Skip to content

Commit 972212e

Browse files
andrewldesousaandrewldesousa
authored
Merge pull request #30 from microsoft/adesousa_microsoft/delete-unused-code
delete unused scripts and delete unused unit tests
2 parents 06b9421 + cae9fbc commit 972212e

16 files changed

+2
-1405
lines changed

‎backend/settings.py‎

Lines changed: 0 additions & 330 deletions
Original file line numberDiff line numberDiff line change
@@ -341,315 +341,6 @@ def construct_payload_configuration(
341341
}
342342

343343

344-
class _AzureCosmosDbMongoVcoreSettings(
345-
BaseSettings,
346-
DatasourcePayloadConstructor
347-
):
348-
model_config = SettingsConfigDict(
349-
env_prefix="AZURE_COSMOSDB_MONGO_VCORE_",
350-
env_file=DOTENV_PATH,
351-
extra="ignore",
352-
env_ignore_empty=True
353-
)
354-
_type: Literal["azure_cosmosdb"] = PrivateAttr(default="azure_cosmosdb")
355-
top_k: int = Field(default=5, serialization_alias="top_n_documents")
356-
strictness: int = 3
357-
enable_in_domain: bool = Field(default=True, serialization_alias="in_scope")
358-
query_type: Literal['vector'] = "vector"
359-
connection_string: str = Field(exclude=True)
360-
index: str = Field(serialization_alias="index_name")
361-
database: str = Field(serialization_alias="database_name")
362-
container: str = Field(serialization_alias="container_name")
363-
content_columns: Optional[List[str]] = Field(default=None, exclude=True)
364-
vector_columns: Optional[List[str]] = Field(default=None, exclude=True)
365-
title_column: Optional[str] = Field(default=None, exclude=True)
366-
url_column: Optional[str] = Field(default=None, exclude=True)
367-
filename_column: Optional[str] = Field(default=None, exclude=True)
368-
369-
# Constructed fields
370-
authentication: Optional[dict] = None
371-
embedding_dependency: Optional[dict] = None
372-
fields_mapping: Optional[dict] = None
373-
374-
@field_validator('content_columns', 'vector_columns', mode="before")
375-
@classmethod
376-
def split_columns(cls, comma_separated_string: str) -> List[str]:
377-
if isinstance(comma_separated_string, str) and len(comma_separated_string) > 0:
378-
return parse_multi_columns(comma_separated_string)
379-
380-
return None
381-
382-
@model_validator(mode="after")
383-
def construct_authentication(self) -> Self:
384-
self.authentication = {
385-
"type": "connection_string",
386-
"connection_string": self.connection_string
387-
}
388-
return self
389-
390-
@model_validator(mode="after")
391-
def set_fields_mapping(self) -> Self:
392-
self.fields_mapping = {
393-
"content_fields": self.content_columns,
394-
"title_field": self.title_column,
395-
"url_field": self.url_column,
396-
"filepath_field": self.filename_column,
397-
"vector_fields": self.vector_columns
398-
}
399-
return self
400-
401-
def construct_payload_configuration(
402-
self,
403-
*args,
404-
**kwargs
405-
):
406-
self.embedding_dependency = \
407-
self._settings.azure_openai.extract_embedding_dependency()
408-
parameters = self.model_dump(exclude_none=True, by_alias=True)
409-
parameters.update(self._settings.search.model_dump(exclude_none=True, by_alias=True))
410-
return {
411-
"type": self._type,
412-
"parameters": parameters
413-
}
414-
415-
416-
class _ElasticsearchSettings(BaseSettings, DatasourcePayloadConstructor):
417-
model_config = SettingsConfigDict(
418-
env_prefix="ELASTICSEARCH_",
419-
env_file=DOTENV_PATH,
420-
extra="ignore",
421-
env_ignore_empty=True
422-
)
423-
_type: Literal["elasticsearch"] = PrivateAttr(default="elasticsearch")
424-
top_k: int = Field(default=5, serialization_alias="top_n_documents")
425-
strictness: int = 3
426-
enable_in_domain: bool = Field(default=True, serialization_alias="in_scope")
427-
endpoint: str
428-
encoded_api_key: str = Field(exclude=True)
429-
index: str = Field(serialization_alias="index_name")
430-
query_type: Literal['simple', 'vector'] = "simple"
431-
content_columns: Optional[List[str]] = Field(default=None, exclude=True)
432-
vector_columns: Optional[List[str]] = Field(default=None, exclude=True)
433-
title_column: Optional[str] = Field(default=None, exclude=True)
434-
url_column: Optional[str] = Field(default=None, exclude=True)
435-
filename_column: Optional[str] = Field(default=None, exclude=True)
436-
embedding_model_id: Optional[str] = Field(default=None, exclude=True)
437-
438-
# Constructed fields
439-
authentication: Optional[dict] = None
440-
embedding_dependency: Optional[dict] = None
441-
fields_mapping: Optional[dict] = None
442-
443-
@field_validator('content_columns', 'vector_columns', mode="before")
444-
@classmethod
445-
def split_columns(cls, comma_separated_string: str) -> List[str]:
446-
if isinstance(comma_separated_string, str) and len(comma_separated_string) > 0:
447-
return parse_multi_columns(comma_separated_string)
448-
449-
return None
450-
451-
@model_validator(mode="after")
452-
def set_authentication(self) -> Self:
453-
self.authentication = {
454-
"type": "encoded_api_key",
455-
"encoded_api_key": self.encoded_api_key
456-
}
457-
458-
return self
459-
460-
@model_validator(mode="after")
461-
def set_fields_mapping(self) -> Self:
462-
self.fields_mapping = {
463-
"content_fields": self.content_columns,
464-
"title_field": self.title_column,
465-
"url_field": self.url_column,
466-
"filepath_field": self.filename_column,
467-
"vector_fields": self.vector_columns
468-
}
469-
return self
470-
471-
def construct_payload_configuration(
472-
self,
473-
*args,
474-
**kwargs
475-
):
476-
self.embedding_dependency = \
477-
{"type": "model_id", "model_id": self.embedding_model_id} if self.embedding_model_id else \
478-
self._settings.azure_openai.extract_embedding_dependency()
479-
480-
parameters = self.model_dump(exclude_none=True, by_alias=True)
481-
parameters.update(self._settings.search.model_dump(exclude_none=True, by_alias=True))
482-
483-
return {
484-
"type": self._type,
485-
"parameters": parameters
486-
}
487-
488-
489-
class _PineconeSettings(BaseSettings, DatasourcePayloadConstructor):
490-
model_config = SettingsConfigDict(
491-
env_prefix="PINECONE_",
492-
env_file=DOTENV_PATH,
493-
extra="ignore",
494-
env_ignore_empty=True
495-
)
496-
_type: Literal["pinecone"] = PrivateAttr(default="pinecone")
497-
top_k: int = Field(default=5, serialization_alias="top_n_documents")
498-
strictness: int = 3
499-
enable_in_domain: bool = Field(default=True, serialization_alias="in_scope")
500-
environment: str
501-
api_key: str = Field(exclude=True)
502-
index_name: str
503-
query_type: Literal["vector"] = "vector"
504-
content_columns: Optional[List[str]] = Field(default=None, exclude=True)
505-
vector_columns: Optional[List[str]] = Field(default=None, exclude=True)
506-
title_column: Optional[str] = Field(default=None, exclude=True)
507-
url_column: Optional[str] = Field(default=None, exclude=True)
508-
filename_column: Optional[str] = Field(default=None, exclude=True)
509-
510-
# Constructed fields
511-
authentication: Optional[dict] = None
512-
embedding_dependency: Optional[dict] = None
513-
fields_mapping: Optional[dict] = None
514-
515-
@field_validator('content_columns', 'vector_columns', mode="before")
516-
@classmethod
517-
def split_columns(cls, comma_separated_string: str) -> List[str]:
518-
if isinstance(comma_separated_string, str) and len(comma_separated_string) > 0:
519-
return parse_multi_columns(comma_separated_string)
520-
521-
return None
522-
523-
@model_validator(mode="after")
524-
def set_authentication(self) -> Self:
525-
self.authentication = {
526-
"type": "api_key",
527-
"api_key": self.api_key
528-
}
529-
530-
return self
531-
532-
@model_validator(mode="after")
533-
def set_fields_mapping(self) -> Self:
534-
self.fields_mapping = {
535-
"content_fields": self.content_columns,
536-
"title_field": self.title_column,
537-
"url_field": self.url_column,
538-
"filepath_field": self.filename_column,
539-
"vector_fields": self.vector_columns
540-
}
541-
return self
542-
543-
def construct_payload_configuration(
544-
self,
545-
*args,
546-
**kwargs
547-
):
548-
self.embedding_dependency = \
549-
self._settings.azure_openai.extract_embedding_dependency()
550-
parameters = self.model_dump(exclude_none=True, by_alias=True)
551-
parameters.update(self._settings.search.model_dump(exclude_none=True, by_alias=True))
552-
553-
return {
554-
"type": self._type,
555-
"parameters": parameters
556-
}
557-
558-
559-
class _AzureMLIndexSettings(BaseSettings, DatasourcePayloadConstructor):
560-
model_config = SettingsConfigDict(
561-
env_prefix="AZURE_MLINDEX_",
562-
env_file=DOTENV_PATH,
563-
extra="ignore",
564-
env_ignore_empty=True
565-
)
566-
_type: Literal["azure_ml_index"] = PrivateAttr(default="azure_ml_index")
567-
top_k: int = Field(default=5, serialization_alias="top_n_documents")
568-
strictness: int = 3
569-
enable_in_domain: bool = Field(default=True, serialization_alias="in_scope")
570-
name: str
571-
version: str
572-
project_resource_id: str = Field(validation_alias="AZURE_ML_PROJECT_RESOURCE_ID")
573-
content_columns: Optional[List[str]] = Field(default=None, exclude=True)
574-
vector_columns: Optional[List[str]] = Field(default=None, exclude=True)
575-
title_column: Optional[str] = Field(default=None, exclude=True)
576-
url_column: Optional[str] = Field(default=None, exclude=True)
577-
filename_column: Optional[str] = Field(default=None, exclude=True)
578-
579-
# Constructed fields
580-
fields_mapping: Optional[dict] = None
581-
582-
@field_validator('content_columns', 'vector_columns', mode="before")
583-
@classmethod
584-
def split_columns(cls, comma_separated_string: str) -> List[str]:
585-
if isinstance(comma_separated_string, str) and len(comma_separated_string) > 0:
586-
return parse_multi_columns(comma_separated_string)
587-
588-
return None
589-
590-
@model_validator(mode="after")
591-
def set_fields_mapping(self) -> Self:
592-
self.fields_mapping = {
593-
"content_fields": self.content_columns,
594-
"title_field": self.title_column,
595-
"url_field": self.url_column,
596-
"filepath_field": self.filename_column,
597-
"vector_fields": self.vector_columns
598-
}
599-
return self
600-
601-
def construct_payload_configuration(
602-
self,
603-
*args,
604-
**kwargs
605-
):
606-
parameters = self.model_dump(exclude_none=True, by_alias=True)
607-
parameters.update(self._settings.search.model_dump(exclude_none=True, by_alias=True))
608-
609-
return {
610-
"type": self._type,
611-
"parameters": parameters
612-
}
613-
614-
615-
class _AzureSqlServerSettings(BaseSettings, DatasourcePayloadConstructor):
616-
model_config = SettingsConfigDict(
617-
env_prefix="AZURE_SQL_SERVER_",
618-
env_file=DOTENV_PATH,
619-
extra="ignore"
620-
)
621-
_type: Literal["azure_sql_server"] = PrivateAttr(default="azure_sql_server")
622-
623-
connection_string: str = Field(exclude=True)
624-
table_schema: str
625-
schema_max_row: Optional[int] = None
626-
top_n_results: Optional[int] = None
627-
628-
# Constructed fields
629-
authentication: Optional[dict] = None
630-
631-
@model_validator(mode="after")
632-
def construct_authentication(self) -> Self:
633-
self.authentication = {
634-
"type": "connection_string",
635-
"connection_string": self.connection_string
636-
}
637-
return self
638-
639-
def construct_payload_configuration(
640-
self,
641-
*args,
642-
**kwargs
643-
):
644-
parameters = self.model_dump(exclude_none=True, by_alias=True)
645-
#parameters.update(self._settings.search.model_dump(exclude_none=True, by_alias=True))
646-
647-
return {
648-
"type": self._type,
649-
"parameters": parameters
650-
}
651-
652-
653344
class _BaseSettings(BaseSettings):
654345
model_config = SettingsConfigDict(
655346
env_file=DOTENV_PATH,
@@ -700,27 +391,6 @@ def set_datasource_settings(self) -> Self:
700391
if self.base_settings.datasource_type == "AzureCognitiveSearch":
701392
self.datasource = _AzureSearchSettings(settings=self, _env_file=DOTENV_PATH)
702393
logging.debug("Using Azure Cognitive Search")
703-
704-
elif self.base_settings.datasource_type == "AzureCosmosDB":
705-
self.datasource = _AzureCosmosDbMongoVcoreSettings(settings=self, _env_file=DOTENV_PATH)
706-
logging.debug("Using Azure CosmosDB Mongo vcore")
707-
708-
elif self.base_settings.datasource_type == "Elasticsearch":
709-
self.datasource = _ElasticsearchSettings(settings=self, _env_file=DOTENV_PATH)
710-
logging.debug("Using Elasticsearch")
711-
712-
elif self.base_settings.datasource_type == "Pinecone":
713-
self.datasource = _PineconeSettings(settings=self, _env_file=DOTENV_PATH)
714-
logging.debug("Using Pinecone")
715-
716-
elif self.base_settings.datasource_type == "AzureMLIndex":
717-
self.datasource = _AzureMLIndexSettings(settings=self, _env_file=DOTENV_PATH)
718-
logging.debug("Using Azure ML Index")
719-
720-
elif self.base_settings.datasource_type == "AzureSqlServer":
721-
self.datasource = _AzureSqlServerSettings(settings=self, _env_file=DOTENV_PATH)
722-
logging.debug("Using SQL Server")
723-
724394
else:
725395
self.datasource = None
726396
logging.warning("No datasource configuration found in the environment -- calls will be made to Azure OpenAI without grounding data.")

‎scripts/aml_config.json‎

Lines changed: 0 additions & 12 deletions
This file was deleted.

‎scripts/chunk_documents.yml‎

Lines changed: 0 additions & 25 deletions
This file was deleted.

0 commit comments

Comments
 (0)