#!/usr/bin/env python

# coding=utf-8

import argparse

import bisect

import datetime

import json

import os.path

import re

import sys

import uuid  # @UnresolvedImport

import requests

import dateutil.tz

from dateutil.parser import parse as parse_date_raw

from dateutil.tz import tzutc

from iri_tweet.utils import get_logger, set_logging, set_logging_options

from lxml import etree

LDT_CONTENT_REST_API_PATH = "api/ldt/1.0/contents/"

LDT_PROJECT_REST_API_PATH = "api/ldt/1.0/projects/"

def parse_date(datestr, default_tz, default=None):

    res = parse_date_raw(datestr, default=default)

    if res.tzinfo is None:

        res = res.replace(tzinfo=default_tz)

    return res

def re_fn(expr, item):

    reg = re.compile(expr, re.I)

    res = reg.search(item)

    if res:

        get_logger().debug("re_fn : " + repr(expr) + "~" + repr(item)) #@UndefinedVariable

    return res is not None

def parse_polemics_1(tw_text, extended_mode):

    """

    parse polemics in text and return a list of polemic code. None if not polemic found

    """

    polemics = {}

    for m in re.finditer(r"(\+\+|\-\-|\?\?|\=\=)",tw_text):

        pol_link = {

            '++' : 'OK',

            '--' : 'KO',

            '??' : 'Q',

            '==' : 'REF'}[m.group(1)]

        polemics[pol_link] = pol_link

    if extended_mode:

        if "?" in tw_text:

            polemics["Q"] = "Q"

    if len(polemics) > 0:

        return polemics.keys()

    else:

        return None

def parse_polemics_2(tw_text, extended_mode):

    """

    parse polemics in text and return a list of polemic code. None if not polemic found

    """

    polemics = {}

    for m in re.finditer(r"(\+\+|\!\!|\?\?|\=\=)",tw_text):

        pol_link = {

            '++' : 'OK',

            '!!' : 'KO',

            '??' : 'Q',

            '==' : 'REF'}[m.group(1)]

        polemics[pol_link] = pol_link

    if extended_mode:

        if "?" in tw_text:

            polemics["Q"] = "Q"

    if len(polemics) > 0:

        return polemics.keys()

    else:

        return None

def parse_polemics_3(tw_text, extended_mode):

    """

    parse polemics in text and return a list of polemic code. None if not polemic found

    """

    polemics = {}

    for m in re.finditer(r"(\+\+|\?\?|\*\*|\=\=)",tw_text):

        pol_link = {

            '++' : 'OK',

            '??' : 'KO',

            '**' : 'REF',

            '==' : 'Q'}[m.group(1)]

        polemics[pol_link] = pol_link

    if len(polemics) > 0:

        return polemics.keys()

    else:

        return None

protocol_version_map = {

    "1" : parse_polemics_1,

    "2" : parse_polemics_2,

    "3" : parse_polemics_3

}

def get_options():

    parser = argparse.ArgumentParser(description="All date should be given using iso8601 format. If no timezone is used, the date is considered as UTC")

    parser.add_argument("-f", "--file", dest="filename",

                      help="write export to file", metavar="FILE", default="project.ldt")

    parser.add_argument("-d", "--chat-database", dest="database",

                      help="Input chat file", metavar="CHAT_DATABASE")

    parser.add_argument("-a", "--annotation-protocol", dest="protocol_version",

                      help="annotation protocol version", metavar="PROTOCOL_VERSION",

                      default="2")

    parser.add_argument("-s", "--start-date", dest="start_date",

                      help="start date", metavar="START_DATE", default=None)

    parser.add_argument("-e", "--end-date", dest="end_date",

                      help="end date", metavar="END_DATE", default=None)

    parser.add_argument("-I", "--content-file", dest="content_file",

                      help="Content file", metavar="CONTENT_FILE")

    parser.add_argument("-c", "--content", dest="content",

                      help="Content url", metavar="CONTENT")

    parser.add_argument("-V", "--video-url", dest="video",

                      help="video url", metavar="VIDEO")

    parser.add_argument("-i", "--content-id", dest="content_id",

                      help="Content id", metavar="CONTENT_ID")

    parser.add_argument("-x", "--exclude", dest="exclude",

                      help="file containing the id to exclude", metavar="EXCLUDE")

    parser.add_argument("-C", "--color", dest="color",

                      help="Color code", metavar="COLOR", default="16763904")

    parser.add_argument("-H", "--hashtag", dest="hashtag",

                      help="Hashtag", metavar="HASHTAG", default=[], action="append")

    parser.add_argument("-D", "--duration", dest="duration", type=int,

                      help="Duration", metavar="DURATION", default=None)

    parser.add_argument("-n", "--name", dest="name",

                      help="Cutting name", metavar="NAME", default="Chats")

    parser.add_argument("-R", "--replace", dest="replace", action="store_true",

                      help="Replace tweet ensemble", default=False)

    parser.add_argument("-m", "--merge", dest="merge", action="store_true",

                      help="merge tweet ensemble, choose the first ensemble", default=False)

    parser.add_argument("-L", "--list-conf", dest="listconf",

                      help="list of file to process", metavar="LIST_CONF", default=None)

    parser.add_argument("-E", "--extended", dest="extended_mode", action="store_true",

                      help="Trigger polemic extended mode", default=False)

    parser.add_argument("-b", "--base-url", dest="base_url",

                      help="base URL of the platform", metavar="BASE_URL", default="http://ldt.iri.centrepompidou.fr/ldtplatform/")

    parser.add_argument("-p", "--project", dest="project_id",

                      help="Project id", metavar="PROJECT_ID", default=None)

    parser.add_argument("-P", "--post-param", dest="post_param",

                      help="Post param", metavar="POST_PARAM", default=None)

    parser.add_argument("--user-whitelist", dest="user_whitelist", action="store",

                      help="A list of user screen name", metavar="USER_WHITELIST",default=None)

    parser.add_argument("--cut", dest="cuts", action="append",

                      help="A cut with the forma <ts in ms>::<duration>", metavar="CUT", default=[])

    parser.add_argument("-Z","--tz", dest="timezone",

                      help="The timezone of the timestamps", metavar="TZ", default="UTC")

    set_logging_options(parser)

    return (parser.parse_args(), parser)

def find_delta(deltas, ts):

    i = bisect.bisect_right(deltas, (ts+1,0))

    if i:

        return deltas[i-1]

    return (0,0)

def parse_duration(s):

    try:

        return int(s)

    except ValueError:

        parts = s.split(":")

        if len(parts) < 2:

            raise ValueError("Bad duration format")

        time_params = {

            'hours': int(parts[0]),

            'minutes': int(parts[1]),

            'seconds': int(parts[2]) if len(parts)>2 else 0

        }

        return int(round(datetime.timedelta(**time_params).total_seconds()*1000))

CHAT_REGEXP = re.compile(r"^(?P<created_at>\d{2}:\d{2}:\d{2})\t\sFrom\s{2}(?P<user>.+?)\s:\s(?P<text>.*)$", re.DOTALL)

CHAT_LINE_REGEXP = re.compile(r"^\d{2}:\d{2}:\d{2}\t\sFrom\s{2}.+?\s:")

CHAT_DM_REGEXP = re.compile(r"\(Direct Message\)", re.IGNORECASE)

def parse_chat_line(chat_id, chat_line):

    if (m := CHAT_REGEXP.match(chat_line)) is not None:

        res = {k: v.replace('\r','\n') if k == 'text' else v for k,v in m.groupdict().items()}

        res['id'] = chat_id

        res['tags'] = re.findall('#(\w+)',res['text'])

        return res

    else:

        return {}

def read_chat_file(chat_file_path):

    current_line = ""

    chat_content = []

    with open(chat_file_path, "r") as chat_file:

        for chat_line in chat_file:

            if CHAT_LINE_REGEXP.match(chat_line) is not None:

                if current_line and CHAT_DM_REGEXP.search(current_line) is None:

                    chat_content.append(current_line)

                current_line = chat_line

            else:

                current_line = current_line + "\n" + chat_line

    if current_line and CHAT_DM_REGEXP.search(current_line) is None:

        chat_content.append(current_line)

    return chat_content

if __name__ == "__main__" :

    (options, parser) = get_options()

    set_logging(options)

    get_logger().debug("OPTIONS : " + repr(options)) #@UndefinedVariable

    deltas = [(0,0)]

    total_delta = 0

    if options.cuts:

        cuts_raw = sorted([tuple([parse_duration(s) for s in c.split("::")]) for c in options.cuts])

        for c, d in cuts_raw:

            deltas.append((c+total_delta, -1))

            total_delta += d

            deltas.append((c+total_delta, total_delta))

    if len(sys.argv) == 1 or options.database is None:

        parser.print_help()

        sys.exit(1)

    user_whitelist_file = options.user_whitelist

    user_whitelist = None

    if options.listconf:

        parameters = []

        confdoc = etree.parse(options.listconf)

        for node in confdoc.xpath("/zoom_export/file"):

            params = {}

            for snode in node:

                if snode.tag == "path":

                    params['content_file'] = snode.text

                    params['content_file_write'] = snode.text

                elif snode.tag == "project_id":

                    params['content_file'] = options.base_url + LDT_PROJECT_REST_API_PATH + snode.text + "/?format=json"

                    params['content_file_write'] = options.base_url + LDT_PROJECT_REST_API_PATH + snode.text + "/?format=json"

                    params['project_id'] = snode.text

                elif snode.tag == "start_date":

                    params['start_date'] = snode.text

                elif snode.tag == "end_date":

                    params['end_date'] = snode.text

                elif snode.tag == "duration":

                    params['duration'] = int(snode.text)

                elif snode.tag == "hashtags":

                    params['hashtags'] = [snode.text]

            if options.hashtag or 'hashtags' not in params :

                params['hashtags'] = options.hashtag

            parameters.append(params)

    else:

        if options.project_id:

            content_file = options.base_url + LDT_PROJECT_REST_API_PATH + options.project_id + "/?format=json"

        else:

            content_file = options.content_file

        parameters = [{

            'start_date': options.start_date,

            'end_date' : options.end_date,

            'duration' : options.duration,

            'content_file' : content_file,

            'content_file_write' : content_file,

            'hashtags' : options.hashtag,

            'project_id' : options.project_id

        }]

    post_param = {}

    if options.post_param:

        post_param = json.loads(options.post_param)

    item_tz = dateutil.tz.UTC

    if options.timezone:

        item_tz = dateutil.tz.gettz(options.timezone)

        get_logger().debug("TIMEZONE " + options.timezone + " PARSED :: " + repr(item_tz))

    if item_tz is None:

        get_logger().error("Timezone '%s' not recognized.", options.timezone)

        print("Error: Timezone '%s' not recognized." % options.timezone)

        parser.print_help()

        sys.exit(1)

    display_content_node = None

    for params in parameters:

        get_logger().debug("PARAMETERS " + repr(params)) #@UndefinedVariable

        start_date_str = params.get("start_date",None)

        end_date_str = params.get("end_date", None)

        duration = params.get("duration", None)

        content_file = params.get("content_file", None)

        content_file_write = params.get("content_file_write", None)

        hashtags = list(set(params.get('hashtags', [])))

        if user_whitelist_file:

            with open(user_whitelist_file, 'r+') as f:

                user_whitelist = list(set([s.strip() for s in f]))

        start_date = None

        if start_date_str:

            start_date = parse_date(start_date_str, item_tz)

        root = None

        ensemble_parent = None

        project = None

        #to do : analyse situation ldt or iri ? filename set or not ?

        if content_file and content_file.find("http") == 0:

            get_logger().debug("url : " + content_file) #@UndefinedVariable

            r = requests.get(content_file, params=post_param)

            get_logger().debug("url response " + repr(r) + " content " + repr(r.text)) #@UndefinedVariable

            project = r.json()

            text_match = re.match(r"\<\?\s*xml.*?\?\>(.*)", project['ldt'], re.I|re.S)

            root = etree.fromstring(text_match.group(1) if text_match else project['ldt'])

        elif content_file and os.path.exists(content_file):

            doc = etree.parse(content_file)

            root = doc.getroot()

            for child in root:

                if child.tag == "project":

                    project = child

                    break

            if project is None:

                root = None

        content_id = None

        if root is None:

            root = etree.Element("iri")

            project = etree.SubElement(root, "project", {"abstract":"Polemics Chat","title":"Polemic Chat", "user":"IRI Web", "id":str(uuid.uuid4())})

            medias = etree.SubElement(root, "medias")

            media = etree.SubElement(medias, "media", {"pict":"", "src":options.content, "video":options.video, "id":options.content_id, "extra":""})

            annotations = etree.SubElement(root, "annotations")

            content = etree.SubElement(annotations, "content", {"id":options.content_id})

            ensemble_parent = content

            content_id = options.content_id

        if ensemble_parent is None:

            file_type = None

            for node in root:

                if node.tag == "project":

                    file_type = "ldt"

                    break

                elif node.tag == "head":

                    file_type = "iri"

                    break

            if file_type == "ldt":

                media_nodes = root.xpath("//media")