Update setup processor according to new result schema #1448
Conversation
sfc-gh-bdufour
force-pushed
the
bdufour-sqlgen-m2
branch
3 times, most recently
from
69a3096 to
5c6f613
Compare
sfc-gh-bdufour
force-pushed
the
bdufour-sqlgen-m2
branch
from
5c6f613 to
9414704
Compare
| @@ -34,7 +36,7 @@ | |||
| ) | |||
|
|
|||
| SNOWPARK_PROCESSOR = "snowpark" | |||
| NA_SETUP_PROCESSOR = "native-app-setup" | |||
| NA_SETUP_PROCESSOR = "native app setup" | |||
There was a problem hiding this comment.
Somehow I keep forgetting that natural strings like this are more natural in YAML.
| processor_ctx = copy.copy(self._bundle_ctx) | ||
| processor_subdirectory = re.sub(r"[^a-zA-Z0-9_$]", "_", processor_name) |
There was a problem hiding this comment.
Changed the logic so that each processor uses its own subdirectory for its artifacts, enforced by the framework. Right now each processor was responsible for doing this, so I just centralized the same logic to facilitate the implementation of new processors.
sfc-gh-bdufour
force-pushed
the
bdufour-sqlgen-m2
branch
from
9414704 to
6460a9f
Compare
| processor_ctx = copy.copy(self._bundle_ctx) | ||
| processor_subdirectory = re.sub(r"[^a-zA-Z0-9_$]", "_", processor_name) | ||
| processor_ctx.bundle_root = ( | ||
| self._bundle_ctx.bundle_root / processor_subdirectory |
There was a problem hiding this comment.
if each processor gets its own directory, how do we chain processors?
There was a problem hiding this comment.
I only made that change for the bundle root (which is the processor's sandbox) and the generated directory, which is similarly processor-specific. The deploy root remains unchanged, as processors should always be reading and writing from/to the deploy root so they can be chained.
sfc-gh-bdufour
force-pushed
the
bdufour-sqlgen-m2
branch
from
6460a9f to
e22c160
Compare
| @@ -40,6 +46,16 @@ | |||
| DEFAULT_TIMEOUT = 30 | |||
| DRIVER_PATH = Path(__file__).parent / "setup_driver.py.source" | |||
|
|
|||
| log = logging.getLogger(__name__) | |||
There was a problem hiding this comment.
wish we had named this logger instead of log everywhere :')
src/snowflake/cli/_plugins/nativeapp/codegen/setup/native_app_setup_processor.py
Show resolved
Hide resolved
sfc-gh-bdufour
dismissed stale reviews from sfc-gh-pjafari and sfc-gh-melnacouzi
via
2788e2e
| for inst in mod.get("instructions", []): | ||
| if inst.get("type") == "set": | ||
| payload = inst.get("payload") | ||
| if payload: | ||
| key = payload.get("key") | ||
| value = payload.get("value") |
There was a problem hiding this comment.
nit: too much nesting here, would be nice to factor this out if possible.
There was a problem hiding this comment.
sure, I'll keep this in mind when I rework this logic in the near future.
sfc-gh-bdufour
force-pushed
the
bdufour-sqlgen-m2
branch
from
2788e2e to
407b5f6
Compare
sfc-gh-bdufour
force-pushed
the
bdufour-sqlgen-m2
branch
from
407b5f6 to
3d3a52d
Compare
Pre-review checklist
Changes description
Updating the setup processor driver to match ToT in the snowflake.app package
Testing
Since the snowflake.app package is still private, this was tested manually using an env var override. Integ and E2E tests will be added when the feature is closer to PrPr.