<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On 13 June 2016 at 18:00, Aaron M. Ucko <span dir="ltr"><<a href="mailto:amu@alum.mit.edu" target="_blank">amu@alum.mit.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;border-left-color:rgb(204,204,204);padding-left:1ex">Justification: fails to build from source (but built successfully in the past)<br>
<br>
Thanks for taking care of the twolc errors I reported in #826659. The<br>
twolc test now succeeds on little-endian systems, and no longer hangs<br>
anywhere, but still fails on big-endian architectures (mips, powerpc,<br>
s390x, and several non-release architectures). I don't have further<br>
details, but perhaps you can reproduce the problem on a porterbox.<br>
Could you please take a look?</blockquote><div><br></div><div>Can reproduce. Looking into it upstream:Â <a href="https://github.com/hfst/hfst/issues/328">https://github.com/hfst/hfst/issues/328</a></div><div><br></div><div>While it did successfully build in the past, that was only because the test suite was disabled until recently. The tests revealed the unsigned char issue which was easy to fix, and now the endian issue which will not be as easy.</div><div><br></div><div>-- Tino Didriksen</div></div></div></div>