Skip to content
This repository has been archived by the owner on Jul 13, 2023. It is now read-only.

Ensure router_type is present in _router_completed #549

Closed
bbangert opened this issue Jul 23, 2016 · 1 comment · Fixed by #576 or #577
Closed

Ensure router_type is present in _router_completed #549

bbangert opened this issue Jul 23, 2016 · 1 comment · Fixed by #576 or #577
Assignees
@jrconlin
Labels
bug p1
Milestone
PUSHSVC-0: quality

Comments

@bbangert
Copy link
Member

bbangert commented Jul 23, 2016 •
edited
Loading

A previous PR fixed the return handing in _router_completed, this fix was not applied to the _router_completed in web/webpush.py however.
@bbangert bbangert added the bug label Jul 23, 2016
@bbangert bbangert added this to the PUSHSVC-0: quality milestone Jul 23, 2016
@bbangert bbangert added the p1 label Jul 25, 2016
@jrconlin jrconlin self-assigned this Jul 25, 2016
jrconlin added a commit that referenced this issue Aug 1, 2016
@jrconlin
bug: Do not remove router_type from UAID info
d74c00c 
Webpush web handler should not remove router_type from UAID information.

Also fixed error message formatting and added appropriate test to ensure
that router_type persists.

closes #549
@jrconlin
Copy link
Member

jrconlin commented Aug 1, 2016

The problem is that "router_type" is being mistakenly deleted from UAID information, this would cause endpoints to potentially lose routing, since each router type would define how messages should be delivered. 9e3fedb only partially fixed all paths.

@jrconlin jrconlin mentioned this issue Aug 1, 2016
Merged
jrconlin added a commit that referenced this issue Aug 2, 2016
@jrconlin
fix: Normalize endpoint and web _router_completed calls and tests
53df3f3 
closes #549
@jrconlin jrconlin mentioned this issue Aug 2, 2016
Merged
jrconlin added a commit that referenced this issue Aug 2, 2016
@jrconlin
fix: Normalize endpoint and web _router_completed calls and tests
aba8dd0 
closes #549
jrconlin added a commit that referenced this issue Aug 2, 2016
@jrconlin
fix: Normalize endpoint and web _router_completed calls and tests
c05ecbd 
closes #549
jrconlin added a commit that referenced this issue Aug 3, 2016
@jrconlin
bug: Allow old GCM senderID
aa1ce41 
Some android clients have an older SenderID. The server needs to allow
and send to that older SenderID since it is extremely difficult to
change this value on the client.

In addtion:
* drops user on bridge update with empty new routing data (there's
nothing to route to, so drop the user)
* fixes comments and var names to be more expicit about what's
happening.
*

closes #549, #579
jrconlin added a commit that referenced this issue Aug 3, 2016
@jrconlin
bug: Allow old GCM senderID
91af092 
Some android clients have an older SenderID. The server needs to allow
and send to that older SenderID since it is extremely difficult to
change this value on the client.

In addtion:
* drops user on bridge update with empty new routing data (there's
nothing to route to, so drop the user)
* fixes comments and var names to be more expicit about what's
happening.
*

closes #549, #579
jrconlin added a commit that referenced this issue Aug 3, 2016
@jrconlin
bug: Allow old GCM senderID
e0d75dd 
Some android clients have an older SenderID. The server needs to allow
and send to that older SenderID since it is extremely difficult to
change this value on the client.

In addtion:
* drops user on bridge update with empty new routing data (there's
nothing to route to, so drop the user)
* fixes comments and var names to be more expicit about what's
happening.
*

closes #549, #579
jrconlin added a commit that referenced this issue Aug 3, 2016
@jrconlin
bug: Allow old GCM senderID
f4272db 
Some android clients have an older SenderID. The server needs to allow
and send to that older SenderID since it is extremely difficult to
change this value on the client.

In addtion:
* drops user on bridge update with empty new routing data (there's
nothing to route to, so drop the user)
* fixes comments and var names to be more expicit about what's
happening.
*

closes #549, #579
jrconlin added a commit that referenced this issue Aug 3, 2016
@jrconlin
bug: Allow old GCM senderID
2592c46 
Some android clients have an older SenderID. The server needs to allow
and send to that older SenderID since it is extremely difficult to
change this value on the client.

In addtion:
* drops user on bridge update with empty new routing data (there's
nothing to route to, so drop the user)
* fixes comments and var names to be more expicit about what's
happening.
*

closes #549, #579
jrconlin added a commit that referenced this issue Aug 3, 2016
@jrconlin
bug: Allow old GCM senderID
15930a9 
Some android clients have an older SenderID. The server needs to allow
and send to that older SenderID since it is extremely difficult to
change this value on the client.

In addtion:
* drops user on bridge update with empty new routing data (there's
nothing to route to, so drop the user)
* fixes comments and var names to be more expicit about what's
happening.
*

closes #549, #579
jrconlin added a commit that referenced this issue Aug 4, 2016
@jrconlin
bug: Allow old GCM senderID
fe0d19c 
Some android clients have an older SenderID. The server needs to allow
and send to that older SenderID since it is extremely difficult to
change this value on the client.

In addtion:
* drops user on bridge update with empty new routing data (there's
nothing to route to, so drop the user)
* fixes comments and var names to be more expicit about what's
happening.
*

closes #549, #579
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees

@jrconlin jrconlin

Labels
bug p1
Projects
None yet
Milestone
PUSHSVC-0: quality
2 participants
@jrconlin @bbangert