neptune/core/db/__init__.py

import random
import string
import time
from abc import ABC, abstractmethod
from math import floor, log10

import orjson
from django.conf import settings
from siphashc import siphash

from core import r
from core.db.processing import annotate_results
from core.util import logs


def remove_defaults(query_params):
    for field, value in list(query_params.items()):
        if field in settings.DRILLDOWN_DEFAULT_PARAMS:
            if value == settings.DRILLDOWN_DEFAULT_PARAMS[field]:
                del query_params[field]


def add_defaults(query_params):
    for field, value in settings.DRILLDOWN_DEFAULT_PARAMS.items():
        if field not in query_params:
            query_params[field] = value


def dedup_list(data, check_keys):
    """
    Remove duplicate dictionaries from list.
    """
    seen = set()
    out = []

    dup_count = 0
    for x in data:
        dedupeKey = tuple(x[k] for k in check_keys if k in x)
        if dedupeKey in seen:
            dup_count += 1
            continue
        if dup_count > 0:
            out.append({"type": "control", "hidden": dup_count})
            dup_count = 0
        out.append(x)
        seen.add(dedupeKey)
    if dup_count > 0:
        out.append({"type": "control", "hidden": dup_count})
    return out


class StorageBackend(ABC):
    def __init__(self, name):
        self.log = logs.get_logger(name)
        self.log.info(f"Initialising storage backend {name}")

        self.initialise_caching()
        self.initialise()

    @abstractmethod
    def initialise(self, **kwargs):
        pass

    def initialise_caching(self):
        hash_key = r.get("cache_hash_key")
        if not hash_key:
            letters = string.ascii_lowercase
            hash_key = "".join(random.choice(letters) for i in range(16))
            self.log.debug(f"Created new hash key: {hash_key}")
            r.set("cache_hash_key", hash_key)
        else:
            hash_key = hash_key.decode("ascii")
            self.log.debug(f"Decoded hash key: {hash_key}")
        self.hash_key = hash_key

    @abstractmethod
    def construct_query(self, **kwargs):
        pass

    def parse_query(self, query_params, tags, size, custom_query, add_bool, **kwargs):
        query_created = False
        if "query" in query_params:
            query = query_params["query"]
            search_query = self.construct_query(query, size, **kwargs)
            query_created = True
        else:
            if custom_query:
                search_query = custom_query
            else:
                search_query = self.construct_query(None, size, blank=True, **kwargs)

        if tags:
            # Get a blank search query
            if not query_created:
                search_query = self.construct_query(None, size, blank=True, **kwargs)
                query_created = True
            for item in tags:
                for tagname, tagvalue in item.items():
                    add_bool.append({tagname: tagvalue})

        bypass_check = kwargs.get("bypass_check", False)
        if not bypass_check:
            valid = self.check_valid_query(query_params, custom_query, **kwargs)
            if isinstance(valid, dict):
                return valid

        return search_query

    def check_valid_query(self, query_params, custom_query):
        required_any = ["query", "tags"]
        if not any([field in query_params.keys() for field in required_any]):
            if not custom_query:
                message = "Empty query!"
                message_class = "warning"
                return {"message": message, "class": message_class}

    @abstractmethod
    def run_query(self, **kwargs):
        pass

    def filter_blacklisted(self, user, response):
        """
        Low level filter to take the raw search response and remove
        objects from it we want to keep secret.
        Does not return, the object is mutated in place.
        """
        response["redacted"] = 0
        response["exemption"] = None
        if user.is_superuser:
            response["exemption"] = True
        # is_anonymous = isinstance(user, AnonymousUser)
        # For every hit from ES
        for index, item in enumerate(list(response["hits"]["hits"])):
            # For every blacklisted type
            for blacklisted_type in settings.ELASTICSEARCH_BLACKLISTED.keys():
                # Check this field we are matching exists
                if "_source" in item.keys():
                    data_index = "_source"
                elif "fields" in item.keys():
                    data_index = "fields"
                else:
                    return False
                if blacklisted_type in item[data_index].keys():
                    content = item[data_index][blacklisted_type]
                    # For every item in the blacklisted array for the type
                    for blacklisted_item in settings.BLACKLISTED[blacklisted_type]:
                        if blacklisted_item == str(content):
                            # Remove the item
                            if item in response["hits"]["hits"]:
                                # Let the UI know something was redacted
                                if (
                                    "exemption"
                                    not in response["hits"]["hits"][index][data_index]
                                ):
                                    response["redacted"] += 1
                                # Anonymous
                                if user.is_anonymous:
                                    # Just set it to none so the index is not off
                                    response["hits"]["hits"][index] = None
                                else:
                                    if not user.has_perm("core.bypass_blacklist"):
                                        response["hits"]["hits"][index] = None
                                    else:
                                        response["hits"]["hits"][index][data_index][
                                            "exemption"
                                        ] = True

        # Actually get rid of all the things we set to None
        response["hits"]["hits"] = [hit for hit in response["hits"]["hits"] if hit]

    def query(self, user, search_query, **kwargs):
        # For time tracking
        start = time.process_time()
        if settings.CACHE:
            # Sort the keys so the hash is the same
            query_normalised = orjson.dumps(search_query, option=orjson.OPT_SORT_KEYS)
            hash = siphash(self.hash_key, query_normalised)
            cache_hit = r.get(f"query_cache.{user.id}.{hash}")
            if cache_hit:
                response = orjson.loads(cache_hit)
                time_took = (time.process_time() - start) * 1000
                # Round to 3 significant figures
                time_took_rounded = round(
                    time_took, 3 - int(floor(log10(abs(time_took)))) - 1
                )
                return {
                    "object_list": response,
                    "took": time_took_rounded,
                    "cache": True,
                }
        response = self.run_query(user, search_query, **kwargs)

        # For Elasticsearch
        if isinstance(response, Exception):
            message = f"Error: {response.info['error']['root_cause'][0]['type']}"
            message_class = "danger"
            return {"message": message, "class": message_class}
        if len(response["hits"]["hits"]) == 0:
            message = "No results."
            message_class = "danger"
            return {"message": message, "class": message_class}

        # For Druid
        if "error" in response:
            if "errorMessage" in response:
                context = {
                    "message": response["errorMessage"],
                    "class": "danger",
                }
                return context
            else:
                return response
        if "took" in response:
            if response["took"] is None:
                return None

        # Removed for now, no point given we have restricted indexes
        # self.filter_blacklisted(user, response)

        # Parse the response
        response_parsed = self.parse(response)

        # Write cache
        if settings.CACHE:
            to_write_cache = orjson.dumps(response_parsed)
            r.set(f"query_cache.{user.id}.{hash}", to_write_cache)
            r.expire(f"query_cache.{user.id}.{hash}", settings.CACHE_TIMEOUT)

        time_took = (time.process_time() - start) * 1000
        # Round to 3 significant figures
        time_took_rounded = round(time_took, 3 - int(floor(log10(abs(time_took)))) - 1)
        return {"object_list": response_parsed, "took": time_took_rounded}

    @abstractmethod
    def query_results(self, **kwargs):
        pass

    def process_results(self, response, **kwargs):
        if kwargs.get("annotate"):
            annotate_results(response)
        if kwargs.get("reverse"):
            response.reverse()
        if kwargs.get("dedup"):
            dedup_fields = kwargs.get("dedup_fields")
            if not dedup_fields:
                dedup_fields = ["msg", "nick", "ident", "host", "net", "channel"]
            response = dedup_list(response, dedup_fields)
        return response

    @abstractmethod
    def parse(self, response):
        pass