dmart 1.4.17__py3-none-any.whl

data_adapters/file/health_check.py

@@ -0,0 +1,560 @@
+#!/usr/bin/env -S BACKEND_ENV=config.env python3
+
+import argparse
+import asyncio
+import json
+import os
+import shutil
+import sys
+import time
+from datetime import datetime
+from pathlib import Path
+
+from jsonschema.validators import Draft7Validator
+from redis.commands.search.query import Query
+from redis.commands.search.result import Result
+
+from api.managed.router import serve_request
+from models.core import Folder
+from utils import repository
+from data_adapters.adapter import data_adapter as db
+from data_adapters.file.custom_validations import get_schema_path
+from utils.helpers import camel_case
+from data_adapters.file.redis_services import RedisServices
+from models import core, api
+from models.enums import ContentType, RequestType, ResourceType
+from utils.settings import settings
+
+duplicated_entries : dict= {}
+
+key_entries: dict = {}
+MAX_INVALID_SIZE = 100
+
+# {"space_name": {"schema_name": SCHEMA_DATA_DICT}}
+spaces_schemas: dict[str, dict[str, dict]] = {}
+
+
+async def main(health_type: str, space_param: str, schemas_param: list):
+    try:
+        spaces = await db.get_spaces()
+        if space_param != "all" and space_param not in spaces:
+            print("space name is not found")
+            return
+        if space_param == "all":
+            for space in spaces:
+                await main(health_type, space, schemas_param)
+                return
+        space_obj = core.Space.model_validate_json(spaces[space_param])
+        if not space_obj.check_health:
+            print(f"EARLY EXIT, health check disabled for space {space_param}")
+            return
+        
+        await cleanup_spaces()
+        is_full: bool = True if not args.space or args.space == 'all' else False
+        print_header()
+        if health_type == 'soft':
+            print("Running soft healthcheck")
+            if not schemas_param and not is_full:
+                print('Add the space name and at least one schema')
+                return
+            if is_full:
+                params = await load_spaces_schemas(space_param)
+            else:
+                params = {space_param: schemas_param}
+            for space in params:
+                print(f'>>>> Processing {space:<10} <<<<')
+                before_time = time.time()
+                health_check : dict[str, list|dict] | None = {'invalid_folders': [], 'folders_report': {}}
+                for schema in params.get(space, []):
+                    health_check_res = await soft_health_check(space, schema)
+                    if health_check_res and health_check and isinstance(health_check['folders_report'], dict):
+                        health_check['folders_report'].update(health_check_res.get('folders_report', {}))
+                print_health_check(health_check)
+                await save_health_check_entry(health_check, space)
+                print(f'Completed in: {"{:.2f}".format(time.time() - before_time)} sec')
+
+        elif not health_type or health_type == 'hard':
+            print("Running hard healthcheck")
+            spaces = {space_param : {}}
+            if is_full:
+                spaces = await db.get_spaces()
+            for space in spaces:
+                print(f'>>>> Processing {space:<10} <<<<')
+                before_time = time.time()
+                health_check = await hard_health_check(space)
+                if health_check:
+                    await save_health_check_entry(health_check, space)
+                print_health_check(health_check)
+                print(f'Completed in: {"{:.2f}".format(time.time() - before_time)} sec')
+        else:
+            print("Wrong mode specify [soft or hard]")
+            return
+        await save_duplicated_entries()
+    finally:
+        await RedisServices().close_pool()
+
+
+def print_header() -> None:
+    print("{:<32} {:<6} {:<6}".format(
+        'subpath',
+        'valid',
+        'invalid')
+    )
+
+
+def print_health_check(health_check):
+    if health_check:
+        for schema_path, val in health_check.get('folders_report', {}).items():
+            valid = val.get('valid_entries', 0)
+            invalid = len(val.get('invalid_entries', []))
+            print("{:<32} {:<6} {:<6}".format(
+                schema_path,
+                valid,
+                invalid)
+            )
+            for one in val.get("invalid_entries", []):
+                print(f"\t\t\t\tInvalid item/issues: {one.get('shortname', 'n/a')}/"
+                      f"{','.join(one.get('issues', []))} - {one.get('exception','')}")
+    if health_check.get('invalid_folders'):
+        print('Invalid folders :')
+    for val in health_check.get('invalid_folders'):
+        print(f"\t\t\t\t {val}")
+
+
+async def load_spaces_schemas(for_space: str | None = None) -> dict:
+    global spaces_schemas
+    if spaces_schemas:
+        return spaces_schemas
+    spaces = await db.get_spaces()
+    if for_space and for_space != "all" and for_space in spaces:
+        spaces_schemas[for_space] = load_space_schemas(for_space)
+        return spaces_schemas
+    for space_name in spaces:
+        schemas = load_space_schemas(space_name)
+        if schemas:
+            spaces_schemas[space_name] = schemas
+    return spaces_schemas
+
+
+def load_space_schemas(space_name: str) -> dict[str, dict]:
+    schemas: dict[str, dict] = {}
+    schemas_path = Path(settings.spaces_folder / space_name / "schema" / ".dm")
+    if not schemas_path.is_dir():
+        return {}
+    for entry in os.scandir(schemas_path):
+        if not entry.is_dir():
+            continue
+
+        schema_path_meta = Path(schemas_path / entry.name / "meta.schema.json")
+        if not schema_path_meta.is_file():
+            continue
+        
+        schema_meta = json.loads(schema_path_meta.read_text())
+        if schema_meta.get("payload", {}).get('body'):
+            schema_path_body = get_schema_path(
+                space_name=space_name,
+                schema_shortname=schema_meta.get("payload").get('body'),
+            )
+            if schema_path_body.is_file():
+                schemas[schema_meta['shortname']] = json.loads(schema_path_body.read_text())
+    return schemas
+
+
+async def soft_health_check(
+        space_name: str,
+        schema_name: str,
+):
+    global spaces_schemas
+    if space_name not in spaces_schemas:
+        await load_spaces_schemas(space_name)
+        
+    schemas = spaces_schemas[space_name]
+    
+    limit = 1000
+    offset = 0
+    folders_report : dict = {}
+    async with RedisServices() as redis:
+        try:
+            ft_index = redis.ft(f"{space_name}:{schema_name}")
+            await ft_index.info()
+        except Exception:
+            if 'meta_schema' not in schema_name:
+                print(f"can't find index: `{space_name}:{schema_name}`")
+            return None
+        while True:
+            search_query = Query(query_string="*")
+            search_query.paging(offset, limit)
+            offset += limit
+            x = await ft_index.search(query=search_query)  # type: ignore
+            if x and isinstance(x, dict) and "results" in x:
+                res_data : list = [one["extra_attributes"]["$"] for one in x["results"] if "extra_attributes" in one]
+                if not res_data:
+                    break
+            else: 
+                break
+            for redis_doc_dict in res_data:
+                redis_doc_dict = json.loads(redis_doc_dict)
+                subpath = redis_doc_dict['subpath']
+                meta_doc_content = {}
+                payload_doc_content = {}
+                resource_class = getattr(
+                    sys.modules["models.core"],
+                    camel_case(redis_doc_dict["resource_type"]),
+                )
+                system_attributes = [
+                    "payload_string",
+                    "query_policies",
+                    "subpath",
+                    "resource_type",
+                    "meta_doc_id",
+                    "payload_doc_id",
+                ]
+                class_fields = resource_class.model_fields.keys()
+                for key, value in redis_doc_dict.items():
+                    if key in class_fields:
+                        meta_doc_content[key] = value
+                    elif key not in system_attributes:
+                        payload_doc_content[key] = value
+
+                if not payload_doc_content and redis_doc_dict.get("payload_doc_id"):
+                    payload_redis_doc = await redis.get_doc_by_id(
+                        redis_doc_dict["payload_doc_id"]
+                    )
+                    if payload_redis_doc:
+                        not_payload_attr = system_attributes + list(class_fields)
+                        for key, value in payload_redis_doc.items():
+                            if key not in not_payload_attr:
+                                payload_doc_content[key] = value
+
+                meta_doc_content["created_at"] = datetime.fromtimestamp(
+                    meta_doc_content["created_at"]
+                )
+                meta_doc_content["updated_at"] = datetime.fromtimestamp(
+                    meta_doc_content["updated_at"]
+                )
+                if not folders_report.get(subpath):
+                    folders_report[subpath] = {}
+
+                meta = None
+                status = {
+                    'is_valid': True,
+                    "invalid": {
+                        "issues": [],
+                        "uuid": redis_doc_dict.get("uuid"),
+                        "shortname": redis_doc_dict.get("shortname"),
+                        "resource_type": redis_doc_dict["resource_type"],
+                        "exception": ""
+                    }
+                }
+                try:
+                    meta = resource_class.model_validate(meta_doc_content)
+                except Exception as ex:
+                    status['is_valid'] = False
+                    if not isinstance(status, dict) and isinstance(status["invalid"], dict):
+                        status['invalid']['exception'] = str(ex)
+                        status['invalid']['issues'].append('meta')
+                if meta:
+                    try:
+                        if meta.payload and meta.payload.schema_shortname and payload_doc_content is not None:
+                            schema_dict: dict = schemas.get(meta.payload.schema_shortname, {})
+                            if schema_dict:
+                                Draft7Validator(schema_dict).validate(payload_doc_content)
+                            else:
+                                continue
+                        if(
+                            meta.payload.checksum and 
+                            meta.payload.client_checksum and 
+                            meta.payload.checksum != meta.payload.client_checksum
+                        ): 
+                            raise Exception(
+                                f"payload.checksum not equal payload.client_checksum {subpath}/{meta.shortname}"
+                            )
+                        if folders_report[subpath].get('valid_entries'):
+                            folders_report[subpath]['valid_entries'] += 1
+                        else:
+                            folders_report[subpath]['valid_entries'] = 1
+                        status['is_valid'] = True
+                    except Exception as ex:
+                        status['is_valid'] = False
+                        if not isinstance(status, dict) and isinstance(status["invalid"], dict):
+                            status['invalid']['exception'] = str(ex)
+                            status['invalid']['issues'].append('payload')
+
+                if not status['is_valid']:
+                    if not folders_report.get(subpath, {}).get('invalid_entries'):
+                        folders_report[subpath]['invalid_entries'] = []
+                    if meta_doc_content["shortname"] not in folders_report[redis_doc_dict['subpath']]["invalid_entries"]:
+                        if len(folders_report[redis_doc_dict['subpath']]["invalid_entries"]) >= MAX_INVALID_SIZE:
+                            break
+                        folders_report[redis_doc_dict['subpath']]["invalid_entries"].append(status.get('invalid'))
+
+                # uuid = redis_doc_dict['uuid'][:8]
+                # await collect_duplicated_with_key('uuid', uuid)
+                # if redis_doc_dict.get('slug'):
+                #     await collect_duplicated_with_key('slug', redis_doc_dict.get('slug'))
+
+    return {"invalid_folders": [], "folders_report": folders_report}
+
+
+async def collect_duplicated_with_key(key, value) -> None:
+    spaces = await db.get_spaces()
+    async with RedisServices() as redis:
+        for space_name, space_data in spaces.items():
+            space_data = json.loads(space_data)
+            try:
+                ft_index = redis.ft(f"{space_name}:meta")
+                await ft_index.info()
+            except Exception:
+                continue
+            search_query = Query(query_string=f"@{key}:{value}*")
+            search_query.paging(0, 1000)
+            x = await ft_index.search(query=search_query) # type: ignore
+            if x and isinstance(x, Result):
+                res_data: Result = x
+                for redis_doc_dict in res_data.docs:
+                    redis_doc_dict = json.loads(redis_doc_dict.json)
+                    if isinstance(redis_doc_dict, dict):
+                        if redis_doc_dict['subpath'] == '/':
+                            redis_doc_dict['subpath'] = ''
+                        loc = space_name + "/" + redis_doc_dict['subpath']
+                        if key not in key_entries:
+                            key_entries[key] = {}
+                        if not key_entries[key].get(value):
+                            key_entries[key][value] = loc
+                        else:
+                            if not duplicated_entries.get(key):
+                                duplicated_entries[key] = {}
+                            if not duplicated_entries[key].get(value) or \
+                                    key_entries[key][value] not in duplicated_entries[key][value]['loc']:
+                                duplicated_entries[key][value] = {}
+                                duplicated_entries[key][value]['loc'] = [key_entries[key][value]]
+                                duplicated_entries[key][value]['total'] = 1
+                            if loc not in duplicated_entries[key][value]['loc']:
+                                duplicated_entries[key][value]['total'] += 1
+                                duplicated_entries[key][value]['loc'].append(loc)
+
+
+async def hard_health_check(space_name: str):
+    spaces = await db.get_spaces()
+    if space_name not in spaces:
+        print("space name is not found")
+        return None
+    # remove checking check_health value from meta space
+    # space_obj = core.Space.model_validate_json(spaces[space_name])
+    # if not space_obj.check_health:
+    #     print(f"EARLY EXIT, health check disabled for space {space_name}")
+    #     return None
+
+    invalid_folders : list = []
+    folders_report: dict[str, dict] = {}
+    meta_folders_health: list = []
+
+    path = settings.spaces_folder / space_name
+
+    subpaths = os.scandir(path)
+    # print(f"{path=} {subpaths=}")
+    for subpath in subpaths:
+        if subpath.is_file():
+            continue
+
+        await repository.validate_subpath_data(
+            space_name=space_name,
+            subpath=subpath.path,
+            user_shortname='dmart',
+            invalid_folders=invalid_folders,
+            folders_report=folders_report,
+            meta_folders_health=meta_folders_health,
+            max_invalid_size=MAX_INVALID_SIZE
+        )
+    res = {"invalid_folders": invalid_folders, "folders_report": folders_report}
+    if meta_folders_health:
+        res['invalid_folders'] = meta_folders_health
+    return res
+
+
+async def save_health_check_entry(health_check, space_name: str):
+    meta_path = Path(settings.spaces_folder / "management/health_check/.dm" / space_name)
+    entry_path = Path(settings.spaces_folder / "management/health_check" / f"{space_name}.json")
+    if meta_path.is_dir():
+        shutil.rmtree(meta_path)
+    if entry_path.is_file():
+        os.remove(entry_path)
+    await serve_request(
+        request=api.Request(
+            space_name="management",
+            request_type=RequestType.create,
+            records=[
+                core.Record(
+                    resource_type=ResourceType.content,
+                    shortname=space_name,
+                    subpath="/health_check",
+                    attributes={
+                        "is_active": True,
+                        "updated_at": str(datetime.now()),
+                        "payload": {
+                            "schema_shortname": "health_check",
+                            "content_type": ContentType.json,
+                            "body": health_check
+                        }
+                    },
+                )
+            ],
+        ),
+        owner_shortname='dmart',
+    )
+
+
+async def save_duplicated_entries() -> None:
+    print('>>>> Processing UUID duplication <<<<')
+    before_time = time.time()
+    uuid_scanned_entries: set = set()
+    uuid_duplicated_entries: dict = {}
+    
+    slug_scanned_entries = set()
+    slug_duplicated_entries: dict = {}
+    spaces : dict = await db.get_spaces()
+    async with RedisServices() as redis:
+        for space_name, space_data in spaces.items():
+            space_data = json.loads(space_data)
+            try:
+                ft_index = redis.ft(f"{space_name}:meta")
+                index_info = await ft_index.info() # type: ignore
+            except Exception:
+                continue
+            for i in range(0, int(index_info["num_docs"]), 10000):  # type: ignore
+                search_query = Query(query_string="*")
+                search_query.paging(i, 10000)
+                x = await ft_index.search(query=search_query) # type: ignore
+                if x and isinstance(x, dict) and "results" in x:
+                    res_data : list = [ one["extra_attributes"]["$"] for one in x["results"] if 'extra_attributes' in one]
+                    for redis_doc_dict in res_data:
+                        redis_doc_dict = json.loads(redis_doc_dict)
+                        if isinstance(redis_doc_dict, dict):
+                            if "uuid" in redis_doc_dict:
+                                # Handle UUID
+                                if "uuid" in redis_doc_dict and redis_doc_dict["uuid"] in uuid_scanned_entries:
+                                    short_uuid = redis_doc_dict["uuid"][:8]
+                                    uuid_duplicated_entries.setdefault(
+                                        short_uuid, {"loc": [], "total": 0}
+                                    )
+                                    uuid_duplicated_entries[short_uuid]["loc"].append(
+                                        space_name + "/" + redis_doc_dict['subpath'] + "/" + redis_doc_dict['shortname']
+                                    )
+                                    uuid_duplicated_entries[short_uuid]["total"]+=1
+                                else:
+                                    uuid_scanned_entries.add(redis_doc_dict["uuid"])
+                            else:
+                                print ("UUID is missing", redis_doc_dict)
+                        
+                            # Handle Slug
+                            if "slug" in redis_doc_dict and redis_doc_dict["slug"] in slug_scanned_entries:
+                                slug_duplicated_entries.setdefault(
+                                    "slug", {"loc": [], "total": 0}
+                                )
+                                slug_duplicated_entries["slug"]["loc"].append(
+                                    space_name + "/" + redis_doc_dict['subpath'] + "/" + redis_doc_dict['shortname']
+                                )
+                                slug_duplicated_entries["slug"]["total"]+=1
+                            elif "slug" in redis_doc_dict:
+                                slug_scanned_entries.add(redis_doc_dict["slug"])
+                        else:
+                            print("Loaded document is not a proper dictionary")
+                        
+                        
+
+
+    entry_path = Path(settings.spaces_folder / "management/health_check/.dm/duplicated_entries/meta.content.json")
+    request_type = RequestType.create
+    if entry_path.is_file():
+        request_type = RequestType.update
+    await serve_request(
+        request=api.Request(
+            space_name=settings.management_space,
+            request_type=request_type,
+            records=[
+                core.Record(
+                    resource_type=ResourceType.content,
+                    shortname="duplicated_entries",
+                    subpath="/health_check",
+                    attributes={
+                        "is_active": True,
+                        "updated_at": str(datetime.now()),
+                        "payload": {
+                            "schema_shortname": "health_check",
+                            "content_type": ContentType.json,
+                            "body": {
+                                "entries": {
+                                    "uuid": uuid_duplicated_entries,
+                                    "slug": slug_duplicated_entries
+                                }
+                            }
+                        }
+                    },
+                )
+            ],
+        ),
+        owner_shortname='dmart',
+    )
+    
+    print(f'Completed in: {"{:.2f}".format(time.time() - before_time)} sec')
+    
+
+async def cleanup_spaces() -> None:
+    spaces = await db.get_spaces()
+    # create health check path meta
+    folder_path = Path(settings.spaces_folder / "management/health_check/.dm")
+    if not os.path.isdir(folder_path):
+        os.makedirs(folder_path)
+    file_path = Path(folder_path / "meta.folder.json")
+    # create meta folder
+    if not os.path.isfile(file_path):
+        meta_obj = Folder(shortname="health_check", is_active=True, owner_shortname='dmart')
+        with open(file_path, "w") as f:
+            f.write(meta_obj.model_dump_json(exclude_none=True))
+    # create health check schema
+    if (
+            not os.path.isfile(Path(settings.spaces_folder / "management/schema/health_check.json")) or
+            not os.path.isfile(Path(settings.spaces_folder / "management/schema/.dm/health_check/meta.schema.json"))
+    ):
+        meta = core.Schema(
+            shortname="health_check",
+            is_active=True,
+            owner_shortname="dmart",
+            payload=core.Payload(
+                content_type=ContentType.json,
+                body="health_check.json",
+            ),
+        )
+        schema = {
+            "type": "object",
+            "title": "health_check",
+            "additionalProperties": True,
+            "properties": {
+            },
+            "required": []
+        }
+        await db.save("management", "schema", meta)
+        await db.save_payload_from_json("management", "schema", meta, schema)
+
+    # clean up entries
+    for folder_name in os.listdir(folder_path):
+        if not os.path.isdir(os.path.join(folder_path, folder_name)):
+            continue
+        if folder_name not in spaces:
+            shutil.rmtree(Path(folder_path / folder_name))
+            os.remove(Path(settings.spaces_folder / "management/health_check" / f"{folder_name}.json"))
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(
+        description="This created for doing health check functionality",
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter,
+    )
+    parser.add_argument("-t", "--type", help="type of health check (soft or hard)")
+    parser.add_argument("-s", "--space", help="hit the target space or pass (all) to make the full health check")
+    parser.add_argument("-m", "--schemas", nargs="*", help="hit the target schema inside the space")
+
+    args = parser.parse_args()
+    before_time = time.time()
+    asyncio.run(main(args.type, args.space or "all", args.schemas))
+    print(f'total time: {"{:.2f}".format(time.time() - before_time)} sec')