Fix cdp-x/y coordinate dimensionality on 2d/3d gathers by removing redundant offset dimension (#751)

* fix cdp x/y on gathers by removing redundant offset dimension

* import minimal compressors

---------

Co-authored-by: Altay Sansal <[email protected]>

Author: tasansal

src/mdio/builder/templates/seismic_2d_cdp.py

@@ -2,6 +2,10 @@
 
 from typing import Any
 
+from mdio.builder.schemas.compressors import Blosc
+from mdio.builder.schemas.compressors import BloscCname
+from mdio.builder.schemas.dtype import ScalarType
+from mdio.builder.schemas.v1.variable import CoordinateMetadata
 from mdio.builder.templates.base import AbstractDatasetTemplate
 from mdio.builder.templates.types import CdpGatherDomain
 from mdio.builder.templates.types import SeismicDataDomain
@@ -30,3 +34,40 @@ def _name(self) -> str:
 
     def _load_dataset_attributes(self) -> dict[str, Any]:
         return {"surveyType": "2D", "gatherType": "cdp"}
+
+    def _add_coordinates(self) -> None:
+        # Add dimension coordinates
+        self._builder.add_coordinate(
+            "cdp",
+            dimensions=("cdp",),
+            data_type=ScalarType.INT32,
+        )
+        self._builder.add_coordinate(
+            self._gather_domain,
+            dimensions=(self._gather_domain,),
+            data_type=ScalarType.INT32,
+            metadata=CoordinateMetadata(units_v1=self.get_unit_by_key(self._gather_domain)),
+        )
+        self._builder.add_coordinate(
+            self.trace_domain,
+            dimensions=(self.trace_domain,),
+            data_type=ScalarType.INT32,
+            metadata=CoordinateMetadata(units_v1=self.get_unit_by_key(self.trace_domain)),
+        )
+
+        # Add non-dimension coordinates
+        compressor = Blosc(cname=BloscCname.zstd)
+        self._builder.add_coordinate(
+            "cdp_x",
+            dimensions=("cdp",),
+            data_type=ScalarType.FLOAT64,
+            compressor=compressor,
+            metadata=CoordinateMetadata(units_v1=self.get_unit_by_key("cdp_x")),
+        )
+        self._builder.add_coordinate(
+            "cdp_y",
+            dimensions=("cdp",),
+            data_type=ScalarType.FLOAT64,
+            compressor=compressor,
+            metadata=CoordinateMetadata(units_v1=self.get_unit_by_key("cdp_y")),
+        )

src/mdio/builder/templates/seismic_3d_cdp.py

@@ -2,6 +2,10 @@
 
 from typing import Any
 
+from mdio.builder.schemas.compressors import Blosc
+from mdio.builder.schemas.compressors import BloscCname
+from mdio.builder.schemas.dtype import ScalarType
+from mdio.builder.schemas.v1.variable import CoordinateMetadata
 from mdio.builder.templates.base import AbstractDatasetTemplate
 from mdio.builder.templates.types import CdpGatherDomain
 from mdio.builder.templates.types import SeismicDataDomain
@@ -30,3 +34,45 @@ def _name(self) -> str:
 
     def _load_dataset_attributes(self) -> dict[str, Any]:
         return {"surveyType": "3D", "gatherType": "cdp"}
+
+    def _add_coordinates(self) -> None:
+        # Add dimension coordinates
+        self._builder.add_coordinate(
+            "inline",
+            dimensions=("inline",),
+            data_type=ScalarType.INT32,
+        )
+        self._builder.add_coordinate(
+            "crossline",
+            dimensions=("crossline",),
+            data_type=ScalarType.INT32,
+        )
+        self._builder.add_coordinate(
+            self._gather_domain,
+            dimensions=(self._gather_domain,),
+            data_type=ScalarType.INT32,
+            metadata=CoordinateMetadata(units_v1=self.get_unit_by_key(self._gather_domain)),
+        )
+        self._builder.add_coordinate(
+            self.trace_domain,
+            dimensions=(self.trace_domain,),
+            data_type=ScalarType.INT32,
+            metadata=CoordinateMetadata(units_v1=self.get_unit_by_key(self.trace_domain)),
+        )
+
+        # Add non-dimension coordinates
+        compressor = Blosc(cname=BloscCname.zstd)
+        self._builder.add_coordinate(
+            "cdp_x",
+            dimensions=("inline", "crossline"),
+            data_type=ScalarType.FLOAT64,
+            compressor=compressor,
+            metadata=CoordinateMetadata(units_v1=self.get_unit_by_key("cdp_x")),
+        )
+        self._builder.add_coordinate(
+            "cdp_y",
+            dimensions=("inline", "crossline"),
+            data_type=ScalarType.FLOAT64,
+            compressor=compressor,
+            metadata=CoordinateMetadata(units_v1=self.get_unit_by_key("cdp_y")),
+        )

tests/unit/v1/templates/test_seismic_2d_cdp.py

@@ -83,7 +83,7 @@ def validate_coordinates_headers_trace_mask(
     cdp_x = validate_variable(
         dataset,
         name="cdp_x",
-        dims=[("cdp", 512), (gather_domain, 36)],
+        dims=[("cdp", 512)],
         coords=["cdp_x"],
         dtype=ScalarType.FLOAT64,
     )
@@ -92,7 +92,7 @@ def validate_coordinates_headers_trace_mask(
     cdp_y = validate_variable(
         dataset,
         name="cdp_y",
-        dims=[("cdp", 512), (gather_domain, 36)],
+        dims=[("cdp", 512)],
         coords=["cdp_y"],
         dtype=ScalarType.FLOAT64,
     )

tests/unit/v1/templates/test_seismic_3d_cdp.py

@@ -91,7 +91,7 @@ def validate_coordinates_headers_trace_mask(
     cdp_x = validate_variable(
         dataset,
         name="cdp_x",
-        dims=[("inline", 512), ("crossline", 768), (gather_domain, 36)],
+        dims=[("inline", 512), ("crossline", 768)],
         coords=["cdp_x"],
         dtype=ScalarType.FLOAT64,
     )
@@ -100,7 +100,7 @@ def validate_coordinates_headers_trace_mask(
     cdp_y = validate_variable(
         dataset,
         name="cdp_y",
-        dims=[("inline", 512), ("crossline", 768), (gather_domain, 36)],
+        dims=[("inline", 512), ("crossline", 768)],
         coords=["cdp_y"],
         dtype=ScalarType.FLOAT64,
     )