W3C home > Mailing lists > Public > public-webrtc-logs@w3.org > January 2018

Re: [webrtc-stats] Is keeping stats around a memory problem?

From: jan-ivar via GitHub <sysbot+gh@w3.org>
Date: Wed, 24 Jan 2018 14:17:50 +0000
To: public-webrtc-logs@w3.org
Message-ID: <issue_comment.created-360149165-1516803468-sysbot+gh@w3.org>
I think the goal is only to report precise data you can't get otherwise. My understanding was this mattered to account for absolute final byte counts etc. related to a track that was sent, for accounting. Even though the stats spec doesn't say so explicitly, it'll be necessary to get these exact numbers as part of the *replaceTrack* algorithm, and I wouldn't want to delay *replaceTrack* with a complete stats call.
 
I still vote for returning a `RTCStatsReport` object for consistency, to handle cases where there's more than one stats object deleted from a single action - though it just occurred to me that `"stream"` stats are never removed, except maybe by transceiver.stop() - like all old ice candidates after a SLD with iceRestart.

-- 
GitHub Notification of comment by jan-ivar
Please view or discuss this issue at https://github.com/w3c/webrtc-stats/issues/235#issuecomment-360149165 using your GitHub account
Received on Wednesday, 24 January 2018 14:17:52 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:21:58 UTC