#789

Committed 18 Sep 2023 04:32PM UTC coverage: 74.685% (-5.7%) from 80.421%

Build # #789

Build Type

push

circle-ci

Committed by Jiashen Cao

Commit Message

reformat

Run Details

8842 of 11839 relevant lines covered (74.69%)

0.75 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

69.05

/evadb/executor/executor_utils.py

# coding=utf-8
# Copyright 2018-2023 EvaDB
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import glob
import os
from pathlib import Path
from typing import TYPE_CHECKING, Generator, List

from evadb.catalog.catalog_utils import xform_column_definitions_to_catalog_entries
from evadb.catalog.models.utils import TableCatalogEntry
from evadb.parser.create_statement import ColumnDefinition

if TYPE_CHECKING:
    from evadb.catalog.catalog_manager import CatalogManager

from evadb.catalog.catalog_type import TableType, VectorStoreType
from evadb.expression.abstract_expression import AbstractExpression
from evadb.expression.function_expression import FunctionExpression
from evadb.models.storage.batch import Batch
from evadb.parser.table_ref import TableInfo
from evadb.parser.types import FileFormatType
from evadb.readers.document.registry import SUPPORTED_TYPES
from evadb.utils.generic_utils import try_to_import_cv2
from evadb.utils.logging_manager import logger


class ExecutorError(Exception):
    pass


def apply_project(
    batch: Batch, project_list: List[AbstractExpression], catalog: "CatalogManager"
):
    if not batch.empty() and project_list:
        batches = [expr.evaluate(batch) for expr in project_list]
        batch = Batch.merge_column_wise(batches)

        # persist stats of function expression
        for expr in project_list:
            for func_expr in expr.find_all(FunctionExpression):
                if func_expr.function_obj and func_expr._stats:
                    function_id = func_expr.function_obj.row_id
                    catalog.upsert_function_cost_catalog_entry(
                        function_id,
                        func_expr.function_obj.name,
                        func_expr._stats.prev_cost,
                    )
    return batch


def apply_predicate(
    batch: Batch, predicate: AbstractExpression, catalog: "CatalogManager"
) -> Batch:
    if not batch.empty() and predicate is not None:
        outcomes = predicate.evaluate(batch)
        batch.drop_zero(outcomes)
        batch.reset_index()

        # persist stats of function expression
        for func_expr in predicate.find_all(FunctionExpression):
            if func_expr.function_obj and func_expr._stats:
                function_id = func_expr.function_obj.row_id
                catalog.upsert_function_cost_catalog_entry(
                    function_id, func_expr.function_obj.name, func_expr._stats.prev_cost
                )
    return batch


def handle_if_not_exists(
    catalog: "CatalogManager", table_info: TableInfo, if_not_exist=False
):
    # Table exists
    if catalog.check_table_exists(
        table_info.table_name,
        table_info.database_name,
    ):
        err_msg = "Table: {} already exists".format(table_info)
        if if_not_exist:
            logger.warn(err_msg)
            return True
        else:
            logger.error(err_msg)
            raise ExecutorError(err_msg)
    # Table does not exist
    else:
        return False


def validate_image(image_path: Path) -> bool:
    try:
        try_to_import_cv2()
        import cv2

        data = cv2.imread(str(image_path))
        return data is not None
    except Exception as e:
        logger.warning(
            f"Unexpected Exception {e} occurred while reading image file {image_path}"
        )
        return False


def iter_path_regex(path_regex: Path) -> Generator[str, None, None]:
    return glob.iglob(os.path.expanduser(path_regex), recursive=True)


def validate_video(video_path: Path) -> bool:
    try:
        try_to_import_cv2()
        import cv2

        vid = cv2.VideoCapture(str(video_path))
        if not vid.isOpened():
            return False
        return True
    except Exception as e:
        logger.warning(
            f"Unexpected Exception {e} occurred while reading video file {video_path}"
        )


def validate_document(doc_path: Path) -> bool:
    return doc_path.suffix in SUPPORTED_TYPES


def validate_pdf(doc_path: Path) -> bool:
    return doc_path.suffix == ".pdf"


def validate_media(file_path: Path, media_type: FileFormatType) -> bool:
    if media_type == FileFormatType.VIDEO:
        return validate_video(file_path)
    elif media_type == FileFormatType.IMAGE:
        return validate_image(file_path)
    elif media_type == FileFormatType.DOCUMENT:
        return validate_document(file_path)
    elif media_type == FileFormatType.PDF:
        return validate_pdf(file_path)
    else:
        raise ValueError(f"Unsupported Media type {str(media_type)}")


def handle_vector_store_params(
    vector_store_type: VectorStoreType, index_path: str
) -> dict:
    """Handle vector store parameters based on the vector store type and index path.

    Args:
        vector_store_type (VectorStoreType): The type of vector store.
        index_path (str): The path to store the index.

    Returns:
        dict: Dictionary containing the appropriate vector store parameters.


    Raises:
        ValueError: If the vector store type in the node is not supported.
    """
    if vector_store_type == VectorStoreType.FAISS:
        return {"index_path": index_path}
    elif vector_store_type == VectorStoreType.QDRANT:
        return {"index_db": str(Path(index_path).parent)}
    else:
        raise ValueError("Unsupported vector store type: {}".format(vector_store_type))


def create_table_catalog_entry_for_native_table(
    table_info: TableInfo, column_list: List[ColumnDefinition]
):
    column_catalog_entires = xform_column_definitions_to_catalog_entries(column_list)

    # Assemble table.
    table_catalog_entry = TableCatalogEntry(
        name=table_info.table_name,
        file_url=None,
        table_type=TableType.NATIVE_DATA,
        columns=column_catalog_entires,
        database_name=table_info.database_name,
    )
    return table_catalog_entry

1	# coding=utf-8
2	# Copyright 2018-2023 EvaDB
3	#
4	# Licensed under the Apache License, Version 2.0 (the "License");
5	# you may not use this file except in compliance with the License.
6	# You may obtain a copy of the License at
7	#
8	# http://www.apache.org/licenses/LICENSE-2.0
9	#
10	# Unless required by applicable law or agreed to in writing, software
11	# distributed under the License is distributed on an "AS IS" BASIS,
12	# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13	# See the License for the specific language governing permissions and
14	# limitations under the License.
15	import glob	1✔
16	import os	1✔
17	from pathlib import Path	1✔
18	from typing import TYPE_CHECKING, Generator, List	1✔
19
20	from evadb.catalog.catalog_utils import xform_column_definitions_to_catalog_entries	1✔
21	from evadb.catalog.models.utils import TableCatalogEntry	1✔
22	from evadb.parser.create_statement import ColumnDefinition	1✔
23
24	if TYPE_CHECKING:
25	from evadb.catalog.catalog_manager import CatalogManager
26
27	from evadb.catalog.catalog_type import TableType, VectorStoreType	1✔
28	from evadb.expression.abstract_expression import AbstractExpression	1✔
29	from evadb.expression.function_expression import FunctionExpression	1✔
30	from evadb.models.storage.batch import Batch	1✔
31	from evadb.parser.table_ref import TableInfo	1✔
32	from evadb.parser.types import FileFormatType	1✔
33	from evadb.readers.document.registry import SUPPORTED_TYPES	1✔
34	from evadb.utils.generic_utils import try_to_import_cv2	1✔
35	from evadb.utils.logging_manager import logger	1✔
36
37
38	class ExecutorError(Exception):	1✔
39	pass	1✔
40
41
42	def apply_project(	1✔
43	batch: Batch, project_list: List[AbstractExpression], catalog: "CatalogManager"
44	):
45	if not batch.empty() and project_list:	1✔
46	batches = [expr.evaluate(batch) for expr in project_list]	1✔
47	batch = Batch.merge_column_wise(batches)	1✔
48
49	# persist stats of function expression
50	for expr in project_list:	1✔
51	for func_expr in expr.find_all(FunctionExpression):	1✔
52	if func_expr.function_obj and func_expr._stats:	1✔
53	function_id = func_expr.function_obj.row_id	1✔
54	catalog.upsert_function_cost_catalog_entry(	1✔
55	function_id,
56	func_expr.function_obj.name,
57	func_expr._stats.prev_cost,
58	)
59	return batch	1✔
60
61
62	def apply_predicate(	1✔
63	batch: Batch, predicate: AbstractExpression, catalog: "CatalogManager"
64	) -> Batch:
65	if not batch.empty() and predicate is not None:	1✔
66	outcomes = predicate.evaluate(batch)	1✔
67	batch.drop_zero(outcomes)	1✔
68	batch.reset_index()	1✔
69
70	# persist stats of function expression
71	for func_expr in predicate.find_all(FunctionExpression):	1✔
72	if func_expr.function_obj and func_expr._stats:	1✔
73	function_id = func_expr.function_obj.row_id	1✔
74	catalog.upsert_function_cost_catalog_entry(	1✔
75	function_id, func_expr.function_obj.name, func_expr._stats.prev_cost
76	)
77	return batch	1✔
78
79
80	def handle_if_not_exists(	1✔
81	catalog: "CatalogManager", table_info: TableInfo, if_not_exist=False
82	):
83	# Table exists
84	if catalog.check_table_exists(	1✔
85	table_info.table_name,
86	table_info.database_name,
87	):
88	err_msg = "Table: {} already exists".format(table_info)	×
89	if if_not_exist:	×
90	logger.warn(err_msg)	×
91	return True	×
92	else:
93	logger.error(err_msg)	×
94	raise ExecutorError(err_msg)
95	# Table does not exist
96	else:
97	return False	1✔
98
99
100	def validate_image(image_path: Path) -> bool:	1✔
101	try:	×
102	try_to_import_cv2()	×
103	import cv2	×
104
105	data = cv2.imread(str(image_path))	×
106	return data is not None	×
107	except Exception as e:
108	logger.warning(
109	f"Unexpected Exception {e} occurred while reading image file {image_path}"
110	)
111	return False
112
113
114	def iter_path_regex(path_regex: Path) -> Generator[str, None, None]:	1✔
115	return glob.iglob(os.path.expanduser(path_regex), recursive=True)	1✔
116
117
118	def validate_video(video_path: Path) -> bool:	1✔
119	try:	1✔
120	try_to_import_cv2()	1✔
121	import cv2	1✔
122
123	vid = cv2.VideoCapture(str(video_path))	1✔
124	if not vid.isOpened():	1✔
125	return False	×
126	return True	1✔
127	except Exception as e:
128	logger.warning(
129	f"Unexpected Exception {e} occurred while reading video file {video_path}"
130	)
131
132
133	def validate_document(doc_path: Path) -> bool:	1✔
134	return doc_path.suffix in SUPPORTED_TYPES	×
135
136
137	def validate_pdf(doc_path: Path) -> bool:	1✔
138	return doc_path.suffix == ".pdf"	×
139
140
141	def validate_media(file_path: Path, media_type: FileFormatType) -> bool:	1✔
142	if media_type == FileFormatType.VIDEO:	1✔
143	return validate_video(file_path)	1✔
144	elif media_type == FileFormatType.IMAGE:	×
145	return validate_image(file_path)	×
146	elif media_type == FileFormatType.DOCUMENT:	×
147	return validate_document(file_path)	×
148	elif media_type == FileFormatType.PDF:	×
149	return validate_pdf(file_path)	×
150	else:
151	raise ValueError(f"Unsupported Media type {str(media_type)}")
152
153
154	def handle_vector_store_params(	1✔
155	vector_store_type: VectorStoreType, index_path: str
156	) -> dict:
157	"""Handle vector store parameters based on the vector store type and index path.
158
159	Args:
160	vector_store_type (VectorStoreType): The type of vector store.
161	index_path (str): The path to store the index.
162
163	Returns:
164	dict: Dictionary containing the appropriate vector store parameters.
165
166
167	Raises:
168	ValueError: If the vector store type in the node is not supported.
169	"""
170	if vector_store_type == VectorStoreType.FAISS:	×
171	return {"index_path": index_path}	×
172	elif vector_store_type == VectorStoreType.QDRANT:	×
173	return {"index_db": str(Path(index_path).parent)}	×
174	else:
175	raise ValueError("Unsupported vector store type: {}".format(vector_store_type))
176
177
178	def create_table_catalog_entry_for_native_table(	1✔
179	table_info: TableInfo, column_list: List[ColumnDefinition]
180	):
181	column_catalog_entires = xform_column_definitions_to_catalog_entries(column_list)	×
182
183	# Assemble table.
184	table_catalog_entry = TableCatalogEntry(	×
185	name=table_info.table_name,
186	file_url=None,
187	table_type=TableType.NATIVE_DATA,
188	columns=column_catalog_entires,
189	database_name=table_info.database_name,
190	)
191	return table_catalog_entry	×

georgia-tech-db / eva / #789

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous