dsi-workflow 1.2.7__tar.gz → 1.2.8__tar.gz

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dsi-workflow
-Version: 1.2.7
+Version: 1.2.8
 Summary: A Data Science Infrastructure Project
 Author-email: Jesus Pulido <pulido@lanl.gov>, James Ahrens <ahrens@lanl.gov>, Divya Banesh <dbanesh@lanl.gov>, Hugh Greenberg <hng@lanl.gov>, Pascal Grosset <pascalgrosset@lanl.gov>, Vedant Iyer <iyer@lanl.gov>, Benjamin Sims <bsims@lanl.gov>, Terece Turton <tlturton@lanl.gov>
 License-Expression: BSD-3-Clause

dsi_workflow-1.2.8/dsi/_version.py

@@ -0,0 +1 @@
+__version__ = "1.2.8"

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/dsi/backends/denodo.py

@@ -1805,8 +1805,36 @@ class Denodo(Webserver):
 
 
     # ----------------------------------------------------------------------
-    # DSI Interface: Find
+    # DSI Interface: Search/Find
     # ----------------------------------------------------------------------
+    def search(self, params, download=False):
+        """
+        Retrieves metadata or data for entries in a specified dataset.
+
+        Parameters
+        ----------
+        `params` : dict
+            Identifies the dataset and, optionally, a specific entry.
+            For example, {"database_id": 1} retrieves metadata for all entries
+            in the dataset, while {"database_id": 1, "entry_id": 1} retrieves
+            data for a specific entry (if supported).
+
+        `download` : bool, optional, default=False
+            If True, downloads specified data to a local directory.
+            If False, returns dataset's entry metadata when no entry is specified, 
+            or references to a specified entry's data for separate download.
+
+        Returns
+        -------
+        pandas.DataFrame or list of pandas.DataFrame
+            Metadata for a dataset's entries or data references for a specified entry.
+            Returns a list of DataFrames when the metadata spans multiple tables.
+
+        None
+            If no metadata found or data directly downloaded with (download=True)
+        """
+        raise NotImplementedError
+
     def find(self, query_object, **kwargs):
         """Search table names, column names and cell values."""
         return (

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/dsi/backends/duckdb.py

@@ -358,15 +358,14 @@ class DuckDB(Filesystem):
             self.cur.execute(create_query)
             units_data = artifacts["dsi_units"]
             for table_val, col_val, unit_val in zip(units_data["table_name"], units_data["column_name"], units_data["unit"]):
-                str_query = f"INSERT INTO dsi_units VALUES ('{table_val}', '{col_val}', '{unit_val}')"
-                unit_result = self.cur.execute(f"""SELECT unit FROM dsi_units 
-                                                WHERE table_name = '{table_val}' AND column_name = '{col_val}';""").fetchone()
+                unit_result = self.cur.execute("""SELECT unit FROM dsi_units WHERE 
+                                                table_name = ? AND column_name = ?;""", [table_val, col_val]).fetchone()
                 if unit_result and unit_result[0] != unit_val: #checks if unit for same table and col exists in db and if units match
                     self.cur.execute("ROLLBACK")
                     raise TypeError(f"Cannot ingest different units for the column {col_val} in {table_val}")
                 elif not unit_result:
                     try:
-                        self.cur.execute(str_query)
+                        self.cur.execute("INSERT INTO dsi_units VALUES (?, ?, ?)", [table_val, col_val, unit_val])
                     except duckdb.Error as e:
                         self.cur.execute("ROLLBACK")
                         self.cur.execute("CHECKPOINT")
@@ -1003,7 +1002,7 @@ class DuckDB(Filesystem):
                         AVG("{col_name}"),
                         STDDEV_SAMP("{col_name}")
                     FROM {table_name}
-                    WHERE "{col_name}" IS NOT NULL
+                    WHERE isfinite("{col_name}")
                 """).fetchone()
             else:
                 min_val = max_val = avg_val = std_dev = None

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/dsi/backends/ndp.py

@@ -1059,8 +1059,36 @@ class NDP(Webserver):
 
 
     # ----------------------------------------------------------------------
-    # Find Methods
+    # Search/Find Methods
     # ----------------------------------------------------------------------
+    def search(self, params, download=False):
+        """
+        Retrieves metadata or data for entries in a specified dataset.
+
+        Parameters
+        ----------
+        `params` : dict
+            Identifies the dataset and, optionally, a specific entry.
+            For example, {"database_id": 1} retrieves metadata for all entries
+            in the dataset, while {"database_id": 1, "entry_id": 1} retrieves
+            data for a specific entry (if supported).
+
+        `download` : bool, optional, default=False
+            If True, downloads specified data to a local directory.
+            If False, returns dataset's entry metadata when no entry is specified, 
+            or references to a specified entry's data for separate download.
+
+        Returns
+        -------
+        pandas.DataFrame or list of pandas.DataFrame
+            Metadata for a dataset's entries or data references for a specified entry.
+            Returns a list of DataFrames when the metadata spans multiple tables.
+
+        None
+            If no metadata found or data directly downloaded with (download=True)
+        """
+        raise NotImplementedError
+
     def find(self, query_object, **kwargs):
         """
         Searches for all instances of query_object across all tables.

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/dsi/backends/oceans11.py

@@ -794,8 +794,36 @@ class Oceans11(Webserver):
         return self._cache
 
     # ------------------------------------------------------------------
-    # Find Methods
+    # Search/Find Methods
     # ------------------------------------------------------------------
+    def search(self, params, download=False):
+        """
+        Retrieves metadata or data for entries in a specified dataset.
+
+        Parameters
+        ----------
+        `params` : dict
+            Identifies the dataset and, optionally, a specific entry.
+            For example, {"database_id": 1} retrieves metadata for all entries
+            in the dataset, while {"database_id": 1, "entry_id": 1} retrieves
+            data for a specific entry (if supported).
+
+        `download` : bool, optional, default=False
+            If True, downloads specified data to a local directory.
+            If False, returns dataset's entry metadata when no entry is specified, 
+            or references to a specified entry's data for separate download.
+
+        Returns
+        -------
+        pandas.DataFrame or list of pandas.DataFrame
+            Metadata for a dataset's entries or data references for a specified entry.
+            Returns a list of DataFrames when the metadata spans multiple tables.
+
+        None
+            If no metadata found or data directly downloaded with (download=True)
+        """
+        raise NotImplementedError
+
     def find(self, query_object, **kwargs):
         """Search for a value across table, column, and cell levels."""
         if not self._loaded:

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/dsi/backends/osti.py

@@ -713,8 +713,36 @@ class OSTI(Webserver):
         )
 
     # ----------------------------------------------------------------------
-    # Find Methods
+    # Search/Find Methods
     # ----------------------------------------------------------------------
+    def search(self, params, download=False):
+        """
+        Retrieves metadata or data for entries in a specified dataset.
+
+        Parameters
+        ----------
+        `params` : dict
+            Identifies the dataset and, optionally, a specific entry.
+            For example, {"database_id": 1} retrieves metadata for all entries
+            in the dataset, while {"database_id": 1, "entry_id": 1} retrieves
+            data for a specific entry (if supported).
+
+        `download` : bool, optional, default=False
+            If True, downloads specified data to a local directory.
+            If False, returns dataset's entry metadata when no entry is specified, 
+            or references to a specified entry's data for separate download.
+
+        Returns
+        -------
+        pandas.DataFrame or list of pandas.DataFrame
+            Metadata for a dataset's entries or data references for a specified entry.
+            Returns a list of DataFrames when the metadata spans multiple tables.
+
+        None
+            If no metadata found or data directly downloaded with (download=True)
+        """
+        raise NotImplementedError
+
     def find(self, query_object, **kwargs):
         """
         Searches for all instances of `query_object` across the table, column, and cell levels.

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/dsi/backends/rcsbpdb.py

@@ -3,65 +3,68 @@ RCSBPDB Webserver Backend for DSI
 
 Read-only metadata-first backend that retrieves rcsb pdb metadata
 and exposes it as in-memory DSI tables.
-"""
-# Access modes
-# ------------
-# 1. Identifier-driven mode
-#    - DOI input, e.g. "10.2210/pdb1cbs/pdb"
-#    - PDB ID input, e.g. "1CBS"
-
-# 2. Query-driven mode, closer to NDP
-#    - params={"keywords": "hemoglobin", "limit": 5}
-#    - find_relation({"keywords": "hemoglobin", "limit": 5})
-#    - find_relation("hemoglobin")
-
-# DOI behavior
-# ------------
-# RCSB-style DOI input is supported through identifiers or params.
-# Only DOIs of the form 10.2210/pdbXXXX/pdb are converted directly into PDB IDs.
-# General publication DOI search is not currently supported.
-
-# REST flow
-# ---------
-# RCSB Search API -> PDB IDs -> RCSB Data API -> normalized DSI tables
-
-# Tables
-# ------
-# - datasets
-# - resources
-# - errors
-
-# Tier mapping
-# ------------
-# self.tables["datasets"] = Tier 1 datasets
-# self.tables["resources"] = Tier 2 normalized resources
-# self.tables["errors"] = failed/skipped lookups
-
-# Current scope
-# -------------
-# - Metadata-first
-# - Read-only
-# - REST APIs only
-# - Exposes mmCIF download URLs
-# - Does not parse raw mmCIF content yet
-
-# identifiers
-# → __init__()
-# → _load_initial_data()
-# → lookup_identifier()
-# → lookup_rcsbpdb()
-# → _request()
-# → GET https://data.rcsb.org/rest/v1/core/entry/{pdb_id}
-
-# params
-# → __init__()
-# → _load_from_params()
-# → _search_rcsb()
-# → _build_search_query()
-# → _post_json()
-# → POST https://search.rcsb.org/rcsbsearch/v2/query
 
 
+Access modes
+------------
+1. Identifier-driven mode
+   - DOI input, e.g. "10.2210/pdb1cbs/pdb"
+   - PDB ID input, e.g. "1CBS"
+
+2. Query-driven mode, closer to NDP
+   - params={"keywords": "hemoglobin", "limit": 5}
+   - find_relation({"keywords": "hemoglobin", "limit": 5})
+   - find_relation("hemoglobin")
+
+DOI behavior
+------------
+RCSB-style DOI input is supported through identifiers or params.
+Only DOIs of the form 10.2210/pdbXXXX/pdb are converted directly into PDB IDs.
+General publication DOI search is not currently supported.
+
+REST flow
+---------
+RCSB Search API -> PDB IDs -> RCSB Data API -> normalized DSI tables
+
+Tables
+------
+- datasets
+- resources
+- errors
+
+Tier mapping
+------------
+self.tables["datasets"] = Tier 1 datasets
+self.tables["resources"] = Tier 2 normalized resources
+self.tables["errors"] = failed/skipped lookups
+
+Current scope
+-------------
+- Metadata-first
+- Read-only
+- REST APIs only
+- Exposes mmCIF download URLs
+- Does not parse raw mmCIF content yet
+
+**identifiers**
+
+- __init__()
+- _load_initial_data()
+- lookup_identifier()
+- lookup_rcsbpdb()
+- _request()
+- GET https://data.rcsb.org/rest/v1/core/entry/{pdb_id}
+
+**params**
+
+- __init__()
+- _load_from_params()
+- _search_rcsb()
+- _build_search_query()
+- _post_json()
+- POST https://search.rcsb.org/rcsbsearch/v2/query
+"""
+
 from __future__ import annotations
 
 import builtins
@@ -1354,6 +1357,34 @@ class RCSBPDB(Webserver):
         self._loaded = True
         return self.tables
 
+    def search(self, params, download=False):
+        """
+        Retrieves metadata or data for entries in a specified dataset.
+
+        Parameters
+        ----------
+        `params` : dict
+            Identifies the dataset and, optionally, a specific entry.
+            For example, {"database_id": 1} retrieves metadata for all entries
+            in the dataset, while {"database_id": 1, "entry_id": 1} retrieves
+            data for a specific entry (if supported).
+
+        `download` : bool, optional, default=False
+            If True, downloads specified data to a local directory.
+            If False, returns dataset's entry metadata when no entry is specified, 
+            or references to a specified entry's data for separate download.
+
+        Returns
+        -------
+        pandas.DataFrame or list of pandas.DataFrame
+            Metadata for a dataset's entries or data references for a specified entry.
+            Returns a list of DataFrames when the metadata spans multiple tables.
+
+        None
+            If no metadata found or data directly downloaded with (download=True)
+        """
+        raise NotImplementedError
+
     def find(self, query_object, **kwargs):
         """
         Searches for all instances of query_object across all loaded tables.

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/dsi/backends/sqlite.py

@@ -323,15 +323,14 @@ class Sqlite(Filesystem):
             self.cur.execute(create_query)
             units_data = artifacts["dsi_units"]
             for table_val, col_val, unit_val in zip(units_data["table_name"], units_data["column_name"], units_data["unit"]):
-                str_query = f"INSERT INTO dsi_units VALUES ('{table_val}', '{col_val}', '{unit_val}')"
-                unit_result = self.cur.execute(f"""SELECT unit FROM dsi_units 
-                                                WHERE table_name = '{table_val}' AND column_name = '{col_val}';""").fetchone()
+                unit_result = self.cur.execute("""SELECT unit FROM dsi_units WHERE 
+                                                table_name = ? AND column_name = ?;""", [table_val, col_val]).fetchone()
                 if unit_result and unit_result[0] != unit_val: #checks if unit for same table and col exists in db and if units match
                     self.con.rollback()
                     raise TypeError(f"Cannot ingest different units for the column {col_val} in {table_val}")
                 elif not unit_result:
                     try:
-                        self.cur.execute(str_query)
+                        self.cur.execute("INSERT INTO dsi_units VALUES (?, ?, ?)", [table_val, col_val, unit_val])
                     except sqlite3.Error as e:
                         self.con.rollback()
                         raise sqlite3.Error(e)

{dsi_workflow-1.2.7 → dsi_workflow-1.2.8}/dsi/backends/tests/test_oceans11.py

@@ -33,25 +33,27 @@ TEST_OSTI_ID = "3022783"
 # =============================================================================
 # Fixtures
 # =============================================================================
-
 @pytest.fixture(scope="session")
 def backend(tmp_path_factory):
     """Create one live Oceans11 backend for the entire test session."""
     workspace = tmp_path_factory.mktemp("oceans11")
 
-    instance = Oceans11(
-        params=[
-            {"osti_id": TEST_OSTI_ID, "rows": 1},
-            {"osti_id": TEST_OSTI_ID, "rows": 1},
-        ],
-        workspace=str(workspace),
-    )
-
-    yield instance
-
-    if instance._loaded:
-        instance.close()
+    try:
+        instance = Oceans11(
+            params=[
+                {"osti_id": TEST_OSTI_ID, "rows": 1},
+                {"osti_id": TEST_OSTI_ID, "rows": 1},
+            ],
+            workspace=str(workspace),
+        )
+    except ConnectionError as e:
+        pytest.skip(f"Oceans11 is unavailable: {e}")
 
+    try:
+        yield instance
+    finally:
+        if instance._loaded:
+            instance.close()
 
 @pytest.fixture(scope="session")
 def record(backend):