dbgraph 0.2.0__tar.gz → 0.2.1__tar.gz

dbgraph-0.2.1/.idea/workspace.xml

@@ -0,0 +1,267 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="AutoImportSettings">
+    <option name="autoReloadType" value="SELECTIVE" />
+  </component>
+  <component name="ChangeListManager">
+    <list default="true" id="263a96da-9a82-40fd-89b2-ba71dd2c194d" name="Changes" comment="fix bugs in builder">
+      <change beforePath="$PROJECT_DIR$/dbgraph/__init__.py" beforeDir="false" afterPath="$PROJECT_DIR$/dbgraph/__init__.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/dbgraph/builder/sql_graph_builder.py" beforeDir="false" afterPath="$PROJECT_DIR$/dbgraph/builder/sql_graph_builder.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/dbgraph/entity/aspect.py" beforeDir="false" afterPath="$PROJECT_DIR$/dbgraph/entity/aspect.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/dbgraph/persistent/json_utils.py" beforeDir="false" afterPath="$PROJECT_DIR$/dbgraph/persistent/json_utils.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/dbgraph/persistent/sql_graph_persistent.py" beforeDir="false" afterPath="$PROJECT_DIR$/dbgraph/persistent/sql_graph_persistent.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/pyproject.toml" beforeDir="false" afterPath="$PROJECT_DIR$/pyproject.toml" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/tests/test_load_graph_sql.py" beforeDir="false" afterPath="$PROJECT_DIR$/tests/test_load_graph_sql.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/tests/test_sql_graph_persistent.py" beforeDir="false" afterPath="$PROJECT_DIR$/tests/test_sql_graph_persistent.py" afterDir="false" />
+    </list>
+    <option name="SHOW_DIALOG" value="false" />
+    <option name="HIGHLIGHT_CONFLICTS" value="true" />
+    <option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
+    <option name="LAST_RESOLUTION" value="IGNORE" />
+  </component>
+  <component name="FileTemplateManagerImpl">
+    <option name="RECENT_TEMPLATES">
+      <list>
+        <option value="Python Unit Test" />
+        <option value="Python Script" />
+      </list>
+    </option>
+  </component>
+  <component name="Git.Settings">
+    <option name="RECENT_BRANCH_BY_REPOSITORY">
+      <map>
+        <entry key="$PROJECT_DIR$" value="packaging" />
+      </map>
+    </option>
+    <option name="RECENT_GIT_ROOT_PATH" value="$PROJECT_DIR$" />
+  </component>
+  <component name="GitHubPullRequestSearchHistory">{
+  &quot;lastFilter&quot;: {
+    &quot;state&quot;: &quot;OPEN&quot;,
+    &quot;assignee&quot;: &quot;minhdenthedev&quot;
+  }
+}</component>
+  <component name="GithubPullRequestsUISettings">{
+  &quot;selectedUrlAndAccountId&quot;: {
+    &quot;url&quot;: &quot;https://github.com/minhdenthedev/dbgraph.git&quot;,
+    &quot;accountId&quot;: &quot;6c0b7cdc-aa0d-4999-86e3-e6f0a1d74a10&quot;
+  },
+  &quot;recentNewPullRequestHead&quot;: {
+    &quot;server&quot;: {
+      &quot;useHttp&quot;: false,
+      &quot;host&quot;: &quot;github.com&quot;,
+      &quot;port&quot;: null,
+      &quot;suffix&quot;: null
+    },
+    &quot;owner&quot;: &quot;minhdenthedev&quot;,
+    &quot;repository&quot;: &quot;dbgraph&quot;
+  }
+}</component>
+  <component name="ProjectColorInfo">{
+  &quot;associatedIndex&quot;: 0,
+  &quot;fromUser&quot;: false
+}</component>
+  <component name="ProjectId" id="3IdnkFbc3mGdbjiFJYWIoYOgRXM" />
+  <component name="ProjectLevelVcsManager">
+    <ConfirmationsSetting value="2" id="Add" />
+  </component>
+  <component name="ProjectViewState">
+    <option name="hideEmptyMiddlePackages" value="true" />
+    <option name="showLibraryContents" value="true" />
+  </component>
+  <component name="PropertiesComponent"><![CDATA[{
+  "keyToString": {
+    "ASKED_SHARE_PROJECT_CONFIGURATION_FILES": "true",
+    "ModuleVcsDetector.initialDetectionPerformed": "true",
+    "Python tests.Python tests for tests.test_dbgraph.TestDBGraph.executor": "Run",
+    "Python tests.Python tests for tests.test_load_graph_sql.TestLoadGraphSQL.test_load.executor": "Run",
+    "Python tests.Python tests for tests.test_sql_graph_builder.TestSQLGraphBuilder.test_build_graph.executor": "Run",
+    "Python tests.Python tests for tests.test_sql_graph_persistent.TestSQLGraphPersistent.executor": "Run",
+    "Python tests.Python tests for tests.test_sql_graph_persistent.TestSQLGraphPersistent.test_load_graph.executor": "Run",
+    "Python tests.Python tests for tests.test_sql_graph_persistent.TestSQLGraphPersistent.test_save_graph.executor": "Run",
+    "Python tests.Python tests in test_load_graph_sql.py.executor": "Run",
+    "Python tests.Python tests in tests.executor": "Run",
+    "RunOnceActivity.ShowReadmeOnStart": "true",
+    "RunOnceActivity.git.unshallow": "true",
+    "RunOnceActivity.typescript.service.memoryLimit.init": "true",
+    "codeWithMe.voiceChat.enabledByDefault": "false",
+    "git-widget-placeholder": "main",
+    "ignore.virus.scanning.warn.message": "true",
+    "nodejs_package_manager_path": "npm",
+    "settings.editor.selected.configurable": "editor.preferences.folding"
+  }
+}]]></component>
+  <component name="RunManager" selected="Python tests.Python tests in tests">
+    <configuration name="Python tests for tests.test_load_graph_sql.TestLoadGraphSQL.test_load" type="tests" factoryName="Autodetect" temporary="true" nameIsGenerated="true">
+      <module name="dbgraph" />
+      <option name="ENV_FILES" value="" />
+      <option name="INTERPRETER_OPTIONS" value="" />
+      <option name="PARENT_ENVS" value="true" />
+      <option name="SDK_HOME" value="" />
+      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
+      <option name="IS_MODULE_SDK" value="true" />
+      <option name="ADD_CONTENT_ROOTS" value="true" />
+      <option name="ADD_SOURCE_ROOTS" value="true" />
+      <option name="DEBUG_JUST_MY_CODE" value="false" />
+      <option name="RUN_TOOL" value="" />
+      <option name="_new_additionalArguments" value="&quot;&quot;" />
+      <option name="_new_target" value="&quot;tests.test_load_graph_sql.TestLoadGraphSQL.test_load&quot;" />
+      <option name="_new_targetType" value="&quot;PYTHON&quot;" />
+      <method v="2" />
+    </configuration>
+    <configuration name="Python tests for tests.test_sql_graph_builder.TestSQLGraphBuilder.test_build_graph" type="tests" factoryName="Autodetect" temporary="true" nameIsGenerated="true">
+      <module name="dbgraph" />
+      <option name="ENV_FILES" value="" />
+      <option name="INTERPRETER_OPTIONS" value="" />
+      <option name="PARENT_ENVS" value="true" />
+      <option name="SDK_HOME" value="" />
+      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
+      <option name="IS_MODULE_SDK" value="true" />
+      <option name="ADD_CONTENT_ROOTS" value="true" />
+      <option name="ADD_SOURCE_ROOTS" value="true" />
+      <option name="DEBUG_JUST_MY_CODE" value="false" />
+      <option name="RUN_TOOL" value="" />
+      <option name="_new_additionalArguments" value="&quot;&quot;" />
+      <option name="_new_target" value="&quot;tests.test_sql_graph_builder.TestSQLGraphBuilder.test_build_graph&quot;" />
+      <option name="_new_targetType" value="&quot;PYTHON&quot;" />
+      <method v="2" />
+    </configuration>
+    <configuration name="Python tests for tests.test_sql_graph_persistent.TestSQLGraphPersistent" type="tests" factoryName="Autodetect" temporary="true" nameIsGenerated="true">
+      <module name="dbgraph" />
+      <option name="ENV_FILES" value="" />
+      <option name="INTERPRETER_OPTIONS" value="" />
+      <option name="PARENT_ENVS" value="true" />
+      <option name="SDK_HOME" value="" />
+      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
+      <option name="IS_MODULE_SDK" value="true" />
+      <option name="ADD_CONTENT_ROOTS" value="true" />
+      <option name="ADD_SOURCE_ROOTS" value="true" />
+      <option name="DEBUG_JUST_MY_CODE" value="false" />
+      <option name="RUN_TOOL" value="" />
+      <option name="_new_additionalArguments" value="&quot;&quot;" />
+      <option name="_new_target" value="&quot;tests.test_sql_graph_persistent.TestSQLGraphPersistent&quot;" />
+      <option name="_new_targetType" value="&quot;PYTHON&quot;" />
+      <method v="2" />
+    </configuration>
+    <configuration name="Python tests in test_load_graph_sql.py" type="tests" factoryName="Autodetect" temporary="true" nameIsGenerated="true">
+      <module name="dbgraph" />
+      <option name="ENV_FILES" value="" />
+      <option name="INTERPRETER_OPTIONS" value="" />
+      <option name="PARENT_ENVS" value="true" />
+      <option name="SDK_HOME" value="" />
+      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
+      <option name="IS_MODULE_SDK" value="true" />
+      <option name="ADD_CONTENT_ROOTS" value="true" />
+      <option name="ADD_SOURCE_ROOTS" value="true" />
+      <option name="DEBUG_JUST_MY_CODE" value="false" />
+      <option name="RUN_TOOL" value="" />
+      <option name="_new_additionalArguments" value="&quot;&quot;" />
+      <option name="_new_target" value="&quot;$PROJECT_DIR$/tests/test_load_graph_sql.py&quot;" />
+      <option name="_new_targetType" value="&quot;PATH&quot;" />
+      <method v="2" />
+    </configuration>
+    <configuration name="Python tests in tests" type="tests" factoryName="Autodetect" temporary="true" nameIsGenerated="true">
+      <module name="dbgraph" />
+      <option name="ENV_FILES" value="" />
+      <option name="INTERPRETER_OPTIONS" value="" />
+      <option name="PARENT_ENVS" value="true" />
+      <option name="SDK_HOME" value="" />
+      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/tests" />
+      <option name="IS_MODULE_SDK" value="true" />
+      <option name="ADD_CONTENT_ROOTS" value="true" />
+      <option name="ADD_SOURCE_ROOTS" value="true" />
+      <option name="DEBUG_JUST_MY_CODE" value="false" />
+      <option name="RUN_TOOL" value="" />
+      <option name="_new_additionalArguments" value="&quot;&quot;" />
+      <option name="_new_target" value="&quot;$PROJECT_DIR$/tests&quot;" />
+      <option name="_new_targetType" value="&quot;PATH&quot;" />
+      <method v="2" />
+    </configuration>
+    <recent_temporary>
+      <list>
+        <item itemvalue="Python tests.Python tests in tests" />
+        <item itemvalue="Python tests.Python tests for tests.test_load_graph_sql.TestLoadGraphSQL.test_load" />
+        <item itemvalue="Python tests.Python tests in test_load_graph_sql.py" />
+        <item itemvalue="Python tests.Python tests for tests.test_sql_graph_persistent.TestSQLGraphPersistent" />
+        <item itemvalue="Python tests.Python tests for tests.test_sql_graph_builder.TestSQLGraphBuilder.test_build_graph" />
+      </list>
+    </recent_temporary>
+  </component>
+  <component name="SharedIndexes">
+    <attachedChunks>
+      <set>
+        <option value="bundled-python-sdk-821f704e66d9-5bd6d13a2a75-com.jetbrains.pycharm.pro.sharedIndexes.bundled-PY-262.9437.214" />
+      </set>
+    </attachedChunks>
+  </component>
+  <component name="TaskManager">
+    <task active="true" id="Default" summary="Default task">
+      <changelist id="263a96da-9a82-40fd-89b2-ba71dd2c194d" name="Changes" comment="" />
+      <created>1788103453676</created>
+      <option name="number" value="Default" />
+      <option name="presentableId" value="Default" />
+      <updated>1788103453676</updated>
+    </task>
+    <task id="LOCAL-00001" summary="update packaging">
+      <option name="closed" value="true" />
+      <created>1788163967330</created>
+      <option name="number" value="00001" />
+      <option name="presentableId" value="LOCAL-00001" />
+      <option name="project" value="LOCAL" />
+      <updated>1788163967330</updated>
+    </task>
+    <task id="LOCAL-00002" summary="update packaging">
+      <option name="closed" value="true" />
+      <created>1788399865892</created>
+      <option name="number" value="00002" />
+      <option name="presentableId" value="LOCAL-00002" />
+      <option name="project" value="LOCAL" />
+      <updated>1788399865892</updated>
+    </task>
+    <task id="LOCAL-00003" summary="fix bugs in builder">
+      <option name="closed" value="true" />
+      <created>1788407330419</created>
+      <option name="number" value="00003" />
+      <option name="presentableId" value="LOCAL-00003" />
+      <option name="project" value="LOCAL" />
+      <updated>1788407330419</updated>
+    </task>
+    <option name="localTasksCounter" value="4" />
+    <servers />
+  </component>
+  <component name="TypeScriptGeneratedFilesManager">
+    <option name="version" value="3" />
+  </component>
+  <component name="Vcs.Log.Tabs.Properties">
+    <option name="TAB_STATES">
+      <map>
+        <entry key="MAIN">
+          <value>
+            <State />
+          </value>
+        </entry>
+      </map>
+    </option>
+  </component>
+  <component name="VcsManagerConfiguration">
+    <MESSAGE value="update packaging" />
+    <MESSAGE value="fix bugs in builder" />
+    <option name="LAST_COMMIT_MESSAGE" value="fix bugs in builder" />
+  </component>
+  <component name="XDebuggerManager">
+    <breakpoint-manager>
+      <breakpoints>
+        <line-breakpoint enabled="true" suspend="THREAD" type="python-line">
+          <url>file://$PROJECT_DIR$/dbgraph/persistent/sql_graph_persistent.py</url>
+          <line>330</line>
+          <option name="timeStamp" value="4" />
+        </line-breakpoint>
+        <line-breakpoint enabled="true" suspend="THREAD" type="python-line">
+          <url>file://$PROJECT_DIR$/dbgraph/persistent/sql_graph_persistent.py</url>
+          <line>351</line>
+          <option name="timeStamp" value="5" />
+        </line-breakpoint>
+      </breakpoints>
+    </breakpoint-manager>
+  </component>
+</project>

{dbgraph-0.2.0 → dbgraph-0.2.1}/PKG-INFO

@@ -1,9 +1,9 @@
 Metadata-Version: 2.5
 Name: dbgraph
-Version: 0.2.0
+Version: 0.2.1
 Summary: dbgraph is a database profiling tool which model the database as a graph
 Requires-Python: >=3.11
-Requires-Dist: bm25s[all]>=0.3.9
+Requires-Dist: bm25s>=0.3.9
 Requires-Dist: build>=1.6.0
 Requires-Dist: grip>=4.6.2
 Requires-Dist: langchain-openai>=1.6.0

{dbgraph-0.2.0 → dbgraph-0.2.1}/dbgraph/__init__.py

@@ -17,5 +17,5 @@ __all__ = [
     "SQLGraphBuilder",
     "SQLGraphPersistent",
     "LangchainGraphDescriptor",
-    "MarkdownRenderer"
+    "MarkdownRenderer",
 ]

{dbgraph-0.2.0 → dbgraph-0.2.1}/dbgraph/builder/sql_graph_builder.py

@@ -13,7 +13,22 @@ from sqlalchemy import (
     func,
     inspect,
     select,
+    SmallInteger,
+    Integer,
+    BigInteger,
+    Float,
+    DECIMAL,
+    Text,
+    Unicode,
+    UnicodeText,
+    CHAR,
+    NCHAR,
+    BLOB,
+    DATE,
+    DATETIME,
+    distinct,
 )
+from sqlalchemy.orm import Session
 
 from dbgraph.builder.graph_builder import GraphBuilder
 from dbgraph.entity.aspect import (
@@ -24,6 +39,7 @@ from dbgraph.entity.aspect import (
     RNumericalStatistics,
     RTableSchemaAspect,
     RTableStatisticsAspect,
+    RTemporalStatistics,
 )
 from dbgraph.entity.asset import Asset
 from dbgraph.entity.asset_type import AssetType
@@ -97,21 +113,40 @@ class SQLGraphBuilder(GraphBuilder):
         value_counts = {str(row[0]): int(row[1]) for row in rows}
         return RCategoricalStatistics(value_counts)
 
+    def _get_temp_aspect(self, col: Column) -> RTemporalStatistics:
+        min_max_nunique_stmt = select(
+            func.min(col), func.max(col), func.count(distinct(col))
+        )
+        mode_stmt = select(col).group_by(col).order_by(func.count(col).desc()).limit(1)
+        with Session(self.engine) as session:
+            min_time, max_time, nunique = session.execute(min_max_nunique_stmt).one()
+            mode = session.execute(mode_stmt).scalar()
+        return RTemporalStatistics(min_time, max_time, mode, nunique)
+
     def _get_stats_aspect(self, col: Column) -> RColumnStatisticsAspect:
         non_null_count, null_count = self._get_null_info(col)
-        if isinstance(col.type, Numeric):
+        num_aspect = None
+        cat_aspect = None
+        temp_aspect = None
+        if isinstance(
+            col.type, (Numeric, Integer, SmallInteger, BigInteger, Float, DECIMAL)
+        ):
             num_aspect = self._get_numerical_stats_aspect(col)
-            cat_aspect = None
-        elif isinstance(col.type, String):
-            num_aspect = None
+        elif isinstance(col.type, (String, Text, Unicode, UnicodeText, CHAR, NCHAR)):
             cat_aspect = self._get_cat_stats_aspect(col)
+        elif isinstance(col.type, (DATE, DATETIME)):
+            temp_aspect = self._get_temp_aspect(col)
+        elif isinstance(col.type, (BLOB,)):
+            pass
         else:
-            num_aspect = None
-            cat_aspect = None
+            raise NotImplementedError(
+                f"Only support Numeric, String, Date, and BLOB. Got {col.type} for {col.name}"
+            )
         return RColumnStatisticsAspect(
             name=f"{col.name}_column_stats",
             numerical_stats=num_aspect,
             categorical_stats=cat_aspect,
+            temporal_stats=temp_aspect,
             non_null_count=non_null_count,
             null_count=null_count,
         )

{dbgraph-0.2.0 → dbgraph-0.2.1}/dbgraph/entity/aspect.py

@@ -1,4 +1,5 @@
 from dataclasses import dataclass
+from datetime import datetime, timedelta
 from enum import StrEnum
 
 
@@ -103,6 +104,16 @@ class RCategoricalStatistics:
     value_counts: dict[str, int]
 
 
+@dataclass
+class RTemporalStatistics:
+    """Statistics for temporal columns"""
+
+    min_time: datetime | None = None
+    max_time: datetime | None = None
+    mode_time: datetime | None = None
+    num_uniques: int | None = None
+
+
 @dataclass
 class RColumnStatisticsAspect(Aspect):
     """Group of properties relevant to RDB's column statistics
@@ -118,6 +129,7 @@ class RColumnStatisticsAspect(Aspect):
     null_count: int
     numerical_stats: RNumericalStatistics | None = None
     categorical_stats: RCategoricalStatistics | None = None
+    temporal_stats: RTemporalStatistics | None = None
 
 
 class FKBehavior(StrEnum):

dbgraph-0.2.1/dbgraph/persistent/json_utils.py

@@ -0,0 +1,34 @@
+import datetime
+import json
+from typing import Any, Callable
+
+
+class DateTimeEncoder(json.JSONEncoder):
+    """Custom encoder for datetime"""
+
+    def default(self, o: Any) -> Any:
+        if isinstance(o, (datetime.date, datetime.datetime)):
+            return o.isoformat()
+        return super().default(o)
+
+
+def datetime_parser(obj):
+    for key, value in obj.items():
+        if isinstance(value, str):
+            try:
+                # try parsing ISO format
+                obj[key] = datetime.datetime.fromisoformat(value)
+            except ValueError:
+                try:
+                    # if there is only date
+                    obj[key] = datetime.date.fromisoformat(value)
+                except ValueError:
+                    pass
+    return obj
+
+
+class DateTimeDecoder(json.JSONDecoder):
+    """Custom decoder for datetime"""
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(object_hook=datetime_parser, *args, **kwargs)

{dbgraph-0.2.0 → dbgraph-0.2.1}/dbgraph/persistent/models.py

@@ -181,8 +181,8 @@ class AssetAspectModel(Base):
         nullable=False,
     )
 
-    json_aspect: Mapped[dict] = mapped_column(
-        JSON,
+    json_aspect: Mapped[str] = mapped_column(
+        Text,
         nullable=False,
     )
 
@@ -212,8 +212,8 @@ class LinkAspectModel(Base):
         nullable=False,
     )
 
-    json_aspect: Mapped[dict] = mapped_column(
-        JSON,
+    json_aspect: Mapped[str] = mapped_column(
+        Text,
         nullable=False,
     )
 

{dbgraph-0.2.0 → dbgraph-0.2.1}/dbgraph/persistent/sql_graph_persistent.py

@@ -1,3 +1,4 @@
+import json
 from dataclasses import asdict, dataclass
 
 from sqlalchemy import create_engine, delete, select, update
@@ -13,6 +14,7 @@ from dbgraph.entity.aspect import (
     RTableSchemaAspect,
     RTableStatisticsAspect,
     SemanticAspect,
+    RTemporalStatistics,
 )
 from dbgraph.entity.asset import Asset
 from dbgraph.entity.asset_type import AssetType
@@ -20,6 +22,7 @@ from dbgraph.entity.dbgraph import DatabaseGraph
 from dbgraph.entity.link import Link
 from dbgraph.entity.link_type import LinkType
 from dbgraph.persistent.graph_persistent import GraphPersistent
+from dbgraph.persistent.json_utils import DateTimeEncoder, DateTimeDecoder
 from dbgraph.persistent.models import (
     AssetAspectModel,
     AssetModel,
@@ -69,7 +72,7 @@ class SQLGraphPersistent(GraphPersistent):
                     asset_id=asset.asset_id,
                     name=aspect.name,
                     type=aspect_type,
-                    json_aspect=asdict(aspect),
+                    json_aspect=json.dumps(asdict(aspect), cls=DateTimeEncoder),
                 )
                 aspects.append(aspect)
         with Session(self.engine) as session:
@@ -110,7 +113,7 @@ class SQLGraphPersistent(GraphPersistent):
                     link_id=link.link_id,
                     name=aspect.name,
                     type=aspect_type,
-                    json_aspect=asdict(aspect),
+                    json_aspect=json.dumps(asdict(aspect), cls=DateTimeEncoder),
                 )
                 aspects.append(link_aspect)
         with Session(self.engine) as session:
@@ -201,7 +204,7 @@ class SQLGraphPersistent(GraphPersistent):
                 name=aspect.name,
                 type=aspect_type,
                 asset_id=asset_id,
-                json_aspect=asdict(aspect),
+                json_aspect=json.dumps(asdict(aspect), cls=DateTimeEncoder),
             )
             for aspect_type, aspect in aspects.items()
         ]
@@ -216,7 +219,7 @@ class SQLGraphPersistent(GraphPersistent):
                 name=aspect.name,
                 type=aspect_type,
                 link_id=link_id,
-                json_aspect=asdict(aspect),
+                json_aspect=json.dumps(asdict(aspect), cls=DateTimeEncoder),
             )
             for aspect_type, aspect in aspects.items()
         ]
@@ -228,7 +231,7 @@ class SQLGraphPersistent(GraphPersistent):
         self, model: AssetAspectModel, asset_type: AssetType
     ) -> Aspect:
         """Turn SQL Alchemy model to Aspect"""
-        json_aspect = model.json_aspect
+        json_aspect = json.loads(model.json_aspect, cls=DateTimeDecoder)
         match model.type:
             case "schema_properties":
                 if asset_type == AssetType.RTABLE:
@@ -241,6 +244,7 @@ class SQLGraphPersistent(GraphPersistent):
                 if asset_type == AssetType.RTABLE:
                     aspect = RTableStatisticsAspect(**json_aspect)
                 elif asset_type == AssetType.RCOLUMN:
+                    numerical_aspect = categorical_aspect = temp_aspect = None
                     if json_aspect["numerical_stats"] is not None:
                         numerical_aspect = RNumericalStatistics(
                             **json_aspect["numerical_stats"]
@@ -251,9 +255,9 @@ class SQLGraphPersistent(GraphPersistent):
                             **json_aspect["categorical_stats"]
                         )
                         numerical_aspect = None
-                    else:
-                        raise RuntimeError(
-                            "Neither categorical or numerical stats exists"
+                    elif json_aspect["temporal_stats"] is not None:
+                        temp_aspect = RTemporalStatistics(
+                            **json_aspect["temporal_stats"]
                         )
                     aspect = RColumnStatisticsAspect(
                         name=json_aspect["name"],
@@ -261,6 +265,7 @@ class SQLGraphPersistent(GraphPersistent):
                         null_count=json_aspect["null_count"],
                         categorical_stats=categorical_aspect,
                         numerical_stats=numerical_aspect,
+                        temporal_stats=temp_aspect,
                     )
                 else:
                     raise NotImplementedError()
@@ -289,7 +294,7 @@ class SQLGraphPersistent(GraphPersistent):
 
     def _parse_link_aspect_model(self, model: LinkAspectModel) -> Aspect:
         """Turn SQL Alchemy model to Link Aspect"""
-        json_aspect = model.json_aspect
+        json_aspect = json.loads(model.json_aspect, cls=DateTimeDecoder)
         match model.type:
             case "foreign_key_properties":
                 aspect = RForeignKeyAspect(**json_aspect)
@@ -315,7 +320,7 @@ class SQLGraphPersistent(GraphPersistent):
         with Session(self.engine) as session:
             models = session.scalars(
                 select(AssetModel).where(AssetModel.graph_id == graph_id)
-            )
+            ).all()
             assets = []
             for model in models:
                 aspects = {}
@@ -336,7 +341,7 @@ class SQLGraphPersistent(GraphPersistent):
         with Session(self.engine) as session:
             models = session.scalars(
                 select(LinkModel).where(LinkModel.graph_id == graph_id)
-            )
+            ).all()
             links = []
             for model in models:
                 aspects = {}

{dbgraph-0.2.0 → dbgraph-0.2.1}/pyproject.toml

@@ -4,12 +4,12 @@ build-backend = "hatchling.build"
 
 [project]
 name = "dbgraph"
-version = "0.2.0"
+version = "0.2.1"
 description = "dbgraph is a database profiling tool which model the database as a graph"
 readme = "README.md"
 requires-python = ">=3.11"
 dependencies = [
-    "bm25s[all]>=0.3.9",
+    "bm25s>=0.3.9",
     "build>=1.6.0",
     "grip>=4.6.2",
     "langchain>=1.3.16",

dbgraph-0.2.1/tests/test_load_graph_sql.py

@@ -0,0 +1,21 @@
+import unittest
+
+from dbgraph import SQLGraphPersistent, SQLGraphBuilder
+
+
+class TestLoadGraphSQL(unittest.TestCase):
+    def setUp(self) -> None:
+        self.graph_builder = SQLGraphBuilder("sqlite:///data/northwind.db")
+        self.persistent = SQLGraphPersistent("sqlite:///:memory:")
+
+    def test_load(self):
+        target_graph_id = "test-graph"
+        target_graph = self.graph_builder.build_graph()
+        self.persistent.save_graph(target_graph, target_graph_id)
+        graph = self.persistent.load_graph(target_graph_id)
+        self.assertEqual(len(target_graph.assets), len(graph.assets))
+        self.assertEqual(len(target_graph.links), len(graph.links))
+
+
+if __name__ == "__main__":
+    unittest.main()

{dbgraph-0.2.0 → dbgraph-0.2.1}/tests/test_sql_graph_builder.py

@@ -14,7 +14,7 @@ def get_kv_memory():
 
 class TestSQLGraphBuilder(unittest.TestCase):
     def setUp(self) -> None:
-        self.graph_builder = SQLGraphBuilder("northwind", "sqlite:///data/northwind.db")
+        self.graph_builder = SQLGraphBuilder("sqlite:///data/northwind.db")
 
     def test_build_graph(self):
         from time import perf_counter_ns

{dbgraph-0.2.0 → dbgraph-0.2.1}/tests/test_sql_graph_persistent.py

@@ -22,7 +22,7 @@ from dbgraph.persistent.sql_graph_persistent import SQLGraphPersistent
 
 class TestSQLGraphPersistent(unittest.TestCase):
     def setUp(self) -> None:
-        self.persistent = SQLGraphPersistent("sqlite:///data/graph.db")
+        self.persistent = SQLGraphPersistent("sqlite:///:memory:")
         assets = [
             Asset(
                 asset_id="1",
@@ -298,10 +298,6 @@ class TestSQLGraphPersistent(unittest.TestCase):
         self.graph_name = "graph-1"
         self.persistent.create_graph(self.graph_id, self.graph_name)
 
-    def tearDown(self) -> None:
-        self.persistent.delete_graph(self.graph_id)
-        os.remove("data/graph.db")
-
     def test_asset(self):
         target_asset = self.graph.get_asset("1")
         self.persistent.insert_asset(target_asset, self.graph_id)
@@ -440,6 +436,11 @@ class TestSQLGraphPersistent(unittest.TestCase):
         graph = self.persistent.load_graph(self.graph_id)
         self.assertEqual(graph, self.graph)
 
+    def test_save_graph(self):
+        self.persistent.save_graph(self.graph, self.graph_id)
+        graph = self.persistent.load_graph(self.graph_id)
+        self.assertEqual(self.graph, graph)
+
     def test_get_graph_info(self):
         graph_info = self.persistent.get_graph_info("1")
         self.assertEqual(graph_info[1], "graph-1")

{dbgraph-0.2.0 → dbgraph-0.2.1}/uv.lock

@@ -284,7 +284,7 @@ wheels = [
 
 [[package]]
 name = "dbgraph"
-version = "0.1.0"
+version = "0.2.1"
 source = { editable = "." }
 dependencies = [
     { name = "bm25s" },
@@ -312,7 +312,7 @@ dependencies = [
 
 [package.metadata]
 requires-dist = [
-    { name = "bm25s", extras = ["all"], specifier = ">=0.3.9" },
+    { name = "bm25s", specifier = ">=0.3.9" },
     { name = "build", specifier = ">=1.6.0" },
     { name = "grip", specifier = ">=4.6.2" },
     { name = "langchain", specifier = ">=1.3.16" },

dbgraph-0.2.0/.idea/workspace.xml

@@ -1,168 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-<project version="4">
-  <component name="AutoImportSettings">
-    <option name="autoReloadType" value="SELECTIVE" />
-  </component>
-  <component name="ChangeListManager">
-    <list default="true" id="263a96da-9a82-40fd-89b2-ba71dd2c194d" name="Changes" comment="update packaging">
-      <change beforePath="$PROJECT_DIR$/pyproject.toml" beforeDir="false" afterPath="$PROJECT_DIR$/pyproject.toml" afterDir="false" />
-      <change beforePath="$PROJECT_DIR$/uv.lock" beforeDir="false" afterPath="$PROJECT_DIR$/uv.lock" afterDir="false" />
-    </list>
-    <option name="SHOW_DIALOG" value="false" />
-    <option name="HIGHLIGHT_CONFLICTS" value="true" />
-    <option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
-    <option name="LAST_RESOLUTION" value="IGNORE" />
-  </component>
-  <component name="FileTemplateManagerImpl">
-    <option name="RECENT_TEMPLATES">
-      <list>
-        <option value="Python Script" />
-      </list>
-    </option>
-  </component>
-  <component name="Git.Settings">
-    <option name="RECENT_BRANCH_BY_REPOSITORY">
-      <map>
-        <entry key="$PROJECT_DIR$" value="packaging" />
-      </map>
-    </option>
-    <option name="RECENT_GIT_ROOT_PATH" value="$PROJECT_DIR$" />
-  </component>
-  <component name="GitHubPullRequestSearchHistory">{
-  &quot;lastFilter&quot;: {
-    &quot;state&quot;: &quot;OPEN&quot;,
-    &quot;assignee&quot;: &quot;minhdenthedev&quot;
-  }
-}</component>
-  <component name="GithubPullRequestsUISettings">{
-  &quot;selectedUrlAndAccountId&quot;: {
-    &quot;url&quot;: &quot;https://github.com/minhdenthedev/dbgraph.git&quot;,
-    &quot;accountId&quot;: &quot;6c0b7cdc-aa0d-4999-86e3-e6f0a1d74a10&quot;
-  },
-  &quot;recentNewPullRequestHead&quot;: {
-    &quot;server&quot;: {
-      &quot;useHttp&quot;: false,
-      &quot;host&quot;: &quot;github.com&quot;,
-      &quot;port&quot;: null,
-      &quot;suffix&quot;: null
-    },
-    &quot;owner&quot;: &quot;minhdenthedev&quot;,
-    &quot;repository&quot;: &quot;dbgraph&quot;
-  }
-}</component>
-  <component name="ProjectColorInfo">{
-  &quot;associatedIndex&quot;: 0,
-  &quot;fromUser&quot;: false
-}</component>
-  <component name="ProjectId" id="3IdnkFbc3mGdbjiFJYWIoYOgRXM" />
-  <component name="ProjectLevelVcsManager">
-    <ConfirmationsSetting value="2" id="Add" />
-  </component>
-  <component name="ProjectViewState">
-    <option name="hideEmptyMiddlePackages" value="true" />
-    <option name="showLibraryContents" value="true" />
-  </component>
-  <component name="PropertiesComponent">{
-  &quot;keyToString&quot;: {
-    &quot;ASKED_SHARE_PROJECT_CONFIGURATION_FILES&quot;: &quot;true&quot;,
-    &quot;ModuleVcsDetector.initialDetectionPerformed&quot;: &quot;true&quot;,
-    &quot;Python tests.Python tests for tests.test_dbgraph.TestDBGraph.executor&quot;: &quot;Run&quot;,
-    &quot;Python tests.Python tests in tests.executor&quot;: &quot;Run&quot;,
-    &quot;RunOnceActivity.ShowReadmeOnStart&quot;: &quot;true&quot;,
-    &quot;RunOnceActivity.git.unshallow&quot;: &quot;true&quot;,
-    &quot;RunOnceActivity.typescript.service.memoryLimit.init&quot;: &quot;true&quot;,
-    &quot;codeWithMe.voiceChat.enabledByDefault&quot;: &quot;false&quot;,
-    &quot;git-widget-placeholder&quot;: &quot;main&quot;,
-    &quot;ignore.virus.scanning.warn.message&quot;: &quot;true&quot;,
-    &quot;nodejs_package_manager_path&quot;: &quot;npm&quot;
-  }
-}</component>
-  <component name="RunManager" selected="Python tests.Python tests for tests.test_dbgraph.TestDBGraph">
-    <configuration name="Python tests for tests.test_dbgraph.TestDBGraph" type="tests" factoryName="Autodetect" temporary="true" nameIsGenerated="true">
-      <module name="dbgraph" />
-      <option name="ENV_FILES" value="" />
-      <option name="INTERPRETER_OPTIONS" value="" />
-      <option name="PARENT_ENVS" value="true" />
-      <option name="SDK_HOME" value="" />
-      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
-      <option name="IS_MODULE_SDK" value="true" />
-      <option name="ADD_CONTENT_ROOTS" value="true" />
-      <option name="ADD_SOURCE_ROOTS" value="true" />
-      <option name="DEBUG_JUST_MY_CODE" value="false" />
-      <option name="RUN_TOOL" value="" />
-      <option name="_new_additionalArguments" value="&quot;&quot;" />
-      <option name="_new_target" value="&quot;tests.test_dbgraph.TestDBGraph&quot;" />
-      <option name="_new_targetType" value="&quot;PYTHON&quot;" />
-      <method v="2" />
-    </configuration>
-    <configuration name="Python tests in tests" type="tests" factoryName="Autodetect" temporary="true" nameIsGenerated="true">
-      <module name="dbgraph" />
-      <option name="ENV_FILES" value="" />
-      <option name="INTERPRETER_OPTIONS" value="" />
-      <option name="PARENT_ENVS" value="true" />
-      <option name="SDK_HOME" value="" />
-      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/tests" />
-      <option name="IS_MODULE_SDK" value="true" />
-      <option name="ADD_CONTENT_ROOTS" value="true" />
-      <option name="ADD_SOURCE_ROOTS" value="true" />
-      <option name="DEBUG_JUST_MY_CODE" value="false" />
-      <option name="RUN_TOOL" value="" />
-      <option name="_new_additionalArguments" value="&quot;&quot;" />
-      <option name="_new_target" value="&quot;$PROJECT_DIR$/tests&quot;" />
-      <option name="_new_targetType" value="&quot;PATH&quot;" />
-      <method v="2" />
-    </configuration>
-    <recent_temporary>
-      <list>
-        <item itemvalue="Python tests.Python tests in tests" />
-        <item itemvalue="Python tests.Python tests for tests.test_dbgraph.TestDBGraph" />
-        <item itemvalue="Python tests.Python tests for tests.test_dbgraph.TestDBGraph" />
-        <item itemvalue="Python tests.Python tests in tests" />
-      </list>
-    </recent_temporary>
-  </component>
-  <component name="SharedIndexes">
-    <attachedChunks>
-      <set>
-        <option value="bundled-python-sdk-821f704e66d9-5bd6d13a2a75-com.jetbrains.pycharm.pro.sharedIndexes.bundled-PY-262.9437.214" />
-      </set>
-    </attachedChunks>
-  </component>
-  <component name="TaskManager">
-    <task active="true" id="Default" summary="Default task">
-      <changelist id="263a96da-9a82-40fd-89b2-ba71dd2c194d" name="Changes" comment="" />
-      <created>1788103453676</created>
-      <option name="number" value="Default" />
-      <option name="presentableId" value="Default" />
-      <updated>1788103453676</updated>
-    </task>
-    <task id="LOCAL-00001" summary="update packaging">
-      <option name="closed" value="true" />
-      <created>1788163967330</created>
-      <option name="number" value="00001" />
-      <option name="presentableId" value="LOCAL-00001" />
-      <option name="project" value="LOCAL" />
-      <updated>1788163967330</updated>
-    </task>
-    <option name="localTasksCounter" value="2" />
-    <servers />
-  </component>
-  <component name="TypeScriptGeneratedFilesManager">
-    <option name="version" value="3" />
-  </component>
-  <component name="Vcs.Log.Tabs.Properties">
-    <option name="TAB_STATES">
-      <map>
-        <entry key="MAIN">
-          <value>
-            <State />
-          </value>
-        </entry>
-      </map>
-    </option>
-  </component>
-  <component name="VcsManagerConfiguration">
-    <MESSAGE value="update packaging" />
-    <option name="LAST_COMMIT_MESSAGE" value="update packaging" />
-  </component>
-</project>