Expand governed Dataflow editor and node library

This commit is contained in:
2026-07-28 11:14:01 +02:00
parent df468a2bd8
commit dee8380631
22 changed files with 3563 additions and 290 deletions
@@ -0,0 +1,305 @@
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Any, Literal
NodeCategory = Literal["load", "combine", "filter", "transform", "output"]
SqlSupport = Literal["full", "partial", "none"]
@dataclass(frozen=True, slots=True)
class NodePortDefinition:
id: str
label: str
required: bool = True
multiple: bool = False
minimum_connections: int = 1
@dataclass(frozen=True, slots=True)
class NodeConfigField:
id: str
label: str
kind: str
required: bool = False
description: str | None = None
options: tuple[tuple[str, str], ...] = ()
@dataclass(frozen=True, slots=True)
class NodeTypeDefinition:
type: str
category: NodeCategory
label: str
description: str
icon: str
input_ports: tuple[NodePortDefinition, ...] = ()
output_ports: tuple[NodePortDefinition, ...] = (
NodePortDefinition(id="output", label="Output"),
)
config_fields: tuple[NodeConfigField, ...] = ()
default_config: dict[str, Any] = field(default_factory=dict)
sql_support: SqlSupport = "full"
NODE_LIBRARY = (
NodeTypeDefinition(
type="source.inline",
category="load",
label="Inline data",
description="Enter a small JSON table directly in the pipeline.",
icon="braces",
config_fields=(
NodeConfigField(id="source_name", label="SQL source name", kind="text", required=True),
NodeConfigField(id="rows", label="Rows", kind="json_rows", required=True),
),
default_config={"source_name": "inline_source", "rows": []},
),
NodeTypeDefinition(
type="source.reference",
category="load",
label="Connector source",
description="Load a bounded, fingerprinted table exposed by Connectors.",
icon="database",
config_fields=(
NodeConfigField(id="source_ref", label="Source", kind="tabular_source", required=True),
NodeConfigField(id="source_name", label="SQL source name", kind="text", required=True),
NodeConfigField(id="expected_fingerprint", label="Expected fingerprint", kind="readonly"),
),
default_config={
"source_ref": "",
"source_name": "connector_source",
"expected_fingerprint": "",
},
),
NodeTypeDefinition(
type="combine.union",
category="combine",
label="Append rows",
description="Append two or more inputs by column name.",
icon="combine",
input_ports=(
NodePortDefinition(
id="input",
label="Inputs",
multiple=True,
minimum_connections=2,
),
),
config_fields=(
NodeConfigField(
id="mode",
label="Duplicates",
kind="select",
options=(("all", "Keep all"), ("distinct", "Remove duplicates")),
),
),
default_config={"mode": "all"},
sql_support="partial",
),
NodeTypeDefinition(
type="combine.join",
category="combine",
label="Join tables",
description="Match two inputs using one or more key columns.",
icon="git-merge",
input_ports=(
NodePortDefinition(id="left", label="Left"),
NodePortDefinition(id="right", label="Right"),
),
config_fields=(
NodeConfigField(
id="join_type",
label="Join type",
kind="select",
options=(
("inner", "Matching rows"),
("left", "All left rows"),
("right", "All right rows"),
("full", "All rows"),
),
),
NodeConfigField(id="left_keys", label="Left keys", kind="column_list", required=True),
NodeConfigField(id="right_keys", label="Right keys", kind="column_list", required=True),
NodeConfigField(id="right_prefix", label="Right-column prefix", kind="text", required=True),
),
default_config={
"join_type": "inner",
"left_keys": [""],
"right_keys": [""],
"right_prefix": "right_",
},
sql_support="partial",
),
NodeTypeDefinition(
type="filter",
category="filter",
label="Filter rows",
description="Keep rows that satisfy a comparison.",
icon="filter",
input_ports=(NodePortDefinition(id="input", label="Input"),),
config_fields=(
NodeConfigField(id="column", label="Column", kind="column", required=True),
NodeConfigField(
id="operator",
label="Operator",
kind="select",
required=True,
options=(
("eq", "Equals"),
("ne", "Does not equal"),
("gt", "Greater than"),
("gte", "Greater than or equal"),
("lt", "Less than"),
("lte", "Less than or equal"),
("contains", "Contains"),
("is_null", "Is empty"),
("not_null", "Is not empty"),
),
),
NodeConfigField(id="value", label="Value", kind="value"),
),
default_config={"column": "", "operator": "eq", "value": ""},
),
NodeTypeDefinition(
type="distinct",
category="filter",
label="Remove duplicates",
description="Keep the first row for each selected key.",
icon="list-filter",
input_ports=(NodePortDefinition(id="input", label="Input"),),
config_fields=(
NodeConfigField(
id="columns",
label="Key columns",
kind="column_list",
description="Leave empty to compare complete rows.",
),
),
default_config={"columns": []},
sql_support="partial",
),
NodeTypeDefinition(
type="select",
category="transform",
label="Select columns",
description="Choose, order, and rename output columns.",
icon="columns-3",
input_ports=(NodePortDefinition(id="input", label="Input"),),
config_fields=(
NodeConfigField(id="fields", label="Fields", kind="field_mapping", required=True),
),
default_config={"fields": [{"column": "", "alias": ""}]},
),
NodeTypeDefinition(
type="derive",
category="transform",
label="Derive column",
description="Create a column through a constrained reusable operation.",
icon="variable",
input_ports=(NodePortDefinition(id="input", label="Input"),),
config_fields=(
NodeConfigField(id="target_column", label="Output column", kind="text", required=True),
NodeConfigField(
id="operation",
label="Operation",
kind="select",
required=True,
options=(
("copy", "Copy"),
("upper", "Uppercase"),
("lower", "Lowercase"),
("trim", "Trim whitespace"),
("concat", "Concatenate"),
("coalesce", "First non-empty"),
("add", "Add"),
("subtract", "Subtract"),
("multiply", "Multiply"),
("divide", "Divide"),
),
),
NodeConfigField(id="source_columns", label="Source columns", kind="column_list", required=True),
NodeConfigField(id="separator", label="Separator", kind="text"),
),
default_config={
"target_column": "",
"operation": "copy",
"source_columns": [""],
"separator": " ",
},
sql_support="partial",
),
NodeTypeDefinition(
type="aggregate",
category="transform",
label="Aggregate",
description="Group rows and calculate counts or numeric summaries.",
icon="sigma",
input_ports=(NodePortDefinition(id="input", label="Input"),),
config_fields=(
NodeConfigField(id="group_by", label="Group by", kind="column_list"),
NodeConfigField(id="aggregates", label="Calculations", kind="aggregates", required=True),
),
default_config={
"group_by": [],
"aggregates": [{"function": "count", "column": "*", "alias": "row_count"}],
},
),
NodeTypeDefinition(
type="sort",
category="transform",
label="Sort rows",
description="Order rows by one or more columns.",
icon="arrow-up-down",
input_ports=(NodePortDefinition(id="input", label="Input"),),
config_fields=(NodeConfigField(id="fields", label="Sort fields", kind="sort_fields", required=True),),
default_config={"fields": [{"column": "", "direction": "asc"}]},
),
NodeTypeDefinition(
type="limit",
category="transform",
label="Limit rows",
description="Keep only the first number of rows.",
icon="list-end",
input_ports=(NodePortDefinition(id="input", label="Input"),),
config_fields=(NodeConfigField(id="count", label="Rows", kind="number", required=True),),
default_config={"count": 100},
),
NodeTypeDefinition(
type="output",
category="output",
label="Preview output",
description="Expose the terminal table for preview or publication.",
icon="panel-top-open",
input_ports=(NodePortDefinition(id="input", label="Input"),),
output_ports=(),
default_config={},
),
)
NODE_TYPES = {definition.type: definition for definition in NODE_LIBRARY}
CATEGORY_LABELS = {
"load": "Load",
"combine": "Combine",
"filter": "Filter",
"transform": "Transform",
"output": "Output",
}
def node_definition(node_type: str) -> NodeTypeDefinition | None:
return NODE_TYPES.get(node_type)
__all__ = [
"CATEGORY_LABELS",
"NODE_LIBRARY",
"NODE_TYPES",
"NodeCategory",
"NodeConfigField",
"NodePortDefinition",
"NodeTypeDefinition",
"SqlSupport",
"node_definition",
]