifically sent the patch using git-email then replied to my post to
add more detail than the commit message. This is the process I use on
kernel mailing lists and never had a complaint. If this is not
acceptable please let me know, but I have a feeling that you read my
reply and not the first email that came through....
>
> To be fair, I should mention that I hav no capacities for testing this
> myself these days. Hopefully someone else does. Increased chances for
> easy patch access ;-)
Obviously I have tested the patch, but given that it falls back to the
original behavior (also tested) it's not a very dangerous patch to apply
even without extensive testing imho.
Thanks,
Zero
08-24-2012, 06:44 PM
Richard Farina
Support snapshot.tar.xz format
On 08/24/2012 02:24 PM, Sebastian Pipping wrote:
> On 08/24/2012 07:48 PM, Sebastian Pipping wrote:
>> On 08/22/2012 06:13 PM, Richard Farina wrote:
>>> This version of the patch addresses the concerns from Matt (of me
>>> failing to copy and paste everything) as well as updates the commit
>>> message to show that I have no verified this should work on all gentoo
>>> supported platforms.
>>>
>>> Please accept.
>>
>> Quoted patches are no fun. Attach the patch or make the whole mail work
>> with git-am, please please please.
>
> Ignore my conmplaint if the patch was identical to the previous version
> and you were just adding another note.
>
yeah that's it, sorry for the confusion you had me really concerned for
a minute.
-Zero
> Best,
>
>
>
>
> Sebastian
>
>
08-25-2012, 07:09 PM
"Jorge Manuel B. S. Vicetto"
Support snapshot.tar.xz format
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 24-08-2012 18:38, Richard Farina wrote:
> On 08/24/2012 01:48 PM, Sebastian Pipping wrote:
>> On 08/22/2012 06:13 PM, Richard Farina wrote:
>>> This version of the patch addresses the concerns from Matt (of
>>> me failing to copy and paste everything) as well as updates the
>>> commit message to show that I have no verified this should work
>>> on all gentoo supported platforms.
>>>
>>> Please accept.
>>
>> Quoted patches are no fun. Attach the patch or make the whole
>> mail work with git-am, please please please.
>
> I specifically sent the patch using git-email then replied to my
> post to add more detail than the commit message. This is the
> process I use on kernel mailing lists and never had a complaint.
> If this is not acceptable please let me know, but I have a feeling
> that you read my reply and not the first email that came
> through....
>>
>> To be fair, I should mention that I hav no capacities for testing
>> this myself these days. Hopefully someone else does. Increased
>> chances for easy patch access ;-)
Richard (Zero) is testing catalyst / working on patches with me on
#gentoo-releng
> Obviously I have tested the patch, but given that it falls back to
> the original behavior (also tested) it's not a very dangerous patch
> to apply even without extensive testing imho.
I'll commit the patch and test it before doing a new release.
> Thanks, Zero
>
- --
Regards,
Jorge Vicetto (jmbsvicetto) - jmbsvicetto at gentoo dot org
Gentoo- forums / Userrel / Devrel / KDE / Elections / RelEng
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/