Merge pull request #40 from astronomy-commons/issue/38/read-hipscat-i…

…ndex-column-from-hipscat

Import HiPSCat index column

benchmarks/asv.conf.json

@@ -14,6 +14,9 @@
     "branches": [
         "HEAD"
     ],
+    "install_command": [
+        "python -mpip install -r requirements.txt {wheel_file}"
+    ],
     "build_command": [
         "python -m build --wheel -o {build_cache_dir} {build_dir}"
     ],

requirements.txt

@@ -0,0 +1 @@
+git+https://github.com/astronomy-commons/hipscat.git@main

src/lsdb/core/crossmatch/kdtree_match.py

@@ -1,6 +1,7 @@
 import healpy as hp
 import numpy as np
 import pandas as pd
+from hipscat.pixel_math.hipscat_id import HIPSCAT_ID_COLUMN
 from sklearn.neighbors import KDTree
 
 from lsdb.core.crossmatch.abstract_crossmatch_algorithm import AbstractCrossmatchAlgorithm
@@ -46,7 +47,7 @@ def crossmatch(
         self._rename_columns_with_suffix(self.right, self.suffixes[1])
 
         # concat dataframes together
-        self.left.index.name = "_hipscat_index"
+        self.left.index.name = HIPSCAT_ID_COLUMN
         left_join_part = self.left.iloc[left_ids_filtered].reset_index()
         right_join_part = self.right.iloc[right_ids_filtered].reset_index(drop=True)
         out = pd.concat(
@@ -56,7 +57,7 @@ def crossmatch(
             ],
             axis=1,
         )
-        out.set_index("_hipscat_index", inplace=True)
+        out.set_index(HIPSCAT_ID_COLUMN, inplace=True)
         out["_DIST"] = distances
 
         return out

src/lsdb/dask/crossmatch_catalog_data.py

@@ -8,7 +8,7 @@
 import pandas as pd
 from dask.delayed import Delayed
 from hipscat.pixel_math import HealpixPixel
-from hipscat.pixel_math.hipscat_id import healpix_to_hipscat_id
+from hipscat.pixel_math.hipscat_id import HIPSCAT_ID_COLUMN, healpix_to_hipscat_id
 from hipscat.pixel_tree import PixelAlignment, PixelAlignmentType, align_trees
 
 from lsdb.core.crossmatch.abstract_crossmatch_algorithm import AbstractCrossmatchAlgorithm
@@ -156,7 +156,7 @@ def crossmatch_catalog_data(
         meta[name + suffixes[1]] = pd.Series(dtype=col_type)
     meta["_DIST"] = pd.Series(dtype=np.dtype("float64"))
     meta_df = pd.DataFrame(meta)
-    meta_df.index.name = "_hipscat_index"
+    meta_df.index.name = HIPSCAT_ID_COLUMN
 
     # create dask df from delayed partitions
     ddf = dd.from_delayed(joined_partitions, meta=meta_df)

src/lsdb/loaders/dataframe/dataframe_catalog_loader.py

@@ -10,7 +10,7 @@
 from hipscat.catalog import CatalogType
 from hipscat.catalog.catalog_info import CatalogInfo
 from hipscat.pixel_math import HealpixPixel, generate_histogram
-from hipscat.pixel_math.hipscat_id import compute_hipscat_id, healpix_to_hipscat_id
+from hipscat.pixel_math.hipscat_id import HIPSCAT_ID_COLUMN, compute_hipscat_id, healpix_to_hipscat_id
 from typing_extensions import TypeAlias
 
 from lsdb.catalog.catalog import Catalog, DaskDFPixelMap
@@ -25,7 +25,6 @@ class DataframeCatalogLoader:
     """Creates a HiPSCat formatted Catalog from a Pandas Dataframe"""
 
     DEFAULT_THRESHOLD = 100_000
-    HIPSCAT_INDEX_COLUMN = "_hipscat_index"
 
     def __init__(
             self,
@@ -107,12 +106,12 @@ def load_catalog(self) -> Catalog:
 
     def _set_hipscat_index(self):
         """Generates the hipscat indices for each data point and assigns
-        the _hipscat_index column as the Dataframe index."""
-        self.df[self.HIPSCAT_INDEX_COLUMN] = compute_hipscat_id(
+        the hipscat index column as the Dataframe index."""
+        self.df[HIPSCAT_ID_COLUMN] = compute_hipscat_id(
             ra_values=self.df[self.catalog_info.ra_column],
             dec_values=self.df[self.catalog_info.dec_column],
         )
-        self.df.set_index(self.HIPSCAT_INDEX_COLUMN, inplace=True)
+        self.df.set_index(HIPSCAT_ID_COLUMN, inplace=True)
 
     def _compute_pixel_map(self) -> Dict[HealpixPixel, HealpixInfo]:
         """Compute object histogram and generate the mapping between

tests/lsdb/catalog/test_crossmatch.py

@@ -1,5 +1,6 @@
 import pandas as pd
 import pytest
+from hipscat.pixel_math.hipscat_id import HIPSCAT_ID_COLUMN
 
 from lsdb.core.crossmatch.abstract_crossmatch_algorithm import AbstractCrossmatchAlgorithm
 
@@ -80,7 +81,7 @@ def crossmatch(self, mock_results: pd.DataFrame = None):
             ],
             axis=1,
         )
-        out.set_index("_hipscat_index", inplace=True)
+        out.set_index(HIPSCAT_ID_COLUMN, inplace=True)
         out["_DIST"] = mock_results["dist"].to_numpy()
 
         return out