| 1 |
2009/5/28 Marijn Schouten (hkBst) <hkBst@g.o>: |
| 2 |
> -----BEGIN PGP SIGNED MESSAGE----- |
| 3 |
> Hash: SHA1 |
| 4 |
> |
| 5 |
> Piotr Jaroszyński wrote: |
| 6 |
>> 2009/5/28 Ulrich Mueller <ulm@g.o>: |
| 7 |
>>>>>>>> On Thu, 28 May 2009, Tiziano Müller wrote: |
| 8 |
>>>>> ${PORTDIR}/app-misc/foo/foo-1a_live.ebuild |
| 9 |
>>>>> ${PORTDIR}/app-misc/foo-1a/foo-1a-live.ebuild |
| 10 |
>>>> you probably mean: |
| 11 |
>>>> ${PORTDIR}/app-misc/foo-1a/foo-1a.live.ebuild |
| 12 |
>>> No, I mean what I had written, namely to use an underscore as |
| 13 |
>>> separator, i.e., "_live". But when the version is just "live" alone, |
| 14 |
>>> one would suppress the underscore for aesthetic reasons, i.e. instead |
| 15 |
>>> of "foo-1a-_live" it would be "foo-1a-live". |
| 16 |
>>> |
| 17 |
>>>> but how would their vdb or binpkg names be unique? |
| 18 |
>>>> vdb for example: |
| 19 |
>>>> app-misc/foo-1a_live for app-misc/foo |
| 20 |
>>> PN=foo, PV=1a_live => app-misc/foo-1a_live |
| 21 |
>>> |
| 22 |
>>>> app-misc/foo-1a_live for app-misc/foo-1a |
| 23 |
>>> PN=foo-1a, PV=live => app-misc/foo-1a-live |
| 24 |
>>> |
| 25 |
>>>> am I missing something? |
| 26 |
>>> Everything is easy, if you keep the following rule in mind: |
| 27 |
>>> |
| 28 |
>>>>> With our current versioning scheme the rule is very simple: ${P} is |
| 29 |
>>>>> split into ${PN} and ${PV} at the last hyphen. This can be done in |
| 30 |
>>>>> a straight forward way by regexp matching, and I would really hate |
| 31 |
>>>>> to lose this nice property. |
| 32 |
>>>> I don't understand why this property is important. Can you please |
| 33 |
>>>> explain? |
| 34 |
>>> See above, it automatically avoids any ambiguities in splitting P into |
| 35 |
>>> PN and PV. And look at function "pkgsplit" in Portage: It can just |
| 36 |
>>> treat PV as an opaque string. |
| 37 |
>>> |
| 38 |
>>> What would be the advantage to use a hyphen instead of an underscore? |
| 39 |
>> |
| 40 |
>> Mainly the thing you observed yourself - foo_live is a bit |
| 41 |
>> inconsistent with current versions. |
| 42 |
> |
| 43 |
> Ulrich is proposing foo-live if live is the entire version, foo_live is not a |
| 44 |
> legal `package name and version'. (It could be a package name though.) |
| 45 |
|
| 46 |
I know, it's also inconsistent. Anyway that's really not that |
| 47 |
important. We could forbid package names that end with a -$PV where |
| 48 |
$PV is a valid version spec. That would make package names like foo-1b |
| 49 |
invalid (didn't see anything named like that in the tree anyway). |
| 50 |
|
| 51 |
>> The case you mention can be avoided with another restriction in PMS. |
| 52 |
>> Buut we might as well go all the way and change the version separator |
| 53 |
>> to -- or something, which would be the most flexible. |
| 54 |
> |
| 55 |
> That would also be a good solution though we don't seem to need it yet. It would |
| 56 |
> also entail compatibility issues. |
| 57 |
|
| 58 |
Not with g55. |
| 59 |
|
| 60 |
-- |
| 61 |
Best Regards, |
| 62 |
Piotr Jaroszyński |
Archives