Contents of /xml/htdocs/proj/en/glep/glep-0037.txt

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1.2 - (hide annotations) (download)
Tue Sep 5 20:54:30 2006 UTC (12 years, 4 months ago) by g2boojum
Branch: MAIN
Changes since 1.1: +11 -4 lines
File MIME type: text/plain

1 g2boojum 1.1 GLEP: 37
2     Title: Virtuals Deprecation
3 g2boojum 1.2 Version: $Revision: 1.1 $
4     Last-Modified: $Date: 2005/04/30 22:35:58 $
5 g2boojum 1.1 Author: Jason Stubbs <jstubbs@gentoo.org>
6 g2boojum 1.2 Status: deferred
7 g2boojum 1.1 Type: Standards Track
8     Content-Type: text/x-rst
9     Created: 30-April-2005
10 g2boojum 1.2 Post-History: 30-April-2005, 5-Sep-2006
11 g2boojum 1.1
13 g2boojum 1.2 Status
14     ======
16     What has been implemented so far diverges somewhat from what is
17     stated in this GLEP. As such, this GLEP (in its current form)
18     has been marked "deferred".
20 g2boojum 1.1 Credits
21     =======
23     Most ideas in this GLEP came out of discussion with Thomas de Grenier de
24     Latour. Ciaran McCreesh, Brian Harring and Stephen Bennett have also provided
25     help in fleshing out the idea.
28     Abstract
29     ========
31     This GLEP covers the pitfalls of the current virtuals system, the benefits of
32     using regular ebuilds to serve the purpose of virtuals and what needs to be
33     supported to make it viable.
36     Motivation
37     ==========
39     The current virtuals system is decentralized; that is there is no way to find
40     information about a specific virtual other than to scan all packages for what
41     they provide. There is also no way to tell whether an atom is a virtual or
42     not - yes, the "virtual/" prefix could have been used but it isn't which has
43     led to its abuse.
45     What this means is that portage must scan all installed packages for the
46     virtuals they provide, that profiles must provide a default for every single
47     virtual that portage might encounter and that every single atom that portage
48     processes must be checked against the list of virtuals. Needless to say that
49     this causes quite a performance decrease.
51     The current virtuals system also has some other major shortcomings. The most
52     well known case is virtual/jdk and kaffe. Kaffe-1.1.4 implements the Java 1.4
53     API but can not satisfy a package that requires >=virtual/jdk-1.4 because
54     kaffe's versioning scheme differs. (ED: Need to add some more here. ;)
57     Specification
58     =============
60     This GLEP recommends that virtuals become no different to regular packages.
61     Specifically, the standard virtual would include the DESCRIPTION, KEYWORDS,
62     IUSE and RDEPEND metadata. An example would be something like this::
64     DESCRIPTION="Java Development Kit 1.4"
65     KEYWORDS="amd64 hppa ia64 ppc ppc64 sparc x86"
66     RDEPEND="|| (
67     =dev-java/blackdown-jdk-1.4\*
68     =dev-java/ibm-jdk-bin-1.4*
69     =dev-java/jrockit-jdk-bin-1.4*
70     =dev-java/kaffe-1.1.4*
71     =dev-java/sun-jdk-1.4*
72     )"
73     IUSE=""
75     However, there are some issues that have been brought up with doing this.
78     Consistency
79     -----------
81     Presently, it is very easy to remove packages even if others are dependent
82     on them, which can lead to broken emerges when packages rely on indirect
83     dependencies. For example, if kdelibs is merged bringing in qt and then
84     qt is unmerged, attempting to merge kdebase will likely fail. This becomes
85     a much bigger problem with virtuals as packages because the dependencies
86     are always indirect.
88     The resolution for this issue will be to add full dependency tracking and
89     verification to portage. The details of how it will be done are outside the
90     scope of this GLEP, but essentially this means that portage will need to be
91     forced to unmerge a package that is depended on by another and will also be
92     able to scan and fix any broken dependencies.
95     Overrides
96     ---------
98     Profiles currently specify the default provider of each virtual and users are
99     able to override these defaults using /etc/portage/profile/virtuals. If
100     virtuals are replaced by regular packages and thus able to have arbitrarily
101     complex DEPENDs, the current method of overriding default virtuals can not
102     be extended to support this.
104     Before looking at a solution, lets look at how the current system works. When
105     portage initializes, it searches installed packages for available virtuals.
106     It then searches profiles and user overrides and adds them to the available
107     providers list and/or changes the order of the providers so that overrides are
108     listed earlier. Portage then expands any virtual atom it finds into an OR
109     list using the order decided upon at initialization.
111     To keep this behaviour available, this GLEP proposes a new file named
112     package.prefer. In its basic form, this is just a list of package names
113     ordered by preference. Portage would use it by reordering the atoms of any
114     OR list it processes to fit the order given by package.prefer. For example,
115     if package.prefer contained "dev-java/kaffe" then:
117     ::
119     || (
120     dev-java/blackdown-jdk
121     dev-java/sun-jdk
122     dev-java/kaffe
123     )
125     would be processed as:
127     ::
129     || (
130     dev-java/kaffe
131     dev-java/blackdown-jdk
132     dev-java/sun-jdk
133     )
135     In its basic form, package.prefer already covers profile and user overrides.
136     However, this GLEP proposes that any type of atom be usable. This will be
137     accomplished by checking for intersections of the atoms in the OR list and
138     atoms in the preferred list. When an intersection is found, both atoms
139     would be specified in a sublist, which would then be treated as a ranged dep.
140     For example, if package.prefer contained "<=dev-java/sun-jdk-1.4" then:
142     ::
144     || (
145     >=dev-java/blackdown-jdk-1.3
146     >=dev-java/sun-jdk-1.3
147     )
149     would be processed as:
151     ::
153     || (
154     (
155     <=dev-java/sun-jdk-1.4
156     >=dev-java/sun-jdk-1.3
157     )
158     >=dev-java/blackdown-jdk-1.3
159     >=dev-java/sun-jdk-1.3
160     )
162     Ranged deps are outside of the scope of this GLEP.
165     Rationale
166     =========
168     The number one advantage is that it offers more power to both the user and
169     the developer. Flexibility of virtuals is far greater in this scheme and
170     fulfills requirements that exist already. It also means that the maintainers
171     of profiles will not need to list a default for every virtual. The user
172     benefits by being able to easily gather a list of providers of a virtual as
173     well as their control being extended to allow selection where there is a
174     choice within any package.
176     Portage code also benefits from this scheme as virtuals will no longer
177     require special handling or dual implementations of essentially the same
178     feature, for example USE-based PROVIDEs. This scheme is also much easier to
179     optimize which will benefit the processing of all packages. It also means
180     that any additions to the DEPEND vocabulary become available for use in the
181     definitions of virtuals.
184     Backwards Compatibility
185     =======================
187     Compatibility will begin by making treat unknown virtuals like
188     regular packages. When the tree is stripped of PROVIDEs and "virtuals"
189     override files, the only virtuals that these portages will use are those that
190     the user has specified and those gleaned from installed packages. Any
191     unknown virtual will be treated like a regular package and looked for in the
192     tree.
194     The next major version of portage (2.1.0) will support consistency
195     checking. The only remaining issue is that of user overrides. The old
196     method will work even with new style virtuals. The only catch is that
197     complex virtuals - that is virtuals that will install more than one package -
198     may not be overridable satisfactorally.
200     Dropping of support of current style virtuals is planned for the following
201     major version of portage (2.2.0). When the time comes to release it, scripts
202     will be written to create packages from the existing virtuals system as well
203     as to create appropriate package.prefer overrides within the profiles.
206     Copyright
207     =========
209     This document has been placed in the public domain.

  ViewVC Help
Powered by ViewVC 1.1.20