<div dir="ltr">Control: reassign -1 ruby-em-http-request<div><div>I took the liberty to reassign it to it. Feel free to reassign to</div><div>whichever you see it would be better assigned.</div><div><br></div><div>Please try this patch on ruby-em-http-request.</div><div><br></div><div><div>--- a/lib/em-http/http_connection.rb</div><div>+++ b/lib/em-http/http_connection.rb</div><div>@@ -131,7 +131,7 @@</div><div> end</div><div><br></div><div> @p.on_message_complete = proc do</div><div>- if !client.continue?</div><div>+ if client and !client.continue?</div><div> c = @clients.shift</div><div> c.state = :finished</div><div> c.on_request_complete</div></div><div><br></div><div><br></div><div><br></div><div><br></div><div>Also i tried this but if fails:</div><div><br></div><div><div>--- a/lib/em-http/http_connection.rb</div><div>+++ b/lib/em-http/http_connection.rb</div><div>@@ -131,7 +131,7 @@</div><div> end</div><div><br></div><div> @p.on_message_complete = proc do</div><div>- if !client.continue?</div><div>+ if !client or !client.continue?</div><div> c = @clients.shift</div><div> c.state = :finished</div><div> c.on_request_complete</div></div><div><br></div><div>"c" is Nil here: the change with http-parser 2.7 makes</div><div>"on_message_complete" called in a case when it wasn't before.</div><div><br></div><div>Bug or feature ? nodejs is using http-parser and has a thorough test suite,</div><div>so i suppose we're on the safe side fixing this package (but my experience</div><div>tells me i'm 80% wrong when i suppose things like that).</div><div><br></div><div>Jérémy</div><div><br></div><div><br><div><br></div></div></div></div><div class="gmail_extra"><br><div class="gmail_quote">2018-04-04 0:16 GMT+02:00 Jérémy Lal <span dir="ltr"><<a href="mailto:jerry@eda.sarl" target="_blank">jerry@eda.sarl</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote"><div><div class="h5">2018-04-03 23:22 GMT+02:00 Christoph Biedl <span dir="ltr"><<a href="mailto:debian.axhn@manchmal.in-ulm.de" target="_blank">debian.axhn@manchmal.in-ulm.<wbr>de</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Jérémy Lal wrote...<br>
<br>
> Control: reassign -1 ruby-http-parser.rb<br>
<br>
Thanks for taking care of that one. However ...<br>
<span><br>
> it seems ruby-em-http-request uses a fork of http_parser.rb that has not<br>
> been updated for a while. However, upstream seems to have fixed some<br>
> issues with more recents http-parser developments:<br>
><br>
> <a href="https://github.com/tmm1/http_parser.rb/commit/7ef407e35247750e369746dec4ed4ed6c2265f73" rel="noreferrer" target="_blank">https://github.com/tmm1/http_p<wbr>arser.rb/commit/7ef407e3524775<wbr>0e369746dec4ed4ed6c2265f73</a><br>
><br>
> So it'd rather be ruby-http-parser.rb debian package that needs a patch here.<br>
<br>
</span>The *ruby-http-parser.rb* package has been fixed in #881627 (yes, that<br>
was me). The failing package however is *ruby-em-http-request*, and<br>
while you state it's is fork of the first, I cannot quite see a lot of<br>
common code.<br>
<br>
Are you sure the assignment to ruby-http-parser.rb is correct?<br>
<br>
Christoph, still trying to understand why the tests fail<br></blockquote><div><br></div></div></div><div>I'm looking at it more thoroughly. Sorry for the thoughtless decision to move the bug,</div><div>i'll put it back to where it belongs if necesssary.</div><span class="HOEnZb"><font color="#888888"><div><br></div><div>Jérémy</div></font></span></div><br></div></div>
</blockquote></div><br></div>