kodexa 7.0.12399109365__py3-none-any.whl → 7.4.5a13228665254__py3-none-any.whl

kodexa/dataclasses/__init__.py

@@ -54,21 +54,35 @@ class LLMDataAttribute(BaseModel):
         self.node_uuid_list = source.node_uuid_list
         self.page_number = source.page_number
 
-    def create_exception(
-            self,
-            exception_type_id: str,
-            exception_type: str,
-            normalized_text: str,
-            message: str,
-            exception_detail: str,
-    ):
-        content_exception = ContentException(
-            exception_type=exception_type,
-            exception_detail=exception_detail,
-            message=message,
-            tag_uuid=self.tag_uuid,
-        )
-        self.exceptions.append(content_exception)
+    def process_exceptions(self, document: "KodexaDocumentLLMWrapper"):
+        # Lets make sure we add all the content exceptions
+        if self.exceptions is not None:
+            for exception in self.exceptions:
+                # We have two types of exception, one in the API and one in the
+                # document
+                from kodexa.model import ContentException as KodexaContentException
+                internal_exception = KodexaContentException(
+                    tag=exception.tag,
+                    exception_type=exception.exception_type,
+                    message=exception.message,
+                    exception_details=exception.exception_details,
+                    severity=exception.severity,
+                    group_uuid=self.group_uuid,
+                    tag_uuid=self.tag_uuid,
+                )
+                document.doc.add_exception(internal_exception)
+
+    def to_dict(self, taxonomy: Taxonomy) -> dict:
+        """Convert attribute to JSON with normalized value"""
+
+        target_taxon = taxonomy.get_taxon_by_path(self.taxon_path)
+        if target_taxon is None:
+            return {}
+
+        taxon_external_name = target_taxon.external_name
+        return {
+            taxon_external_name: self.normalized_text if self.normalized_text else self.value
+        }
 
 
 class LLMDataObject(BaseModel):
@@ -99,6 +113,24 @@ class LLMDataObject(BaseModel):
     class Config:
         arbitrary_types_allowed = True
 
+    def process_exceptions(self, document: "KodexaDocumentLLMWrapper"):
+        # Lets make sure we add all the content exceptions
+        if self.exceptions is not None:
+            for exception in self.exceptions:
+                # We have two types of exception, one in the API and one in the
+                # document
+                from kodexa.model import ContentException as KodexaContentException
+                internal_exception = KodexaContentException(
+                    tag=exception.tag,
+                    exception_type=exception.exception_type,
+                    message=exception.message,
+                    exception_details=exception.exception_details,
+                    severity=exception.severity,
+                    group_uuid=self.group_uuid,
+                    tag_uuid=self.tag_uuid,
+                )
+                document.doc.add_exception(internal_exception)
+
     def get_all_review_pages(self):
         """
         Returns a list of unique page numbers that would be included in the review.
@@ -106,7 +138,7 @@ class LLMDataObject(BaseModel):
         :return: list of unique page numbers
         """
         pages = set()
-        for field in self.__fields__:
+        for field in self.model_fields:
             pages.update(self._get_field_pages(field))
         return sorted(list(pages))
 
@@ -154,6 +186,35 @@ class LLMDataObject(BaseModel):
         if 'normalized_text' in field_data:
             attr.normalized_text = field_data['normalized_text']
 
+    def to_dict(self, taxonomy: Taxonomy) -> dict:
+        """Convert data object to JSON using normalized values and taxon paths"""
+        result = {}
+        for field in self.model_fields:
+            value = getattr(self, field)
+
+            if isinstance(value, list) and len(value) > 0:
+                if isinstance(value[0], LLMDataObject):
+                    # We need to find the first field of the object that is a LLMDataAttribute
+                    # and use that to derive the taxon path of the LLMDataObject
+                    data_attribute = None
+                    for child_field in value[0].model_fields:
+                        child_attr = getattr(value[0], child_field)
+                        if isinstance(child_attr, LLMDataAttribute):
+                            data_attribute = child_attr
+                            break
+                    if data_attribute is not None:
+                        taxon_path = data_attribute.taxon_path.rsplit('/', 1)[0]
+                        target_taxon = taxonomy.get_taxon_by_path(taxon_path)
+                        if target_taxon is not None:
+                            result[target_taxon.external_name] = [item.to_dict(taxonomy) for item in value if isinstance(item, (LLMDataObject, LLMDataAttribute))]
+            elif isinstance(value, LLMDataAttribute):
+                result.update(value.to_dict(taxonomy))
+            elif isinstance(value, LLMDataObject):
+                target_taxon = taxonomy.get_taxon_by_path(value.taxon_path)
+                if target_taxon is not None:
+                    result[target_taxon.external_name] = value.to_dict(taxonomy)
+        return result
+
     def to_review(self, page_number=None):
         """
         Build a representation of the data object and its data attributes that is a dict that includes the
@@ -164,7 +225,7 @@ class LLMDataObject(BaseModel):
         :return: dict of this data object and children for the specified page
         """
         review = {}
-        for field in self.__fields__:
+        for field in self.model_fields:
             review_field = self._build_review(field, page_number)
             if review_field:
                 review[field] = review_field
@@ -193,26 +254,6 @@ class LLMDataObject(BaseModel):
 
         return None
 
-    def create_exception(
-            self,
-            exception_type_id: str,
-            exception_type: str,
-            message: str,
-            exception_detail: str,
-            severity: str = "ERROR",
-    ):
-        content_exception = ContentException(
-            exception_type=exception_type,
-            exception_details=exception_detail,
-            message=message,
-            group_uuid=self.group_uuid,
-            severity=severity,
-        )
-        if self.exceptions is None:
-            self.exceptions = []
-
-        self.exceptions.append(content_exception)
-
     def apply_labels(
             self, document: "KodexaDocumentLLMWrapper", parent_group_uuid: str = None,
             assistant: Optional["Assistant"] = None
@@ -234,24 +275,11 @@ class LLMDataObject(BaseModel):
         """
 
         # Lets make sure we add all the content exceptions
-        if self.exceptions is not None:
-            for exception in self.exceptions:
-                # We have two types of exception, one in the API and one in the
-                # document
-                from kodexa.model import ContentException as KodexaContentException
-                internal_exception = KodexaContentException(
-                    exception_type=exception.exception_type,
-                    message=exception.message,
-                    exception_details=exception.exception_details,
-                    severity=exception.severity,
-                    group_uuid=exception.group_uuid,
-                    tag_uuid=exception.tag_uuid,
-                )
-                document.doc.add_exception(internal_exception)
+        self.process_exceptions(document)
 
         # Let's go through this data object and find all the attributes that have a value
         # then we will apply the labels to the document
-        for field in self.__fields__:
+        for field in self.model_fields:
             logger.info(f"Processing field {field}")
             value = getattr(self, field)
 
@@ -270,8 +298,6 @@ class LLMDataObject(BaseModel):
             # We need to add the label to the document for this attribute
 
             tag = value.taxon_path
-
-            # TODO need to work out why we are missing them?
             logger.info(f"Value: {value.normalized_text}, node_uuid_list: {value.node_uuid_list}")
             if value.node_uuid_list is None:
                 value.node_uuid_list = value.line_ids
@@ -320,31 +346,16 @@ class LLMDataObject(BaseModel):
                         current_value.append(new_tag)
                         node.remove_feature("tag", tag)
                         node.add_feature("tag", tag, current_value, single=False)
-                        # try:
-                        #     if value.data_type == 'Derived':
-                        #         logger.info(f"Node already has tag {tag} - Tagging something nearby {node.get_all_content()}")
-                        #         nearby_node = find_nearby_word_to_tag(node, tag)
-                        #         nearby_node.tag(
-                        #             tag_to_apply=tag,
-                        #             value=value.normalized_text,
-                        #             tag_uuid=tag_uuid,
-                        #             cell_index=self.cell_index,
-                        #             selector="//word",
-                        #             confidence=-1,
-                        #             group_uuid=self.group_uuid,
-                        #             parent_group_uuid=parent_group_uuid,
-                        #             owner_uri=f"assistant://{assistant.id}" if assistant else f"model://taxonomy-llm",
-                        #         )
-                        #     else:
-                        #         logger.info(f"Node already has tag {tag} - Skipping.")
-                        # except:
-                        #     logger.error(f"Error tagging nearby node with tag {tag}")
 
             logger.info(f"Applied label {tag} to {len(nodes_to_label)} nodes")
+
+            # Lets make sure we add all the content exceptions
+            self.process_exceptions(document)
+
         if isinstance(value, LLMDataObject):
             # We need to apply the labels to the document for this object
             value.apply_labels(document, parent_group_uuid=self.group_uuid)
-            # logger.info(f"Applied labels to data object {value.group_uuid}")
+            logger.info(f"Applied labels to data object {value.group_uuid}")
 
 
 def find_nearby_word_to_tag(node, tag):

kodexa/model/entities/product.py

@@ -1,13 +1,17 @@
-from typing import Optional, List
+from decimal import Decimal
+from typing import Optional, List, Set
 
 from pydantic import BaseModel, ConfigDict, Field
+
 from kodexa.model.base import StandardDateTime
 from kodexa.platform.client import EntityEndpoint, PageEndpoint, EntitiesEndpoint
+from .product_group import ProductGroup
+from ..objects import ProjectTemplate
 
 
-class Product(BaseModel):
+class ProjectTemplateMetadata(BaseModel):
     """
-
+    A project template metadata entity
     """
     model_config = ConfigDict(
         populate_by_name=True,
@@ -15,11 +19,41 @@ class Product(BaseModel):
         arbitrary_types_allowed=True,
         protected_namespaces=("model_config",),
     )
+
+    id: str
+
+
+class ProductProjectTemplate(BaseModel):
     """
-    A product
+    A product project template entity representing the relationship between products and project templates
     """
+    model_config = ConfigDict(
+        populate_by_name=True,
+        use_enum_values=True,
+        arbitrary_types_allowed=True,
+        protected_namespaces=("model_config",),
+    )
+
+    id: Optional[str] = None
+    uuid: Optional[str] = None
+    change_sequence: Optional[int] = Field(None, alias="changeSequence")
+    created_on: Optional[StandardDateTime] = Field(None, alias="createdOn")
+    updated_on: Optional[StandardDateTime] = Field(None, alias="updatedOn")
+    display_order: Optional[int] = Field(None, alias="displayOrder")
+    project_template_metadata: Optional[ProjectTemplateMetadata] = Field(None, alias="projectTemplateMetadata")
 
 
+class Product(BaseModel):
+    """
+    A product entity representing a product in the Kodexa platform
+    """
+    model_config = ConfigDict(
+        populate_by_name=True,
+        use_enum_values=True,
+        arbitrary_types_allowed=True,
+        protected_namespaces=("model_config",),
+    )
+
     id: Optional[str] = None
     uuid: Optional[str] = None
     change_sequence: Optional[int] = Field(None, alias="changeSequence")
@@ -28,6 +62,26 @@ class Product(BaseModel):
     name: str
     description: Optional[str] = None
     overview_markdown: Optional[str] = Field(None, alias="overviewMarkdown")
+    product_group: ProductGroup = Field(..., alias="productGroup")
+    parent: Optional['Product'] = None
+    image_url: Optional[str] = Field(None, alias="imageUrl")
+    price_id: Optional[str] = Field(None, alias="priceId")
+    price: Optional[Decimal] = None
+    number_of_credits: Optional[int] = Field(None, alias="numberOfCredits")
+    price_suffix: Optional[str] = Field(None, alias="priceSuffix")
+    has_quantity: bool = Field(False, alias="hasQuantity")
+    active: bool = True
+    order: Optional[int] = None
+    promoted: Optional[bool] = None
+    project_templates: Optional[Set[ProjectTemplate]] = Field(None, alias="projectTemplates")
+    search_text: Optional[str] = None
+
+    def update_search_text(self):
+        """Updates the search text for the product"""
+        if self.product_group:
+            self.search_text = f"{self.name.lower()} {self.product_group.name.lower()}"
+        else:
+            self.search_text = self.name.lower()
 
 
 class ProductEndpoint(Product, EntityEndpoint):
@@ -53,7 +107,7 @@ class ProductEndpoint(Product, EntityEndpoint):
 
 class PageProduct(BaseModel):
     """
-
+    Represents a paginated list of products
     """
     model_config = ConfigDict(
         populate_by_name=True,
@@ -66,7 +120,6 @@ class PageProduct(BaseModel):
     size: Optional[int] = None
     content: Optional[List[Product]] = None
     number: Optional[int] = None
-
     number_of_elements: Optional[int] = Field(None, alias="numberOfElements")
     first: Optional[bool] = None
     last: Optional[bool] = None

kodexa/model/entities/product_group.py

@@ -0,0 +1,126 @@
+from typing import Optional, List
+
+from pydantic import BaseModel, ConfigDict, Field
+from kodexa.model.base import StandardDateTime
+from kodexa.platform.client import EntityEndpoint, PageEndpoint, EntitiesEndpoint
+
+
+class ProductGroup(BaseModel):
+    """
+
+    """
+    model_config = ConfigDict(
+        populate_by_name=True,
+        use_enum_values=True,
+        arbitrary_types_allowed=True,
+        protected_namespaces=("model_config",),
+    )
+    """
+    A product group
+    """
+
+
+    id: Optional[str] = None
+    uuid: Optional[str] = None
+    change_sequence: Optional[int] = Field(None, alias="changeSequence")
+    created_on: Optional[StandardDateTime] = Field(None, alias="createdOn")
+    updated_on: Optional[StandardDateTime] = Field(None, alias="updatedOn")
+    name: str
+    description: Optional[str] = None
+    overview_markdown: Optional[str] = Field(None, alias="overviewMarkdown")
+
+
+class ProductGroupEndpoint(ProductGroup, EntityEndpoint):
+    """Handles the endpoint for a product group
+
+    This class is a combination of DataException and EntityEndpoint. It is used
+    to manage the endpoint for data exceptions.
+
+    Methods:
+        get_type: Returns the type of the endpoint.
+    """
+
+    def get_type(self) -> str:
+        """Gets the type of the endpoint.
+
+        This method returns the type of the endpoint which is "exceptions".
+
+        Returns:
+            str: The type of the endpoint.
+        """
+        return "product-groups"
+
+
+class PageProductGroup(BaseModel):
+    """
+
+    """
+    model_config = ConfigDict(
+        populate_by_name=True,
+        use_enum_values=True,
+        arbitrary_types_allowed=True,
+        protected_namespaces=("model_config",),
+    )
+    total_pages: Optional[int] = Field(None, alias="totalPages")
+    total_elements: Optional[int] = Field(None, alias="totalElements")
+    size: Optional[int] = None
+    content: Optional[List[ProductGroup]] = None
+    number: Optional[int] = None
+
+    number_of_elements: Optional[int] = Field(None, alias="numberOfElements")
+    first: Optional[bool] = None
+    last: Optional[bool] = None
+    empty: Optional[bool] = None
+
+
+class PageProductGroupEndpoint(PageProductGroup, PageEndpoint):
+    def get_type(self) -> Optional[str]:
+        return "product-group"
+
+
+class ProductGroupsEndpoint(EntitiesEndpoint):
+    """Represents the product groups endpoint
+
+    This class is used to represent the product groups endpoint in the system.
+
+    Attributes:
+        object_dict: A dictionary containing the object data.
+    """
+
+    """Represents a assistants endpoint"""
+
+    def get_type(self) -> str:
+        """Get the type of the endpoint
+
+        This method is used to get the type of the endpoint.
+
+        Returns:
+            str: The type of the endpoint.
+        """
+        return "product-groups"
+
+    def get_instance_class(self, object_dict=None):
+        """Get the instance class of the endpoint
+
+        This method is used to get the instance class of the endpoint.
+
+        Args:
+            object_dict (dict, optional): A dictionary containing the object data.
+
+        Returns:
+            AssistantEndpoint: The instance class of the endpoint.
+        """
+        return ProductGroupEndpoint
+
+    def get_page_class(self, object_dict=None):
+        """Get the page class of the endpoint
+
+        This method is used to get the page class of the endpoint.
+
+        Args:
+            object_dict (dict, optional): A dictionary containing the object data.
+
+        Returns:
+            PageAssistantEndpoint: The page class of the endpoint.
+        """
+        return PageProductGroupEndpoint

kodexa/model/model.py

@@ -2443,11 +2443,14 @@ class Document(object):
     def get_exceptions(self) -> List[ContentException]:
         return self._persistence_layer.get_exceptions()
 
-    def get_external_data(self) -> dict:
-        return self._persistence_layer.get_external_data()
+    def get_external_data(self, key="default") -> dict:
+        return self._persistence_layer.get_external_data(key)
 
-    def set_external_data(self, external_data:dict):
-        return self._persistence_layer.set_external_data(external_data)
+    def get_external_data_keys(self) -> list[str]:
+        return self._persistence_layer.get_external_data_keys()
+
+    def set_external_data(self, external_data:dict, key="default"):
+        return self._persistence_layer.set_external_data(external_data, key)
 
     def get_steps(self) -> list[ProcessingStep]:
         return self._persistence_layer.get_steps()