[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: [OASIS Issue Tracker] (MQTT-203) MQTT URI Scheme
[ https://issues.oasis-open.org/browse/MQTT-203?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=61686#comment-61686 ]
Nicholas O'Leary edited comment on MQTT-203 at 1/27/16 11:40 AM:
-----------------------------------------------------------------
Following discussion at the Jan 2016 face-to-face TC meeting, the following comments were made:
- This should reference RFC 3986 for URI structure > https://www.ietf.org/rfc/rfc3986.txt
- It should get registered as an IANA URI Scheme
- Publish first as a Committee note, but in proper IANA language to enable its registration (can be picked up as a (non?)normative chapter in a future spec)
- The 'query parameters' should be full specified for all connect options (not just clientid as in the current proposal)
- The URI could include a topic as part of the 'path' component
- including a topic raises questions over encoding:
- Standard uri parsing libraries will clean up path components to remove double slashes and relative components ('/a/../b' -> '/b')
- topics that begin with a '/' are tricky as they will naturally need to have a double slash in the uri
- This presumes a raw TCP connection. How might it extend to identifying the transport is websocket based (or ano underlying transport in the future)
Next steps:
- Nick to come back with an updated proposal that addresses these points
was (Author: knolleary):
Following discussion at the Jan 2016 face-to-face TC meeting, the following comments were made:
- This should reference RFC 3986 for URI structure > https://www.ietf.org/rfc/rfc3986.txt
- It should get registered as an IANA URI Scheme
- Publish first as a Committee note, but in proper IANA language to enable its registration (can be picked up as a (non?)normative chapter in a future spec)
- The 'query parameters' should be full specified for all connect options (not just clientid as in the current proposal)
- The URI could include a topic as part of the 'path' component
- including a topic raises questions over encoding:
- Standard uri parsing libraries will clean up path components to remove double slashes and relative components ('/a/../b' -> '/b')
- topics that begin with a '/' are tricky as they will naturally need to have a double slash in the uri
- This presumes a raw TCP connection. How might it extend to identifying the transport is websocket based (or ano underlying transport in the future)
> MQTT URI Scheme
> ---------------
>
> Key: MQTT-203
> URL: https://issues.oasis-open.org/browse/MQTT-203
> Project: OASIS Message Queuing Telemetry Transport (MQTT) TC
> Issue Type: New Feature
> Components: futures
> Reporter: Nicholas O'Leary
> Priority: Critical
>
> There have been a number of discussions in the community on how to specify a broker's connection details in a uri.
> The result of the community discussion is captured here - https://github.com/mqtt/mqtt.github.io/wiki/URI-Scheme
> In summary, the following uri format is proposed:
> mqtt://[username][:password]@host.domain[:port][?clientid=clientid]
> The uri scheme could alternative be 'mqtts' to specify a secure TLS connection should be used.
> Other connection-time options (clean session etc) could also be expressed as query parameters.
> A path portion of the uri could be used to specify a specific topic.
> This JIRA can be addressed by way of Committee Note, rather than as an addition to the spec itsef.
--
This message was sent by Atlassian JIRA
(v6.2.2#6258)
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]