NXxas pydantic model

setup.cfg

@@ -23,6 +23,9 @@ package_dir=
 packages=find:
 python_requires = >=3.8
 install_requires = 
+    gitpython
+    xmlschema
+    pydantic
 
 [options.packages.find]
 where=src

src/pynxxas/nxdl/models.py

@@ -0,0 +1,329 @@
+"""NXDL models"""
+
+from typing import Optional, Union, List
+
+import pydantic
+
+
+class Item(pydantic.BaseModel):
+
+    class Config:
+        extra = "forbid"
+
+
+class EnumerationItem(Item):
+    value: str
+    doc: Optional[str] = None
+
+
+class Enumeration(Item):
+    item: List[EnumerationItem]
+
+
+class DimensionItem(Item):
+    index: int
+    required: bool
+    value: Optional[str] = None
+    ref: Optional[str] = None
+
+
+class DocDimensionItem(Item):
+    index: str
+    value: str
+
+
+class DocDimensions(Item):
+    dim: List[DocDimensionItem]
+
+
+class Dimensions(Item):
+    rank: Optional[Union[int, str]] = None
+    dim: Optional[List[DimensionItem]] = None
+    doc: Optional[Union[DocDimensions, str]] = None
+
+
+class Attribute(Item):
+    name: str
+    type: str
+    doc: Optional[str] = None
+    recommended: Optional[bool] = None
+    optional: Optional[bool] = None
+    dimensions: Optional[Dimensions] = None
+    enumeration: Optional[Enumeration] = None
+    deprecated: Optional[str] = None
+
+
+class Field(Item):
+    name: str
+    type: str
+    nameType: Optional[str] = None
+    units: Optional[str] = None
+    signal: Optional[int] = None
+    axis: Optional[int] = None
+    primary: Optional[int] = None
+    axes: Optional[str] = None
+    doc: Optional[str] = None
+    recommended: Optional[bool] = None
+    optional: Optional[bool] = None
+    minOccurs: Optional[int] = None
+    maxOccurs: Optional[Union[int, str]] = None
+    stride: Optional[bool] = None
+    data_offset: Optional[bool] = None
+    dimensions: Optional[Dimensions] = None
+    enumeration: Optional[Enumeration] = None
+    attribute: Optional[List[Attribute]] = None
+    deprecated: Optional[str] = None
+
+    class Config:
+        extra = "forbid"
+
+
+class Group(Item):
+    type: str
+    name: Optional[str] = None  # type[2:].upper()
+    doc: Optional[List[Optional[str]]] = None  # NXmirror returns [None]
+    recommended: Optional[bool] = None
+    optional: Optional[bool] = None
+    minOccurs: Optional[int] = None
+    maxOccurs: Optional[Union[int, str]] = None
+    attribute: Optional[List[Attribute]] = None
+    field: Optional[List[Field]] = None
+    group: Optional[List["Group"]] = None
+    link: Optional[List["Link"]] = None
+    deprecated: Optional[str] = None
+
+    @pydantic.model_validator(mode="after")
+    def name_from_type(self) -> "Group":
+        if self.name is None:
+            self.name = self.type[2:].upper()
+        return self
+
+
+class Link(Item):
+    name: str
+    target: str
+    doc: Optional[str] = None
+
+    class Config:
+        extra = "forbid"
+
+
+class Choice(Item):
+    name: str
+    group: List[Group]
+
+    class Config:
+        extra = "forbid"
+
+
+class Symbol(Item):
+    name: str
+    doc: str
+
+
+class Symbols(Item):
+    symbol: Optional[List[Symbol]] = None
+    doc: Optional[str] = None
+
+
+class Definition(Item):
+    xmlns: str
+    xmlns_xsi: str
+    name: str
+    type: str
+    category: str
+    xsi_schemaLocation: str
+    ignoreExtraGroups: bool
+    ignoreExtraFields: bool
+    ignoreExtraAttributes: bool
+    xmlns_xs: Optional[str] = None
+    xmlns_ns: Optional[str] = None
+    extends: Optional[str] = None
+    deprecated: Optional[str] = None
+    doc: Optional[str] = None
+    symbols: Optional[Symbols] = None
+    attribute: Optional[List[Attribute]] = None
+    field: Optional[List[Field]] = None
+    group: Optional[List[Group]] = None
+    link: Optional[List[Link]] = None
+    choice: Optional[List[Choice]] = None
+
+    class Config:
+        extra = "forbid"
+
+
+def load_enumeration_item(enum_item: dict) -> dict:
+    data = dict()
+    for key, value in enum_item.items():
+        if key.startswith("@"):
+            key = key[1:]
+        data[key] = value
+    return data
+
+
+def load_enumeration(enumeration: dict) -> dict:
+    data = dict()
+    for key, value in enumeration.items():
+        data[key] = [load_enumeration_item(item) for item in value]
+    return data
+
+
+def load_doc_dimension_item(dim_item: dict) -> dict:
+    data = dict()
+    for key, value in dim_item.items():
+        if key.startswith("@"):
+            key = key[1:]
+        data[key] = value
+    return data
+
+
+def load_dimension_item(dim_item: dict) -> dict:
+    data = dict()
+    for key, value in dim_item.items():
+        if key.startswith("@"):
+            key = key[1:]
+        data[key] = value
+    return data
+
+
+def load_doc_dimensions(dimensions: dict) -> dict:
+    data = dict()
+    for key, value in dimensions.items():
+        if key.startswith("@"):
+            key = key[1:]
+        elif key == "dim":
+            value = [load_doc_dimension_item(item) for item in value]
+        data[key] = value
+    return data
+
+
+def load_dimensions(dimensions: dict) -> dict:
+    data = dict()
+    for key, value in dimensions.items():
+        if key.startswith("@"):
+            key = key[1:]
+        elif key == "dim":
+            value = [load_dimension_item(item) for item in value]
+        elif key == "doc" and isinstance(value, dict):
+            value = load_doc_dimensions(value)
+        data[key] = value
+    return data
+
+
+def load_link(link: dict) -> dict:
+    data = dict()
+    for key, value in link.items():
+        if key.startswith("@"):
+            key = key[1:]
+        data[key] = value
+    return data
+
+
+def load_choice(choice: dict) -> dict:
+    data = dict()
+    for key, value in choice.items():
+        if key.startswith("@"):
+            key = key[1:]
+        elif key == "group":
+            value = [load_group(group) for group in value]
+        data[key] = value
+    return data
+
+
+def load_attribute(attr: dict) -> dict:
+    data = dict()
+    for key, value in attr.items():
+        if key.startswith("@"):
+            key = key[1:]
+        elif key == "enumeration":
+            value = load_enumeration(value)
+        elif key == "dimensions":
+            value = load_dimensions(value)
+        data[key] = value
+    return data
+
+
+def load_field(field: dict) -> dict:
+    data = dict()
+    for key, value in field.items():
+        if key.startswith("@"):
+            key = key[1:]
+        elif key == "attribute":
+            value = [load_attribute(attr) for attr in value]
+        elif key == "enumeration":
+            value = load_enumeration(value)
+        elif key == "dimensions":
+            value = load_dimensions(value)
+        data[key] = value
+    return data
+
+
+def load_group(group: dict) -> dict:
+    data = dict()
+    for key, value in group.items():
+        if key.startswith("@"):
+            key = key[1:]
+        elif key == "attribute":
+            value = [load_attribute(attr) for attr in value]
+        elif key == "field":
+            value = [load_field(field) for field in value]
+        elif key == "group":
+            value = [load_group(group) for group in value]
+        elif key == "link":
+            value = [load_link(attr) for attr in value]
+        data[key] = value
+    return data
+
+
+def load_symbol(symbol: dict) -> dict:
+    data = dict()
+    for key, value in symbol.items():
+        if key.startswith("@"):
+            key = key[1:]
+        data[key] = value
+    return data
+
+
+def load_symbols(symbols: dict) -> dict:
+    data = dict()
+    for key, value in symbols.items():
+        if key.startswith("@"):
+            key = key[1:]
+        elif key == "symbol":
+            value = [load_symbol(attr) for attr in value]
+        data[key] = value
+    return data
+
+
+def load_definition(definition: str) -> Definition:
+    data = dict()
+
+    for key, value in definition.items():
+        if key.startswith("@"):
+            key = key[1:].replace(":", "_")
+        elif key == "attribute":
+            value = [load_attribute(attr) for attr in value]
+        elif key == "field":
+            value = [load_field(attr) for attr in value]
+        elif key == "group":
+            value = [load_group(attr) for attr in value]
+        elif key == "link":
+            value = [load_link(attr) for attr in value]
+        elif key == "choice":
+            value = [load_choice(attr) for attr in value]
+        elif key == "symbols":
+            value = load_symbols(value)
+        data[key] = value
+
+    return Definition(**data)
+
+
+if __name__ == "__main__":
+    from . import repo
+
+    names = repo.get_nxdl_class_names(
+        url="https://github.com/XraySpectroscopy/nexus_definitions.git"
+    )
+
+    for name in names:
+        load_definition(repo.get_nxdl_class(name))