diff options
Diffstat (limited to 'doc/src/sgml/html/regress-variant.html')
| -rw-r--r-- | doc/src/sgml/html/regress-variant.html | 77 |
1 files changed, 77 insertions, 0 deletions
diff --git a/doc/src/sgml/html/regress-variant.html b/doc/src/sgml/html/regress-variant.html new file mode 100644 index 0000000..6326d82 --- /dev/null +++ b/doc/src/sgml/html/regress-variant.html @@ -0,0 +1,77 @@ +<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>33.3. Variant Comparison Files</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="regress-evaluation.html" title="33.2. Test Evaluation" /><link rel="next" href="regress-tap.html" title="33.4. TAP Tests" /></head><body id="docContent" class="container-fluid col-10"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">33.3. Variant Comparison Files</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="regress-evaluation.html" title="33.2. Test Evaluation">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="regress.html" title="Chapter 33. Regression Tests">Up</a></td><th width="60%" align="center">Chapter 33. Regression Tests</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 15.5 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="regress-tap.html" title="33.4. TAP Tests">Next</a></td></tr></table><hr /></div><div class="sect1" id="REGRESS-VARIANT"><div class="titlepage"><div><div><h2 class="title" style="clear: both">33.3. Variant Comparison Files</h2></div></div></div><p> + Since some of the tests inherently produce environment-dependent + results, we have provided ways to specify alternate <span class="quote">“<span class="quote">expected</span>”</span> + result files. Each regression test can have several comparison files + showing possible results on different platforms. There are two + independent mechanisms for determining which comparison file is used + for each test. + </p><p> + The first mechanism allows comparison files to be selected for + specific platforms. There is a mapping file, + <code class="filename">src/test/regress/resultmap</code>, that defines + which comparison file to use for each platform. + To eliminate bogus test <span class="quote">“<span class="quote">failures</span>”</span> for a particular platform, + you first choose or make a variant result file, and then add a line to the + <code class="filename">resultmap</code> file. + </p><p> + Each line in the mapping file is of the form +</p><pre class="synopsis"> +testname:output:platformpattern=comparisonfilename +</pre><p> + The test name is just the name of the particular regression test + module. The output value indicates which output file to check. For the + standard regression tests, this is always <code class="literal">out</code>. The + value corresponds to the file extension of the output file. + The platform pattern is a pattern in the style of the Unix + tool <code class="command">expr</code> (that is, a regular expression with an implicit + <code class="literal">^</code> anchor at the start). It is matched against the + platform name as printed by <code class="command">config.guess</code>. + The comparison file name is the base name of the substitute result + comparison file. + </p><p> + For example: some systems lack a working <code class="literal">strtof</code> function, + for which our workaround causes rounding errors in the + <code class="filename">float4</code> regression test. + Therefore, we provide a variant comparison file, + <code class="filename">float4-misrounded-input.out</code>, which includes + the results to be expected on these systems. To silence the bogus + <span class="quote">“<span class="quote">failure</span>”</span> message on <span class="systemitem">HP-UX 10</span> + platforms, <code class="filename">resultmap</code> includes: +</p><pre class="programlisting"> +float4:out:hppa.*-hp-hpux10.*=float4-misrounded-input.out +</pre><p> + which will trigger on any machine where the output of + <code class="command">config.guess</code> matches <code class="literal">hppa.*-hp-hpux10.*</code>. + Other lines in <code class="filename">resultmap</code> select the variant comparison + file for other platforms where it's appropriate. + </p><p> + The second selection mechanism for variant comparison files is + much more automatic: it simply uses the <span class="quote">“<span class="quote">best match</span>”</span> among + several supplied comparison files. The regression test driver + script considers both the standard comparison file for a test, + <code class="literal"><em class="replaceable"><code>testname</code></em>.out</code>, and variant files named + <code class="literal"><em class="replaceable"><code>testname</code></em>_<em class="replaceable"><code>digit</code></em>.out</code> + (where the <em class="replaceable"><code>digit</code></em> is any single digit + <code class="literal">0</code>-<code class="literal">9</code>). If any such file is an exact match, + the test is considered to pass; otherwise, the one that generates + the shortest diff is used to create the failure report. (If + <code class="filename">resultmap</code> includes an entry for the particular + test, then the base <em class="replaceable"><code>testname</code></em> is the substitute + name given in <code class="filename">resultmap</code>.) + </p><p> + For example, for the <code class="literal">char</code> test, the comparison file + <code class="filename">char.out</code> contains results that are expected + in the <code class="literal">C</code> and <code class="literal">POSIX</code> locales, while + the file <code class="filename">char_1.out</code> contains results sorted as + they appear in many other locales. + </p><p> + The best-match mechanism was devised to cope with locale-dependent + results, but it can be used in any situation where the test results + cannot be predicted easily from the platform name alone. A limitation of + this mechanism is that the test driver cannot tell which variant is + actually <span class="quote">“<span class="quote">correct</span>”</span> for the current environment; it will just pick + the variant that seems to work best. Therefore it is safest to use this + mechanism only for variant results that you are willing to consider + equally valid in all contexts. + </p></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="regress-evaluation.html" title="33.2. Test Evaluation">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="regress.html" title="Chapter 33. Regression Tests">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="regress-tap.html" title="33.4. TAP Tests">Next</a></td></tr><tr><td width="40%" align="left" valign="top">33.2. Test Evaluation </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 15.5 Documentation">Home</a></td><td width="40%" align="right" valign="top"> 33.4. TAP Tests</td></tr></table></div></body></html>
\ No newline at end of file |