- Adds new streams
talk_phone_numbers
andticket_metric_events
#111
- Adds backoff/retry for
ProtocolError
andChunkedEncodingError
#131
- Incremental Exports API implementation for User's stream #127
- Fix Infinite Loop for Users #103
- Request Timeout Implementation #79
- 503 error code retry is taken care in ths release #95
- 524 and 520 error codes are taken care in this release to retry 93
- Reverted back API access change login during discover mode 90
- Removed Buffer System 77
- Check API access in discovery mode 74
- Comprehensive Error Messaging 69
- Other small enhancements including request timeout, following best practicesfor integration testing 78
- Fixing the via.from field on the tickets, and ticket_comments stream #75
- Remove usage of
zenpy
library for tickets stream and all sub-streams
- Remove usage of
- Revert Organizations Stream back to the Incremental Search endpoint #70
- Limit the Satisfaction Ratings Stream by the bookmarked updated at value #67
- Log the id for the users found outside of the queried window #65
Changes introduced in #64
- Add to setup.py a "test" extra_requires so that CI doesn't have to install ipdb
- Add to setup.py dependencies on backoff and requests
- Add retry logic to the requests made for the Organizations, Satisfaction Ratings, Ticket Fields, Tags, and GroupMemberships streams
- Add unit tests around our new cursor based pagination logic
- Update singer-python from 5.2.1 to 5.12.2
- Update zenpy from 2.0.0 to 2.0.24
- Update the Organizations, Satisfaction Ratings, Ticket Fields, Tags, GroupMemberships, and Tickets to use cursor-based pagination
- Tickets leans on zenpy to do pagination
- The other streams do not
- Fix Circle Config to install "test" dependencies instead of "dev"
- Log Request URL (and URL params), Response ETag, and Response 'X-Request-Id' header to help with troubleshooting #63
- Break out of infinite loop with users stream and bookmark on date window end #46
- Add retry logic for users stream #45
- Add error message to go along with assert #44
- Add date windowing to users stream and satisfaction ratings stream #42
- Use
start_time
query parameter for satisfaction_ratings stream #37
- Reduce the number of
Transformer
log messages from 1 per record to 1 per stream #28
- Add zendesk partner headers to all requests #27
- Add
ticket_id
field toticket_comments
stream #22
- Make some fields for the
sla_policies
stream nullable
- Adds the SLA Policies stream #21
- Add ability to auth with API token #20
- The Following changes included in #18:
- Fixes a bug where some
group_membership
records were being returned without anupdated_at
field. The solution is to check for theupdated_at
, and if it is not present, log an info message and sync it. This means records without anupdated_at
field will be sync'd on every run. - Fixes schema refs by adding
shared/
- Fixes a bug where some
- Add
report_csv
tousers
stream
- Include shared schema refs in package
- Add
satisfaction_ratings
andticket_comment
streams #17
- Adds more JSON Schema to the tickets and ticket_audits schemas #16
- Add
ticket_forms
stream #15
- Version bump for initial release
- Catch RecordNotFound exceptions in ticket_audits and ticket_metrics
- Bug fix to populate tickets bookmark if replication_key changes
- The following changes are included in #14:
- Made ticket_audits and ticket_metrics substreams of tickets stream. This allows audits and metrics to be retrieved for archived tickets
- Changed replication key to
generated_timestamp
for tickets, ticket_audits, and ticket_metrics - Added
problem_id
,forum_topic_id
, andsatisfaction_rating
fields to tickets schema (addresses #10) - Added
system_field_options
andsub_type_id
to ticket_fields schema (addresses #11)
- Adding additional fields to the tickets schema #13
- Bumped
singer-python
to 5.1.5 to get fix for emptyproperties
subschema - Added
deleted_at
to organizations schema
- Changed the '-' to a '_' in stream names #9
- Fixed a bug in metadata generation for the Tags stream #8
- Sets replication-keys to 'automatic' inclusion
- Removes "fields" because they duplicate "custom-fields"
- Skips tickets that come back by accident (out of the date range)
- Updates to schemas #7
- Update streams' schemas to reflect testing data
- Add discovery of custom fields for organizations and users (fixes error when syncing these streams due to incorrect schema)
- Add audits stream
- Update schemas for other streams
- Misc bug fixes
- Initial end-to-end with most streams, bookmarking, and discovery