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/DAMD-4 Data Model <p>Hash of calculate_pfsVisitHash and calculate_pfsConfigId need to be 8byte.<br/> &gt; # convert to int and truncate to 8 bytes<br/> &gt; sha1 = int(m.hexdigest(), 16) &amp; 0xffffffff</p> <p>I think 0xffffffff is 4byte....?</p> DAMD-4

Hash need to be 8byte

Bug Major Done Done rhl shimono Thu, 16 Jun 2016 04:15:15 +0000 Mon, 6 May 2019 17:31:24 +0000 Sun, 14 Jan 2018 23:16:13 +0000 0 2 <p>Sorry this took so long to resolve (I thought I had real work to do!)</p> <p>Yes, you're right, but the data model specifies 8 hex digits, so it's a typo in the comment. I've fixed it.</p> <p>I didn't merge the <a href="https://pfspipe.ipmu.jp/jira/browse/DAMD-4" title="Hash need to be 8byte" class="issue-link" data-issue-key="DAMD-4"><del>DAMD-4</del></a> code in utils.py, and when I came to finally merge the issue I realised that we do indeed have a problem. Shimono-san correctly fixed the SHA truncation, but he didn't address the filenames.</p> <p>I specified 64-bit SHA1s; I think that that's plenty and anyway that's the biggest number that we can easily put into fits tables so let's stick to that. Unfortunately, I confused bytes with hexadecimal digits and specified %08x formats, whereas I should have said %016x.</p> <p>This is a pain. I'd simply truncate the SHA1s to 4 bytes for filenames, except that some of the filenames include objIds, and they are also 64 bits and we can't simply truncate them as they are very far from random.</p> <p>I think we have to face reality and write out all 16 hex digits in both places &#8211; ugly. What do you think?</p> <p>OK, I went with the 0x016x format and that is what is implemented in the data model and DRP-2D</p> Relates DAMD-56 Development Rank 0|ii00fb: