refactor: modify type of field_specs

pygwalker/__init__.py

@@ -10,7 +10,7 @@
 from pygwalker.services.global_var import GlobalVarManager
 from pygwalker.services.kaggle import show_tips_user_kaggle as __show_tips_user_kaggle
 
-__version__ = "0.4.8a0"
+__version__ = "0.4.8a1"
 __hash__ = __rand_str()
 
 from pygwalker.api.jupyter import walk, render, table

pygwalker/api/gradio.py

@@ -1,4 +1,4 @@
-from typing import Union, Dict, Optional
+from typing import Union, List, Optional
 from typing_extensions import Literal
 
 from .pygwalker import PygWalker
@@ -18,7 +18,7 @@ def get_html_on_gradio(
     dataset: Union[DataFrame, Connector],
     gid: Union[int, str] = None,
     *,
-    field_specs: Optional[Dict[str, FieldSpec]] = None,
+    field_specs: Optional[List[FieldSpec]] = None,
     theme_key: Literal['vega', 'g2'] = 'g2',
     dark: Literal['media', 'light', 'dark'] = 'media',
     spec: str = "",
@@ -36,7 +36,7 @@ def get_html_on_gradio(
 
     Kargs:
         - env: (Literal['Jupyter' | 'Streamlit'], optional): The enviroment using pygwalker. Default as 'Jupyter'
-        - field_specs (Dict[str, FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
+        - field_specs (List[FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
         - theme_key ('vega' | 'g2'): theme type.
         - dark (Literal['media' | 'light' | 'dark']): 'media': auto detect OS theme.
         - spec (str): chart config data. config id, json, remote file url
@@ -50,7 +50,7 @@ def get_html_on_gradio(
     walker = PygWalker(
         gid=gid,
         dataset=dataset,
-        field_specs=field_specs if field_specs is not None else {},
+        field_specs=field_specs if field_specs is not None else [],
         spec=spec,
         source_invoke_code="",
         theme_key=theme_key,

pygwalker/api/html.py

@@ -1,4 +1,4 @@
-from typing import Union, Dict, Optional, Any
+from typing import Union, Dict, Optional, Any, List
 import logging
 
 from typing_extensions import Literal
@@ -20,7 +20,7 @@ def _to_html(
     gid: Union[int, str] = None,
     *,
     spec: str = "",
-    field_specs: Optional[Dict[str, FieldSpec]] = None,
+    field_specs: Optional[List[FieldSpec]] = None,
     theme_key: Literal['vega', 'g2'] = 'g2',
     dark: Literal['media', 'light', 'dark'] = 'media',
     default_tab: Literal["data", "vis"] = "vis",
@@ -31,14 +31,24 @@ def _to_html(
 ) -> str:
     """
     Generate embeddable HTML code of Graphic Walker with data of `df`.
+
+    Args:
+        - df (pl.DataFrame | pd.DataFrame, optional): dataframe.
+        - gid (Union[int, str], optional): GraphicWalker container div's id ('gwalker-{gid}')
+
+    Kargs:
+        - field_specs (List[FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
+        - spec (str): chart config data. config id, json, remote file url
+        - theme_key ('vega' | 'g2'): theme type.
+        - dark ('media' | 'light' | 'dark'): 'media': auto detect OS theme.
     """
     check_expired_params(kwargs)
 
     if gid is None:
         gid = generate_hash_code()
 
     if field_specs is None:
-        field_specs = {}
+        field_specs = []
 
     walker = PygWalker(
         gid=gid,
@@ -68,7 +78,7 @@ def to_html(
     gid: Union[int, str] = None,
     *,
     spec: str = "",
-    field_specs: Optional[Dict[str, FieldSpec]] = None,
+    field_specs: Optional[List[FieldSpec]] = None,
     theme_key: Literal['vega', 'g2'] = 'g2',
     dark: Literal['media', 'light', 'dark'] = 'media',
     default_tab: Literal["data", "vis"] = "vis",
@@ -121,7 +131,7 @@ def to_table_html(
         df,
         None,
         spec="",
-        field_specs={},
+        field_specs=[],
         theme_key=theme_key,
         dark=dark,
         gw_mode="table",
@@ -151,7 +161,7 @@ def to_render_html(
         df,
         None,
         spec=spec,
-        field_specs={},
+        field_specs=[],
         theme_key=theme_key,
         dark=dark,
         gw_mode="filter_renderer",

pygwalker/api/jupyter.py

@@ -1,4 +1,4 @@
-from typing import Union, Dict, Optional
+from typing import Union, List, Optional
 import inspect
 
 from typing_extensions import Literal
@@ -18,7 +18,7 @@ def walk(
     gid: Union[int, str] = None,
     *,
     env: Literal['Jupyter', 'JupyterWidget'] = 'JupyterWidget',
-    field_specs: Optional[Dict[str, FieldSpec]] = None,
+    field_specs: Optional[List[FieldSpec]] = None,
     theme_key: Literal['vega', 'g2'] = 'g2',
     dark: Literal['media', 'light', 'dark'] = 'media',
     spec: str = "",
@@ -37,7 +37,7 @@ def walk(
 
     Kargs:
         - env: (Literal['Jupyter' | 'JupyterWidget'], optional): The enviroment using pygwalker. Default as 'JupyterWidget'
-        - field_specs (Dict[str, FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
+        - field_specs (List[FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
         - theme_key ('vega' | 'g2'): theme type.
         - dark (Literal['media' | 'light' | 'dark']): 'media': auto detect OS theme.
         - spec (str): chart config data. config id, json, remote file url
@@ -49,7 +49,7 @@ def walk(
     check_expired_params(kwargs)
 
     if field_specs is None:
-        field_specs = {}
+        field_specs = []
 
     source_invoke_code = get_formated_spec_params_code_from_frame(
         inspect.stack()[1].frame
@@ -121,7 +121,7 @@ def render(
     walker = PygWalker(
         gid=None,
         dataset=dataset,
-        field_specs={},
+        field_specs=[],
         spec=spec,
         source_invoke_code="",
         theme_key=theme_key,
@@ -163,7 +163,7 @@ def table(
     walker = PygWalker(
         gid=None,
         dataset=dataset,
-        field_specs={},
+        field_specs=[],
         spec="",
         source_invoke_code="",
         theme_key=theme_key,

pygwalker/api/kanaries_cloud.py

@@ -1,4 +1,4 @@
-from typing import Dict, Optional, Union
+from typing import List, Optional, Union
 from datetime import datetime
 
 from pygwalker.data_parsers.base import FieldSpec
@@ -7,7 +7,6 @@
 from pygwalker.data_parsers.database_parser import Connector
 from pygwalker.services.cloud_service import CloudService
 from pygwalker.services.data_parsers import get_parser
-from pygwalker.services.global_var import GlobalVarManager
 
 
 def create_cloud_dataset(
@@ -44,7 +43,7 @@ def create_cloud_walker(
     *,
     chart_name: str,
     workspace_name: str,
-    field_specs: Optional[Dict[str, FieldSpec]] = None,
+    field_specs: Optional[List[FieldSpec]] = None,
     kanaries_api_key: str = ""
 ) -> str:
     """
@@ -63,7 +62,7 @@ def create_cloud_walker(
         str: pygwalker url in kanaries cloud
     """
     if field_specs is None:
-        field_specs = {}
+        field_specs = []
 
     cloud_service = CloudService(kanaries_api_key)
     data_parser = get_parser(dataset, False, field_specs)

pygwalker/api/pygwalker.py

@@ -1,4 +1,4 @@
-from typing import List, Dict, Any, Optional, Union
+from typing import List, Dict, Any, Optional, Union, Tuple
 import html as m_html
 import urllib
 import json
@@ -9,7 +9,7 @@
 import pandas as pd
 
 from pygwalker._typing import DataFrame
-from pygwalker.data_parsers.base import BaseDataParser
+from pygwalker.data_parsers.base import BaseDataParser, FieldSpec
 from pygwalker.data_parsers.database_parser import Connector
 from pygwalker.utils.display import display_html
 from pygwalker.utils.randoms import rand_str
@@ -52,7 +52,7 @@ def __init__(
         *,
         gid: Optional[Union[int, str]],
         dataset: Union[DataFrame, Connector, str],
-        field_specs: Dict[str, Any],
+        field_specs: List[FieldSpec],
         spec: str,
         source_invoke_code: str,
         theme_key: Literal['vega', 'g2'],
@@ -120,7 +120,7 @@ def _get_data_parser(
         self,
         *,
         dataset: Union[DataFrame, Connector, str],
-        field_specs: Dict[str, Any],
+        field_specs: List[FieldSpec],
         use_cloud_calc: bool,
         kanaries_api_key: str,
         cloud_service: CloudService
@@ -153,7 +153,7 @@ def _get_parse_dsl_type(self, data_parser: BaseDataParser) -> Literal["server",
             return "server"
         return "client"
 
-    def _init_spec(self, spec: Dict[str, Any], field_specs: List[Dict[str, Any]]):
+    def _init_spec(self, spec: Dict[str, Any], field_specs: List[FieldSpec]):
         spec_obj, spec_type = get_spec_json(spec)
         self._update_vis_spec(spec_obj["config"] and fill_new_fields(spec_obj["config"], field_specs))
         self.spec_type = spec_type

pygwalker/api/streamlit.py

@@ -56,7 +56,7 @@ def __init__(
         dataset: Union[DataFrame, Connector],
         gid: Union[int, str] = None,
         *,
-        field_specs: Optional[Dict[str, FieldSpec]] = None,
+        field_specs: Optional[List[FieldSpec]] = None,
         theme_key: Literal['vega', 'g2'] = 'g2',
         dark: Literal['media', 'light', 'dark'] = 'media',
         spec: str = "",
@@ -74,7 +74,7 @@ def __init__(
             - gid (Union[int, str], optional): GraphicWalker container div's id ('gwalker-{gid}')
 
         Kargs:
-            - field_specs (Dict[str, FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
+            - field_specs (List[FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
             - theme_key ('vega' | 'g2'): theme type.
             - dark (Literal['media' | 'light' | 'dark']): 'media': auto detect OS theme.
             - spec (str): chart config data. config id, json, remote file url
@@ -90,7 +90,7 @@ def __init__(
         self.walker = PygWalker(
             gid=gid,
             dataset=dataset,
-            field_specs=field_specs if field_specs is not None else {},
+            field_specs=field_specs if field_specs is not None else [],
             spec=spec,
             source_invoke_code="",
             theme_key=theme_key,
@@ -264,7 +264,7 @@ def get_streamlit_html(
     dataset: Union[DataFrame, Connector],
     gid: Union[int, str] = None,
     *,
-    field_specs: Optional[Dict[str, FieldSpec]] = None,
+    field_specs: Optional[List[FieldSpec]] = None,
     theme_key: Literal['vega', 'g2'] = 'g2',
     dark: Literal['media', 'light', 'dark'] = 'media',
     spec: str = "",
@@ -283,7 +283,7 @@ def get_streamlit_html(
         - gid (Union[int, str], optional): GraphicWalker container div's id ('gwalker-{gid}')
 
     Kargs:
-        - field_specs (Dict[str, FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
+        - field_specs (List[FieldSpec], optional): Specifications of some fields. They'll been automatically inferred from `df` if some fields are not specified.
         - theme_key ('vega' | 'g2'): theme type.
         - dark (Literal['media' | 'light' | 'dark']): 'media': auto detect OS theme.
         - spec (str): chart config data. config id, json, remote file url
@@ -293,7 +293,7 @@ def get_streamlit_html(
         - default_tab (Literal["data", "vis"]): default tab to show. Default to "vis"
     """
     if field_specs is None:
-        field_specs = {}
+        field_specs = []
 
     renderer = StreamlitRenderer(
         gid=gid,

pygwalker/data_parsers/base.py

@@ -1,11 +1,12 @@
-from typing import Generic, Dict, List, Any, Optional, NamedTuple
+from typing import Generic, Dict, List, Any, Optional
 from typing_extensions import Literal
 from functools import lru_cache
 from datetime import datetime, date
 from datetime import timedelta
 import abc
 import io
 
+from pydantic import BaseModel
 import duckdb
 import arrow
 import pytz
@@ -16,21 +17,21 @@
 
 
 # pylint: disable=broad-except
-class FieldSpec(NamedTuple):
+class FieldSpec(BaseModel):
     """Field specification.
 
     Args:
-    - semanticType: '?' | 'nominal' | 'ordinal' | 'temporal' | 'quantitative'. default to '?'.
-    - analyticType: '?' | 'dimension' | 'measure'. default to '?'.
+    - fname: str. The field name.
+    - semantic_type: '?' | 'nominal' | 'ordinal' | 'temporal' | 'quantitative'. default to '?'.
+    - analytic_type: '?' | 'dimension' | 'measure'. default to '?'.
     - display_as: str. The field name displayed. None means using the original column name.
     """
-    semanticType: Literal['?', 'nominal', 'ordinal', 'temporal', 'quantitative'] = '?'
-    analyticType: Literal['?', 'dimension', 'measure'] = '?'
+    fname: str
+    semantic_type: Literal['?', 'nominal', 'ordinal', 'temporal', 'quantitative'] = '?'
+    analytic_type: Literal['?', 'dimension', 'measure'] = '?'
     display_as: str = None
 
 
-default_field_spec = FieldSpec()
-
 INFINITY_DATA_SIZE = 1 << 62
 
 
@@ -41,7 +42,7 @@ class BaseDataParser(abc.ABC):
     def __init__(
         self,
         data: Any,
-        field_specs: Dict[str, FieldSpec],
+        field_specs: List[FieldSpec],
         infer_string_to_date: bool,
         infer_number_to_dimension: bool,
         other_params: Dict[str, Any]
@@ -118,7 +119,7 @@ class BaseDataFrameDataParser(Generic[DataFrame], BaseDataParser):
     """DataFrame property getter"""
     def __init__(
         self, df: DataFrame,
-        field_specs: Dict[str, FieldSpec],
+        field_specs: List[FieldSpec],
         infer_string_to_date: bool,
         infer_number_to_dimension: bool,
         other_params: Dict[str, Any]
@@ -149,7 +150,7 @@ def raw_fields(self) -> List[Dict[str, str]]:
         ]
 
     def _infer_prop(
-        self, col: str, field_specs: Dict[str, FieldSpec] = None
+        self, col: str, field_specs: List[FieldSpec] = None
     ) -> Dict[str, str]:
         """get IMutField
 
@@ -158,9 +159,12 @@ def _infer_prop(
         """
         s = self._example_df[col]
         orig_fname = col
-        field_spec = field_specs.get(orig_fname, default_field_spec)
-        semantic_type = self._infer_semantic(s, orig_fname) if field_spec.semanticType == '?' else field_spec.semanticType
-        analytic_type = self._infer_analytic(s, orig_fname) if field_spec.analyticType == '?' else field_spec.analyticType
+
+        field_spec_map = {field_spec.fname: field_spec for field_spec in field_specs}
+
+        field_spec = field_spec_map.get(orig_fname, FieldSpec(fname=orig_fname))
+        semantic_type = self._infer_semantic(s, orig_fname) if field_spec.semantic_type == '?' else field_spec.semantic_type
+        analytic_type = self._infer_analytic(s, orig_fname) if field_spec.analytic_type == '?' else field_spec.analytic_type
         fname = orig_fname if field_spec.display_as is None else field_spec.display_as
         return {
             'fid': col,

pygwalker/data_parsers/cloud_dataset_parser.py

@@ -19,7 +19,7 @@ class CloudDatasetParser(BaseDataParser):
     def __init__(
         self,
         dataset_id: str,
-        field_specs: Dict[str, FieldSpec],
+        field_specs: List[FieldSpec],
         infer_string_to_date: bool,
         infer_number_to_dimension: bool,
         other_params: Dict[str, Any]

pygwalker/data_parsers/database_parser.py

@@ -99,7 +99,7 @@ class DatabaseDataParser(BaseDataParser):
     def __init__(
         self,
         conn: Connector,
-        field_specs: Dict[str, FieldSpec],
+        field_specs: List[FieldSpec],
         infer_string_to_date: bool,
         infer_number_to_dimension: bool,
         other_params: Dict[str, Any]