<div dir="ltr"><div>I think it coil have been a complete replacement since a long time ago :)</div><div><br></div><div>About your questions:</div><div><br></div><div>1. In current community version, you get directly the socket from the pool, so it depends if the pool is overloaded or not but that should be fast. In a coming version reusing a socket is local so it doesn't go back and forth.</div><div>2. Yes. This is actually how is handled socks5 or the proxy.</div><div><br></div><div>Feel free to ping me in private if you want more info</div><div><br></div><div>Benoît</div><div><br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, Dec 20, 2020 at 10:29 AM Max Lapshin <<a href="mailto:max.lapshin@gmail.com">max.lapshin@gmail.com</a>> wrote:<br></div><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">Wow, so with pool management it is a complete replacement for lhttpc?<br>
<br>
<br>
Short question: I have several specific hacks to lhttpc:<br>
<br>
1. very detailed timing response: how much time were waiting for<br>
socket, for connecting, for first byte, for headers, etc.<br>
2. our own internal socket source. It is a normal Socket, but we have<br>
some communication before HTTP. Is it possible to provide some kind<br>
of transport module to hackney?<br>
</blockquote></div>