view release on metacpan or  search on metacpan

README.html  view on Meta::CPAN

</dd>
<dt id="NEW5">NEW</dt>
<dd>

<p><code>Fsdb::Support::Freds</code> provides a thread-like abstraction over forking, with some nice support for callbacks in the parent upon child termination.</p>

</dd>
<dt id="ENHANCEMENT23">ENHANCEMENT</dt>
<dd>

<p>Details about removing threads: <code>dbpipeline</code> is thread free, and new tests to verify each of its parts. The easy cases are <code>dbcolpercentile</code>, <code>dbcolstats</code>, <code>dbfilepivot</code>, <code>dbjoin</code>, and <code>d...

</dd>
<dt id="BUG-FIX56">BUG FIX</dt>
<dd>

<p>When running with user-only output (<code>-n</code>), <a>dbroweval</a> now resets the output vector <code>$ofref</code> after it has been output.</p>

</dd>
<dt id="NEW6">NEW</dt>
<dd>

<p><a>dbcolcreate</a> will create all columns at the head of each row with the <code>--first</code> option.</p>

</dd>
<dt id="NEW7">NEW</dt>
<dd>

<p><a>dbfilecat</a> will concatenate two files, verifying that they have the same schema.</p>

</dd>
<dt id="ENHANCEMENT24">ENHANCEMENT</dt>
<dd>

<p><a>dbmapreduce</a> now passes comments through, rather than eating them as before.</p>

<p>Also, <a>dbmapreduce</a> now supports a <code>--</code> option to prevent misinterpreting sub-program parameters as for dbmapreduce.</p>

</dd>
<dt id="INCOMPATIBLE-CHANGE7">INCOMPATIBLE CHANGE</dt>
<dd>

<p><a>dbmapreduce</a> no longer figures out if it needs to add the key to the output. For multi-key-aware reducers, it never does (and cannot). For non-multi-key-aware reducers, it defaults to add the key and will now fail if the reducer adds the key...

</dd>
<dt id="INCOMPATIBLE-CHANGE8">INCOMPATIBLE CHANGE</dt>
<dd>

<p><a>dbmapreduce</a> no longer copies the input field separator by default. For multi-key-aware reducers, it never does (and cannot). For non-multi-key-aware reducers, it defaults to <i>not</i> copying the field separator, but it will copy it (the o...

</dd>
</dl>

<h2 id="cleanup-from-de-thread-ification">2.45, 2013-10-07 cleanup from de-thread-ification</h2>

<dl>

<dt id="BUG-FIX57">BUG FIX</dt>
<dd>

<p>Corrected a fast busy-wait in <a>dbmerge</a>.</p>

</dd>
<dt id="ENHANCEMENT25">ENHANCEMENT</dt>
<dd>

<p>Endgame mode enabled in <a>dbmerge</a>; it (and also large cases of <a>dbsort</a>) should now exploit greater parallelism.</p>

</dd>
<dt id="BUG-FIX58">BUG FIX</dt>
<dd>

<p>Test case with <code>Fsdb::BoundedQueue</code> (gone since 2.44) now removed.</p>

</dd>
</dl>

<h2 id="continuing-cleanup-of-our-no-threads-version">2.46, 2013-10-08 continuing cleanup of our no-threads version</h2>

<dl>

<dt id="BUG-FIX59">BUG FIX</dt>
<dd>

<p>Fixed some packaging details. (Really, threads are no longer required, missing tests in the MANIFEST.)</p>

</dd>
<dt id="IMPROVEMENT30">IMPROVEMENT</dt>
<dd>

<p><a>dbsort</a> now better communicates with the merge process to avoid bursty parallelism.</p>

<p><a>Fsdb::IO::Writer</a> now can take <code>-autoflush =</code> 1&gt; for line-buffered IO.</p>

</dd>
</dl>

<h2 id="test-suite-cleanup-for-non-threaded-perls">2.47, 2013-10-12 test suite cleanup for non-threaded perls</h2>

<dl>

<dt id="BUG-FIX60">BUG FIX</dt>
<dd>

<p>Removed some stray &quot;use threads&quot; in some test cases. We didn&#39;t need them, and these were breaking non-threaded perls.</p>

</dd>
<dt id="BUG-FIX61">BUG FIX</dt>
<dd>

<p>Better handling of Fred cleanup; should fix intermittent <a>dbmapreduce</a> failures on BSD.</p>

</dd>
<dt id="ENHANCEMENT26">ENHANCEMENT</dt>
<dd>

<p>Improved test framework to show output when tests fail. (This time, for real.)</p>

</dd>
</dl>