On 10/21/19 8:06 AM, Michał Górny wrote:
> Change FETCHCOMMAND_RSYNC to use '-Lt' over '-a'. Notably, this
> replaces --links with --copy-links option, i.e. makes rsync copy
> underlying files when symlinks are met. This is important since
> we do not transfer symlink targets, therefore '-l'
On Mon, 2019-10-21 at 17:25 +0200, Ulrich Mueller wrote:
> > > > > > On Mon, 21 Oct 2019, Michał Górny wrote:
> > This also removes most of the other options that are irrelevant or even
> > undesirable to distfile fetching, that is:
> > - '-r' since we always fetch a single file, so recursive
> On Mon, 21 Oct 2019, Michał Górny wrote:
> This also removes most of the other options that are irrelevant or even
> undesirable to distfile fetching, that is:
> - '-r' since we always fetch a single file, so recursive operation is
> unnecessary
> - '-p', '-o', '-g' since we want to
Change FETCHCOMMAND_RSYNC to use '-Lt' over '-a'. Notably, this
replaces --links with --copy-links option, i.e. makes rsync copy
underlying files when symlinks are met. This is important since
we do not transfer symlink targets, therefore '-l' ends up creating
dangling symlinks.
This also
On 10/21/19 1:43 AM, Michał Górny wrote:
> Since DeletionIterator needs to stat the distfile and therefore find
> one working path for it, pass it to DeletionTask instead of recomputing
> it there. This also fixes wrongly assuming that first layout will
> always be correct.
>
> Bug:
On 10/21/19 2:16 AM, Michał Górny wrote:
> On Mon, 2019-10-21 at 02:10 -0700, Zac Medico wrote:
>> On 10/21/19 1:43 AM, Michał Górny wrote:
>>> Use real os.walk() when getting filenames for FlatLayout. Unlike
>>> the wrapped Portage module, it return str output for str path parameter,
>>> so we
On Mon, 2019-10-21 at 02:10 -0700, Zac Medico wrote:
> On 10/21/19 1:43 AM, Michał Górny wrote:
> > Use real os.walk() when getting filenames for FlatLayout. Unlike
> > the wrapped Portage module, it return str output for str path parameter,
> > so we don't have to recode it back and forth.
> >
On 10/21/19 1:43 AM, Michał Górny wrote:
> Use real os.walk() when getting filenames for FlatLayout. Unlike
> the wrapped Portage module, it return str output for str path parameter,
> so we don't have to recode it back and forth.
>
> Signed-off-by: Michał Górny
> ---
>
Since DeletionIterator needs to stat the distfile and therefore find
one working path for it, pass it to DeletionTask instead of recomputing
it there. This also fixes wrongly assuming that first layout will
always be correct.
Bug: https://bugs.gentoo.org/697890
Signed-off-by: Michał Górny
---
Use real os.walk() when getting filenames for FlatLayout. Unlike
the wrapped Portage module, it return str output for str path parameter,
so we don't have to recode it back and forth.
Signed-off-by: Michał Górny
---
lib/portage/package/ebuild/fetch.py | 3 ++-
1 file changed, 2 insertions(+),
On 10/21/19 1:02 AM, Michał Górny wrote:
> Bug: https://bugs.gentoo.org/697906
> Signed-off-by: Michał Górny
> ---
> lib/portage/_emirrordist/DeletionIterator.py | 11 ---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/lib/portage/_emirrordist/DeletionIterator.py
>
Bug: https://bugs.gentoo.org/697906
Signed-off-by: Michał Górny
---
lib/portage/_emirrordist/DeletionIterator.py | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/lib/portage/_emirrordist/DeletionIterator.py
b/lib/portage/_emirrordist/DeletionIterator.py
index
On Mon, 2019-10-21 at 00:44 -0700, Zac Medico wrote:
> On 10/20/19 3:23 AM, Michał Górny wrote:
> > Bug: https://bugs.gentoo.org/697906
> > Signed-off-by: Michał Górny
> > ---
> > lib/portage/_emirrordist/DeletionIterator.py | 21 +---
> > 1 file changed, 18 insertions(+), 3
On 10/20/19 3:23 AM, Michał Górny wrote:
> Bug: https://bugs.gentoo.org/697906
> Signed-off-by: Michał Górny
> ---
> lib/portage/_emirrordist/DeletionIterator.py | 21 +---
> 1 file changed, 18 insertions(+), 3 deletions(-)
>
> diff --git
On 10/20/19 2:54 AM, Michał Górny wrote:
> When DeletionIterator fails, report all stat() errors. Reporting
> just the first one results in confusing logs, suggesting that one
> of the location did not exist but the other existed and was removed.
> ---
>
15 matches
Mail list logo