dbdocs 0.0.0__tar.gz → 1.0.0__tar.gz

{dbdocs-0.0.0 → dbdocs-1.0.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dbdocs
-Version: 0.0.0
+Version: 1.0.0
 Summary: Alternative dbt docs site: Catalog + ERD + column-level lineage
 Project-URL: Homepage, https://github.com/datnguye/dbt-docs
 Project-URL: Repository, https://github.com/datnguye/dbt-docs
@@ -27,7 +27,7 @@ Description-Content-Type: text/markdown
 <p align="center"><b>An alternative dbt docs site — catalog + ERD + column-level lineage, baked into one file.</b></p>
 
 <p align="center">
-  <a href="https://dbdocs.datnguye.me/demo/"><img src="https://img.shields.io/badge/live-demo-FF694A?style=flat&logo=rocket&logoColor=white" alt="live demo"></a>
+  <a href="https://dbdocs.datnguye.me/latest/demo/"><img src="https://img.shields.io/badge/live-demo-FF694A?style=flat&logo=rocket&logoColor=white" alt="live demo"></a>
   <a href="https://dbdocs.datnguye.me/"><img src="https://img.shields.io/badge/docs-visit%20site-blue?style=flat&logo=gitbook&logoColor=white" alt="docs"></a>
   <a href="https://pypi.org/project/dbdocs/"><img src="https://badge.fury.io/py/dbdocs.svg" alt="PyPI version"></a>
   <img src="https://img.shields.io/badge/CLI-Python-FFCE3E?labelColor=14354C&logo=python&logoColor=white" alt="python-cli">

{dbdocs-0.0.0 → dbdocs-1.0.0}/README.md

@@ -5,7 +5,7 @@
 <p align="center"><b>An alternative dbt docs site — catalog + ERD + column-level lineage, baked into one file.</b></p>
 
 <p align="center">
-  <a href="https://dbdocs.datnguye.me/demo/"><img src="https://img.shields.io/badge/live-demo-FF694A?style=flat&logo=rocket&logoColor=white" alt="live demo"></a>
+  <a href="https://dbdocs.datnguye.me/latest/demo/"><img src="https://img.shields.io/badge/live-demo-FF694A?style=flat&logo=rocket&logoColor=white" alt="live demo"></a>
   <a href="https://dbdocs.datnguye.me/"><img src="https://img.shields.io/badge/docs-visit%20site-blue?style=flat&logo=gitbook&logoColor=white" alt="docs"></a>
   <a href="https://pypi.org/project/dbdocs/"><img src="https://badge.fury.io/py/dbdocs.svg" alt="PyPI version"></a>
   <img src="https://img.shields.io/badge/CLI-Python-FFCE3E?labelColor=14354C&logo=python&logoColor=white" alt="python-cli">

{dbdocs-0.0.0 → dbdocs-1.0.0}/dbdocs/extract/_sqlglot_lineage.py

@@ -44,19 +44,19 @@ class Node:
             yield from d.walk()
 
 
-def lineage(
-    column: str | exp.Column,
+def prepare_scope(
     sql: str | exp.Expression,
     schema: dict | Schema | None = None,
     sources: t.Mapping[str, str | exp.Query] | None = None,
     dialect: DialectType = None,
-    scope: Scope | None = None,
-    trim_selects: bool = True,
     **kwargs,
-) -> Node:
-    """Build the lineage graph for a column of a SQL query."""
+) -> Scope:
+    """Parse, qualify, and build the optimizer scope for a SQL query once.
+
+    The expensive half of :func:`lineage`. Callers tracing many columns of one
+    query build the scope here and reuse it across :func:`lineage` calls.
+    """
     expression = maybe_parse(sql, dialect=dialect)
-    column = normalize_identifiers.normalize_identifiers(column, dialect=dialect).name
 
     if sources:
         expression = exp.expand(
@@ -65,22 +65,42 @@ def lineage(
             dialect=dialect,
         )
 
+    expression = qualify.qualify(
+        expression,
+        dialect=dialect,
+        schema=schema,
+        **{
+            "validate_qualify_columns": False,
+            "identify": False,
+            "allow_partial_qualification": True,
+            **kwargs,
+        },
+    )
+    scope = build_scope(expression)
     if not scope:
-        expression = qualify.qualify(
-            expression,
-            dialect=dialect,
-            schema=schema,
-            **{
-                "validate_qualify_columns": False,
-                "identify": False,
-                "allow_partial_qualification": True,
-                **kwargs,
-            },
-        )
-        scope = build_scope(expression)
+        raise SqlglotError("Cannot build lineage, sql must be SELECT")
+    return scope
+
+
+def lineage(
+    column: str | exp.Column,
+    sql: str | exp.Expression,
+    schema: dict | Schema | None = None,
+    sources: t.Mapping[str, str | exp.Query] | None = None,
+    dialect: DialectType = None,
+    scope: Scope | None = None,
+    trim_selects: bool = True,
+    **kwargs,
+) -> Node:
+    """Build the lineage graph for a column of a SQL query.
+
+    Pass a prebuilt ``scope`` (from :func:`prepare_scope`) to skip the per-call
+    parse/qualify/build_scope when tracing many columns of the same query.
+    """
+    column = normalize_identifiers.normalize_identifiers(column, dialect=dialect).name
 
     if not scope:
-        raise SqlglotError("Cannot build lineage, sql must be SELECT")
+        scope = prepare_scope(sql, schema=schema, sources=sources, dialect=dialect, **kwargs)
 
     select_names_original = {select.alias_or_name for select in scope.expression.selects}
     select_names_lower = {name.lower(): name for name in select_names_original}

{dbdocs-0.0.0 → dbdocs-1.0.0}/dbdocs/extract/column_lineage.py

@@ -20,7 +20,7 @@ from sqlglot.errors import SqlglotError
 from dbdocs.core.artifacts import db_schema, node_name
 from dbdocs.core.exceptions import LineageError
 from dbdocs.core.log import logger
-from dbdocs.extract._sqlglot_lineage import Node, lineage
+from dbdocs.extract._sqlglot_lineage import Node, lineage, prepare_scope
 
 #: dbt adapter_type → sqlglot dialect, when the names differ. Most match 1:1.
 _DIALECT_ALIASES = {
@@ -80,9 +80,14 @@ class ColumnLineageExtractor:
             output_columns = (
                 list(getattr(catalog_node, "columns", {}) or {}) if catalog_node else []
             )
+        if not output_columns:
+            return
+        # Build the scope once per model, then trace each column against it —
+        # qualify is the expensive part and is identical for every column.
+        scope = prepare_scope(compiled, schema=self.schema, dialect=self.dialect)
         for column in output_columns:
             try:
-                root = lineage(column, compiled, schema=self.schema, dialect=self.dialect)
+                root = lineage(column, compiled, dialect=self.dialect, scope=scope)
             except SqlglotError:
                 # One unresolvable column shouldn't drop the rest of the model.
                 continue

{dbdocs-0.0.0 → dbdocs-1.0.0}/dbdocs/site/builder.py

@@ -115,8 +115,10 @@ class ReportBuilder:
         data = self.build_data()
         index = out / "index.html"
         index.write_text(inject(index.read_text(encoding="utf-8"), data), encoding="utf-8")
+        # Compact, not indented — keeps the debug dump cheap on large projects.
         (out / "dbdocs-data.json").write_text(
-            json.dumps(data, indent=2, default=self._json_default), encoding="utf-8"
+            json.dumps(data, separators=(",", ":"), default=self._json_default),
+            encoding="utf-8",
         )
 
         logger.info(

{dbdocs-0.0.0 → dbdocs-1.0.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "dbdocs"
-version = "0.0.0"
+dynamic = ["version"]
 description = "Alternative dbt docs site: Catalog + ERD + column-level lineage"
 readme = "README.md"
 license = { text = "MIT" }
@@ -29,9 +29,12 @@ Repository = "https://github.com/datnguye/dbt-docs"
 dbdocs = "dbdocs.main:main"
 
 [build-system]
-requires = ["hatchling"]
+requires = ["hatchling", "hatch-vcs"]
 build-backend = "hatchling.build"
 
+[tool.hatch.version]
+source = "vcs"
+
 [tool.hatch.build.targets.wheel]
 packages = ["dbdocs"]
 artifacts = [