searx-qt/searxqt/core/searx.py

########################################################################
#  Searx-Qt - Lightweight desktop application for Searx.
#  Copyright (C) 2020-2022  CYBERDEViL
#
#  This file is part of Searx-Qt.
#
#  Searx-Qt is free software: you can redistribute it and/or modify
#  it under the terms of the GNU General Public License as published by
#  the Free Software Foundation, either version 3 of the License, or
#  (at your option) any later version.
#
#  Searx-Qt is distributed in the hope that it will be useful,
#  but WITHOUT ANY WARRANTY; without even the implied warranty of
#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#  GNU General Public License for more details.
#
#  You should have received a copy of the GNU General Public License
#  along with this program.  If not, see <https://www.gnu.org/licenses/>.
#
########################################################################

import time
import urllib.parse
from bs4 import BeautifulSoup

from searxqt.core.requests import JsonResult, ErrorType, Schemas
from searxqt.core.handler import HandlerProto, NetworkTypes
from searxqt.utils.string import parseFilesize

from searxqt.translations import _


## API result (format=json)
class SearchResult(JsonResult):
    Schema = Schemas['searxng_query']

    def __init__(self, url, response, err="", errType=ErrorType.Success):
        JsonResult.__init__(self, url, response, err=err, errType=errType)

    def verifyFurther(self):
        JsonResult.verifyFurther(self)

        # One of the following keys has to be not empty, else we count it as
        # no (usable) result.
        validKeys = [
            'results',
            'answers',
            'corrections',
            'infoboxes',
            'suggestions'
        ]

        if self._errType == ErrorType.Success:
            data = self.json()
            valid = False
            for key in validKeys:
                if len(data.get(key, [])):
                    valid = True
                    break

            if not valid:
                self._errType = ErrorType.NoResults
                self._err = f"NoResults: got: `{self.json()}`"


def fixUrlScheme(url):
    """! Adds 'https://' when the scheme is missing."""
    parsedUrl = urllib.parse.urlparse(url)
    if not parsedUrl.scheme:
        return f"https://{url}"
    return url


## HTML result that will be parsed into JSON
class SearchResult2(SearchResult):
    Schema = Schemas['searxng_query']

    def __init__(self, url, response, err="", errType=ErrorType.Success):
        self.__json = {}
        SearchResult.__init__(self, url, response, err=err, errType=errType)

    def verifyFurther(self):
        self.__json = self.parse()
        SearchResult.verifyFurther(self)

    def json(self):
        return self.__json

    def makeUrlAbsolute(self, url):
        """! Returns a absolute URL. It will add the SearXNG instance its
             schema and location in front when they are missing."""
        parsedUrl = urllib.parse.urlparse(url)
        instanceUrl = urllib.parse.urlparse(self.url())
        if not parsedUrl.netloc:
            url = f"{instanceUrl.netloc}{url}"
        if not parsedUrl.scheme:
            url = f"{instanceUrl.scheme}://{url}"
        return url

    def parseImagesResult(self, result):
        """! Parse image results from HTML."""
        """Example HTML:

        <article class="result result-images category-images">
            <a href="https://wallup.net/wp-content/uploads/2019/09/441567-landscapes-nature-wallpaper.jpg" rel="noreferrer">
                <img alt="landscapes, Nature, Wallpaper Wallpapers HD / Desktop and Mobile ..." class="image_thumbnail" height="200" loading="lazy" rel="noreferrer" src="/image_proxy?url=https%3A%2F%2Fs2.qwant.com%2Fthumbr%2F474x315%2Ff%2F1%2F5fe20d297b0af77d40641a1c2d1a0a430b235e0f98e1584d580cf7931b28f9%2Fth.jpg%3Fu%3Dhttps%253A%252F%252Ftse.mm.bing.net%252Fth%253Fid%253DOIP.bLDwvUIZXCd5HCilSOxKCAHaE7%2526pid%253DApi%26q%3D0%26b%3D1%26p%3D0%26a%3D0&amp;h=HASH ..." width="200"/>

                <span class="title">
                    landscapes, Nature, Wallpaper Wallpapers HD / Desktop and Mobile ...
                </span>
                <span class="source">wallup.net</span>
            </a>
            <div class="detail">
                <a class="result-detail-close" href="#">
                    <svg SVG_STUFF ...></svg>
                </a>
                <a class="result-detail-previous" href="#">
                    <svg SVG_STUFF ...></svg>
                </a>
                <a class="result-images-source" href="https://wallup.net/wp-content/uploads/2019/09/441567-landscapes-nature-wallpaper.jpg" rel="noreferrer">
                    <img alt="landscapes, Nature, Wallpaper Wallpapers HD / Desktop and Mobile ..." data-src="/image_proxy?url=https%3A%2F%2Fwallup.net%2Fwp-content%2Fuploads%2F2019%2F09%2F441567-landscapes-nature-wallpaper.jpg&amp;h=HASH ..." src=""/>
                </a>
                <div class="result-images-labels">
                    <h4>landscapes, Nature, Wallpaper Wallpapers HD / Desktop and Mobile ...</h4>
                    <p class="result-content"> </p>
                    <hr/>
                    <p class="result-author"> </p>
                    <p class="result-format"> </p>
                    <p class="result-source"> </p>
                    <p class="result-engine">
                        <span>Engine:</span>qwant images
                    </p>
                    <p class="result-url">
                        <span>View source:</span>
                        <a href="https://wallup.net/landscapes-nature-wallpaper-69/" rel="noreferrer">https://wallup.net/landscapes-nature-wallpaper-69/</a>
                    </p>
                </div>
            </div>
        </article>
        """
        title = ''         # image title
        url = ''           # url to the website of the image
        content = ''       # probably same as the title
        engines = []       # see img_src
        #publishedDate = ''
        img_format = ''    # size/format of the image in string format
        img_src = ''       # source if the image (engine)
        thumbnail_src = '' # url to thumbnail
        source = ''        # where does the image come from?
        category = 'images'

        # !! GET Title
        try:
            title = result.a.img.get('alt')
        except AttributeError:
            log.debug("Failed to get img title", self)

        # !! GET thumbnail_src
        try:
            thumbnail_src = result.a.img.get('src')
        except AttributeError:
            log.debug("Failed to get img thumbnail url", self)
        # Make sure the thumbnail url is absolute
        thumbnail_src = self.makeUrlAbsolute(thumbnail_src)

        # !! GET url
        felem = result.find("p", {"class": "result-url"})
        if felem:
            try:
                url = felem.a.get('href')
            except AttributeError:
                log.debug("Failed to get img url (1)", self)
        else:
            log.debug("Failed to get img url (2)", self)

        # !! GET img_src
        felem = result.find("a", {"class": "result-images-source"})
        if felem:
            img_src = felem.get('href')
            img_src = fixUrlScheme(img_src)  # Make sure it has a scheme
        else:
            log.debug("Failed to get img_src", self)

        ## !! GET content
        # p class=result-content
        felem = result.find("p", {"class": "result-content"})
        if felem:
            content = felem.get_text()
        else:
            log.debug("Failed to get img content", self)

        # !! GET img_format
        # p class=result-format
        felem = result.find("p", {"class": "result-format"})
        if felem:
            img_format = felem.get_text()
        else:
            log.debug("Failed to get img format", self)

        # !! GET source
        felem = result.find("span", {"class": "source"})
        if felem:
            source = felem.get_text()
        else:
            log.debug("Failed to get img source", self)

        # !! GET engines
        #<p class="result-engine">
        felem = result.find("p", {"class": "result-engine"})
        if felem:
            for engine in felem.find_all("span"):
                engines.append(engine.nextSibling.get_text().replace(' ', '-'))
        else:
            log.debug("Failed to get img source", self)

        return {
            'title': title,
            'url': url,
            'content': content,
            'engines': [engine for engine in engines],
            'img_format': img_format,
            'img_src': img_src,
            'thumbnail_src': thumbnail_src,
            'source': source,
            'category': category
        }

    def parse(self):
        if self.errorType() != ErrorType.Success:
            return {}

        jsonResult = {
            'results': [],
            'answers': [],
            'corrections': [],
            'infoboxes': [],
            'suggestions': [],
            'unresponsive_engines': []
        }

        soup = BeautifulSoup(self.content(), "html.parser")

        #######################################################################
        ## 'results' key
        #######################################################################
        def _getResults():
            # Because the element may be a 'article' or 'div', also depending
            # on the category of the result.
            for result in soup.find_all("article", {"class": "result"}):
                yield result
            for result in soup.find_all("div", {"class": "result"}):
                yield result

        for result in _getResults():
            # Image results
            if "result-images" in result.get("class"):
                jsonResult['results'].append(
                    self.parseImagesResult(result)
                )
                continue

            # Normal search results
            """
            <article class="result result-default category-general qwant duckduckgo google">
              <a href="https://linuxize.com/post/curl-post-request/" class="url_wrapper" rel="noreferrer">
                <span class="url_o1">
                  <span class="url_i1">https://linuxize.com</span>
                </span>
                <span class="url_o2">
                  <span class="url_i2"> › post › curl-post-request</span>
                </span>
              </a>
              <h3>
                <a href="https://linuxize.com/post/curl-post-request/" rel="noreferrer">
                  How to make a <span class="highlight">POST</span>
                  <span class="highlight">request</span>
                  with <span class="highlight">curl</span>
                </a>
              </h3>
              <p class="content">
                Learn how to use <span class="highlight">curl</span>, a command-line utility for transferring data from or to a remote server, to make <span class="highlight">POST</span> requests. See examples of sending data, files, and JSON data with <span class="highlight">curl</span> options and options.
              </p>
              <div class="engines">
                <span>qwant</span>
                <span>duckduckgo</span>
                <span>google</span>
                <a href="https://web.archive.org/web/https://linuxize.com/post/curl-post-request/" class="cache_link" rel="noreferrer">
                  <svg SVG_STUFF .../></svg>
                  cached
                </a>
                &lrm;
              </div>
              <div class="break"></div>
            </article>
            """

            """
            <article class="result result-torrent category-files solidtorrents">
              <a href="https://solidtorrents.to/torrents/STUFF .../" class="url_wrapper" rel="noreferrer">
                <span class="url_o1">
                  <span class="url_i1">https://solidtorrents.to</span>
                </span>
                <span class="url_o2">
                  <span class="url_i2"> › torrents › SOME_NAME › SOME_HASH</span>
                </span>
              </a>
              <h3>
                <a href="https://solidtorrents.to/torrents/SOME_NAME/SOME_HASH/" rel="noreferrer">
                  <span class="highlight">SOME</span>-<span class="highlight">NAME</span>
                </a>
              </h3>
              <time class="published_date" datetime="2018-10-20 00:00:00" >Oct 20, 2018</time>
              <div class="highlight">Other/Archive</div>
              <p class="altlink">
                &bull;
                <a href="magnet:MAGNET_LINK ..." class="magnetlink" rel="noreferrer"><svg SVG_STUFF .../></svg>magnet link</a>
              </p>
              <p class="altlink">
                &bull;
                <a href="https://itorrents.org/torrent/TORRENT_LINK ..." class="torrentfile" rel="noreferrer">torrent file</a>
              </p>
              <p class="stat">
                &bull;  Seeder
                <span class="badge">407</span>
                &bull; Leecher 
                <span class="badge">748</span>
              </p>
              <p class="stat"> Filesize
                <span class="badge">2.88 GiB</span>
              </p>
              <div class="engines">
                <span>solidtorrents</span>
                <a href="https://web.archive.org/web/https://solidtorrents.to/torrents/TORRENT_STUFF ..." class="cache_link" rel="noreferrer"><svg SVG_STUFF .../></svg>cached</a>
                &lrm;
              </div>
              <div class="break"></div>
            </article>
            """

            title = ''
            url = ''
            content = ''
            engines = []
            publishedDate = ''
            magnetlink = ''
            torrentfile = ''
            filesize = 0
            files = 0  # TODO unused for now
            seed = None
            leech = None

            # !! GET Title
            try:
                title = result.h3.a.get_text().lstrip().rstrip()
            except AttributeError:
                print("Failed to get title")

            # !! GET URL
            try:
                url = result.h3.a.get("href")
            except AttributeError:
                print("Failed to get url")

            # !! GET Content
            felem = result.find("p", {"class": "content"})
            if felem:
                content = felem.get_text().lstrip().rstrip()

            # !! GET Engines
            felem = result.find("div", {"class": "engines"})
            if felem:
                for engine in felem.find_all("span"):
                    engines.append(engine.get_text().rstrip().lstrip())

            ## !! Get publishDate
            felem = result.find("time", {"class": "published_date"})
            if felem:
                publishedDate = felem.get("datetime", "")

            ## !! Get magnetlink
            felem = result.find("a", {"class": "magnetlink"})
            if felem:
                magnetlink = felem.get('href')

            ## !! Get torrentfile
            felem = result.find("a", {"class": "torrentfile"})
            if felem:
                torrentfile = felem.get('href')

            ## !! Get filesize
            for felem in result.find_all("span", {"class": "badge"}):
                if felem.previousSibling:
                    precedingText = felem.previousSibling
                    if "Filesize" in precedingText:
                        filesize = parseFilesize(felem.get_text().rstrip().lstrip())
                    elif "Seeder" in precedingText:
                        seed = felem.get_text()
                    elif "Leecher" in precedingText:
                        leech = felem.get_text()

            # !! Add result
            resultData = {
                'title': title,
                'url': url,
                'content': content,
                'engines': [engine for engine in engines],
                # Optional
                'publishedDate': publishedDate,
                # File attributes
                'magnetlink': magnetlink,
                'torrentfile': torrentfile,
                'filesize': filesize,
                'files': files,
                'img_format': ''  # TODO
            }
            if seed is not None:
                resultData.update({'seed': seed})
            if leech is not None:
                resultData.update({'leech': leech})
            jsonResult['results'].append(resultData)

        ##########################################################################
        ## 'suggestions' key
        ##########################################################################
        """
        <div id="sidebar">
        <div id="suggestions" role="complementary" aria-labelledby="suggestions-title">
        <details class="sidebar-collapsable">
        <summary class="title" id="suggestions-title">Suggestions</summary>
        <div class="wrapper">
        <form method="POST" action="/search">
          <input type="hidden" name="q" value="curl post request json">
          <input type="hidden" name="category_general" value="1">
          <input type="hidden" name="language" value="auto">
          <input type="hidden" name="time_range" value="">
          <input type="hidden" name="safesearch" value="0">
          <input type="hidden" name="theme" value="simple">
          <input type="submit" class="suggestion" role="link" value="&bull; curl post request json">
        """

        felem = soup.find("div", {"id": "suggestions"})
        if felem:
            for suggestion in felem.find_all("input", {"name": "q"}):
                jsonResult['suggestions'].append(suggestion.get("value"))


        ##########################################################################
        ## 'answers' key
        ##########################################################################
        """
        <h4 class="title" id="answers-title">Answers : </h4>
        <div class="answer">
            <span>LONG TEXT ...</span>
            <a href="some url ..." class="answer-url">url text ...</a>
        </div>
        """
        for answer in soup.find_all("div", {"class": "answer"}):
            felem = answer.find("span")
            if felem:
                jsonResult['answers'].append(felem.get_text())

        ##########################################################################
        ## 'corrections' key
        ##########################################################################
        """ TODO """

        ##########################################################################
        ## 'infoboxes' key
        ##########################################################################
        """
        <details open="" class="sidebar-collapsable">
            <summary class="title">Info</summary>
            <aside class="infobox" aria-label="Banana">
                <h2 class="title"><bdi>Banana</bdi></h2>
                <img src="/image_proxy?url=long_image_url" title="Banana" alt="Banana">
                <p><bdi>LOGNG TEXT HERE ...</bdi></p>
                <div class="urls">
                    <ul>
                        <li class="url"><bdi><a href="https://en.wikipedia.org/wiki/Banana" rel="noreferrer">Wikipedia</a></bdi></li>
                        <li class="url"><bdi><a href="http://www.wikidata.org/entity/Q503" rel="noreferrer">Wikidata</a></bdi></li>
                    </ul>
                </div>
            </aside>
        </details>
        """

        """
        <details open="" class="sidebar-collapsable">
            <summary class="title">Info</summary>
            <aside class="infobox" aria-label="Water">
                <h2 class="title"><bdi>Water</bdi></h2>
                <img src="/image_proxy?url=long url .." title="Water" alt="Water">
                <p><bdi>LONG TEXT ...</bdi></p>
                <div class="attributes">
                    <dl>
                        <dt><bdi>Chemical formula :</bdi></dt>
                        <dd><bdi>H₂O</bdi></dd>
                    </dl>
                </div>
                <div class="urls">
                    <ul>
                        <li class="url"><bdi><a href="https://en.wikipedia.org/wiki/Water" rel="noreferrer">Wikipedia</a></bdi></li>
                        <li class="url"><bdi><a href="http://www.wikidata.org/entity/Q283" rel="noreferrer">Wikidata</a></bdi></li>
                    </ul>
                </div>
            </aside>
        </details>
        """

        """
        infoboxes = []
        ibox = {
            'infobox': 'str',
            'id': 'uri',
            'content': 'str',
            'img_src': 'uri' | null
            'urls': [
                {
                    'title': 'str',
                    'url': 'uri',
                    'entity': 'str',
                    'official': true
                }
            ],
            'attributes': [
                {
                    'label': 'str',
                    'value': 'str',
                    'entity': 'str'
                }
            ],
            'engines': ['str'],
            'engine': 'str'
        }
        """
        for infobox in soup.find_all("aside", {"class": "infobox"}):
            title = ""
            id = ""
            content = ""
            img_src = ""
            urls = []
            attributes = []
            engines = []

            # Title
            felem = infobox.find("h2", {"class": "title"})
            if felem:
                title = felem.get_text().rstrip().lstrip()

            # ID
            # TODO

            # Content
            felem = infobox.find("p")
            if felem:
                felem = felem.find("bdi")
                if felem:
                    content = felem.get_text().rstrip().lstrip()

            # Image
            felem = infobox.find("img")
            if felem:
                img_src = self.makeUrlAbsolute(felem.get("src"))

            # URLs
            for felem in infobox.find_all("li", {"class": "url"}):
                felem = felem.find("a")
                if felem:
                    urls.append({
                        'title': felem.get_text().lstrip().rstrip(),
                        'url': felem.get("href", ""),
                        'entity': '',  # TODO
                        'official': False  # TODO
                    })

            # Attributes
            """
            <div class="attributes">
                <dl>
                    <dt><bdi>Chemical formula :</bdi></dt>
                    <dd><bdi>H₂O</bdi></dd>
                </dl>
            </div>
            """
            felem = infobox.find("div", {"class": "attributes"})
            if felem:
                for item in felem.find_all("dl"):
                    label = ""
                    value = ""
                    entity = ""  # TODO

                    try:
                        label = item.dt.bdi.get_text().rstrip().lstrip()
                        value = item.dd.bdi.get_text().rstrip().lstrip()
                    except AttributeError:
                        continue

                    attributes.append({
                        "label": label,
                        "value": value,
                        "entity": entity
                    })

            # Engines
            for url in urls:
                engines.append(url['title'].lower())

            jsonResult['infoboxes'].append({
                "infobox": title,
                "id": id,
                "content": content,
                "img_src": img_src,
                "urls": urls,
                "attributes": attributes,
                "engines": engines
            })

        ##########################################################################
        ## 'unresponsive_engines' key
        ##########################################################################
        """
        <div id="engines_msg">
            <details class="sidebar-collapsable" open="">
                <summary class="title" id="engines_msg-title">Messages from the search engines</summary>
                <div class="dialog-error" role="alert">
                    <svg class="ion-icon-big" etc..></svg>
                    <div>
                        <p>
                            <strong>Error!</strong>
                            Engines cannot retrieve results:
                        </p>
                        <p>
                            brave (<a href="/stats?engine=brave" title="View error logs and submit a bug report">Suspended: too many requests</a>)
                        </p>
                        <p>
                            qwant (<a href="/stats?engine=qwant" title="View error logs and submit a bug report">Suspended: too many requests</a>)
                        </p>
                    </div>
                </div>
            </details>
        </div>
        """
        felem = soup.find("div", {"id": "engines_msg"})
        if felem:
            for errDialog in felem.find_all("div", {"class": "dialog-error"}):
                for p in errDialog.find_all("p"):
                    a = p.find("a")
                    if not a:
                        continue

                    engine, msg = p.get_text().split(" ", 1)
                    jsonResult['unresponsive_engines'].append([engine, msg])


        return jsonResult


class SearxConfigResult(JsonResult):
    Schema = Schemas['searxng_config']

    def __init__(self, url, response, err="", errType=ErrorType.Success):
        JsonResult.__init__(self, url, response, err=err, errType=errType)


class Categories:
    types = {
        'general': (_('General'), 'category_general'),
        'files': (_('Files'), 'category_files'),
        'images': (_('Images'), 'category_images'),
        'videos': (_('Videos'), 'category_videos'),
        'it': (_('IT'), 'category_it'),
        'map': (_('Location'), 'category_map'),
        'music': (_('Music'), 'category_music'),
        'news': (_('News'), 'category_news'),
        'science': (_('Science'), 'category_science'),
        'social media': (_('Social'), 'category_social media'),
        'onions': (_('Onions'), 'category_onions'),
        'shopping': (_('Shopping'), 'category_shopping')
    }

    def __init__(self):
        self._options = {}
        self.__makeOptions()

    def __makeOptions(self):
        self._options.clear()
        for key, t in self.types.items():
            self._options.update({key: False})

    def reset(self):
        self.__makeOptions()

    def get(self, key):
        return self._options[key]

    def set(self, key, state):
        """
        @param key: One of the keys in Categories.types
        @type key: str

        @param state: Enabled / disabled state
        @type state: bool
        """
        self._options[key] = state

    def dict(self):
        newDict = {}
        for key, state in self._options.items():
            if state:
                newDict.update({self.types[key][1]: 'on'})
        return newDict

    def enabledKeys(self):
        """ Returns a list with enabled engine strings (key from
        Categories.types)
        """
        return [key for key, state in self._options.items() if state]


class Engines(list):
    def __init__(self):
        list.__init__(self)

    def dict(self):
        if not self:
            return {}
        return {
            'engines': ",".join(self)
        }


class SearX:
    Periods = {
        '': _('Anytime'),
        'day': _('Last day'),
        'week': _('Last week'),
        'month': _('Last month'),
        'year': _('Last year')
    }

    # https://github.com/asciimoo/searx/blob/master/searx/languages.py
    Languages = {
        '': _('No language'),
        'all': _('Default language'),
        'af-NA': 'Afrikaans - af-NA',
        'ca-AD': 'Català - ca-AD',
        'da-DK': 'Dansk - da-DK',
        'de': 'Deutsch - de',
        'de-AT': 'Deutsch (Österreich) - de-AT',
        'de-CH': 'Deutsch (Schweiz) - de-CH',
        'de-DE': 'Deutsch (Deutschland) - de-DE',
        'et-EE': 'Eesti - et-EE',
        'en': 'English - en',
        'en-AU': 'English (Australia) - en-AU',
        'en-CA': 'English (Canada) - en-CA',
        'en-GB': 'English (United Kingdom) - en-GB',
        'en-IE': 'English (Ireland) - en-IE',
        'en-IN': 'English (India) - en-IN',
        'en-NZ': 'English (New Zealand) - en-NZ',
        'en-PH': 'English (Philippines) - en-PH',
        'en-SG': 'English (Singapore) - en-SG',
        'en-US': 'English (United States) - en-US',
        'es': 'Español - es',
        'es-AR': 'Español (Argentina) - es-AR',
        'es-CL': 'Español (Chile) - es-CL',
        'es-ES': 'Español (España) - es-ES',
        'es-MX': 'Español (México) - es-MX',
        'fr': 'Français - fr',
        'fr-BE': 'Français (Belgique) - fr-BE',
        'fr-CA': 'Français (Canada) - fr-CA',
        'fr-CH': 'Français (Suisse) - fr-CH',
        'fr-FR': 'Français (France) - fr-FR',
        'hr-HR': 'Hrvatski - hr-HR',
        'id-ID': 'Indonesia - id-ID',
        'it-IT': 'Italiano - it-IT',
        'sw-KE': 'Kiswahili - sw-KE',
        'lv-LV': 'Latviešu - lv-LV',
        'lt-LT': 'Lietuvių - lt-LT',
        'hu-HU': 'Magyar - hu-HU',
        'ms-MY': 'Melayu - ms-MY',
        'nl': 'Nederlands - nl',
        'nl-BE': 'Nederlands (België) - nl-BE',
        'nl-NL': 'Nederlands (Nederland) - nl-NL',
        'nb-NO': 'Norsk Bokmål - nb-NO',
        'pl-PL': 'Polski - pl-PL',
        'pt': 'Português - pt',
        'pt-BR': 'Português (Brasil) - pt-BR',
        'pt-PT': 'Português (Portugal) - pt-PT',
        'ro-RO': 'Română - ro-RO',
        'sk-SK': 'Slovenčina - sk-SK',
        'sl-SI': 'Slovenščina - sl-SI',
        'sr-RS': 'Srpski - sr-RS',
        'fi-FI': 'Suomi - fi-FI',
        'sv-SE': 'Svenska - sv-SE',
        'vi-VN': 'Tiếng Việt - vi-VN',
        'tr-TR': 'Türkçe - tr-TR',
        'is-IS': 'Íslenska - is-IS',
        'cs-CZ': 'Čeština - cs-CZ',
        'el-GR': 'Ελληνικά - el-GR',
        'be-BY': 'Беларуская - be-BY',
        'bg-BG': 'Български - bg-BG',
        'ru-RU': 'Русский - ru-RU',
        'uk-UA': 'Українська - uk-UA',
        'hy-AM': 'Հայերեն - hy-AM',
        'he-IL': 'עברית - he-IL',
        'ar-SA': 'العربية - ar-SA',
        'fa-IR': 'فارسی - fa-IR',
        'th-TH': 'ไทย - th-TH',
        'zh': '中文 - zh',
        'zh-CN': '中文 (中国) - zh-CN',
        'zh-TW': '中文 (台灣) - zh-TW',
        'ja-JP': '日本語 - ja-JP',
        'ko-KR': '한국어 - ko-KR'
    }

    def __init__(self, requestHandler):
        self._requestHandler = requestHandler

        self._url = ""
        self._categories = Categories()
        self._engines = Engines()

        self._query = ""
        self._lang = ""
        self._pageno = ""  # int formatted as string
        self._timeRange = ""  # '', 'day', 'week', 'month' or 'year'
        self._safesearch = False

        self._parseHtml = True

    @property
    def categories(self): return self._categories

    @property
    def engines(self): return self._engines

    @property
    def url(self):
        """
        @return: Instance url
        @rtype: str
        """
        return self._url

    @url.setter
    def url(self, url):
        """
        @param url: Instance url
        @type url: str
        """
        self._url = url

    @property
    def query(self):
        """
        @return: Search query
        @rtype: str
        """
        return self._query

    @query.setter
    def query(self, q):
        """
        @param q: Search query
        @type q: str
        """
        self._query = q

    @property
    def lang(self):
        """
        @return: Language code
        @rtype: str
        """
        return self._lang

    @lang.setter
    def lang(self, lang):
        """
        @param lang: Language code
        @type lang: str
        """
        self._lang = lang

    @property
    def pageno(self):
        """
        @return: Page number
        @rtype: int
        """
        return int(self._pageno)

    @pageno.setter
    def pageno(self, i):
        """
        @param i: Page number
        @type i: int
        """
        self._pageno = str(i)

    @property
    def timeRange(self):
        """
        @return: Search time range ('', 'day', 'week', 'month' or 'year')
        @rtype: str
        """
        return self._timeRange

    @timeRange.setter
    def timeRange(self, value):
        """
        @param value: Key from SearX.Periods
        @type value: str
        """
        self._timeRange = value

    @property
    def safeSearch(self):
        """
        @return: Whether safe search is enabled or not.
        @rtype: bool
        """
        return self._safesearch

    @safeSearch.setter
    def safeSearch(self, state):
        """
        @param state: Enable/disable safe search.
        @type state: bool
        """
        self._safesearch = state

    @property
    def parseHtml(self):
        """
        @return: Whether parsing HTML is enabled, this will not use the
                 JSON API when it returns True.
        @rtype: bool
        """
        return self._parseHtml

    @parseHtml.setter
    def parseHtml(self, state):
        """
        @param value: Enable/disable parsing HTML instead of using the JSON API
        @type value: bool
        """
        self._parseHtml = state

    @property
    def requestKwargs(self):
        """ Returns current data that will be send with the POST
        request used for the search operation. The search query,
        language, page-number and enabled categories/engines.

        @rtype: dict
        """
        data = {
            "q": self.query,
            "safesearch": "1" if self.safeSearch else "0"
        }

        # Choose what resource to use (JSON API or HTML parser)
        if self.parseHtml:
            data.update({"theme": "simple"})
        else:
            data.update({"format": "json"})

        # After testing found that searx will honor only engines when
        # both engines and categories are set.
        if self.engines:
            data.update(self.engines.dict())
        elif self.categories:
            data.update(self.categories.dict())

        if self.lang:
            data.update({"language": self.lang})
        if self.pageno:
            data.update({"pageno": self.pageno})
        if self.timeRange:
            data.update({"time_range": self.timeRange})

        return data

    def reset(self):
        self.url = ""
        self.timeRange = ""
        self.lang = ""
        self.pageno = 1
        self.categories.reset()
        self.engines.clear()
        self.query = ""

    def search(self):
        """ Preform search operation with current set values.

        @returns: The result of this search.
        @rtype: SearchResult
        """
        rtype = SearchResult
        if self.parseHtml:
            rtype = SearchResult2
        return self._requestHandler.get(
            self.url,
            data=self.requestKwargs,
            ResultType=rtype
        )


class SearxConfigHandler(HandlerProto):
    def __init__(self, requestsHandler):
        HandlerProto.__init__(self, requestsHandler)

    def updateInstance(self, url):
        newUrl = urllib.parse.urljoin(url, "/config")
        result = self.requestsHandler.get(newUrl, ResultType=SearxConfigResult)

        if result:
            instance = self.instances[url]
            j = result.json()

            """ Update instance version
            """
            instance.update({
                "version": j.get("version", "")
            })

            """ Update instance network_type to use our own network type
            definitions as class NetworkTypes (core/handler.py)
            """
            instance.update({"network_type": NetworkTypes.netTypeFromUrl(url)})

            """ Update Engines

            What we get:
                "engines": [
                  categories (list, str)
                  enabled (bool)
                  language_support (bool)
                  name (str)
                  paging (bool)
                  safesearch (bool)
                  shortcut (str)
                  supported_languages (list, str)
                  time_range_support (bool)
                  timeout (float)
                ]

            What instanceModel wants
                "engines" : {
                    "not evil": {
                        "error_rate": 15,
                        "errors": [
                            0
                        ]
                    }
                }

            What enginesModel wants
                "engines": {
                    "1337x": {
                        "categories": [
                            "videos"
                        ],
                        "language_support": true,
                        "paging": true,
                        "safesearch": false,
                        "shortcut": "1337x",
                        "time_range_support": false
                    },
            """
            newInstanceEngines = {}
            newEnginesEngines = {}
            for engine in j.get('engines', []):
                name = engine.get('name', "")
                if not name:
                    continue

                newInstanceEngines.update({
                    name: {}
                })

                if name not in self.engines:
                    newEnginesEngines.update({
                        name: {
                            "categories": list(engine.get('categories', [])),
                            "language_support": engine.get(
                                'language_support',
                                False
                            ),
                            "paging": engine.get('paging', False),
                            "safesearch": engine.get('safesearch', False),
                            "shortcut": engine.get('shortcut', ""),
                            "time_range_support": engine.get(
                                'time_range_support',
                                False
                            )
                        }
                    })

            instance.update({
                "engines": dict(newInstanceEngines)
            })

            self.engines.update(newEnginesEngines)

            """ Update instance lastUpdated
            """
            instance.update({
                "lastUpdated": time.time()
            })

            return True
        return False

    def addInstance(self, url):
        if url not in self.instances:
            self._instances[url] = {}
            return True
        return False

    def removeInstance(self, url):
        """
        @param url: url of the instance to remove.
        @type url: str
        """
        del self._instances[url]

    def removeMultiInstances(self, urls):
        """ Remove instance(s) by url without emitting changed for every
        instance that got removed.

        @param urls: list with urls of instances to remove.
        @type urls: list
        """
        for url in urls:
            del self._instances[url]