limitations.html revision b9cc48a43ed984587c939d02fba5316bf5c0df6e 
1<!doctype html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
2<html>
3<head>
4<meta http-equiv="content-type" content="text/html; charset=iso-8859-1">
5<meta http-equiv="content-style-type" content="text/css">
6<link rel="stylesheet" type="text/css" href="style.css">
7<title>ProGuard Limitations</title>
8<script type="text/javascript" language="JavaScript">
9<!--
10if (window.self==window.top)
11  window.top.location.replace("/index.html#"+window.location.pathname+window.location.hash);
12else {
13  var hash="#"+window.location.pathname.replace(window.top.location.pathname.replace("index.html", ""), "");
14  if (window.top.location.hash!=hash)
15    window.top.location.hash=hash;
16}
17//-->
18</script>
19</head>
20<body>
21
22<h2>Limitations</h2>
23
24When using ProGuard, you should be aware of a few technical issues, all of
25which are easily avoided or resolved:
26<p>
27<ul class="spacious">
28
29<li>For best results, ProGuard's optimization algorithms assume that the
30    processed code never <b>intentionally throws NullPointerExceptions</b> or
31    ArrayIndexOutOfBoundsExceptions, or even OutOfMemoryErrors or
32    StackOverflowErrors, in order to achieve something useful. For instance,
33    it may remove a method call <code>myObject.myMethod()</code> if that call
34    wouldn't have any effect. It ignores the possibility that
35    <code>myObject</code> might be null, causing a NullPointerException. In
36    some way this is a good thing: optimized code may throw fewer exceptions.
37    Should this entire assumption be false, you'll have to switch off
38    optimization using the <code>-dontoptimize</code> option.</li>
39
40<li>ProGuard's optimization algorithms currently also assume that the
41    processed code never creates <b>busy-waiting loops</b> without at least
42    testing on a volatile field. Again, it may remove such loops. Should this
43    assumption be false, you'll have to switch off optimization using
44    the <code>-dontoptimize</code> option.</li>
45
46<li>If an input jar and a library jar contain classes in the <b>same
47    package</b>, the obfuscated output jar may contain class names that
48    overlap with class names in the library jar. This is most likely if the
49    library jar has been obfuscated before, as it will then probably contain
50    classes named 'a', 'b', etc. Packages should therefore never be split
51    across input jars and library jars.</li>
52
53<li>When obfuscating, ProGuard writes out class files named
54    "<code>a.class</code>", "<code>b.class</code>", etc. If a package contains
55    a large number of classes, ProGuard may also write out
56    <b>"<code>aux.class</code>"</b>. Inconveniently, Windows refuses to create
57    files with this reserved name (among a few other names). It's generally
58    better to write the output to a jar, in order to avoid such problems.</li>
59
60</ul>
61
62<hr />
63<noscript><div><a target="_top" href="/index.html" class="button">Show menu</a></div></noscript>
64<address>
65Copyright &copy; 2002-2013
66<a target="other" href="http://www.lafortune.eu/">Eric Lafortune</a>.
67</address>
68</body>
69</html>
70