Zendesk v2

from collections import deque
from collections.abc import Callable, Iterable, Iterator
from copy import copy
from functools import lru_cache

import demistomock as demisto
from CommonServerPython import *
from requests import Response
from requests.exceptions import HTTPError
from urllib3 import disable_warnings

STR_OR_STR_LIST = str | list[str]
MAX_PAGE_SIZE = 100
USER_CONTEXT_PATH = "Zendesk.User"
USERS_HEADERS = ["id", "name", "email", "role", "active", "external_id", "created_at", "updated_at"]
ORGANIZATIONS_HEADERS = ["id", "name", "domain_names", "tags", "external_id", "created_at", "updated_at"]
TICKETS_HEADERS = ["id", "subject", "description", "priority", "status", "assignee_id", "created_at", "updated_at", "external_id"]
COMMENTS_HEADERS = ["id", "body", "created_at", "public", "attachments"]
ATTACHMENTS_HEADERS = ["id", "file_name", "content_url", "size", "content_type"]
GROUP_USER_HEADERS = ["id", "name", "email", "role", "created_at"]
GROUP_HEADERS = ["id", "name", "is_public", "created_at", "updated_at"]
ARTICLES_HEADERS = ["body"]
ROLES = ["end-user", "admin", "agent"]
ROLE_TYPES = {
    "custom_agent": 0,
    "light_agent": 1,
    "chat_agent": 2,
    "chat_agent_contributor": 3,
    "admin": 4,
    "billing_admin": 5,
}
TICKET_FILTERS = ["assigned", "requested", "ccd", "followed", "recent"]
CURSOR_SORTS = {
    "id_asc": "id",
    "status_asc": "status",
    "updated_at_asc": "updated_at",
    "id_desc": "-id",
    "status_desc": "-status",
    "updated_at_desc": "-updated_at",
}
TICKET_TYPE = ["problem", "incident", "question", "task"]
TICKET_STATUS = ["open", "pending", "hold", "solved", "closed"]
TICKET_PRIORITY = ["urgent", "high", "normal", "low"]
PRIORITY_MAP = {
    "urgent": IncidentSeverity.CRITICAL,
    "high": IncidentSeverity.HIGH,
    "normal": IncidentSeverity.MEDIUM,
    "low": IncidentSeverity.LOW,
}
params = demisto.params()  # pylint: disable=W9016
MIRROR_USER_AGENT = "XSOAR mirror"
MIRROR_DIRECTION = {"Incoming": "In", "Outgoing": "Out", "Incoming And Outgoing": "Both"}.get(params.get("mirror_direction"))
FIELDS_TO_REMOVE_FROM_MIROR_IN = ["url", "id", "created_at"]
DEFAULT_UPLOAD_FILES_COMMENT = "Uploaded from XSOAR."
MIRROR_TAGS = params.get("mirror_tags") or []
CLOSE_INCIDENT = argToBoolean(params.get("close_incident", False))
INTEGRATION_INSTANCE = demisto.integrationInstance()
CACHE = None
ZENDESK_FETCH_TIME_FORMAT = "%Y-%m-%dT%H:%M:00Z"


class CacheManager:
    def __init__(self, zendesk_client):
        self._data = None
        self._zendesk_client: ZendeskClient = zendesk_client

    @staticmethod
    def zendesk_clear_cache(**kwargs):
        demisto.setIntegrationContext({})
        return "Cache clear done."

    def save(self):
        if self._data:
            demisto.setIntegrationContext(self._data)

    def replace_ids_change(self, data: dict, organization_fields: list[str] = [], user_fields: list[str] = []):
        for fields, get_func in [(organization_fields, self.organization), (user_fields, self.user)]:
            for field in fields:
                obj_id = data.get(field)
                if obj_id:
                    field = field.replace("_id", "")
                    if isinstance(obj_id, list):
                        data[field] = list(map(get_func, obj_id))
                    else:
                        data[field] = get_func(obj_id)
        return data

    @property
    def data(self):
        if self._data is None:
            self._data = demisto.getIntegrationContext()
        return self._data

    @lru_cache
    def user(self, user_id: int) -> str:
        return self._generic_get_by_id("users", user_id, self._zendesk_client._get_user_by_id, "email")

    @lru_cache
    def organization(self, organization_id: int) -> str:
        return self._generic_get_by_id("organizations", organization_id, self._zendesk_client._get_organization_by_id, "name")

    @lru_cache
    def organization_name(self, organization_name: str) -> int | dict:
        organizations = self._zendesk_client._get_organizations_by_name(organization_name)
        ids = ",".join(str(x["id"]) for x in organizations)
        assert len(organizations) == 1, f"found {len(organizations)} organizations with name {organization_name} and ids {ids}"

        return organizations[0]["id"]

    def _generic_get_by_id(self, data_type: str, obj_id: int, data_get: Callable, val_field: str):
        self.data[data_type] = self.data.get(data_type, {})
        try:
            return self.data[data_type][obj_id]
        except KeyError:
            pass

        try:
            user_email = data_get(obj_id)[val_field] or obj_id
            self.data[data_type][obj_id] = user_email
            return user_email
        except:  # noqa # lgtm[py/]
            return obj_id


def datetime_to_iso(date: datetime) -> str:
    return date.strftime("%Y-%m-%dT%H:%M:%SZ")


def prepare_kwargs(
    kwargs: dict[str, Any],
    ignore_args: STR_OR_STR_LIST = [],
    str_args: STR_OR_STR_LIST = [],
    list_args: STR_OR_STR_LIST = [],
    bool_args: STR_OR_STR_LIST = [],
    int_args: STR_OR_STR_LIST = [],
    json_args: STR_OR_STR_LIST = [],
) -> dict[str, Any]:
    return_kwargs = {}

    for arg in ignore_args if isinstance(ignore_args, list) else [ignore_args]:
        if arg in kwargs:
            return_kwargs[arg] = kwargs[arg]

    for arg in str_args if isinstance(str_args, list) else [str_args]:
        if arg in kwargs:
            return_kwargs[arg] = str(kwargs[arg])

    for arg in list_args if isinstance(list_args, list) else [list_args]:
        if arg in kwargs:
            return_kwargs[arg] = argToList(kwargs[arg])

    for arg in bool_args if isinstance(bool_args, list) else [bool_args]:
        if arg in kwargs:
            return_kwargs[arg] = argToBoolean(kwargs[arg])

    for arg in int_args if isinstance(int_args, list) else [int_args]:
        if arg in kwargs:
            return_kwargs[arg] = int(kwargs[arg])

    for arg in json_args if isinstance(json_args, list) else [json_args]:
        if arg in kwargs:
            return_kwargs[arg] = kwargs[arg] if isinstance(kwargs[arg], dict) else json.loads(kwargs[arg])

    return return_kwargs


def error_entry(error_msg: str) -> dict[str, Any]:
    return {
        "Type": EntryType.ERROR,
        "ContentsFormat": EntryFormat.TEXT,
        "Contents": error_msg,
    }


def close_entry(reason: str) -> dict[str, Any]:
    return {
        "Type": EntryType.NOTE,
        "Contents": {"dbotIncidentClose": True, "closeReason": reason},
        "ContentsFormat": EntryFormat.JSON,
    }


class Validators:
    @staticmethod
    def _validate(val: Any, arg_name: str, allowed: Iterable[Any]):
        copy_value = argToList(val)
        for value in copy_value:
            assert value in allowed, f"'{val}' is not a valid {arg_name}.\nallowed {arg_name}s are '{','.join(allowed)}'"

    @staticmethod
    def validate_role(role: STR_OR_STR_LIST):
        Validators._validate(role, "role", ROLES)

    @staticmethod
    def validate_role_type(role_type: str):
        Validators._validate(role_type, "role type", ROLE_TYPES.keys())

    @staticmethod
    def validate_ticket_filter(ticket_filter: str):
        Validators._validate(ticket_filter, "filter", TICKET_FILTERS)

    @staticmethod
    def validate_ticket_sort(ticket_sort: str):
        Validators._validate(ticket_sort, "sort", CURSOR_SORTS.keys())

    @staticmethod
    def validate_ticket_type(ticket_type: str):
        Validators._validate(ticket_type, "type", TICKET_TYPE)

    @staticmethod
    def validate_ticket_status(ticket_status: str):
        Validators._validate(ticket_status, "status", TICKET_STATUS)

    @staticmethod
    def validate_ticket_priority(ticket_priority: str):
        Validators._validate(ticket_priority, "priority", TICKET_PRIORITY)


class TicketEvents:
    def __init__(self, zendesk_client, after_cursor: str = None, tickets_list: list = []):
        self._client = zendesk_client
        self._demisto_params = demisto.params()  # pylint: disable=W9016
        self._tickets_list = tickets_list
        self._after_cursor = after_cursor
        self._last_fetch: dict = {}

    def _get_all(self, **kwargs):
        return self._client._http_request("GET", url_suffix="incremental/tickets/cursor", params=kwargs)

    def next_run(self):
        next_run = {
            "after_cursor": self._last_fetch.get("after_cursor", self._after_cursor),
        }
        if self._tickets_list:
            next_run["tickets"] = self._tickets_list
        return next_run

    @abstractmethod
    def query_params(self):
        if after_cursor := self._last_fetch.get("after_cursor", self._after_cursor):
            return {"cursor": after_cursor}
        return {}

    def tickets(self, limit: int = 1000, params: dict | None = {}):
        yielded = 0
        if self._tickets_list:
            for _ in range(min(limit, len(self._tickets_list))):
                yield self._tickets_list.pop(0)
                yielded += 1
            if yielded >= limit:
                return

        res = self._get_all(**(self.query_params() | params))
        self._tickets_list = res.get("tickets", [])
        while True:
            self._after_cursor = res.get("after_cursor") or self._after_cursor
            for _ in range(min(limit - yielded, len(self._tickets_list))):
                yield self._tickets_list.pop(0)
                yielded += 1

            if res["end_of_stream"]:
                return
            res = self._get_all(**(self.query_params() | params))


class UpdatedTickets(TicketEvents):
    def __init__(self, zendesk_client, last_update: int, last_run_data: dict = {}):
        self._last_update = last_update
        super().__init__(zendesk_client, last_run_data.get("after_cursor"))

    def query_params(self):
        params = super().query_params()
        if not params:
            params["start_time"] = self._last_update
        return params

    def tickets(self):  # type: ignore[override]
        def filter_created_ticket(ticket: dict):
            return ticket["created_at"] != ticket["updated_at"]

        yield from filter(filter_created_ticket, super().tickets())


class ZendeskClient(BaseClient):
    def __init__(
        self, base_url: str, username: str | None = None, password: str | None = None, proxy: bool = False, verify: bool = True
    ):
        base_url = urljoin(base_url, "/api/v2/")
        auth = headers = None
        if username and password:
            # auth = (f'{username}/token', password)
            auth = (username, password)
        elif password:
            headers = {"Authorization": f"Bearer {password}"}

        super().__init__(base_url, auth=auth, proxy=proxy, verify=verify, headers=headers)

    @staticmethod
    def error_handler(res: Response) -> None:
        if res.text:
            raise DemistoException(f"Error occurred in Zendesk API: {res.text}")
        res.raise_for_status()

    def _http_request(  # type: ignore[override]
        self,
        method: str,
        url_suffix: str = "",
        full_url: str | None = None,  # type: ignore[override]
        json_data: dict | None = None,
        params: dict = None,
        data: dict = None,
        content: bytes = None,
        resp_type: str = "json",
        return_empty_response: bool = False,
        **kwargs,
    ):
        if params:
            final_params_list = []
            for k, v in params.items():
                if isinstance(v, list):
                    for singel_v in v:
                        final_params_list.append(f"{k}[]={singel_v}")
                else:
                    final_params_list.append(f"{k}={v}")
            params_str = f'?{"&".join(final_params_list)}'
            if url_suffix:
                url_suffix = f"{url_suffix}{params_str}"
            if full_url:
                full_url = f"{full_url}{params_str}"

        return super()._http_request(
            method,
            url_suffix=url_suffix,
            full_url=full_url,
            json_data=json_data,
            data=data or content,
            return_empty_response=return_empty_response,
            resp_type=resp_type,
            error_handler=self.error_handler,
            **kwargs,
        )

    def __cursor_pagination(
        self, url_suffix: str, data_field_name: str, params: dict | None = None, limit: int = 50
    ) -> Iterator[dict]:
        # API docs here https://developer.zendesk.com/api-reference/ticketing/introduction/#using-cursor-pagination
        page_size = min(limit, MAX_PAGE_SIZE)
        next_link_section = "next"
        count_data = 1
        paged_params = copy(params) if params is not None else {}
        paged_params["page[size]"] = page_size
        res = self._http_request("GET", url_suffix=url_suffix, params=paged_params)
        while True:
            for i in res[data_field_name]:
                yield i
                count_data += 1
                if count_data > limit:
                    return

            if not dict_safe_get(res, ["meta", "has_more"]):
                break

            res = self._http_request("GET", full_url=res["links"][next_link_section])

    def __get_spesific_page(
        self, url_suffix: str, data_field_name: str, page_size: int, page_number: int, params: dict | None = None
    ) -> Iterator[dict]:
        # API docs here https://developer.zendesk.com/api-reference/ticketing/introduction/#using-offset-pagination
        page_size = min(page_size, MAX_PAGE_SIZE)
        paged_params = copy(params) if params is not None else {}
        paged_params["per_page"] = page_size
        paged_params["page"] = page_number
        yield from self._http_request("GET", url_suffix=url_suffix, params=paged_params)[data_field_name]

    def _paged_request(
        self,
        url_suffix: str,
        data_field_name: str,
        params: dict | None = None,
        limit: int = 50,
        page_size: int | None = None,
        page_number: int | None = None,
    ) -> Iterator[dict]:
        # validate parameters
        if page_size is not None and page_number is not None:
            return self.__get_spesific_page(
                url_suffix=url_suffix,
                data_field_name=data_field_name,
                params=params,
                page_size=int(page_size),
                page_number=int(page_number),
            )
        elif page_size is not None or page_number is not None:
            raise AssertionError("you need to specify both 'page_size' and 'page_number'.")
        else:
            return self.__cursor_pagination(
                url_suffix=url_suffix, data_field_name=data_field_name, params=params, limit=int(limit)
            )

    # ---- user related functions ---- #

    @staticmethod
    def __command_results_zendesk_users(users: list[dict]):
        role_types_reverse = {int_k: str_k for str_k, int_k in ROLE_TYPES.items()}

        def _iter_context(user: dict):
            user = CACHE.replace_ids_change(user, ["organization_id"])  # type: ignore
            role_type = role_types_reverse.get(user.get("role_type"))  # type: ignore
            if role_type:
                user["role_type"] = role_type
            return user

        raw_results = copy(users)
        context = list(map(_iter_context, users))
        readable_outputs = tableToMarkdown(
            name="Zendek users:", t=context, headers=USERS_HEADERS, headerTransform=camelize_string
        )
        return CommandResults(
            outputs_prefix=USER_CONTEXT_PATH, outputs=context, readable_output=readable_outputs, raw_response=raw_results
        )

    def _get_user_by_id(self, user_id: str):
        return self._http_request("GET", f"users/{user_id}")["user"]

    def zendesk_user_list(
        self, user_id: STR_OR_STR_LIST | None = None, user_name: str | None = None, role: list[str] | str | None = None, **kwargs
    ):
        users_field_name = "users"
        results = []
        error_msgs = []

        if user_id is not None:
            users_list = []
            for single_user in argToList(user_id):
                try:
                    users_list.append(self._get_user_by_id(single_user))
                except Exception as e:
                    demisto.error(f"could not retrieve user: {single_user}\n{traceback.format_exc()}")
                    error_msgs.append(f"could not retrieve user: {single_user}\n{e}")
        elif user_name is not None:
            users_list = self._http_request("GET", "users/autocomplete", params={"name": user_name})[users_field_name]
        else:
            params = prepare_kwargs(kwargs=kwargs, str_args="external_id")
            if role:
                role_list = argToList(role)
                Validators.validate_role(role_list)
                params["role"] = role_list[0] if len(role_list) == 1 else role_list
            users_list = list(self._paged_request("users", "users", params=params, **kwargs))

        if users_list:
            results.append(self.__command_results_zendesk_users(users_list))

        if error_msgs:
            results.append(error_entry("\n".join(error_msgs)))

        return results if results else "No outputs."

    @staticmethod
    def _handle_role_argument(role: str | None = None, role_type: str | None = None) -> dict[str, Any]:
        role_params: dict[str, str | int] = {}
        if role:
            Validators.validate_role(role)
            role_params["role"] = role
            if role_type is not None:
                assert role == "agent", "You cannot use the 'role_type' argument if the selected role is not 'agent'"
                Validators.validate_role_type(role_type)
                role_params["role_type"] = ROLE_TYPES[role_type]
        return role_params

    def zendesk_user_create(
        self,
        name: str,
        email: str,
        role: str | None = None,
        role_type: str | None = None,
        check_if_user_exists: bool = False,
        **kwargs,
    ):
        url_suffix = "users/create" if argToBoolean(check_if_user_exists) else "users/create_or_update"

        user_body = {"name": name, "email": email}

        if "organization_name" in kwargs:
            assert "organization_id" not in kwargs, "you can specify 'organization_id' or 'organization_name' not both."
            kwargs["organization_id"] = CACHE.organization_name(kwargs.pop("organization_name"))  # type: ignore

        user_body.update(
            prepare_kwargs(
                kwargs=kwargs,
                str_args=["phone", "notes", "details", "external_id", "locale", "alias"],
                list_args="tags",
                int_args=["organization_id", "default_group_id", "custom_role_id"],
                bool_args="verified",
                json_args=["identities", "user_fields"],
            )
        )
        user_body.update(self._handle_role_argument(role=role, role_type=role_type))

        return self.__command_results_zendesk_users(
            [self._http_request("POST", url_suffix=url_suffix, json_data={"user": user_body})["user"]]
        )

    def zendesk_user_update(self, user_id: str, role: str | None = None, role_type: str | None = None, **kwargs):
        if "organization_name" in kwargs:
            assert "organization_id" not in kwargs, "you can specify 'organization_id' or 'organization_name' not both."
            kwargs["organization_id"] = CACHE.organization_name(kwargs.pop("organization_name"))  # type: ignore

        user_body = prepare_kwargs(
            kwargs=kwargs,
            str_args=["name", "email", "phone", "notes", "details", "external_id", "locale", "alias"],
            list_args="tags",
            int_args=["organization_id", "default_group_id", "custom_role_id"],
            bool_args=["verified", "suspended"],
            json_args=["identities", "user_fields"],
        )
        user_body.update(self._handle_role_argument(role=role or "agent", role_type=role_type))

        return self.__command_results_zendesk_users(
            [self._http_request("PUT", url_suffix=f"users/{user_id}", json_data={"user": user_body})["user"]]
        )

    def zendesk_user_delete(self, user_id: str, **kwargs):  # pragma: no cover
        self._http_request("DELETE", url_suffix=f"users/{user_id}")
        return f"User deleted. (id: {user_id})"

    # ---- organization related functions ---- #

    @staticmethod
    def __command_results_zendesk_organizations(organizations: list[dict]):  # pragma: no cover
        readable_outputs = tableToMarkdown(
            name="Zendek organizations:", t=organizations, headers=ORGANIZATIONS_HEADERS, headerTransform=camelize_string
        )
        return CommandResults(outputs_prefix="Zendesk.Organization", outputs=organizations, readable_output=readable_outputs)

    def _get_organization_by_id(self, organization_id: str) -> dict[str, Any]:
        return self._http_request("GET", f"organizations/{organization_id}")["organization"]

    def _get_organizations_by_name(self, organization_name: str) -> list[dict[str, Any]]:
        return self._http_request("GET", "organizations/autocomplete", params={"name": organization_name})["organizations"]

    def zendesk_organization_list(self, organization_id: str | None = None, **kwargs):
        if organization_id:
            organizations = [self._get_organization_by_id(organization_id)]
        else:
            organizations = list(self._paged_request(url_suffix="organizations", data_field_name="organizations", **kwargs))

        return self.__command_results_zendesk_organizations(organizations)

    # ---- group related functions ---- #
    @staticmethod
    def __command_results_zendesk_group_users(users: list[dict]):  # pragma: no cover
        readable_outputs = tableToMarkdown(
            name="Zendesk Group Users:", t=users, headers=GROUP_USER_HEADERS, headerTransform=camelize_string
        )
        return CommandResults(outputs_prefix="Zendesk.UserGroup", outputs=users, readable_output=readable_outputs)

    def list_group_users(self, group_id: int, **kwargs):
        users = list(self._paged_request(url_suffix=f"groups/{group_id}/users", data_field_name="users", **kwargs))
        return self.__command_results_zendesk_group_users(users)

    @staticmethod
    def __command_results_zendesk_groups(groups):  # pragma: no cover
        readable_outputs = tableToMarkdown(
            name="Zendesk groups:", t=groups, headers=GROUP_HEADERS, headerTransform=camelize_string
        )
        return CommandResults(outputs_prefix="Zendesk.Group", outputs=groups, readable_output=readable_outputs)

    def list_groups(self, **kwargs):
        groups = list(self._paged_request(url_suffix="groups", data_field_name="groups", **kwargs))
        return self.__command_results_zendesk_groups(groups)

    # ---- ticket related functions ---- #

    @staticmethod
    def __ticket_context(ticket: dict[str, Any]):
        return CACHE.replace_ids_change(  # type: ignore
            ticket,
            organization_fields=["organization_id"],  # type: ignore
            user_fields=["assignee_id", "collaborator_ids", "email_cc_ids", "follower_ids", "requester_id", "submitter_id"],
        )

    @staticmethod
    def __command_results_zendesk_tickets(tickets: list[dict]):
        raw = tickets
        context = list(map(ZendeskClient.__ticket_context, tickets))
        readable_outputs = tableToMarkdown(
            name="Zendek tickets:", t=context, headers=TICKETS_HEADERS, headerTransform=camelize_string
        )
        return CommandResults(
            outputs_prefix="Zendesk.Ticket", outputs=tickets, readable_output=readable_outputs, raw_response=raw
        )

    def _get_ticket_by_id(self, ticket_id: str, **kwargs):
        return self._http_request("GET", f"tickets/{ticket_id}", **kwargs)["ticket"]

    @staticmethod
    def __get_sort_params(sort: str, cursor_paging: bool = False):
        Validators.validate_ticket_sort(sort)
        if not cursor_paging:
            # using the offset paged request
            sort_list = sort.split("_")
            sort, order = "_".join(sort_list[:-1]), sort_list[-1]
            return {"sort_by": sort, "sort_order": order}
            # using the cursor paged request
        return {"sort": CURSOR_SORTS[sort]}

    @staticmethod
    def __get_tickets_url_suffix(filter: str, user_id: str | int | None = None) -> str:
        match filter:
            case None:
                return "tickets"
            case "recent":  # lgtm[py/unreachable-statement]
                return "tickets/recent"
            case _:  # lgtm[py/unreachable-statement]
                assert user_id, f"user_id is required when using '{filter}' as filter."
                Validators.validate_ticket_filter(filter)
                return f"/users/{user_id}/tickets/{filter}"

    def zendesk_ticket_list(
        self,
        ticket_id: STR_OR_STR_LIST | None = None,
        query: str | None = None,
        user_id: str | None = None,
        sort: str | None = None,
        page_number: int | None = None,
        **kwargs,
    ):
        filter_ = kwargs.pop("filter", None)
        error_msgs = []
        command_results = []
        if query is not None:
            assert ticket_id is None, "please provide either 'query' or 'ticket_id' not both."
            ticket_filter = "type:ticket"
            query = query if query.startswith(ticket_filter) else f"{ticket_filter} {query}"
            ticket_id = [
                x["id"]
                for x in filter(
                    lambda x: x["result_type"] == "ticket",
                    self.__zendesk_search_results(query=query, page_number=page_number, **kwargs),
                )
            ]
        if ticket_id is not None:
            tickets = []
            for single_ticket in argToList(ticket_id):
                try:
                    tickets.append(self._get_ticket_by_id(single_ticket))
                except Exception as e:
                    demisto.error(f"could not retrieve ticket: {single_ticket}\n{traceback.format_exc()}")
                    error_msgs.append(f"could not retrieve ticket: {single_ticket}\n{e}")
        else:
            can_use_cursor_paging = page_number is None
            sort_params = self.__get_sort_params(sort, can_use_cursor_paging) if sort else None
            url_suffix = self.__get_tickets_url_suffix(filter_, user_id)
            tickets = list(
                self._paged_request(
                    url_suffix=url_suffix, data_field_name="tickets", params=sort_params, page_number=page_number, **kwargs
                )
            )

        if tickets:
            command_results.append(self.__command_results_zendesk_tickets(tickets))
        if error_msgs:
            command_results.append(error_entry("\n".join(error_msgs)))
        return command_results if command_results else "No outputs."

    class Ticket:
        def __init__(
            self,
            type: str | None = None,
            collaborators: str | None = None,
            comment: str | None = None,
            html_comment: str | None = None,
            public: str | bool | None = None,
            email_ccs: str | None = None,
            priority: str | None = None,
            followers: list[str] | str | None = None,
            status: str | None = None,
            **kwargs,
        ):
            self._data: dict[str, Any] = {}

            if type:
                Validators.validate_ticket_type(type)
                self._data["type"] = type
                if type != "incident":
                    assert "problem_id" not in kwargs, "you can't use 'problem_id' if the ticket type is not 'incident'"
            if priority:
                Validators.validate_ticket_priority(priority)
                self._data["priority"] = priority
            if comment:
                self._data["comment"] = {"body": comment}
                if public:
                    self._data["comment"]["public"] = argToBoolean(public)
            if html_comment:
                self._data["comment"] = {"html_body": html_comment}
                if public:
                    self._data["comment"]["public"] = argToBoolean(public)
            if status:
                Validators.validate_ticket_status(status)
                self._data["status"] = status
            if collaborators:
                self._data["collaborators"] = list(map(self.try_int, argToList(collaborators)))
            if followers:
                self._data["followers"] = list(map(self.follower_and_email_cc_parse, argToList(followers)))
            if email_ccs:
                self._data["email_ccs"] = list(map(self.follower_and_email_cc_parse, argToList(email_ccs)))

            self._data.update(
                prepare_kwargs(
                    kwargs=kwargs,
                    str_args=[
                        "subject",
                        "requester",
                        "assignee_email",
                        "recipient",
                        "priority",
                        "external_id",
                        "due_at",
                        "comment",
                    ],
                    list_args="tags",
                    int_args=[
                        "forum_topic_id",
                        "group_id",
                        "organization_id",
                        "via_followup_source_id",
                        "brand_id",
                        "problem_id",
                    ],
                    json_args="custom_fields",
                )
            )

        def __iter__(self):
            yield from self._data.items()

        @staticmethod
        def try_int(value: str) -> int | str:
            try:
                return int(value)
            except (ValueError, TypeError):
                return value

        @staticmethod
        def follower_and_email_cc_parse(user_and_action: str):
            follower_or_email_cc = {}
            splited_user_and_action = user_and_action.split(":")
            try:
                follower_or_email_cc["user_id"] = str(int(splited_user_and_action[0]))
            except ValueError:
                follower_or_email_cc["user_email"] = splited_user_and_action[0]

            if len(splited_user_and_action) == 2:  # action included
                follower_or_email_cc["action"] = splited_user_and_action[1]

            return follower_or_email_cc

    def zendesk_ticket_create(self, **kwargs):  # pragma: no cover
        for argument in ["subject", "type", "requester", "description"]:
            assert argument in kwargs, f"'{argument}' is a required argument."
        kwargs["comment"] = kwargs.pop("description")
        return self.__command_results_zendesk_tickets(
            [self._http_request("POST", url_suffix="tickets", json_data={"ticket": dict(self.Ticket(**kwargs))})["ticket"]]
        )

    def zendesk_ticket_update(
        self,
        ticket_id: str,
        results: bool | None = True,  # pragma: no cover
        is_mirror: bool = False,
        **kwargs,
    ):
        headers = {"user-agent": MIRROR_USER_AGENT} if is_mirror else {}
        res = self._http_request(
            "PUT", url_suffix=f"tickets/{ticket_id}", json_data={"ticket": dict(self.Ticket(**kwargs))}, headers=headers
        )
        if results:
            return self.__command_results_zendesk_tickets([res["ticket"]])
        return None

    def zendesk_ticket_delete(self, ticket_id: str, **kwargs):  # pragma: no cover
        self._http_request("DELETE", url_suffix=f"tickets/{ticket_id}", return_empty_response=True)
        return f"ticket: {ticket_id} deleted."

    @staticmethod
    def _map_comment_attachments(comment: dict):
        if not comment.get("attachments"):
            return comment

        copy_comment = copy(comment)
        copy_comment["attachments"] = []
        for attachment in comment["attachments"]:
            copy_comment["attachments"].append(
                {
                    "file name": attachment["file_name"],
                    "content url": attachment["content_url"],
                    "id": attachment["id"],
                }
            )
        return copy_comment

    @staticmethod
    def __command_results_zendesk_ticket_comments(comments: list[dict]):
        readable_pre_proces = list(map(ZendeskClient._map_comment_attachments, comments))
        readable_outputs = tableToMarkdown(
            name="Zendek comments:",
            t=readable_pre_proces,
            headers=COMMENTS_HEADERS,
            headerTransform=camelize_string,
            is_auto_json_transform=True,
        )
        return CommandResults(outputs_prefix="Zendesk.Ticket.Comment", outputs=comments, readable_output=readable_outputs)

    def _get_comments(self, ticket_id: str, **kwargs) -> list:  # type:ignore
        for comment in self._paged_request(url_suffix=f"tickets/{ticket_id}/comments", data_field_name="comments", **kwargs):
            for attachment in comment.get("attachments", []):
                attachment.pop("thumbnails", None)
            yield CACHE.replace_ids_change(comment, user_fields=["author_id"])  # type: ignore

    def zendesk_ticket_comment_list(self, ticket_id: str, **kwargs):
        return self.__command_results_zendesk_ticket_comments(list(self._get_comments(ticket_id, **kwargs)))

    # ---- attachment related functions ---- #

    def zendesk_ticket_attachment_add(
        self,
        file_id: STR_OR_STR_LIST,
        ticket_id: int,
        comment: str,
        filename: STR_OR_STR_LIST | None = None,
        is_mirror: bool = False,
        **kwargs,
    ):
        headers = {"Content-Type": "application/binary"}
        if is_mirror:
            headers["user-agent"] = MIRROR_USER_AGENT
        file_id = argToList(file_id)
        file_name_list = argToList(filename) if filename else [None] * len(file_id)
        file_tokens = []
        uploaded_files = []
        for single_file_id, single_file_name in zip(file_id, file_name_list):
            single_file = demisto.getFilePath(single_file_id)
            display_name = single_file_name or single_file["name"]
            with open(single_file["path"], "rb") as file_to_upload:
                file_tokens.append(
                    self._http_request(
                        "POST",
                        url_suffix="uploads",
                        params={"filename": display_name},
                        headers=headers,
                        content=file_to_upload.read(),
                    )["upload"]["token"]
                )
            uploaded_files.append(display_name)

        self._http_request(
            "PUT", url_suffix=f"tickets/{ticket_id}", json_data={"ticket": {"comment": {"uploads": file_tokens, "body": comment}}}
        )

        return f'file: {", ".join(uploaded_files)} attached to ticket: {ticket_id}'

    def zendesk_attachment_get(self, attachment_id):
        attachments = [
            self._http_request("GET", url_suffix=f"attachments/{single_attachent_id}")["attachment"]
            for single_attachent_id in argToList(attachment_id)
        ]

        def filter_thumbnails(attachment: dict):
            attachment.pop("thumbnails")
            return attachment

        attachments = list(map(filter_thumbnails, attachments))

        return attachments

    def get_file_entries(self, attachments):
        results = []
        for attachment_link, attachment_name in ((x["content_url"], x["file_name"]) for x in attachments):
            res = self._http_request("GET", full_url=attachment_link, resp_type="response")
            res.raise_for_status()
            results.append(fileResult(filename=attachment_name, data=res.content, file_type=EntryType.ENTRY_INFO_FILE))
        return results

    def zendesk_attachment_get_command(self, attachment_id: int, **kwargs):
        attachments = self.zendesk_attachment_get(attachment_id)
        readable_output = tableToMarkdown(
            name="Zendesk attachments", t=attachments, headers=ATTACHMENTS_HEADERS, headerTransform=camelize_string
        )
        results = [CommandResults(outputs_prefix="Zendesk.Attachment", outputs=attachments, readable_output=readable_output)]

        file_entries = self.get_file_entries(attachments)
        results.append(file_entries)

        return results

    # ---- search related functions ---- #

    def __zendesk_search_results(
        self, query: str, limit: int = 50, page_number: int | None = None, page_size: int = 50, additional_params: dict = {}
    ):
        params = {"query": query} | additional_params
        results = []
        if page_number:
            results = list(
                self.__get_spesific_page(
                    url_suffix="search.json",
                    params=params,
                    data_field_name="results",
                    page_number=int(page_number),
                    page_size=int(page_size),
                )
            )
        else:
            count = self._http_request("GET", url_suffix="search/count.json", params=params)["count"]
            limit = min(int(limit), count)
            size = min(limit, MAX_PAGE_SIZE)
            current_page = 1
            while len(results) < limit:
                results.extend(
                    self.__get_spesific_page(
                        url_suffix="search.json",
                        params=params,
                        data_field_name="results",
                        page_number=current_page,
                        page_size=size,
                    )
                )
                current_page += 1
            results = results[:limit]

        return results

    def zendesk_search(self, query: str, limit: int = 50, page_number: int | None = None, page_size: int = 50, **kwargs):
        return CommandResults(
            outputs_prefix="Zendesk.Search",
            outputs=self.__zendesk_search_results(query=query, limit=limit, page_number=page_number, page_size=page_size),
        )

    # ---- articles related functions ---- #

    def zendesk_article_list(self, locale: str | None = "", article_id: int | None = None, **kwargs):
        if locale:
            locale = f"{locale}/"
        if article_id:
            articles = [self._http_request("GET", url_suffix=f"help_center/{locale}articles/{article_id}")["article"]]
        else:
            articles = list(self._paged_request(url_suffix=f"help_center/{locale}articles", data_field_name="articles", **kwargs))

        readable_output = ["</h1>Zendesk articles</h1>"]
        for title, body in ((x["title"], x["body"]) for x in articles):
            readable_output.append(f"<h1>{title}</h1>\n{body}")

        return CommandResults(outputs_prefix="Zendesk.Article", outputs=articles, readable_output="\n\n\n".join(readable_output))

    # ---- demisto related functions ---- #

    def test_module(self):  # pragma: no cover
        exception: Exception
        # If one of the endpoints work we will pass the test_module check.
        for data_type in ["tickets", "users", "organizations"]:
            try:
                self._paged_request(url_suffix=data_type, data_field_name=data_type, limit=1)
                return "ok"
            except Exception as e:
                exception = e

        raise exception from None

    def _ticket_to_incident(self, ticket: dict):
        ticket |= {
            "severity": PRIORITY_MAP.get(ticket["priority"]),
            "mirror_instance": INTEGRATION_INSTANCE,
            "mirror_id": str(ticket["id"]),
            "mirror_direction": MIRROR_DIRECTION,
            "mirror_last_sync": datetime.now().strftime("%Y-%m-%dT%H:%M:%SZ"),
            "mirror_tags": MIRROR_TAGS,
        }
        return {
            "rawJSON": json.dumps(ticket),
            "name": ticket["subject"],
            "occurred": ticket["created_at"],
            "attachment": ticket.get("attachments", []),
        }

    @staticmethod
    def _fetch_query_builder(ticket_priority: str = None, ticket_status: str = None, ticket_types: str = None, **_):
        query_parts = []
        if ticket_priority and "all" not in ticket_priority:
            for priority in argToList(ticket_priority):
                query_parts.append(f"priority:{priority}")
        if ticket_status and "all" not in ticket_status:
            for status in argToList(ticket_status):
                query_parts.append(f"status:{status}")
        if ticket_types and "all" not in ticket_types:
            for ticket_type in argToList(ticket_types):
                query_parts.append(f"ticket_type:{ticket_type}")
        return " ".join(query_parts)

    @staticmethod
    def _fetch_args(params, last_run):
        # from params
        time_filter = "updated" if params.get("time_filter") == "updated-at" else "created"
        query = params.get("fetch_query") or ZendeskClient._fetch_query_builder(**params)
        max_fetch = min(100, int(params.get("max_fetch") or 50))
        get_attachments = params.get("get_attachments")

        # from last_run
        fetched_tickets = deque(last_run.get("fetched_tickets") or [])
        if query != last_run.get("query") or time_filter != last_run.get("time_filter"):
            last_run = {"fetched_tickets": fetched_tickets}
        page_number = last_run.get("page_number") or 1
        last_fetch = last_run.get("fetch_time")
        max_fetch = last_run.get("max_fetch") or max_fetch
        if not last_fetch:
            first_fetch = params.get("first_fetch") or "3 days"
            first_fetch_datetime = dateparser.parse(first_fetch, settings={"TIMEZONE": "UTC"})
            if not first_fetch_datetime:
                raise DemistoException(f"invalid first fetch time specified ({first_fetch})")
            last_fetch = first_fetch_datetime.strftime(ZENDESK_FETCH_TIME_FORMAT)

        return fetched_tickets, last_fetch, time_filter, query, max_fetch, page_number, get_attachments

    @staticmethod
    def _next_fetch_args(
        fetched_tickets, search_results_ids, next_run_start_time, query, time_filter, max_fetch, page_number, current_fetch
    ):
        next_run = {
            "fetched_tickets": list(fetched_tickets)[-1000:],
            "fetch_time": next_run_start_time.strftime(ZENDESK_FETCH_TIME_FORMAT),
        }
        if len(search_results_ids) >= max_fetch:
            next_run |= {
                "max_fetch": max_fetch,
                "page_number": page_number + 1,
                "fetch_time": current_fetch,
                "query": query,
                "time_filter": time_filter,
            }

        return next_run

    def get_attachments_ids(self, ticket: dict) -> list[int]:
        """

        Args:
            ticket (dict): The fetched ticket.

        Returns (list): all the attachment ids for a ticket

        """
        attachments_ids = []
        ticket_id = ticket["id"]
        comments_list = self._get_comments(ticket_id=ticket_id)
        for comment in comments_list:
            attachments = comment.get("attachments", [])
            for attachment in attachments:
                attachment_id = attachment.get("id")
                attachments_ids.append(attachment_id)

        return attachments_ids

    def get_attachment_entries(self, ticket: dict) -> list:
        """

        Args:
            ticket (dict): The ticket to get the file entries

        Returns: The attachments entries.

        """
        attachments_ids = self.get_attachments_ids(ticket)
        file_names = []
        attachments = self.zendesk_attachment_get(attachments_ids)
        demisto.debug(f"The fetched attachments - {attachments}")
        attachments_entries = self.get_file_entries(attachments)
        if isinstance(attachments_entries, list):
            for file_result in attachments_entries:
                if file_result["Type"] == entryTypes["error"]:
                    raise Exception(f"Error getting attachment: {file_result.get('Contents', '')!s}")
                file_names.append({"path": file_result.get("FileID", ""), "name": file_result.get("File", "")})
        return file_names

    def fetch_incidents(self, params: dict, lastRun: str | None = None):
        last_run = json.loads(lastRun or "null") or demisto.getLastRun() or {}
        fetched_tickets, last_fetch, time_filter, query, max_fetch, page_number, get_attachments = self._fetch_args(
            params, last_run
        )

        # look back window for tickets
        next_run_start_time = datetime.utcnow() - timedelta(minutes=1)

        query_parts = ["type:ticket", f"{time_filter}>{last_fetch}"]
        query_parts = query_parts + [query] if query else query_parts

        search_results = self.__zendesk_search_results(
            query=" ".join(query_parts),
            limit=max_fetch,
            page_size=max_fetch,
            page_number=page_number,
            additional_params={"sort_by": f"{time_filter}_at", "sort_order": "asc"},
        )
        search_results_ids = [x["id"] for x in search_results]
        filtered_search_results_ids = list(filter(lambda x: x not in fetched_tickets, search_results_ids))
        tickets = (self._get_ticket_by_id(x) for x in filtered_search_results_ids)
        ticket_modified = []
        if get_attachments:
            for ticket in tickets:
                attachments = ZendeskClient.get_attachment_entries(self, ticket)
                ticket.update({"attachments": attachments})
                ticket_modified.append(ticket)

        tickets = ticket_modified if ticket_modified else tickets
        incidents = list(map(self._ticket_to_incident, tickets))

        demisto.incidents(incidents)
        fetched_tickets.extend(filtered_search_results_ids)
        demisto.setLastRun(
            self._next_fetch_args(
                fetched_tickets, search_results_ids, next_run_start_time, query, time_filter, max_fetch, page_number, last_fetch
            )
        )

    def get_modified_remote_data(self, lastUpdate: str | None = None, **kwargs):
        try:
            timestamp = int(dateparser.parse(lastUpdate).timestamp())  # type: ignore
        except (TypeError, AttributeError):
            timestamp = 0
        last_run = get_last_mirror_run() or {}
        updated_tickets = UpdatedTickets(self, timestamp, last_run)
        tickets_ids = [str(x["id"]) for x in updated_tickets.tickets()]
        if tickets_ids:
            return_results(GetModifiedRemoteDataResponse(tickets_ids))
        try:
            set_last_mirror_run(updated_tickets.next_run())
        except json.decoder.JSONDecodeError as e:
            demisto.debug(f"{e}")

    @staticmethod
    def _create_entry_from_comment(comment: dict):
        comment_body = comment.get("body")
        attachments = comment.get("attachments")
        if attachments:
            attachments_table = tableToMarkdown("attachments", attachments, ["file_name", "id"], headerTransform=camelize_string)
            comment_body = f"{comment_body}\n{attachments_table}"

        return {"Type": EntryType.NOTE, "Contents": comment_body, "ContentsFormat": EntryFormat.MARKDOWN, "Note": True}

    def get_remote_data(self, **kwargs):
        try:
            parsed_args = GetRemoteDataArgs(kwargs)
            last_update = datetime_to_iso(dateparser.parse(parsed_args.last_update, settings={"TIMEZONE": "UTC"}))  # type: ignore
            try:
                ticket_data = self._get_ticket_by_id(parsed_args.remote_incident_id)
            except HTTPError as e:
                if e.response.status_code == 404 and CLOSE_INCIDENT:
                    return GetRemoteDataResponse(
                        close_entry(f"ticket {parsed_args.remote_incident_id} deleted."),
                        [close_entry(f"ticket {parsed_args.remote_incident_id} deleted.")],
                    )
                raise e from None

            context = self.__ticket_context(ticket_data)
            context["severity"] = PRIORITY_MAP.get(ticket_data["priority"])
            context["incomming_mirror_error"] = ""
            for field_to_delete in FIELDS_TO_REMOVE_FROM_MIROR_IN:
                if field_to_delete in context:
                    del context[field_to_delete]

            def filter_comments(comment: dict):
                return (
                    comment["created_at"] > last_update
                    and dict_safe_get(comment, ["metadata", "system", "client"]) != MIRROR_USER_AGENT
                )

            ticket_entries = list(
                map(
                    self._create_entry_from_comment,
                    filter(filter_comments, self._get_comments(parsed_args.remote_incident_id, limit=200)),
                )
            )
            if ticket_data.get("status") == "closed" and CLOSE_INCIDENT:
                ticket_entries.append(close_entry(f"ticket {parsed_args.remote_incident_id} closed."))

            return GetRemoteDataResponse(context, ticket_entries)
        except Exception as e:
            return GetRemoteDataResponse(
                {"incomming_mirror_error": f"mirroring failed with error: {e}\n{traceback.format_exc()}"}, []
            )

    def update_remote_system(self, **kwargs):
        args = UpdateRemoteSystemArgs(kwargs)
        files = []
        args.delta = {key: val for key, val in args.delta.items() if val}

        if "severity" in args.delta:
            severity = args.delta.pop("severity")
            severity = max(severity, IncidentSeverity.LOW)
            for priority, severity_val in PRIORITY_MAP.items():
                if severity == severity_val:
                    args.delta["priority"] = priority
                    break

        if (
            args.incident_changed
            and CLOSE_INCIDENT
            and (args.inc_status == IncidentStatus.DONE or (args.data.get("state") == "closed"))
        ):
            args.delta["status"] = "closed"

        def upload_files_and_reset_files_list(files: list):
            while files:
                comment = files[0].get("contents", DEFAULT_UPLOAD_FILES_COMMENT)
                files_to_upload = []
                while files and files[0].get("contents", DEFAULT_UPLOAD_FILES_COMMENT) == comment:
                    files_to_upload.append(files.pop(0)["id"])

                self.zendesk_ticket_attachment_add(
                    file_id=files_to_upload, ticket_id=args.remote_incident_id, comment=comment, is_mirror=True
                )

        try:
            for entry in args.entries or []:
                # Mirroring files as entries
                if entry["type"] in [EntryType.ENTRY_INFO_FILE, EntryType.FILE, EntryType.IMAGE]:
                    files.append(entry)
                else:
                    upload_files_and_reset_files_list(files)
                    # Mirroring comment and work notes as entries
                    self.zendesk_ticket_update(
                        ticket_id=args.remote_incident_id, comment=entry["contents"], results=False, is_mirror=True
                    )

            upload_files_and_reset_files_list(files)
            if args.delta:
                self.zendesk_ticket_update(ticket_id=args.remote_incident_id, results=False, **args.delta)

        except HTTPError as e:
            if e.response.status_code != 404 and CLOSE_INCIDENT:
                raise e from None
            demisto.debug(f"ticket {args.remote_incident_id} deleted.")

        return args.remote_incident_id

    def get_mapping_fields(self, **kwargs):  # pragma: no cover
        zendesk_ticket_scheme = SchemeTypeMapping("Zendesk Ticket")
        zendesk_ticket_scheme.add_field(
            name="type", description='The type of this ticket. Allowed values are "problem", "incident", "question", or "task".'
        )
        zendesk_ticket_scheme.add_field(name="subject", description="The value of the subject field for this ticket.")
        zendesk_ticket_scheme.add_field(name="description", description="The ticket description.")
        zendesk_ticket_scheme.add_field(
            name="priority",
            description="The urgency with which the ticket should be addressed. "
            'Allowed values are "urgent", "high", "normal", or "low".',
        )
        zendesk_ticket_scheme.add_field(
            name="status",
            description='The state of the ticket. Allowed values are "new", "open", "pending", "hold", "solved", or "closed".',
        )
        zendesk_ticket_scheme.add_field(name="recipient", description="The original recipient e-mail address of the ticket.")
        zendesk_ticket_scheme.add_field(name="requester", description="The user who requested this ticket.")
        zendesk_ticket_scheme.add_field(name="assigne", description="The agent currently assigned to the ticket.")
        zendesk_ticket_scheme.add_field(name="organization", description="The organization of the requester.")
        zendesk_ticket_scheme.add_field(name="collaborators", description="The users currently CC'ed on the ticket.")
        zendesk_ticket_scheme.add_field(name="followers", description="The agents currently following the ticket.")
        zendesk_ticket_scheme.add_field(name="email_ccs", description="The agents or end users currently CC'ed on the ticket.")
        zendesk_ticket_scheme.add_field(name="tags", description="The array of tags applied to this ticket.")
        zendesk_ticket_scheme.add_field(
            name="custom_fields",
            description="Custom fields for the ticket (this is a json formatted argument see: https://developer.zendesk.com/documentation/ticketing/managing-tickets/creating-and-updating-tickets#setting-custom-field-values).",
        )  # noqa: E501

        return GetMappingFieldsResponse([zendesk_ticket_scheme])


def main():  # pragma: no cover
    params = demisto.params()
    verify = not params.get("insecure", False)
    if not verify:
        disable_warnings()
    client = ZendeskClient(
        base_url=params["base_url"],
        username=params["credentials"].get("identifier"),
        password=params["credentials"]["password"],
        proxy=params.get("proxy", False),
        verify=verify,
    )
    global CACHE
    CACHE = CacheManager(client)
    try:
        command = demisto.command()
        args = demisto.args()
        commands: dict[str, Callable] = {
            # demisto commands
            "test-module": client.test_module,
            "get-modified-remote-data": client.get_modified_remote_data,
            "get-remote-data": client.get_remote_data,
            "update-remote-system": client.update_remote_system,
            "get-mapping-fields": client.get_mapping_fields,
            "zendesk-clear-cache": CacheManager.zendesk_clear_cache,
            # user commands
            "zendesk-user-list": client.zendesk_user_list,
            "zendesk-user-create": client.zendesk_user_create,
            "zendesk-user-update": client.zendesk_user_update,
            "zendesk-user-delete": client.zendesk_user_delete,
            "zendesk-group-user-list": client.list_group_users,
            # Group commands
            "zendesk-group-list": client.list_groups,
            # organization commands
            "zendesk-organization-list": client.zendesk_organization_list,
            # ticket commands
            "zendesk-ticket-list": client.zendesk_ticket_list,
            "zendesk-ticket-create": client.zendesk_ticket_create,
            "zendesk-ticket-update": client.zendesk_ticket_update,
            "zendesk-ticket-delete": client.zendesk_ticket_delete,
            "zendesk-ticket-comment-list": client.zendesk_ticket_comment_list,
            # attachment commands
            "zendesk-ticket-attachment-add": client.zendesk_ticket_attachment_add,
            "zendesk-attachment-get": client.zendesk_attachment_get_command,
            # search command
            "zendesk-search": client.zendesk_search,
            # articles command
            "zendesk-article-list": client.zendesk_article_list,
        }
        demisto.debug(f"command {command} called")

        if command == "fetch-incidents":
            client.fetch_incidents(params, **args)
        elif command in commands:
            if command_res := commands[command](**args):
                return_results(command_res)
        else:
            raise NotImplementedError(command)
    except Exception as e:
        return_error(f"An error occurred: {e}", error=e)
    finally:
        CACHE.save()


if __name__ in ("__main__", "__builtin__", "builtins"):  # pragma: no cover
    main()