/[gentoo]/xml/htdocs/proj/en/glep/glep-0047.html
Gentoo

Diff of /xml/htdocs/proj/en/glep/glep-0047.html

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

Revision 1.1 Revision 1.9
1<?xml version="1.0" encoding="utf-8" ?> 1<?xml version="1.0" encoding="utf-8" ?>
2<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> 2<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
3<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"> 3<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
4<!-- 4
5This HTML is auto-generated. DO NOT EDIT THIS FILE! If you are writing a new
6PEP, see http://www.python.org/peps/pep-0001.html for instructions and links
7to templates. DO NOT USE THIS HTML FILE AS YOUR TEMPLATE!
8-->
9<head> 5<head>
10 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> 6 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
11 <meta name="generator" content="Docutils 0.3.5: http://docutils.sourceforge.net/" /> 7 <meta name="generator" content="Docutils 0.4: http://docutils.sourceforge.net/" />
12 <title>GLEP 47 -- Creating 'safe' environment variables</title> 8 <title>GLEP 47 -- Creating 'safe' environment variables</title>
13 <link rel="stylesheet" href="tools/glep.css" type="text/css" /> 9 <link rel="stylesheet" href="tools/glep.css" type="text/css" />
14</head> 10</head>
15<body bgcolor="white"> 11<body bgcolor="white">
16<table class="navigation" cellpadding="0" cellspacing="0" 12<table class="navigation" cellpadding="0" cellspacing="0"
20<img src="http://www.gentoo.org/images/gentoo-new.gif" alt="[Gentoo]" 16<img src="http://www.gentoo.org/images/gentoo-new.gif" alt="[Gentoo]"
21 border="0" width="150" height="35" /></a></td> 17 border="0" width="150" height="35" /></a></td>
22<td class="textlinks" align="left"> 18<td class="textlinks" align="left">
23[<b><a href="http://www.gentoo.org/">Gentoo Linux Home</a></b>] 19[<b><a href="http://www.gentoo.org/">Gentoo Linux Home</a></b>]
24[<b><a href="http://www.gentoo.org/proj/en/glep">GLEP Index</a></b>] 20[<b><a href="http://www.gentoo.org/proj/en/glep">GLEP Index</a></b>]
25[<b><a href="./glep-0047.txt">GLEP Source</a></b>] 21[<b><a href="http://www.gentoo.org/proj/en/glep/glep-0047.txt">GLEP Source</a></b>]
26</td></tr></table> 22</td></tr></table>
27<div class="document">
28<table class="rfc2822 field-list" frame="void" rules="none"> 23<table class="rfc2822 docutils field-list" frame="void" rules="none">
29<col class="field-name" /> 24<col class="field-name" />
30<col class="field-body" /> 25<col class="field-body" />
31<tbody valign="top"> 26<tbody valign="top">
32<tr class="field"><th class="field-name">GLEP:</th><td class="field-body">47</td> 27<tr class="field"><th class="field-name">GLEP:</th><td class="field-body">47</td>
33</tr> 28</tr>
34<tr class="field"><th class="field-name">Title:</th><td class="field-body">Creating 'safe' environment variables</td> 29<tr class="field"><th class="field-name">Title:</th><td class="field-body">Creating 'safe' environment variables</td>
35</tr> 30</tr>
36<tr class="field"><th class="field-name">Version:</th><td class="field-body">1.8</td> 31<tr class="field"><th class="field-name">Version:</th><td class="field-body">1.6</td>
37</tr> 32</tr>
38<tr class="field"><th class="field-name">Last-Modified:</th><td class="field-body"><a class="reference" href="http://www.gentoo.org/cgi-bin/viewcvs/xml/htdocs/proj/en/glep/glep-0047.txt?cvsroot=gentoo">2006/01/26 20:56:55</a></td> 33<tr class="field"><th class="field-name">Last-Modified:</th><td class="field-body"><a class="reference" href="http://www.gentoo.org/cgi-bin/viewcvs.cgi/xml/htdocs/proj/en/glep/glep-0047.txt?cvsroot=gentoo">2007/07/22 10:03:18</a></td>
39</tr> 34</tr>
40<tr class="field"><th class="field-name">Author:</th><td class="field-body">Diego Pettenò, Fabian rroffen &lt;{flameeyes,grobian}&#64;gentoo.org&gt;</td> 35<tr class="field"><th class="field-name">Author:</th><td class="field-body">Diego Pettenò, Fabian Groffen</td>
41</tr> 36</tr>
42<tr class="field"><th class="field-name">Status:</th><td class="field-body">Active</td> 37<tr class="field"><th class="field-name">Status:</th><td class="field-body">Rejected</td>
43</tr> 38</tr>
44<tr class="field"><th class="field-name">Type:</th><td class="field-body">Proposal</td> 39<tr class="field"><th class="field-name">Type:</th><td class="field-body">Standards Track</td>
45</tr> 40</tr>
46<tr class="field"><th class="field-name">Content-Type:</th><td class="field-body"><a class="reference" href="glep-0012.html">text/x-rst</a></td> 41<tr class="field"><th class="field-name">Content-Type:</th><td class="field-body"><a class="reference" href="glep-0002.html">text/x-rst</a></td>
47</tr> 42</tr>
48<tr class="field"><th class="field-name">Created:</th><td class="field-body">14-Oct-2005</td> 43<tr class="field"><th class="field-name">Created:</th><td class="field-body">14-Oct-2005</td>
49</tr> 44</tr>
50<tr class="field"><th class="field-name">Post-History:</th><td class="field-body"></td> 45<tr class="field"><th class="field-name">Post-History:</th><td class="field-body">09-Feb-2006</td>
51</tr> 46</tr>
52</tbody> 47</tbody>
53</table> 48</table>
54<hr /> 49<hr />
55<div class="contents topic" id="contents"> 50<div class="contents topic">
56<p class="topic-title first"><a name="contents">Contents</a></p> 51<p class="topic-title first"><a id="contents" name="contents">Contents</a></p>
57<ul class="simple"> 52<ul class="simple">
58<li><a class="reference" href="#credits" id="id9" name="id9">Credits</a></li> 53<li><a class="reference" href="#credits" id="id5" name="id5">Credits</a></li>
59<li><a class="reference" href="#abstract" id="id10" name="id10">Abstract</a></li> 54<li><a class="reference" href="#abstract" id="id6" name="id6">Abstract</a></li>
60<li><a class="reference" href="#motivation" id="id11" name="id11">Motivation</a></li> 55<li><a class="reference" href="#motivation" id="id7" name="id7">Motivation</a></li>
61<li><a class="reference" href="#rationale" id="id12" name="id12">Rationale</a></li> 56<li><a class="reference" href="#rationale" id="id8" name="id8">Rationale</a></li>
62<li><a class="reference" href="#backwards-compatibility" id="id13" name="id13">Backwards Compatibility</a></li> 57<li><a class="reference" href="#backwards-compatibility" id="id9" name="id9">Backwards Compatibility</a></li>
63<li><a class="reference" href="#specification" id="id14" name="id14">Specification</a><ul> 58<li><a class="reference" href="#specification" id="id10" name="id10">Specification</a><ul>
64<li><a class="reference" href="#variables" id="id15" name="id15">Variables</a></li> 59<li><a class="reference" href="#variable-assignment" id="id11" name="id11">Variable Assignment</a></li>
65</ul> 60</ul>
66</li> 61</li>
67<li><a class="reference" href="#references" id="id16" name="id16">References</a></li> 62<li><a class="reference" href="#references" id="id12" name="id12">References</a></li>
68<li><a class="reference" href="#copyright" id="id17" name="id17">Copyright</a></li> 63<li><a class="reference" href="#copyright" id="id13" name="id13">Copyright</a></li>
69</ul> 64</ul>
70</div> 65</div>
71<div class="section" id="credits"> 66<div class="section">
72<h1><a class="toc-backref" href="#id9" name="credits">Credits</a></h1> 67<h1><a class="toc-backref" href="#id5" id="credits" name="credits">Credits</a></h1>
73<p>The text of this GLEP is a result of a discussion and input of the 68<p>The text of this GLEP is a result of a discussion and input of the
74following persons, in no particular order: Mike Frysinger, Diego 69following persons, in no particular order: Mike Frysinger, Diego
75Pettenò, Fabian Groffen and Finn Thain.</p> 70Pettenò, Fabian Groffen and Finn Thain.</p>
76</div> 71</div>
77<div class="section" id="abstract"> 72<div class="section">
78<h1><a class="toc-backref" href="#id10" name="abstract">Abstract</a></h1> 73<h1><a class="toc-backref" href="#id6" id="abstract" name="abstract">Abstract</a></h1>
79<p>In order for ebuilds and eclasses to be able to make host specific 74<p>In order for ebuilds and eclasses to be able to make host specific
80decisions, it is necessary to have a number of environmental variables 75decisions, it is necessary to have a number of environmental variables
81which allow for such decisions. This GLEP introduces some measures that 76which allow for such decisions. This GLEP introduces some measures that
82need to be made to make these decisions 'safe', by making sure the 77need to be made to make these decisions 'safe', by making sure the
83variables the decisions are based on are 'safe'. A small overlap with 78variables the decisions are based on are 'safe'. A small overlap with
84GLEP 22 <a class="footnote-reference" href="#id5" id="id1" name="id1">[1]</a> is being handled in this GLEP where the use of 2-tuple 79GLEP 22 <a class="footnote-reference" href="#id3" id="id1" name="id1">[1]</a> is being handled in this GLEP where the use of 2-tuple
85keywords are being kept instead of 4-tuple keywords. Additionally, the 80keywords are being kept instead of 4-tuple keywords. Additionally, the
86<tt class="literal"><span class="pre">ELIBC</span></tt>, <tt class="literal"><span class="pre">KERNEL</span></tt> and <tt class="literal"><span class="pre">ARCH</span></tt> get auto filled starting from 81<tt class="docutils literal"><span class="pre">ELIBC</span></tt>, <tt class="docutils literal"><span class="pre">KERNEL</span></tt> and <tt class="docutils literal"><span class="pre">ARCH</span></tt> get auto filled starting from
87<tt class="literal"><span class="pre">CHOST</span></tt> and the 2-tuple keyword, instead of solely from they 4-tuple 82<tt class="docutils literal"><span class="pre">CHOST</span></tt> and the 2-tuple keyword, instead of solely from they 4-tuple
88keyword as proposed in GLEP 22.</p> 83keyword as proposed in GLEP 22.</p>
89<p>The destiny of the <tt class="literal"><span class="pre">USERLAND</span></tt> variable is out of the scope of this 84<p>The destiny of the <tt class="docutils literal"><span class="pre">USERLAND</span></tt> variable is out of the scope of this
90GLEP. Depending on its presence in the tree, it may be decided to set 85GLEP. Depending on its presence in the tree, it may be decided to set
91this variable the same way we propose to set <tt class="literal"><span class="pre">ELIBC</span></tt>, <tt class="literal"><span class="pre">KERNEL</span></tt> and 86this variable the same way we propose to set <tt class="docutils literal"><span class="pre">ELIBC</span></tt>, <tt class="docutils literal"><span class="pre">KERNEL</span></tt> and
92<tt class="literal"><span class="pre">ARCH</span></tt>, or alternatively, e.g. via the profiles.</p> 87<tt class="docutils literal"><span class="pre">ARCH</span></tt>, or alternatively, e.g. via the profiles.</p>
93</div> 88</div>
94<div class="section" id="motivation"> 89<div class="section">
95<h1><a class="toc-backref" href="#id11" name="motivation">Motivation</a></h1> 90<h1><a class="toc-backref" href="#id7" id="motivation" name="motivation">Motivation</a></h1>
96<p>The Gentoo/Alt project is in an emerging state to get ready to serve a 91<p>The Gentoo/Alt project is in an emerging state to get ready to serve a
97plethora of 'alternative' configurations such as FreeBSD, NetBSD, 92plethora of 'alternative' configurations such as FreeBSD, NetBSD,
98DragonflyBSD, GNU/kFreeBSD, Mac OS X, (Open)Darwin, (Open)Solaris and so 93DragonflyBSD, GNU/kFreeBSD, Mac OS X, (Open)Darwin, (Open)Solaris and so
99on. As such, the project is in need for a better grip on the actual 94on. As such, the project is in need for a better grip on the actual
100host being built on. This information on the host environment is 95host being built on. This information on the host environment is
101necessary to make proper (automated) decisions on settings that are 96necessary to make proper (automated) decisions on settings that are
102highly dependant on the build environment, such as platform or as in 97highly dependant on the build environment, such as platform or C-library
103<a class="footnote-reference" href="#id6" id="id2" name="id2">[2]</a>.</p> 98implementation.</p>
104</div> 99</div>
105<div class="section" id="rationale"> 100<div class="section">
106<h1><a class="toc-backref" href="#id12" name="rationale">Rationale</a></h1> 101<h1><a class="toc-backref" href="#id8" id="rationale" name="rationale">Rationale</a></h1>
107<p>Gentoo's unique Portage system allows easy installation of applications 102<p>Gentoo's unique Portage system allows easy installation of applications
108from source packages. Compiling sources is prone to many environmental 103from source packages. Compiling sources is prone to many environmental
109settings and availability of certain tools. Only recently the Gentoo 104settings and availability of certain tools. Only recently the Gentoo
110for FreeBSD project has started, as second Gentoo project that operates 105for FreeBSD project has started, as second Gentoo project that operates
111on a foreign host operating system using foreign (non-GNU) C-libraries 106on a foreign host operating system using foreign (non-GNU) C-libraries
116environmental differences, information regarding it should be supplied. 111environmental differences, information regarding it should be supplied.
117Since decisions based on this information can be vital, it is of high 112Since decisions based on this information can be vital, it is of high
118importance that this information can be trusted and the values can be 113importance that this information can be trusted and the values can be
119considered 'safe' and correct.</p> 114considered 'safe' and correct.</p>
120</div> 115</div>
121<div class="section" id="backwards-compatibility"> 116<div class="section">
122<h1><a class="toc-backref" href="#id13" name="backwards-compatibility">Backwards Compatibility</a></h1> 117<h1><a class="toc-backref" href="#id9" id="backwards-compatibility" name="backwards-compatibility">Backwards Compatibility</a></h1>
123<p>The proposed keywording scheme in this GLEP is fully compatible with the 118<p>The proposed keywording scheme in this GLEP is fully compatible with the
124current situation of the portage tree, this in contrast to GLEP 22. The 119current situation of the portage tree, this in contrast to GLEP 22. The
125variables provided by GLEP 22 can't be extracted from the new keyword, 120variables provided by GLEP 22 can't be extracted from the new keyword,
126but since GLEP 22-style keywords aren't in the tree at the moment, that 121but since GLEP 22-style keywords aren't in the tree at the moment, that
127is not a problem. The same information can be extracted from the CHOST 122is not a problem. The same information can be extracted from the CHOST
128variable, if necessary. No modifications to ebuilds will have to be 123variable, if necessary. No modifications to ebuilds will have to be
129made.</p> 124made.</p>
130</div> 125</div>
131<div class="section" id="specification"> 126<div class="section">
132<h1><a class="toc-backref" href="#id14" name="specification">Specification</a></h1> 127<h1><a class="toc-backref" href="#id10" id="specification" name="specification">Specification</a></h1>
133<p>Unlike GLEP 22 the current keyword scheme as used in practice is not 128<p>Unlike GLEP 22 the currently used keyword scheme is not changed.
134changed. Instead of proposing a 4-tuple <a class="footnote-reference" href="#id7" id="id3" name="id3">[3]</a> keyword, a 2-tuple 129Instead of proposing a 4-tuple <a class="footnote-reference" href="#id4" id="id2" name="id2">[2]</a> keyword, a 2-tuple keyword is chosen
135keyword is chosen for archs that require them. Archs for which a 130for archs that require them. Archs for which a 1-tuple keyword
1361-tuple keyword suffices, keep that keyword. Since this doesn't change 131suffices, can keep that keyword. Since this doesn't change anything to
137anything to the current situation in the tree, it is considered to be a 132the current situation in the tree, it is considered to be a big
138big advantage over the 4-tuple keyword from GLEP 22. This GLEP is an 133advantage over the 4-tuple keyword from GLEP 22. This GLEP is an
139official specification of the syntax of the keyword.</p> 134official specification of the syntax of the keyword.</p>
140<p>Keywords will consist out of two parts separated by a hyphen ('-'). The 135<p>Keywords will consist out of two parts separated by a hyphen ('-'). The
141left hand part of the keyword 2-tuple is the architecture, such as 136part up to the first hyphen from the left of the keyword 2-tuple is the
142ppc64, sparc and x86. The right hand part indicates the operating 137architecture, such as ppc64, sparc and x86. Allowed characters for the
138architecture name are in <tt class="docutils literal"><span class="pre">a-z0-9</span></tt>. The remaining part on the right of
139the first hyphen from the left indicates the operating system or
143system or distribution, such as linux, macos, darwin, obsd, etc. If the 140distribution, such as linux, macos, darwin, obsd, et-cetera. If the
144right hand part is omitted, it implies the operating system/distribution 141right hand part is omitted, it implies the operating system/distribution
145type is GNU/Linux. In such case the hyphen is also omitted. Examples 142type is Gentoo GNU/Linux. In such case the hyphen is also omitted, and
146of such keywords are ppc-darwin and x86. This is fully compatible with 143the keyword consists of solely the architecture. The operating system
147the current keywords used in the tree.</p> 144or distribution name can consist out of characters in <tt class="docutils literal"><span class="pre">a-zA-Z0-9_+:-</span></tt>.
145Please note that the hyphen is an allowed character, and therefore the
146separation of the two fields in the keyword is only determinable by
147scanning for the first hyphen character from the start of the keyword
148string. Examples of keywords following this specification are
149ppc-darwin and x86. This is fully compatible with the current use of
150keywords in the tree.</p>
148<p>The variables <tt class="literal"><span class="pre">ELIBC</span></tt>, <tt class="literal"><span class="pre">KERNEL</span></tt> and <tt class="literal"><span class="pre">ARCH</span></tt> are currently set in 151<p>The variables <tt class="docutils literal"><span class="pre">ELIBC</span></tt>, <tt class="docutils literal"><span class="pre">KERNEL</span></tt> and <tt class="docutils literal"><span class="pre">ARCH</span></tt> are currently set in
149the profiles when other than their defaults for a GNU/Linux system. 152the profiles when other than their defaults for a GNU/Linux system.
150They can as such easily be overridden and defined by the user. To 153They can as such easily be overridden and defined by the user. To
151prevent this from happening, the variables should be auto filled by 154prevent this from happening, the variables should be auto filled by
152Portage itself, based on the <tt class="literal"><span class="pre">CHOST</span></tt> variable.</p> 155Portage itself, based on the <tt class="docutils literal"><span class="pre">CHOST</span></tt> variable. While the <tt class="docutils literal"><span class="pre">CHOST</span></tt>
156variable can be as easy as the others set by the user, it still is
157assumed to be 'safe'. This assumption is grounded in the fact that the
158variable itself is being used in various other places with the same
159intention, and that an invalid <tt class="docutils literal"><span class="pre">CHOST</span></tt> will cause major malfunctioning
160of the system. A user that changes the <tt class="docutils literal"><span class="pre">CHOST</span></tt> into something that is
161not valid for the system, is already warned that this might render the
162system unusable. Concluding, the 'safeness' of the <tt class="docutils literal"><span class="pre">CHOST</span></tt> variable
163is based on externally assumed 'safeness', which's discussion falls
164outside this GLEP.</p>
165<p>Current USE-expansion of the variables is being maintained, as this
166results in full backward compatibility. Since the variables themself
167don't change in what they represent, but only how they are being
168assigned, there should be no problem in maintaining them. Using
169USE-expansion, conditional code can be written down in ebuilds, which is
170not different from any existing methods at all:</p>
171<pre class="literal-block">
172...
173RDEPEND=&quot;elibc_FreeBSD? ( sys-libs/com_err )&quot;
174...
175src_compile() {
176 ...
177 use elibc_FreeBSD &amp;&amp; myconf=&quot;${myconf} -Dlibc=/usr/lib/libc.a&quot;
178 ...
179}
180</pre>
181<p>Alternatively, the variables <tt class="docutils literal"><span class="pre">ELIBC</span></tt>, <tt class="docutils literal"><span class="pre">KERNEL</span></tt> and <tt class="docutils literal"><span class="pre">ARCH</span></tt>
182are available in the ebuild evironment and they can be used instead of
183invoking <tt class="docutils literal"><span class="pre">xxx_Xxxx</span></tt> or in switch statements where they are actually
184necessary.</p>
153<p>A map file can be used to have the various <tt class="literal"><span class="pre">CHOST</span></tt> values being 185<p>A map file can be used to have the various <tt class="docutils literal"><span class="pre">CHOST</span></tt> values being
154translated to the correct values for the four variables. This change is 186translated to the correct values for the four variables. This change is
155invisible for ebuilds and eclasses, but allows to rely on these 187invisible for ebuilds and eclasses, but allows to rely on these
156variables as they are based on a 'safe' value -- the <tt class="literal"><span class="pre">CHOST</span></tt> variable. 188variables as they are based on a 'safe' value -- the <tt class="docutils literal"><span class="pre">CHOST</span></tt> variable.
157Ebuilds should not be sensitive to the keyword value, but use the 189Ebuilds should not be sensitive to the keyword value, but use the
158aforementioned four variables instead. They allow specific tests for 190aforementioned four variables instead. They allow specific tests for
159properties. If this is undesirable, the full <tt class="literal"><span class="pre">CHOST</span></tt> variable can be 191properties. If this is undesirable, the full <tt class="docutils literal"><span class="pre">CHOST</span></tt> variable can be
160used to match a complete operating system.</p> 192used to match a complete operating system.</p>
161<p>Current USE-expansion is being maintained, for backwards compatibility. 193<div class="section">
162Since the expansion is based on the variables mentioned above which do
163not change, but only in the way they are generated, there should be no
164problem in maintaining them.</p>
165<div class="section" id="variables">
166<h2><a class="toc-backref" href="#id15" name="variables">Variables</a></h2> 194<h2><a class="toc-backref" href="#id11" id="variable-assignment" name="variable-assignment">Variable Assignment</a></h2>
167<p>The <tt class="literal"><span class="pre">ELIBC</span></tt>, <tt class="literal"><span class="pre">KERNEL</span></tt>, <tt class="literal"><span class="pre">ARCH</span></tt> variables are filled from a profile 195<p>The <tt class="docutils literal"><span class="pre">ELIBC</span></tt>, <tt class="docutils literal"><span class="pre">KERNEL</span></tt>, <tt class="docutils literal"><span class="pre">ARCH</span></tt> variables are filled from a profile
168file. The file can be overlaid, such that the following entries in the 196file. The file can be overlaid, such that the following entries in the
169map file (on the left of the arrow) will result in the assigned 197map file (on the left of the arrow) will result in the assigned
170variables on the right hand side of the arrow:</p> 198variables on the right hand side of the arrow:</p>
171<pre class="literal-block"> 199<pre class="literal-block">
172*-*-linux-* -&gt; KERNEL=&quot;linux&quot; 200*-*-linux-* -&gt; KERNEL=&quot;linux&quot;
173*-*-*-gnu -&gt; ELIBC=&quot;glibc&quot; 201*-*-*-gnu -&gt; ELIBC=&quot;glibc&quot;
174*-*-kfreebsd-gnu -&gt; KERNEL=&quot;FreeBSD&quot; ELIBC=&quot;glibc&quot; 202*-*-kfreebsd-gnu -&gt; KERNEL=&quot;FreeBSD&quot; ELIBC=&quot;glibc&quot;
175*-*-freebsd* -&gt; KERNEL=&quot;FreeBSD&quot; ELIBC=&quot;FreeBSD&quot; 203*-*-freebsd* -&gt; KERNEL=&quot;FreeBSD&quot; ELIBC=&quot;FreeBSD&quot;
176*-*-darwin* -&gt; KERNEL=&quot;Darwin&quot; ELIBC=&quot;Darwin&quot; 204*-*-darwin* -&gt; KERNEL=&quot;Darwin&quot; ELIBC=&quot;Darwin&quot;
177*-*-netbsd* -&gt; KERNEL=&quot;NetBSD&quot; ELIBC=&quot;NetBSD&quot; 205*-*-netbsd* -&gt; KERNEL=&quot;NetBSD&quot; ELIBC=&quot;NetBSD&quot;
178*-*-solaris* -&gt; KERNEL=&quot;Solaris&quot; ELIBC=&quot;Solaris&quot; 206*-*-solaris* -&gt; KERNEL=&quot;Solaris&quot; ELIBC=&quot;Solaris&quot;
179</pre> 207</pre>
180<p>A way to achieve this is proposed by Mike Frysinger <a class="footnote-reference" href="#id8" id="id4" name="id4">[4]</a>, which 208<p>A way to achieve this is proposed by Mike Frysinger, which
181suggests to have a env-map file, for instance filled with:</p> 209suggests to have an env-map file, for instance filled with:</p>
182<pre class="literal-block"> 210<pre class="literal-block">
183% cat env-map 211% cat env-map
184*-linux-* KERNEL=linux 212*-linux-* KERNEL=linux
185*-gnu ELIBC=glibc 213*-gnu ELIBC=glibc
186x86_64-* ARCH=amd64 214x86_64-* ARCH=amd64
187</pre> 215</pre>
188<p>then the following bash script can be used to set the four variables to 216<p>then the following bash script can be used to set the four variables to
189their correct values:</p> 217their correct values:</p>
190<pre class="literal-block"> 218<pre class="literal-block">
191% cat readmap 219% cat readmap
192#!/bin/bash 220#!/bin/bash
193 221
194CBUILD=${CBUILD:-${CHOST=${CHOST:-$1}}} 222CBUILD=${CBUILD:-${CHOST=${CHOST:-$1}}}
195[[ -z ${CHOST} ]] &amp;&amp; echo need chost 223[[ -z ${CHOST} ]] &amp;&amp; echo need chost
196 224
197unset KERNEL ELIBC ARCH 225unset KERNEL ELIBC ARCH
198 226
199while read LINE ; do 227while read LINE ; do
200 set -- ${LINE} 228 set -- ${LINE}
201 targ=$1 229 targ=$1
202 shift 230 shift
203 [[ ${CBUILD} == ${targ} ]] &amp;&amp; eval $&#64; 231 [[ ${CBUILD} == ${targ} ]] &amp;&amp; eval $&#64;
204done &lt; env-map 232done &lt; env-map
205 233
206echo ARCH=${ARCH} KERNEL=${KERNEL} ELIBC=${ELIBC} 234echo ARCH=${ARCH} KERNEL=${KERNEL} ELIBC=${ELIBC}
207</pre> 235</pre>
208<p>Given the example env-map file, this script would result in:</p> 236<p>Given the example env-map file, this script would result in:</p>
209<pre class="literal-block"> 237<pre class="literal-block">
210% ./readmap x86_64-pc-linux-gnu 238% ./readmap x86_64-pc-linux-gnu
211ARCH=amd64 KERNEL=linux ELIBC=glibc 239ARCH=amd64 KERNEL=linux ELIBC=glibc
212</pre> 240</pre>
241<p>The entries in the <tt class="docutils literal"><span class="pre">env-map</span></tt> file will be evaluated in a forward
242linear full scan. A side-effect of this exhaustive search is that the
243variables can be re-assigned if multiple entries match the given
244<tt class="docutils literal"><span class="pre">CHOST</span></tt>. Because of this, the order of the entries does matter.
245Because the <tt class="docutils literal"><span class="pre">env-map</span></tt> file size is assumed not to exceed the block
246size of the file system, the performance penalty of a full scan versus
247'first-hit-stop technique' is assumed to be minimal.</p>
213<p>It should be noted, however, that the bash script is a proof of concept 248<p>It should be noted, however, that the above bash script is a proof of
214implementation. It cannot be used as Portage will need this 249concept implementation. Since Portage is largerly written in Python, it
215information, which is written in Python. Hence, an equivalent of this 250will be more efficient to write an equivalent of this code in Python
216bash script should be written in Python to be able to use it within 251also. Coding wise, this is considered to be a non-issue, but the format
217Portage. This is considered to be a non-issue coding wise.</p> 252of the <tt class="docutils literal"><span class="pre">env-map</span></tt> file, and especially its wildcard characters, might
253not be the best match with Python. For this purpose, the format
254specification of the <tt class="docutils literal"><span class="pre">env-map</span></tt> file is deferred to the Python
255implementation, and only the requirements are given here.</p>
256<p>The <tt class="docutils literal"><span class="pre">env-map</span></tt> file should be capable of encoding a <tt class="docutils literal"><span class="pre">key</span></tt>, <tt class="docutils literal"><span class="pre">value</span></tt>
257pair, where <tt class="docutils literal"><span class="pre">key</span></tt> is a (regular) expression that matches a
258chost-string, and <tt class="docutils literal"><span class="pre">value</span></tt> contains at least one, distinct variable
259assignment for the variables <tt class="docutils literal"><span class="pre">ARCH</span></tt>, <tt class="docutils literal"><span class="pre">KERNEL</span></tt> and <tt class="docutils literal"><span class="pre">ELIBC</span></tt>. The
260interpreter of the <tt class="docutils literal"><span class="pre">env-map</span></tt> file must scan the file linearly and
261continue trying to match the <tt class="docutils literal"><span class="pre">key</span></tt>s and assign variables if
262appropriate until the end of file.</p>
263<p>Since Portage will use the <tt class="docutils literal"><span class="pre">env-map</span></tt> file, the location of the file is
264beyond the scope of this GLEP and up to the Portage implementors.</p>
218</div> 265</div>
219</div> 266</div>
220<div class="section" id="references"> 267<div class="section">
221<h1><a class="toc-backref" href="#id16" name="references">References</a></h1> 268<h1><a class="toc-backref" href="#id12" id="references" name="references">References</a></h1>
222<table class="footnote" frame="void" id="id5" rules="none"> 269<table class="docutils footnote" frame="void" id="id3" rules="none">
223<colgroup><col class="label" /><col /></colgroup> 270<colgroup><col class="label" /><col /></colgroup>
224<tbody valign="top"> 271<tbody valign="top">
225<tr><td class="label"><a class="fn-backref" href="#id1" name="id5">[1]</a></td><td>GLEP 22, New &quot;keyword&quot; system to incorporate various 272<tr><td class="label"><a class="fn-backref" href="#id1" name="id3">[1]</a></td><td>GLEP 22, New &quot;keyword&quot; system to incorporate various
226userlands/kernels/archs, Goodyear, 273userlands/kernels/archs, Goodyear,
227(<a class="reference" href="http://glep.gentoo.org/glep-0022.html">http://glep.gentoo.org/glep-0022.html</a>)</td></tr> 274(<a class="reference" href="http://glep.gentoo.org/glep-0022.html">http://glep.gentoo.org/glep-0022.html</a>)</td></tr>
228</tbody> 275</tbody>
229</table> 276</table>
230<table class="footnote" frame="void" id="id6" rules="none"> 277<table class="docutils footnote" frame="void" id="id4" rules="none">
231<colgroup><col class="label" /><col /></colgroup> 278<colgroup><col class="label" /><col /></colgroup>
232<tbody valign="top"> 279<tbody valign="top">
233<tr><td class="label"><a class="fn-backref" href="#id2" name="id6">[2]</a></td><td><p>For example in the perl ebuild, it is necessary to 280<tr><td class="label"><a class="fn-backref" href="#id2" name="id4">[2]</a></td><td>For the purpose of readability, we will refer to 1, 2 and
234fill in the C-library part, which on a FreeBSD system is other than 2814-tuples, even though tuple in itself suggest a field consisting of
235on a Linux system and currently is handled as follows:</p> 282two values. For clarity: a 1-tuple describes a single value field,
236<pre class="literal-block"> 283while a 4-tuple describes a field consisting out of four values.</td></tr>
237[[ ${ELIBC} == &quot;FreeBSD&quot; ]] &amp;&amp; myconf=&quot;${myconf} -Dlibc=/usr/lib/libc.a&quot;
238</pre>
239</td></tr>
240</tbody> 284</tbody>
241</table> 285</table>
242<table class="footnote" frame="void" id="id7" rules="none">
243<colgroup><col class="label" /><col /></colgroup>
244<tbody valign="top">
245<tr><td class="label"><a class="fn-backref" href="#id3" name="id7">[3]</a></td><td>For the purpose of readability, we will refer to 1, 2 and
2464-tuples, even though tuple in itself suggest a field consisting of
247two values. For clarity: a 1-tuple describes a single value field,
248while a 4-tuple decribes a field consisting out of four values.</td></tr>
249</tbody>
250</table>
251<table class="footnote" frame="void" id="id8" rules="none">
252<colgroup><col class="label" /><col /></colgroup>
253<tbody valign="top">
254<tr><td class="label"><a class="fn-backref" href="#id4" name="id8">[4]</a></td><td><a class="reference" href="mailto:vapier">mailto:vapier</a> [at) gentoo .dot org</td></tr>
255</tbody>
256</table>
257</div> 286</div>
258<div class="section" id="copyright"> 287<div class="section">
259<h1><a class="toc-backref" href="#id17" name="copyright">Copyright</a></h1> 288<h1><a class="toc-backref" href="#id13" id="copyright" name="copyright">Copyright</a></h1>
260<p>This document has been placed in the public domain.</p> 289<p>This document has been placed in the public domain.</p>
261</div> 290</div>
262</div>
263 291
292</div>
293<div class="footer">
264<hr class="footer" /> 294<hr class="footer" />
265<div class="footer">
266<a class="reference" href="glep-0047.txt">View document source</a>. 295<a class="reference" href="glep-0047.txt">View document source</a>.
267Generated on: 2006-02-09 21:32 UTC. 296Generated on: 2007-10-13 13:39 UTC.
268Generated by <a class="reference" href="http://docutils.sourceforge.net/">Docutils</a> from <a class="reference" href="http://docutils.sourceforge.net/rst.html">reStructuredText</a> source. 297Generated by <a class="reference" href="http://docutils.sourceforge.net/">Docutils</a> from <a class="reference" href="http://docutils.sourceforge.net/rst.html">reStructuredText</a> source.
298
269</div> 299</div>
270</body> 300</body>
271</html> 301</html>
272 302

Legend:
Removed from v.1.1  
changed lines
  Added in v.1.9

  ViewVC Help
Powered by ViewVC 1.1.20