redpanda-data / redpanda-data/connect
React on input from GCS for newer files
Nobody has claimed this yet.
- Dominant language
- Go
- Stars
- 8.8k
- Forks
- 969
- Avg merge
- 1d 13h
- Merged PRs (30d)
- 64
Description
At the moment the GCS supports reading from the bucket for all files that are present. However if i have a bucket which is partitioned by the date for example, it would read all the files from all the buckets at every execution
e.g.:
gs://bucket_name/prefix_name/2021-10-20/*.csv
gs://bucket_name/prefix_name/2021-10-19/*.csv
gs://bucket_name/prefix_name/2021-10-18/*.csv
There is the option to delete the objects (i.e. delete_objects after benthos has read them. However this approach is not suitable for all usecase as it is deleting data.
It would be good to have a way to make benthos be aware of the state so it doesn't read the same files again unless specified otherwise.
Or any other idea like for example use a benthos wildcard for the folder names which can be calculated with bloblang.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue points to the GCS input documentation at components/inputs/gcp_cloud_storage; start by reviewing how object discovery and delete_objects currently work. Define how processed-object state or date-based selection should behave without deleting data, and verify that previously read files are not read again unless explicitly requested.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go, google-cloud
- Domain
- cloud, data-engineering, stream-processing
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100