Closed Bug 995767 Opened 11 years ago Closed 11 years ago

Merge wimms into main tokenserver codebase

Categories

(Cloud Services Graveyard :: Server: Token, defect)

x86_64
Windows 7
defect
Not set
normal

Tracking

(Not tracked)

VERIFIED FIXED

People

(Reporter: rfkelly, Assigned: rfkelly)

References

Details

(Whiteboard: [qa+])

Attachments

(1 file)

It's getting unwieldy to maintain wimms as a separate project, and adding migrations per 777650 is to me the final straw. Let's pull out the parts of wimms that we're actually using and put them directly in the tokenserver repo.
Attached patch tokenserver-subsume-wimms.diff (deleted) — Splinter Review
This patch replaces tokenserver.assignment.sqlnode with the matching class from WIMMS, turning it from a single .py file in to a module. In the process it gets rid of a bunch of unused code: * the "ShardedSQLMetadata" wimms backend * the "SecureShardedNodeAssignment" assignment backend We can resurrect these if and when we find we need them, but in the meantime, this will greatly reduce the complexity of working on this repo. The code for SQLNodeAssignment (nee "SQLMetadata" in WIMMS) is unchanged except for the names of a few imports. It could be cleaned up some given the above removals, but that can wait.
Assignee: nobody → rfkelly
Attachment #8405878 - Flags: review?(telliott)
> It could be cleaned up some In particular, we should pull some of the sqlalchemy-handling (mishandling?) stuff into mozsvc for sharing between tokenserver and syncstorage.
Blocks: 993537
No longer blocks: 993537
Comment on attachment 8405878 [details] [diff] [review] tokenserver-subsume-wimms.diff Review of attachment 8405878 [details] [diff] [review]: ----------------------------------------------------------------- Had a momentary panic when I couldn't find the email index, but mostly looks good. ::: tokenserver/assignment/sqlnode/sql.py @@ +252,5 @@ > + # If the current row is marked as replaced, and they haven't > + # been retired, then create them a new node assignment. > + if cur_row.replaced_at is not None: > + if cur_row.generation < MAX_GENERATION: > + user = self.create_user(service, email, can we call this allocate_user or something rather than create_user? create_user is pretty misleading in this context. @@ +256,5 @@ > + user = self.create_user(service, email, > + cur_row.generation, > + cur_row.client_state) > + for old_row in old_rows: > + # Colect any previously-seen client-state values. typo
Attachment #8405878 - Flags: review?(telliott) → review+
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
Whiteboard: [qa+]
ahh... ok So, because of this change, we need to update the syncserver repo to address this bug: https://github.com/mozilla-services/syncserver/issues/14
OK, the fix is in the TokenServer repo now.
Status: RESOLVED → VERIFIED
Product: Cloud Services → Cloud Services Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: