Page MenuHomePhorge

wait until we have at least a tag+command for filtering to commence
ClosedPublic

Authored by seigo on Sep 7 2016, 10:01 AM.
Tags
None
Referenced Files
F11586783: D215.id623.diff
Thu, Mar 28, 3:50 PM
F11586387: D215.id476.diff
Thu, Mar 28, 2:44 PM
F11585262: D215.diff
Thu, Mar 28, 11:12 AM
Unknown Object (File)
Mon, Mar 18, 1:05 AM
Unknown Object (File)
Mon, Mar 18, 12:12 AM
Unknown Object (File)
Wed, Mar 13, 4:36 AM
Unknown Object (File)
Mon, Mar 11, 12:21 PM
Unknown Object (File)
Sun, Mar 10, 10:51 AM
Subscribers
None

Details

Summary

This attempts to work around the interesting behavior seen with iPhones doing
TLS where the tag of the command is sent in a network packet on its own separate from
the command string.

Diff Detail

Repository
rG guam
Lint
Lint Not Applicable
Unit
Tests Not Applicable

Event Timeline

seigo retitled this revision from to wait until we have at least a tag+command for filtering to commence.
seigo updated this object.
seigo edited the test plan for this revision. (Show Details)
seigo added a reviewer: Restricted Project.

Note that only very nominal testing has been done thus far, but I wanted to give others a preview look at what is cooking here.

This also brings a nice little optimization where commands are split into tag/command/data only *once* instead of at least twice per client-side rule application (something that would get even worse as the number of rues grows ...)

This revision was automatically updated to reflect the committed changes.
  • works correctly and one hopes with enough performance now
This revision was automatically updated to reflect the committed changes.