| 1 |
Ciaran McCreesh wrote: |
| 2 |
> On Tue, 24 Feb 2009 18:16:54 +0100 |
| 3 |
> Luca Barbato <lu_zero@g.o> wrote: |
| 4 |
>>> You're doubling the number of files that have to be read for an |
| 5 |
>>> operation that's almost purely i/o bound. On top of that, you're |
| 6 |
>>> introducing a whole bunch of disk seeks in what's otherwise a nice |
| 7 |
>>> linear operation. |
| 8 |
>> I see words, not numbers. |
| 9 |
> |
| 10 |
> Number: double. That's a '2 times'. |
| 11 |
|
| 12 |
given that the simplest thing is hacking ebuild.sh and extract eapi with |
| 13 |
a simple C program (you can use pcre or ragel if you want) exactly |
| 14 |
before the ebuild source: |
| 15 |
|
| 16 |
Index: bin/ebuild.sh |
| 17 |
=================================================================== |
| 18 |
--- bin/ebuild.sh (revision 12704) |
| 19 |
+++ bin/ebuild.sh (working copy) |
| 20 |
@@ -1848,6 +1848,7 @@ |
| 21 |
# eclasses, they need to be unset before this process of |
| 22 |
# interaction begins. |
| 23 |
unset DEPEND RDEPEND PDEPEND IUSE |
| 24 |
+ EAPI=$(eapitool "${EBUILD}") |
| 25 |
source "${EBUILD}" || die "error sourcing ebuild" |
| 26 |
|
| 27 |
if [ "${EBUILD_PHASE}" != "depend" ] ; then |
| 28 |
|
| 29 |
I think your numbers are a bit pessimistic, this is when you get EAPI in |
| 30 |
portage, post source ${EBUILD}, opening the file before source would |
| 31 |
just put in the cache one line earlier. |
| 32 |
|
| 33 |
> You don't know whether the cache is valid until you know the EAPI. It |
| 34 |
> only works currently because EAPIs don't change inherit behaviour. |
| 35 |
|
| 36 |
There were already discussions about switching cache format, if we want |
| 37 |
to change the inherit behaviour we could just switch at the same time |
| 38 |
the cache format and leave dummy entry for compatibility with legacy |
| 39 |
portage. |
| 40 |
|
| 41 |
> So you have patches for Portage? Please show them. |
| 42 |
|
| 43 |
Up there what's enough to check the viability for the solution. |
| 44 |
|
| 45 |
the bash subst I wrote before could be used instead of the call to get |
| 46 |
the eapi in extension behaviour. |
| 47 |
|
| 48 |
>> unknown isn't unsupported? |
| 49 |
> |
| 50 |
> Huh? Please explain what you mean. |
| 51 |
|
| 52 |
mv cat/pkg-version.ebuild cat/pkg-version_foo.ebuild |
| 53 |
|
| 54 |
emerge -vp pkg |
| 55 |
|
| 56 |
portage will warn about not knowing pkg-version_foo and will ignore it. |
| 57 |
|
| 58 |
lu |
| 59 |
|
| 60 |
-- |
| 61 |
|
| 62 |
Luca Barbato |
| 63 |
Gentoo Council Member |
| 64 |
Gentoo/linux Gentoo/PPC |
| 65 |
http://dev.gentoo.org/~lu_zero |
Archives