https://pfspipe.ipmu.jp/jira This file is an XML representation of an issue en-us 8.3.4 803005 13-09-2019 https://pfspipe.ipmu.jp/jira/browse/PIPE2D-26 DRP 2-D Pipeline <p>Write tests for FiberTraceSet using LSST-style wrapping of python's unittest</p> PIPE2D-26

Write tests for FiberTraceSet generation

Story Major Done Done aritter rhl Wed, 3 Sep 2014 14:33:09 +0000 Fri, 29 Jul 2016 05:33:38 +0000 Fri, 29 Jul 2016 05:33:38 +0000 0 3 <p>@rhl: What exactly do you mean by LSST-style wrapping of python's unittest? I have written python unittests for FiberTraces, Spectra, and PSF. Also, this issue is pretty much the same is <a href="https://pfspipe.ipmu.jp/jira/browse/INFRA-20" title="Create unittests" class="issue-link" data-issue-key="INFRA-20"><del>INFRA-20</del></a>? How should I name git branches for issues like <a href="https://pfspipe.ipmu.jp/jira/browse/INFRA-20" title="Create unittests" class="issue-link" data-issue-key="INFRA-20"><del>INFRA-20</del></a> and <a href="https://pfspipe.ipmu.jp/jira/browse/PIPE2D-20" title="Estimate (2-D) PSF given a comb spectrum" class="issue-link" data-issue-key="PIPE2D-20"><del>PIPE2D-20</del></a>? Standard I think is tickets/0020 for both of them...</p> <p>Added tests to check number and length of FiberTraces as now we have kind of realistic simulated images which can stick around and won't be redundant every time we get new simulations</p> <p>As far as I can see, the total amount of test code added as part of this ticket is very small. However, it's a delta on top of a large amount of existing code in <tt>tests/FiberTraces.py</tt>. The issue description isn't really specific enough for me to understand whether the existing code + the new changes fulfil the goals. These changes do check that we extract the correct number of fibre traces from some predefined test data, and then verifies that their size lies within acceptable ranges. That seems like a good addition.</p> <p>I have only minor comments on the new code:</p> <ul> <li>Please try not to add extra, unnecessary white space. (Line 44 of <tt>tests/FiberTraces.py</tt> reads " \n", which is a bit useless.)</li> <li>It would be good to specify why these particular values for the constants <tt>nFiberTraces</tt>, <tt>minLength</tt> and <tt>maxLength</tt> were chosen. The answer is presumably just that these correspond to the particular test data available in the repository, but, if so, say so.</li> <li>In general, the code in this file consists of a few very, very long methods. The new code just makes one of them longer. This is a really bad way to structure tests: much better to break them up into fine-grained chunks. I don't think restructuring the whole file is in scope for this ticket, but consider whether these new lines could be broken out into a separate method.</li> </ul> <p>Added comment about nFiberTraces, minLength, and maxLength as advised by the reviewer. Merged with master.</p> Blocks INFRA-31 Development Epic Link PIPE2D-23 Rank 0|02qpuc:v Reviewers swinbank Sprint 2014-9 2014-12 2014-13 2014-14 2014-15 Story Points 6.0