/[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 1.2
… …
31<tbody valign="top"> 31<tbody valign="top">
32<tr class="field"><th class="field-name">GLEP:</th><td class="field-body">47</td> 32<tr class="field"><th class="field-name">GLEP:</th><td class="field-body">47</td>
33</tr> 33</tr>
34<tr class="field"><th class="field-name">Title:</th><td class="field-body">Creating 'safe' environment variables</td> 34<tr class="field"><th class="field-name">Title:</th><td class="field-body">Creating 'safe' environment variables</td>
35</tr> 35</tr>
36<tr class="field"><th class="field-name">Version:</th><td class="field-body">1.8</td> 36<tr class="field"><th class="field-name">Version:</th><td class="field-body">1.1</td>
37</tr> 37</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> 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">2006/02/09 21:42:57</a></td>
39</tr> 39</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> 40<tr class="field"><th class="field-name">Author:</th><td class="field-body">Diego Pettenò, Fabian Groffen</td>
41</tr> 41</tr>
42<tr class="field"><th class="field-name">Status:</th><td class="field-body">Active</td> 42<tr class="field"><th class="field-name">Status:</th><td class="field-body">Draft</td>
43</tr> 43</tr>
44<tr class="field"><th class="field-name">Type:</th><td class="field-body">Proposal</td> 44<tr class="field"><th class="field-name">Type:</th><td class="field-body">Standards Track</td>
45</tr> 45</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> 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>
47</tr> 47</tr>
48<tr class="field"><th class="field-name">Created:</th><td class="field-body">14-Oct-2005</td> 48<tr class="field"><th class="field-name">Created:</th><td class="field-body">14-Oct-2005</td>
49</tr> 49</tr>
50<tr class="field"><th class="field-name">Post-History:</th><td class="field-body"></td> 50<tr class="field"><th class="field-name">Post-History:</th><td class="field-body">09-Feb-2006</td>
51</tr> 51</tr>
52</tbody> 52</tbody>
53</table> 53</table>
54<hr /> 54<hr />
55<div class="contents topic" id="contents"> 55<div class="contents topic" id="contents">
56<p class="topic-title first"><a name="contents">Contents</a></p> 56<p class="topic-title first"><a name="contents">Contents</a></p>
57<ul class="simple"> 57<ul class="simple">
58<li><a class="reference" href="#credits" id="id9" name="id9">Credits</a></li> 58<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> 59<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> 60<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> 61<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> 62<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> 63<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> 64<li><a class="reference" href="#variable-assignment" id="id11" name="id11">Variable Assignment</a></li>
65</ul> 65</ul>
66</li> 66</li>
67<li><a class="reference" href="#references" id="id16" name="id16">References</a></li> 67<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> 68<li><a class="reference" href="#copyright" id="id13" name="id13">Copyright</a></li>
69</ul> 69</ul>
70</div> 70</div>
71<div class="section" id="credits"> 71<div class="section" id="credits">
72<h1><a class="toc-backref" href="#id9" name="credits">Credits</a></h1> 72<h1><a class="toc-backref" href="#id5" name="credits">Credits</a></h1>
73<p>The text of this GLEP is a result of a discussion and input of the 73<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 74following persons, in no particular order: Mike Frysinger, Diego
75Pettenò, Fabian Groffen and Finn Thain.</p> 75Pettenò, Fabian Groffen and Finn Thain.</p>
76</div> 76</div>
77<div class="section" id="abstract"> 77<div class="section" id="abstract">
78<h1><a class="toc-backref" href="#id10" name="abstract">Abstract</a></h1> 78<h1><a class="toc-backref" href="#id6" name="abstract">Abstract</a></h1>
79<p>In order for ebuilds and eclasses to be able to make host specific 79<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 80decisions, it is necessary to have a number of environmental variables
81which allow for such decisions. This GLEP introduces some measures that 81which allow for such decisions. This GLEP introduces some measures that
82need to be made to make these decisions 'safe', by making sure the 82need to be made to make these decisions 'safe', by making sure the
83variables the decisions are based on are 'safe'. A small overlap with 83variables 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 84GLEP 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 85keywords 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 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
87<tt class="literal"><span class="pre">CHOST</span></tt> and the 2-tuple keyword, instead of solely from they 4-tuple 87<tt class="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> 88keyword 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 89<p>The destiny of the <tt class="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 90GLEP. 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 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
92<tt class="literal"><span class="pre">ARCH</span></tt>, or alternatively, e.g. via the profiles.</p> 92<tt class="literal"><span class="pre">ARCH</span></tt>, or alternatively, e.g. via the profiles.</p>
93</div> 93</div>
94<div class="section" id="motivation"> 94<div class="section" id="motivation">
95<h1><a class="toc-backref" href="#id11" name="motivation">Motivation</a></h1> 95<h1><a class="toc-backref" href="#id7" name="motivation">Motivation</a></h1>
96<p>The Gentoo/Alt project is in an emerging state to get ready to serve a 96<p>The Gentoo/Alt project is in an emerging state to get ready to serve a
97plethora of 'alternative' configurations such as FreeBSD, NetBSD, 97plethora of 'alternative' configurations such as FreeBSD, NetBSD,
98DragonflyBSD, GNU/kFreeBSD, Mac OS X, (Open)Darwin, (Open)Solaris and so 98DragonflyBSD, 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 99on. 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 100host being built on. This information on the host environment is
101necessary to make proper (automated) decisions on settings that are 101necessary to make proper (automated) decisions on settings that are
102highly dependant on the build environment, such as platform or as in 102highly 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> 103implementation.</p>
104</div> 104</div>
105<div class="section" id="rationale"> 105<div class="section" id="rationale">
106<h1><a class="toc-backref" href="#id12" name="rationale">Rationale</a></h1> 106<h1><a class="toc-backref" href="#id8" name="rationale">Rationale</a></h1>
107<p>Gentoo's unique Portage system allows easy installation of applications 107<p>Gentoo's unique Portage system allows easy installation of applications
108from source packages. Compiling sources is prone to many environmental 108from source packages. Compiling sources is prone to many environmental
109settings and availability of certain tools. Only recently the Gentoo 109settings and availability of certain tools. Only recently the Gentoo
110for FreeBSD project has started, as second Gentoo project that operates 110for FreeBSD project has started, as second Gentoo project that operates
111on a foreign host operating system using foreign (non-GNU) C-libraries 111on a foreign host operating system using foreign (non-GNU) C-libraries
… …
117Since decisions based on this information can be vital, it is of high 117Since decisions based on this information can be vital, it is of high
118importance that this information can be trusted and the values can be 118importance that this information can be trusted and the values can be
119considered 'safe' and correct.</p> 119considered 'safe' and correct.</p>
120</div> 120</div>
121<div class="section" id="backwards-compatibility"> 121<div class="section" id="backwards-compatibility">
122<h1><a class="toc-backref" href="#id13" name="backwards-compatibility">Backwards Compatibility</a></h1> 122<h1><a class="toc-backref" href="#id9" name="backwards-compatibility">Backwards Compatibility</a></h1>
123<p>The proposed keywording scheme in this GLEP is fully compatible with the 123<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 124current 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, 125variables 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 126but 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 127is not a problem. The same information can be extracted from the CHOST
128variable, if necessary. No modifications to ebuilds will have to be 128variable, if necessary. No modifications to ebuilds will have to be
129made.</p> 129made.</p>
130</div> 130</div>
131<div class="section" id="specification"> 131<div class="section" id="specification">
132<h1><a class="toc-backref" href="#id14" name="specification">Specification</a></h1> 132<h1><a class="toc-backref" href="#id10" name="specification">Specification</a></h1>
133<p>Unlike GLEP 22 the current keyword scheme as used in practice is not 133<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 134Instead 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 135for archs that require them. Archs for which a 1-tuple keyword
1361-tuple keyword suffices, keep that keyword. Since this doesn't change 136suffices, 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 137the 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 138advantage over the 4-tuple keyword from GLEP 22. This GLEP is an
139official specification of the syntax of the keyword.</p> 139official specification of the syntax of the keyword.</p>
140<p>Keywords will consist out of two parts separated by a hyphen ('-'). The 140<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 141left hand part of the keyword 2-tuple is the architecture, such as
142ppc64, sparc and x86. The right hand part indicates the operating 142ppc64, sparc and x86. The right hand part indicates the operating
143system or distribution, such as linux, macos, darwin, obsd, etc. If the 143system or distribution, such as linux, macos, darwin, obsd, etc. If the
144right hand part is omitted, it implies the operating system/distribution 144right hand part is omitted, it implies the operating system/distribution
145type is GNU/Linux. In such case the hyphen is also omitted. Examples 145type is Gentoo GNU/Linux. In such case the hyphen is also omitted.
146of such keywords are ppc-darwin and x86. This is fully compatible with 146Examples of such keywords are ppc-darwin and x86. This is fully
147the current keywords used in the tree.</p> 147compatible with the current use of keywords 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 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
149the profiles when other than their defaults for a GNU/Linux system. 149the profiles when other than their defaults for a GNU/Linux system.
150They can as such easily be overridden and defined by the user. To 150They can as such easily be overridden and defined by the user. To
151prevent this from happening, the variables should be auto filled by 151prevent 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> 152Portage itself, based on the <tt class="literal"><span class="pre">CHOST</span></tt> variable. While the <tt class="literal"><span class="pre">CHOST</span></tt>
153variable can be as easy as the others set by the user, it still is
154assumed to be 'safe'. This assumption is grounded in the fact that the
155variable itself is being used in various other places with the same
156intention, and that an invalid <tt class="literal"><span class="pre">CHOST</span></tt> will cause major malfunctioning
157of the system. A user that changes the <tt class="literal"><span class="pre">CHOST</span></tt> into something that is
158not valid for the system, is already warned that this might render the
159system unusable. Concluding, the 'safeness' of the <tt class="literal"><span class="pre">CHOST</span></tt> variable
160is based on externally assumed 'safeness', which's discussion falls
161outside this GLEP.</p>
162<p>Current USE-expansion of the variables is being maintained, as this
163results in full backward compatibility. Since the variables themself
164don't change in what they represent, but only how they are being
165assigned, there should be no problem in maintaining them. Using
166USE-expansion, conditional code can be written down in ebuilds, which is
167not different from any existing methods at all:</p>
168<pre class="literal-block">
169...
170RDEPEND=&quot;elibc_FreeBSD? ( sys-libs/com_err )&quot;
171...
172src_compile() {
173 ...
174 use elibc_FreeBSD &amp;&amp; myconf=&quot;${myconf} -Dlibc=/usr/lib/libc.a&quot;
175 ...
176}
177</pre>
178<p>Alternatively, 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>
179are available in the ebuild evironment and they can be used instead of
180invoking <tt class="literal"><span class="pre">xxx_Xxxx</span></tt> or in switch statements where they are actually
181necessary.</p>
153<p>A map file can be used to have the various <tt class="literal"><span class="pre">CHOST</span></tt> values being 182<p>A map file can be used to have the various <tt class="literal"><span class="pre">CHOST</span></tt> values being
154translated to the correct values for the four variables. This change is 183translated to the correct values for the four variables. This change is
155invisible for ebuilds and eclasses, but allows to rely on these 184invisible 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. 185variables as they are based on a 'safe' value -- the <tt class="literal"><span class="pre">CHOST</span></tt> variable.
157Ebuilds should not be sensitive to the keyword value, but use the 186Ebuilds should not be sensitive to the keyword value, but use the
158aforementioned four variables instead. They allow specific tests for 187aforementioned 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 188properties. If this is undesirable, the full <tt class="literal"><span class="pre">CHOST</span></tt> variable can be
160used to match a complete operating system.</p> 189used to match a complete operating system.</p>
161<p>Current USE-expansion is being maintained, for backwards compatibility.
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"> 190<div class="section" id="variable-assignment">
166<h2><a class="toc-backref" href="#id15" name="variables">Variables</a></h2> 191<h2><a class="toc-backref" href="#id11" 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 192<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
168file. The file can be overlaid, such that the following entries in the 193file. 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 194map file (on the left of the arrow) will result in the assigned
170variables on the right hand side of the arrow:</p> 195variables on the right hand side of the arrow:</p>
171<pre class="literal-block"> 196<pre class="literal-block">
… …
175*-*-freebsd* -&gt; KERNEL=&quot;FreeBSD&quot; ELIBC=&quot;FreeBSD&quot; 200*-*-freebsd* -&gt; KERNEL=&quot;FreeBSD&quot; ELIBC=&quot;FreeBSD&quot;
176*-*-darwin* -&gt; KERNEL=&quot;Darwin&quot; ELIBC=&quot;Darwin&quot; 201*-*-darwin* -&gt; KERNEL=&quot;Darwin&quot; ELIBC=&quot;Darwin&quot;
177*-*-netbsd* -&gt; KERNEL=&quot;NetBSD&quot; ELIBC=&quot;NetBSD&quot; 202*-*-netbsd* -&gt; KERNEL=&quot;NetBSD&quot; ELIBC=&quot;NetBSD&quot;
178*-*-solaris* -&gt; KERNEL=&quot;Solaris&quot; ELIBC=&quot;Solaris&quot; 203*-*-solaris* -&gt; KERNEL=&quot;Solaris&quot; ELIBC=&quot;Solaris&quot;
179</pre> 204</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 205<p>A way to achieve this is proposed by Mike Frysinger, which
181suggests to have a env-map file, for instance filled with:</p> 206suggests to have a env-map file, for instance filled with:</p>
182<pre class="literal-block"> 207<pre class="literal-block">
183% cat env-map 208% cat env-map
184*-linux-* KERNEL=linux 209*-linux-* KERNEL=linux
185*-gnu ELIBC=glibc 210*-gnu ELIBC=glibc
… …
208<p>Given the example env-map file, this script would result in:</p> 233<p>Given the example env-map file, this script would result in:</p>
209<pre class="literal-block"> 234<pre class="literal-block">
210% ./readmap x86_64-pc-linux-gnu 235% ./readmap x86_64-pc-linux-gnu
211ARCH=amd64 KERNEL=linux ELIBC=glibc 236ARCH=amd64 KERNEL=linux ELIBC=glibc
212</pre> 237</pre>
238<p>The entries in the <tt class="literal"><span class="pre">env-map</span></tt> file will be evaluated in a forward
239linear full scan. A side-effect of this exhaustive search is that the
240variables can be re-assigned if multiple entries match the given
241<tt class="literal"><span class="pre">CHOST</span></tt>. Because of this, the order of the entries does matter.
242Because the <tt class="literal"><span class="pre">env-map</span></tt> file size is assumed not to exceed the block
243size of the file system, the performance penalty of a full scan versus
244'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 245<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 246concept implementation. Since Portage is largerly written in Python, it
215information, which is written in Python. Hence, an equivalent of this 247will 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 248also. 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> 249of the <tt class="literal"><span class="pre">env-map</span></tt> file, and especially its wildcard characters, might
250not be the best match with Python. For this purpose, the format
251specification of the <tt class="literal"><span class="pre">env-map</span></tt> file is deferred to the Python
252implementation, and only the requirements are given here.</p>
253<p>The <tt class="literal"><span class="pre">env-map</span></tt> file should be capable of encoding a <tt class="literal"><span class="pre">key</span></tt>, <tt class="literal"><span class="pre">value</span></tt>
254pair, where <tt class="literal"><span class="pre">key</span></tt> is a (regular) expression that matches a
255chost-string, and <tt class="literal"><span class="pre">value</span></tt> contains at least one, distinct variable
256assignment for the variables <tt class="literal"><span class="pre">ARCH</span></tt>, <tt class="literal"><span class="pre">KERNEL</span></tt> and <tt class="literal"><span class="pre">ELIBC</span></tt>. The
257interpreter of the <tt class="literal"><span class="pre">env-map</span></tt> file must scan the file linearly and
258continue trying to match the <tt class="literal"><span class="pre">key</span></tt>s and assign variables if
259appropriate until the end of file.</p>
260<p>Since Portage will use the <tt class="literal"><span class="pre">env-map</span></tt> file, the location of the file is
261beyond the scope of this GLEP and up to the Portage implementors.</p>
218</div> 262</div>
219</div> 263</div>
220<div class="section" id="references"> 264<div class="section" id="references">
221<h1><a class="toc-backref" href="#id16" name="references">References</a></h1> 265<h1><a class="toc-backref" href="#id12" name="references">References</a></h1>
222<table class="footnote" frame="void" id="id5" rules="none"> 266<table class="footnote" frame="void" id="id3" rules="none">
223<colgroup><col class="label" /><col /></colgroup> 267<colgroup><col class="label" /><col /></colgroup>
224<tbody valign="top"> 268<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 269<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, 270userlands/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> 271(<a class="reference" href="http://glep.gentoo.org/glep-0022.html">http://glep.gentoo.org/glep-0022.html</a>)</td></tr>
228</tbody> 272</tbody>
229</table> 273</table>
230<table class="footnote" frame="void" id="id6" rules="none"> 274<table class="footnote" frame="void" id="id4" rules="none">
231<colgroup><col class="label" /><col /></colgroup> 275<colgroup><col class="label" /><col /></colgroup>
232<tbody valign="top"> 276<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 277<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 2784-tuples, even though tuple in itself suggest a field consisting of
235on a Linux system and currently is handled as follows:</p> 279two values. For clarity: a 1-tuple describes a single value field,
236<pre class="literal-block"> 280while 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> 281</tbody>
241</table> 282</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> 283</div>
258<div class="section" id="copyright"> 284<div class="section" id="copyright">
259<h1><a class="toc-backref" href="#id17" name="copyright">Copyright</a></h1> 285<h1><a class="toc-backref" href="#id13" name="copyright">Copyright</a></h1>
260<p>This document has been placed in the public domain.</p> 286<p>This document has been placed in the public domain.</p>
261</div> 287</div>
262</div> 288</div>
263 289
264<hr class="footer" /> 290<hr class="footer" />
265<div class="footer"> 291<div class="footer">
266<a class="reference" href="glep-0047.txt">View document source</a>. 292<a class="reference" href="glep-0047.txt">View document source</a>.
267Generated on: 2006-02-09 21:32 UTC. 293Generated on: 2006-02-11 21:38 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. 294Generated 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.
269</div> 295</div>
270</body> 296</body>
271</html> 297</html>
272 298
Legend:
Removed from v.1.1  
changed lines
  Added in v.1.2
  ViewVC Help
Powered by ViewVC 1.1.13