User tests: Successful: Unsuccessful:
code review
Mozilla telemetry indicates that there is less than 0.24% still using 1.0 https://mzl.la/3tFMtkA
httparchive didnt even include 1.0 in its 2020 almanac https://almanac.httparchive.org/en/2020/http#fig-3 but in 2019 they reported less than 0.06%
Status | New | ⇒ | Pending |
Category | ⇒ | Libraries |
Probably but its beyond the scope of this PR which is just getting rid of outdated http/1.0
Feel free to submit a pr there
I have tested this item
shouldn't be backported on j3 ?
are we backporting everything?
i mean this specific "thing"
I have tested this item
Status | Pending | ⇒ | Ready to Commit |
RTC
Status | Ready to Commit | ⇒ | Fixed in Code Base |
Closed_Date | 0000-00-00 00:00:00 | ⇒ | 2021-11-01 13:08:12 |
Closed_By | ⇒ | richard67 | |
Labels |
Added:
?
?
|
Thanks
Thanks
Is it possible to check the protocol supported by the server (or forcing it by a setting) and then use the right HTTP version?
For example using h2 or Quic/h3 for requests, if supported...
We can reduce a lot latencies (with requests or stream multiplexing) and minimize the protocol overhead (HTTP/1.1 has a lot)
For multiple requests we can see a real speed improvement.