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"> |
3 |
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"> |
4 |
<!-- |
5 |
This HTML is auto-generated. DO NOT EDIT THIS FILE! If you are writing a new |
6 |
PEP, see http://www.python.org/peps/pep-0001.html for instructions and links |
7 |
to templates. DO NOT USE THIS HTML FILE AS YOUR TEMPLATE! |
8 |
--> |
9 |
<head> |
10 |
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> |
11 |
<meta name="generator" content="Docutils 0.4: http://docutils.sourceforge.net/" /> |
12 |
<title>GLEP 44 -- Manifest2 format</title> |
13 |
<style type="text/css"> |
14 |
|
15 |
/* |
16 |
:Author: David Goodger |
17 |
:Contact: goodger@users.sourceforge.net |
18 |
:date: $Date: 2003/06/02 17:03:08 $ |
19 |
:version: $Revision: 1.1 $ |
20 |
:copyright: This stylesheet has been placed in the public domain. |
21 |
|
22 |
Default cascading style sheet for the PEP HTML output of Docutils. |
23 |
*/ |
24 |
|
25 |
.first { |
26 |
margin-top: 0 } |
27 |
|
28 |
.last { |
29 |
margin-bottom: 0 } |
30 |
|
31 |
.navigation { |
32 |
width: 100% ; |
33 |
background: #cc99ff ; |
34 |
margin-top: 0px ; |
35 |
margin-bottom: 0px } |
36 |
|
37 |
.navigation .navicon { |
38 |
width: 150px ; |
39 |
height: 35px } |
40 |
|
41 |
.navigation .textlinks { |
42 |
padding-left: 1em ; |
43 |
text-align: left } |
44 |
|
45 |
.navigation td, .navigation th { |
46 |
padding-left: 0em ; |
47 |
padding-right: 0em ; |
48 |
vertical-align: middle } |
49 |
|
50 |
.rfc2822 { |
51 |
margin-top: 0.5em ; |
52 |
margin-left: 0.5em ; |
53 |
margin-right: 0.5em ; |
54 |
margin-bottom: 0em } |
55 |
|
56 |
.rfc2822 td { |
57 |
text-align: left } |
58 |
|
59 |
.rfc2822 th.field-name { |
60 |
text-align: right ; |
61 |
font-family: sans-serif ; |
62 |
padding-right: 0.5em ; |
63 |
font-weight: bold ; |
64 |
margin-bottom: 0em } |
65 |
|
66 |
a.toc-backref { |
67 |
text-decoration: none ; |
68 |
color: black } |
69 |
|
70 |
body { |
71 |
margin: 0px ; |
72 |
margin-bottom: 1em ; |
73 |
padding: 0px } |
74 |
|
75 |
dd { |
76 |
margin-bottom: 0.5em } |
77 |
|
78 |
div.section { |
79 |
margin-left: 1em ; |
80 |
margin-right: 1em ; |
81 |
margin-bottom: 1.5em } |
82 |
|
83 |
div.section div.section { |
84 |
margin-left: 0em ; |
85 |
margin-right: 0em ; |
86 |
margin-top: 1.5em } |
87 |
|
88 |
div.abstract { |
89 |
margin: 2em 5em } |
90 |
|
91 |
div.abstract p.topic-title { |
92 |
font-weight: bold ; |
93 |
text-align: center } |
94 |
|
95 |
div.attention, div.caution, div.danger, div.error, div.hint, |
96 |
div.important, div.note, div.tip, div.warning { |
97 |
margin: 2em ; |
98 |
border: medium outset ; |
99 |
padding: 1em } |
100 |
|
101 |
div.attention p.admonition-title, div.caution p.admonition-title, |
102 |
div.danger p.admonition-title, div.error p.admonition-title, |
103 |
div.warning p.admonition-title { |
104 |
color: red ; |
105 |
font-weight: bold ; |
106 |
font-family: sans-serif } |
107 |
|
108 |
div.hint p.admonition-title, div.important p.admonition-title, |
109 |
div.note p.admonition-title, div.tip p.admonition-title { |
110 |
font-weight: bold ; |
111 |
font-family: sans-serif } |
112 |
|
113 |
div.figure { |
114 |
margin-left: 2em } |
115 |
|
116 |
div.footer, div.header { |
117 |
font-size: smaller } |
118 |
|
119 |
div.footer { |
120 |
margin-left: 1em ; |
121 |
margin-right: 1em } |
122 |
|
123 |
div.system-messages { |
124 |
margin: 5em } |
125 |
|
126 |
div.system-messages h1 { |
127 |
color: red } |
128 |
|
129 |
div.system-message { |
130 |
border: medium outset ; |
131 |
padding: 1em } |
132 |
|
133 |
div.system-message p.system-message-title { |
134 |
color: red ; |
135 |
font-weight: bold } |
136 |
|
137 |
div.topic { |
138 |
margin: 2em } |
139 |
|
140 |
h1 { |
141 |
font-family: sans-serif ; |
142 |
font-size: large } |
143 |
|
144 |
h2 { |
145 |
font-family: sans-serif ; |
146 |
font-size: medium } |
147 |
|
148 |
h3 { |
149 |
font-family: sans-serif ; |
150 |
font-size: small } |
151 |
|
152 |
h4 { |
153 |
font-family: sans-serif ; |
154 |
font-style: italic ; |
155 |
font-size: small } |
156 |
|
157 |
h5 { |
158 |
font-family: sans-serif; |
159 |
font-size: x-small } |
160 |
|
161 |
h6 { |
162 |
font-family: sans-serif; |
163 |
font-style: italic ; |
164 |
font-size: x-small } |
165 |
|
166 |
.section hr { |
167 |
width: 75% } |
168 |
|
169 |
ol.simple, ul.simple { |
170 |
margin-bottom: 1em } |
171 |
|
172 |
ol.arabic { |
173 |
list-style: decimal } |
174 |
|
175 |
ol.loweralpha { |
176 |
list-style: lower-alpha } |
177 |
|
178 |
ol.upperalpha { |
179 |
list-style: upper-alpha } |
180 |
|
181 |
ol.lowerroman { |
182 |
list-style: lower-roman } |
183 |
|
184 |
ol.upperroman { |
185 |
list-style: upper-roman } |
186 |
|
187 |
p.caption { |
188 |
font-style: italic } |
189 |
|
190 |
p.credits { |
191 |
font-style: italic ; |
192 |
font-size: smaller } |
193 |
|
194 |
p.label { |
195 |
white-space: nowrap } |
196 |
|
197 |
p.topic-title { |
198 |
font-family: sans-serif ; |
199 |
font-weight: bold } |
200 |
|
201 |
pre.line-block { |
202 |
font-family: serif ; |
203 |
font-size: 100% } |
204 |
|
205 |
pre.literal-block, pre.doctest-block { |
206 |
margin-left: 2em ; |
207 |
margin-right: 2em ; |
208 |
background-color: #eeeeee } |
209 |
|
210 |
span.classifier { |
211 |
font-family: sans-serif ; |
212 |
font-style: oblique } |
213 |
|
214 |
span.classifier-delimiter { |
215 |
font-family: sans-serif ; |
216 |
font-weight: bold } |
217 |
|
218 |
span.interpreted { |
219 |
font-family: sans-serif } |
220 |
|
221 |
span.option-argument { |
222 |
font-style: italic } |
223 |
|
224 |
span.pre { |
225 |
white-space: pre } |
226 |
|
227 |
span.problematic { |
228 |
color: red } |
229 |
|
230 |
table { |
231 |
margin-top: 0.5em ; |
232 |
margin-bottom: 0.5em } |
233 |
|
234 |
td, th { |
235 |
padding-left: 0.5em ; |
236 |
padding-right: 0.5em ; |
237 |
vertical-align: top } |
238 |
|
239 |
td.num { |
240 |
text-align: right } |
241 |
|
242 |
th.field-name { |
243 |
font-weight: bold ; |
244 |
text-align: left ; |
245 |
white-space: nowrap } |
246 |
|
247 |
h1 tt, h2 tt, h3 tt, h4 tt, h5 tt, h6 tt { |
248 |
font-size: 100% } |
249 |
|
250 |
tt { |
251 |
background-color: #eeeeee } |
252 |
|
253 |
ul.auto-toc { |
254 |
list-style-type: none } |
255 |
|
256 |
</style> |
257 |
</head> |
258 |
<body bgcolor="white"> |
259 |
<table class="navigation" cellpadding="0" cellspacing="0" |
260 |
width="100%" border="0"> |
261 |
<tr><td class="navicon" width="150" height="35"> |
262 |
<a href="http://www.gentoo.org/" title="Gentoo Linux Home Page"> |
263 |
<img src="http://www.gentoo.org/images/gentoo-new.gif" alt="[Gentoo]" |
264 |
border="0" width="150" height="35" /></a></td> |
265 |
<td class="textlinks" align="left"> |
266 |
[<b><a href="http://www.gentoo.org/">Gentoo Linux Home</a></b>] |
267 |
[<b><a href="http://www.gentoo.org/peps">GLEP Index</a></b>] |
268 |
[<b><a href="http://www.gentoo.org/proj/en/glep/glep-0044.txt">GLEP Source</a></b>] |
269 |
</td></tr></table> |
270 |
<table class="rfc2822 docutils field-list" frame="void" rules="none"> |
271 |
<col class="field-name" /> |
272 |
<col class="field-body" /> |
273 |
<tbody valign="top"> |
274 |
<tr class="field"><th class="field-name">GLEP:</th><td class="field-body">44</td> |
275 |
</tr> |
276 |
<tr class="field"><th class="field-name">Title:</th><td class="field-body">Manifest2 format</td> |
277 |
</tr> |
278 |
<tr class="field"><th class="field-name">Version:</th><td class="field-body">1.6</td> |
279 |
</tr> |
280 |
<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-0044.txt?cvsroot=gentoo">2006/09/04 03:06:32</a></td> |
281 |
</tr> |
282 |
<tr class="field"><th class="field-name">Author:</th><td class="field-body">Marius Mauch <genone at gentoo.org>,</td> |
283 |
</tr> |
284 |
<tr class="field"><th class="field-name">Status:</th><td class="field-body">Final</td> |
285 |
</tr> |
286 |
<tr class="field"><th class="field-name">Type:</th><td class="field-body">Standards Track</td> |
287 |
</tr> |
288 |
<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> |
289 |
</tr> |
290 |
<tr class="field"><th class="field-name">Created:</th><td class="field-body">04-Dec-2005</td> |
291 |
</tr> |
292 |
<tr class="field"><th class="field-name">Post-History:</th><td class="field-body">06-Dec-2005, 23-Jan-2006, 3-Sep-2006</td> |
293 |
</tr> |
294 |
</tbody> |
295 |
</table> |
296 |
<hr /> |
297 |
<div class="contents topic"> |
298 |
<p class="topic-title first"><a id="contents" name="contents">Contents</a></p> |
299 |
<ul class="simple"> |
300 |
<li><a class="reference" href="#abstract" id="id9" name="id9">Abstract</a></li> |
301 |
<li><a class="reference" href="#motivation" id="id10" name="id10">Motivation</a></li> |
302 |
<li><a class="reference" href="#specification" id="id11" name="id11">Specification</a><ul> |
303 |
<li><a class="reference" href="#compability-entries" id="id12" name="id12">Compability Entries</a></li> |
304 |
<li><a class="reference" href="#scope" id="id13" name="id13">Scope</a></li> |
305 |
<li><a class="reference" href="#number-of-hashes" id="id14" name="id14">Number of hashes</a></li> |
306 |
</ul> |
307 |
</li> |
308 |
<li><a class="reference" href="#rationale" id="id15" name="id15">Rationale</a><ul> |
309 |
<li><a class="reference" href="#removal-of-digest-files" id="id16" name="id16">Removal of digest files</a></li> |
310 |
<li><a class="reference" href="#reducing-redundancy" id="id17" name="id17">Reducing redundancy</a></li> |
311 |
<li><a class="reference" href="#removal-of-checksum-collisions" id="id18" name="id18">Removal of checksum collisions</a></li> |
312 |
<li><a class="reference" href="#flexible-verification-system" id="id19" name="id19">Flexible verification system</a></li> |
313 |
</ul> |
314 |
</li> |
315 |
<li><a class="reference" href="#backwards-compatibility" id="id20" name="id20">Backwards Compatibility</a></li> |
316 |
<li><a class="reference" href="#other-problems" id="id21" name="id21">Other problems</a><ul> |
317 |
<li><a class="reference" href="#impacts-on-infrastructure" id="id22" name="id22">Impacts on infrastructure</a></li> |
318 |
</ul> |
319 |
</li> |
320 |
<li><a class="reference" href="#reference-implementation" id="id23" name="id23">Reference Implementation</a></li> |
321 |
<li><a class="reference" href="#options" id="id24" name="id24">Options</a></li> |
322 |
<li><a class="reference" href="#credits" id="id25" name="id25">Credits</a></li> |
323 |
<li><a class="reference" href="#references" id="id26" name="id26">References</a></li> |
324 |
<li><a class="reference" href="#copyright" id="id27" name="id27">Copyright</a></li> |
325 |
</ul> |
326 |
</div> |
327 |
<div class="section"> |
328 |
<h1><a class="toc-backref" href="#id9" id="abstract" name="abstract">Abstract</a></h1> |
329 |
<p>This GLEP proposes a new format for the Portage Manifest and digest file system |
330 |
by unifying both filetypes into one to improve functional and non-functional |
331 |
aspects of the Portage Tree.</p> |
332 |
</div> |
333 |
<div class="section"> |
334 |
<h1><a class="toc-backref" href="#id10" id="motivation" name="motivation">Motivation</a></h1> |
335 |
<p>Please see <a class="footnote-reference" href="#reorg-thread" id="id1" name="id1">[1]</a> for a general overview. |
336 |
The main long term goals of this proposal are to:</p> |
337 |
<ul class="simple"> |
338 |
<li>Remove the tiny digest files from the tree. They are a major annoyance as on a |
339 |
typical configuration they waste a lot of disk space and the simple transmission |
340 |
of the names for all digest files during a <tt class="docutils literal"><span class="pre">emerge</span> <span class="pre">--sync</span></tt> needs a substantial |
341 |
amount of bandwidth.</li> |
342 |
<li>Reduce redundancy when multiple hash functions are used</li> |
343 |
<li>Remove potential for checksum collisions if a file is recorded in more than one |
344 |
digest file</li> |
345 |
<li>Difference between filetypes for a more flexible verification system</li> |
346 |
</ul> |
347 |
</div> |
348 |
<div class="section"> |
349 |
<h1><a class="toc-backref" href="#id11" id="specification" name="specification">Specification</a></h1> |
350 |
<p>The new Manifest format would change the existing format in the following ways:</p> |
351 |
<ul> |
352 |
<li><p class="first">Addition of a filetype specifier, currently planned are</p> |
353 |
<ul class="simple"> |
354 |
<li><tt class="docutils literal"><span class="pre">AUX</span></tt> for files directly used by ebuilds (e.g. patches or initscripts), |
355 |
located in the <tt class="docutils literal"><span class="pre">files/</span></tt> subdirectory</li> |
356 |
<li><tt class="docutils literal"><span class="pre">EBUILD</span></tt> for all ebuilds</li> |
357 |
<li><tt class="docutils literal"><span class="pre">MISC</span></tt> for files not directly used by ebuilds like <tt class="docutils literal"><span class="pre">ChangeLog</span></tt> or |
358 |
<tt class="docutils literal"><span class="pre">metadata.xml</span></tt> files</li> |
359 |
<li><tt class="docutils literal"><span class="pre">DIST</span></tt> for release tarballs recorded in the <tt class="docutils literal"><span class="pre">SRC_URI</span></tt> variable of an ebuild, |
360 |
these were previously recorded in the digest files</li> |
361 |
</ul> |
362 |
<p>Future portage improvements might extend this list (for example with types |
363 |
relevant for eclasses or profiles)</p> |
364 |
</li> |
365 |
<li><p class="first">Only have one line per file listing all information instead of one line per |
366 |
file and checksum type</p> |
367 |
</li> |
368 |
<li><p class="first">Remove the separated digest-* files in the <tt class="docutils literal"><span class="pre">files/</span></tt> subdirectory</p> |
369 |
</li> |
370 |
</ul> |
371 |
<p>Each line in the new format has the following format:</p> |
372 |
<pre class="literal-block"> |
373 |
<filetype> <filename> <filesize> <chksumtype1> <chksum1> ... <chksumtypen> <chksumn> |
374 |
</pre> |
375 |
<p>However theses entries will be stored in the existing Manifest files.</p> |
376 |
<p>An <a class="reference" href="glep-0044-extras/manifest2-example.txt">actual example</a> <a class="footnote-reference" href="#id7" id="id8" name="id8">[6]</a> for a (pure) Manifest2 file..</p> |
377 |
<div class="section"> |
378 |
<h2><a class="toc-backref" href="#id12" id="compability-entries" name="compability-entries">Compability Entries</a></h2> |
379 |
<p>To maintain compability with existing portage versions a transition period after |
380 |
is the introduction of the Manifest2 format is required during which portage |
381 |
will not only have to be capable of using existing Manifest and digest files but |
382 |
also generate them in addition to the new entries. |
383 |
Fortunately this can be accomplished by simply mixing old and new style entries |
384 |
in one file for the Manifest files, existing portage versions will simply ignore |
385 |
the new style entries. For the digest files there are no new entries to care |
386 |
about.</p> |
387 |
</div> |
388 |
<div class="section"> |
389 |
<h2><a class="toc-backref" href="#id13" id="scope" name="scope">Scope</a></h2> |
390 |
<p>It is important to note that this proposal only deals with a change of the |
391 |
format of the digest and Manifest system.</p> |
392 |
<p>It does not expand the scope of it to cover eclasses, profiles or anything |
393 |
else not already covered by the Manifest system, it also doesn't affect |
394 |
the Manifest signing efforts in any way (though the implementations of both |
395 |
might be coupled).</p> |
396 |
<p>Also while multiple hash functions will become standard with the proposed |
397 |
implementation they are not a specific feature of this format <a class="footnote-reference" href="#multi-hash-thread" id="id3" name="id3">[2]</a>.</p> |
398 |
</div> |
399 |
<div class="section"> |
400 |
<h2><a class="toc-backref" href="#id14" id="number-of-hashes" name="number-of-hashes">Number of hashes</a></h2> |
401 |
<p>While using multiple hashes for each file is a major feature of this proposal |
402 |
we have to make sure that the number of hashes listed is limited to avoid |
403 |
an explosion of the Manifest size that would revert the main benefit of this proposal |
404 |
(reduzing tree size). Therefore the number of hashes that will be generated |
405 |
will be limited to three different hash functions. For compability though we |
406 |
have to rely on at least one hash function to always be present, this proposal |
407 |
suggest to use SHA1 for this purpose (as it is supposed to be more secure than MD5 |
408 |
and currently only SHA1 and MD5 are directly available in python, also MD5 doesn't |
409 |
have any benefit in terms of compability).</p> |
410 |
</div> |
411 |
</div> |
412 |
<div class="section"> |
413 |
<h1><a class="toc-backref" href="#id15" id="rationale" name="rationale">Rationale</a></h1> |
414 |
<p>The main goals of the proposal have been listed in the <a class="reference" href="#motivation">Motivation</a>, here now |
415 |
the explanation why they are improvements and how the proposed format will |
416 |
accomplish them.</p> |
417 |
<div class="section"> |
418 |
<h2><a class="toc-backref" href="#id16" id="removal-of-digest-files" name="removal-of-digest-files">Removal of digest files</a></h2> |
419 |
<p>Normal users that don't use a "tuned" filesystem for the portage tree are |
420 |
wasting several dozen to a few hundred megabytes of disk space with the current |
421 |
system, largely caused by the digest files. |
422 |
This is due to the filesystem overhead present in most filesystem that |
423 |
have a standard blocksize of four kilobytes while most digest files are under |
424 |
one kilobyte in size, so this results in approximately a waste of three kilobytes |
425 |
per digest file (likely even more). At the time of this writing the tree contains |
426 |
roughly 22.000 digest files, so the overall waste caused by digest files is |
427 |
estimated at about 70-100 megabytes. |
428 |
Furthermore it is assumed that this will also reduce the disk space wasted by |
429 |
the Manifest files as they now contain more content, but this hasn't been |
430 |
verified yet.</p> |
431 |
<p>By unifying the digest files with the Manifest these tiny files are eliminated |
432 |
(in the long run), reducing the apparent tree size by about 20%, benefitting |
433 |
both users and the Gentoo infrastructure.</p> |
434 |
</div> |
435 |
<div class="section"> |
436 |
<h2><a class="toc-backref" href="#id17" id="reducing-redundancy" name="reducing-redundancy">Reducing redundancy</a></h2> |
437 |
<p>When multiple hashes are used with the current system |
438 |
both the filename and filesize are repeated for every checksum type used as each |
439 |
checksum is standalone. However this doesn't add any functionality and is |
440 |
therefore useless, so the new format removes this redundancy. |
441 |
This is a theoretical improvement at this moment as only one hash function is in |
442 |
use, but expected to change soon (see <a class="footnote-reference" href="#multi-hash-thread" id="id4" name="id4">[2]</a>).</p> |
443 |
</div> |
444 |
<div class="section"> |
445 |
<h2><a class="toc-backref" href="#id18" id="removal-of-checksum-collisions" name="removal-of-checksum-collisions">Removal of checksum collisions</a></h2> |
446 |
<p>The current system theoretically allows for a <tt class="docutils literal"><span class="pre">DIST</span></tt> type file to be recorded |
447 |
in multiple digest files with different sizes and/or checksums. In such a case |
448 |
one version of a package would report a checksum violation while another one |
449 |
would not. This could create confusion and uncertainity among users. |
450 |
So far this case hasn't been observed, but it can't be ruled out with the |
451 |
existing system. |
452 |
As the new format lists each file exactly once this would be no longer possible.</p> |
453 |
</div> |
454 |
<div class="section"> |
455 |
<h2><a class="toc-backref" href="#id19" id="flexible-verification-system" name="flexible-verification-system">Flexible verification system</a></h2> |
456 |
<p>Right now portage verifies the checksum of every file listed in the Manifest |
457 |
before using any file of the package and all <tt class="docutils literal"><span class="pre">DIST</span></tt> files of an ebuild |
458 |
before using that ebuild. This is unnecessary in many cases:</p> |
459 |
<ul class="simple"> |
460 |
<li>During the "depend" phase (when the ebuild metadata is generated) only |
461 |
files of type <tt class="docutils literal"><span class="pre">EBUILD</span></tt> are used, so verifying the other types isn't |
462 |
necessary. Theoretically it is possible for an ebuild to include other |
463 |
files like those of type <tt class="docutils literal"><span class="pre">AUX</span></tt> at this phase, but that would be a |
464 |
major QA violation and should never occur, so it can be ignored here. |
465 |
It is also not a security concern as the ebuild is verified before parsing |
466 |
it, so each manipulation would show up.</li> |
467 |
<li>Generally files of type <tt class="docutils literal"><span class="pre">MISC</span></tt> don't need to be verified as they are |
468 |
only used in very specific situations, aren't executed (just parsed at most) |
469 |
and don't affect the package build process.</li> |
470 |
<li>Files of type <tt class="docutils literal"><span class="pre">DIST</span></tt> only need to be verified directly after fetching and |
471 |
before unpacking them (which often will be one step), not every time their |
472 |
associated ebuild is used.</li> |
473 |
</ul> |
474 |
</div> |
475 |
</div> |
476 |
<div class="section"> |
477 |
<h1><a class="toc-backref" href="#id20" id="backwards-compatibility" name="backwards-compatibility">Backwards Compatibility</a></h1> |
478 |
<p>Switching the Manifest system is a task that will need a long transition period |
479 |
like most changes affecting both portage and the tree. In this case the |
480 |
implementation will be rolled out in several phases:</p> |
481 |
<ol class="arabic simple"> |
482 |
<li>Add support for verification of Manifest2 entries in portage</li> |
483 |
<li>Enable generation of Manifest2 entries in addition to the current system</li> |
484 |
<li>Ignore digests during <tt class="docutils literal"><span class="pre">emerge</span> <span class="pre">--sync</span></tt> to get the size-benefit clientside. |
485 |
This step may be ommitted if the following steps are expected to follow soon.</li> |
486 |
<li>Disable generation of entries for the current system</li> |
487 |
<li>Remove all traces of the current system from the tree (serverside)</li> |
488 |
</ol> |
489 |
<p>Each step has its own issues. While 1) and 2) can be implemented without any |
490 |
compability problems all later steps have a major impact:</p> |
491 |
<ul class="simple"> |
492 |
<li>Step 3) can only be implemented when the whole tree is Manifest2 ready |
493 |
(ideally speaking, practically the requirement will be more like 95% coverage |
494 |
with the expectation that for the remaining 5% either bugs will be filed after |
495 |
step 3) is completed or they'll be updated at step 5).</li> |
496 |
<li>Steps 4) and 5) will render all portage versions without Manifest2 support |
497 |
basically useless (users would have to regenerate the digest and Manifest |
498 |
for each package before being able to merge it), so this requires a almost |
499 |
100% coverage of the userbase with Manifest2 capabale portage versions |
500 |
(with step 1) completely implemented).</li> |
501 |
</ul> |
502 |
<p>Another problem is that some steps affect different targets:</p> |
503 |
<ul class="simple"> |
504 |
<li>Steps 1) and 3) target portage versions used by users</li> |
505 |
<li>Steps 2) and 4) target portage versions used by devs</li> |
506 |
<li>Step 5) targets the portage tree on the cvs server</li> |
507 |
</ul> |
508 |
<p>While it is relatively easy to get all devs to use a new portage version this is |
509 |
practically impossible with users as some don't update their systems regulary. |
510 |
While six months are probably sufficient to reach a 95% coverage one year is |
511 |
estimated to reach an almost-complete coverage. All times are relative to the |
512 |
stable-marking of a compatible portage version.</p> |
513 |
<p>No timeframe for implementation is presented here as it is highly dependent |
514 |
on the completion of each step.</p> |
515 |
<p>In summary it can be said that while a full conversion will take over a year |
516 |
to be completed due to compability issues mentioned above some benefits of the |
517 |
system can selectively be used as soon as step 2) is completed.</p> |
518 |
</div> |
519 |
<div class="section"> |
520 |
<h1><a class="toc-backref" href="#id21" id="other-problems" name="other-problems">Other problems</a></h1> |
521 |
<div class="section"> |
522 |
<h2><a class="toc-backref" href="#id22" id="impacts-on-infrastructure" name="impacts-on-infrastructure">Impacts on infrastructure</a></h2> |
523 |
<p>While one long term goal of this proposal is to reduce the size of the tree |
524 |
and therefore make life for the Gentoo Infrastructure easier this will only |
525 |
take effect once the implementation is rolled out completely. In the meantime |
526 |
however it will increase the tree size due to keeping checksums in both formats. |
527 |
It's not possible to give a usable estimate on the degree of the increase as |
528 |
it depends on many variables such as the exact implementation timeframe, |
529 |
propagation of Manifest2 capable portage versions among devs or the update |
530 |
rate of the tree. It has been suggested that Manifest files that are not gpg |
531 |
signed could be mass converted in one step, this could certainly help but only |
532 |
to some degree (according to a recent research <a class="footnote-reference" href="#gpg-numbers" id="id5" name="id5">[3]</a> about 40% of |
533 |
all Manifests in the tree are signed, but this number hasn't been verified).</p> |
534 |
</div> |
535 |
</div> |
536 |
<div class="section"> |
537 |
<h1><a class="toc-backref" href="#id23" id="reference-implementation" name="reference-implementation">Reference Implementation</a></h1> |
538 |
<p>A patch for a prototype implementation of Manifest2 verification and partial |
539 |
generation has been posted at <a class="footnote-reference" href="#manifest2-patch" id="id6" name="id6">[4]</a>, it will be reworked before |
540 |
being considered for inclusion in portage. However it shows that adding support |
541 |
for verification is quite simple, but generation is a bit tricky and will |
542 |
therefore be implemented later.</p> |
543 |
</div> |
544 |
<div class="section"> |
545 |
<h1><a class="toc-backref" href="#id24" id="options" name="options">Options</a></h1> |
546 |
<p>Some things have been considered for this GLEP but aren't part of the proposal |
547 |
yet for various reasons:</p> |
548 |
<ul class="simple"> |
549 |
<li>timestamp field: the author has considered adding a timestamp field for |
550 |
each entry to list the time the entry was created. However so far no practical |
551 |
use for such a feature has been found.</li> |
552 |
<li>convert size field into checksum: Another idea was to treat the size field |
553 |
like any other checksum. But so far no real benefit (other than a slightly |
554 |
more modular implementation) for this has been seen while it has several |
555 |
drawbacks: For once, unlike checksums, the size field is definitely required |
556 |
for all <tt class="docutils literal"><span class="pre">DIST</span></tt> files, also it would slightly increase the length of |
557 |
each entry by adding a <tt class="docutils literal"><span class="pre">SIZE</span></tt> keyword.</li> |
558 |
<li>removal of the <tt class="docutils literal"><span class="pre">MISC</span></tt> type: It has been suggested to completely drop |
559 |
entries of type <tt class="docutils literal"><span class="pre">MISC</span></tt>. This would result in a minor space reduction |
560 |
(its rather unlikely to free any blocks) but completely remove the ability |
561 |
to check these files for integrity. While they don't influence portage |
562 |
or packages directly they can contain viable information for users, so |
563 |
the author has the opinion that at least the option for integrity checks |
564 |
should be kept.</li> |
565 |
</ul> |
566 |
</div> |
567 |
<div class="section"> |
568 |
<h1><a class="toc-backref" href="#id25" id="credits" name="credits">Credits</a></h1> |
569 |
<p>Thanks to the following persons for their input on or related to this GLEP |
570 |
(even though they might not have known it): |
571 |
Ned Ludd (solar), Brian Harring (ferringb), Jason Stubbs (jstubbs), |
572 |
Robin H. Johnson (robbat2), Aron Griffis (agriffis)</p> |
573 |
<p>Also thanks to Nicholas Jones (carpaski) to make the current Manifest system |
574 |
resistent enough to be able to handle this change without too many transition |
575 |
problems.</p> |
576 |
</div> |
577 |
<div class="section"> |
578 |
<h1><a class="toc-backref" href="#id26" id="references" name="references">References</a></h1> |
579 |
<table class="docutils footnote" frame="void" id="reorg-thread" rules="none"> |
580 |
<colgroup><col class="label" /><col /></colgroup> |
581 |
<tbody valign="top"> |
582 |
<tr><td class="label"><a class="fn-backref" href="#id1" name="reorg-thread">[1]</a></td><td><a class="reference" href="http://thread.gmane.org/gmane.linux.gentoo.devel/21920">http://thread.gmane.org/gmane.linux.gentoo.devel/21920</a></td></tr> |
583 |
</tbody> |
584 |
</table> |
585 |
<table class="docutils footnote" frame="void" id="multi-hash-thread" rules="none"> |
586 |
<colgroup><col class="label" /><col /></colgroup> |
587 |
<tbody valign="top"> |
588 |
<tr><td class="label"><a name="multi-hash-thread">[2]</a></td><td><em>(<a class="fn-backref" href="#id3">1</a>, <a class="fn-backref" href="#id4">2</a>)</em> <a class="reference" href="http://thread.gmane.org/gmane.linux.gentoo.devel/33434">http://thread.gmane.org/gmane.linux.gentoo.devel/33434</a></td></tr> |
589 |
</tbody> |
590 |
</table> |
591 |
<table class="docutils footnote" frame="void" id="gpg-numbers" rules="none"> |
592 |
<colgroup><col class="label" /><col /></colgroup> |
593 |
<tbody valign="top"> |
594 |
<tr><td class="label"><a class="fn-backref" href="#id5" name="gpg-numbers">[3]</a></td><td>gentoo-core mailing list, topic "Gentoo key signing practices |
595 |
and official Gentoo keyring", Message-ID <<a class="reference" href="mailto:20051117075838.GB15734@curie-int.vc.shawcable.net">20051117075838.GB15734@curie-int.vc.shawcable.net</a>></td></tr> |
596 |
</tbody> |
597 |
</table> |
598 |
<table class="docutils footnote" frame="void" id="manifest2-patch" rules="none"> |
599 |
<colgroup><col class="label" /><col /></colgroup> |
600 |
<tbody valign="top"> |
601 |
<tr><td class="label"><a class="fn-backref" href="#id6" name="manifest2-patch">[4]</a></td><td><a class="reference" href="http://thread.gmane.org/gmane.linux.gentoo.portage.devel/1374">http://thread.gmane.org/gmane.linux.gentoo.portage.devel/1374</a></td></tr> |
602 |
</tbody> |
603 |
</table> |
604 |
<table class="docutils footnote" frame="void" id="manifest2-example" rules="none"> |
605 |
<colgroup><col class="label" /><col /></colgroup> |
606 |
<tbody valign="top"> |
607 |
<tr><td class="label"><a name="manifest2-example">[5]</a></td><td><a class="reference" href="http://www.gentoo.org/proj/en/glep/glep-0044-extras/manifest2-example">http://www.gentoo.org/proj/en/glep/glep-0044-extras/manifest2-example</a></td></tr> |
608 |
</tbody> |
609 |
</table> |
610 |
<table class="docutils footnote" frame="void" id="id7" rules="none"> |
611 |
<colgroup><col class="label" /><col /></colgroup> |
612 |
<tbody valign="top"> |
613 |
<tr><td class="label"><a class="fn-backref" href="#id8" name="id7">[6]</a></td><td><a class="reference" href="glep-0044-extras/manifest2-example.txt">glep-0044-extras/manifest2-example.txt</a></td></tr> |
614 |
</tbody> |
615 |
</table> |
616 |
</div> |
617 |
<div class="section"> |
618 |
<h1><a class="toc-backref" href="#id27" id="copyright" name="copyright">Copyright</a></h1> |
619 |
<p>This document has been placed in the public domain.</p> |
620 |
</div> |
621 |
|
622 |
</div> |
623 |
<div class="footer"> |
624 |
<hr class="footer" /> |
625 |
<a class="reference" href="glep-0044.txt">View document source</a>. |
626 |
Generated on: 2006-10-10 20:23 UTC. |
627 |
Generated 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. |
628 |
|
629 |
</div> |
630 |
</body> |
631 |
</html> |
632 |
|