X-Git-Url: https://git.lizzy.rs/?a=blobdiff_plain;f=nhentai%2Fparser.py;h=7b9142cd3232b9cead8337a96e13e06471573e05;hb=ba141efba7d31cf46fa90da7802ca59a54319ff3;hp=6f5323a3eae7ee4750c382d2befebabdcac2db5b;hpb=0df51c83e57b3a72bd9c5494ea421fb4e7d1c5ca;p=nhentai.git

diff --git a/nhentai/parser.py b/nhentai/parser.py
index 6f5323a..7b9142c 100644
--- a/nhentai/parser.py
+++ b/nhentai/parser.py
@@ -1,54 +1,74 @@
 # coding: utf-8
-from __future__ import unicode_literals, print_function
 
 import os
 import re
-import threadpool
-import requests
 import time
 from bs4 import BeautifulSoup
 from tabulate import tabulate
 
 import nhentai.constant as constant
+from nhentai.utils import request
 from nhentai.logger import logger
 
 
-def request(method, url, **kwargs):
-    if not hasattr(requests, method):
-        raise AttributeError('\'requests\' object has no attribute \'{0}\''.format(method))
-
-    return requests.__dict__[method](url, proxies=constant.PROXY, verify=False, **kwargs)
-
-
-def login_parser(username, password):
-    s = requests.Session()
-    s.proxies = constant.PROXY
-    s.verify = False
-    s.headers.update({'Referer': constant.LOGIN_URL})
-
-    s.get(constant.LOGIN_URL)
-    content = s.get(constant.LOGIN_URL).content
+def _get_csrf_token(content):
     html = BeautifulSoup(content, 'html.parser')
     csrf_token_elem = html.find('input', attrs={'name': 'csrfmiddlewaretoken'})
-
     if not csrf_token_elem:
         raise Exception('Cannot find csrf token to login')
-    csrf_token = csrf_token_elem.attrs['value']
+    return csrf_token_elem.attrs['value']
+
+
+def login(username, password):
+    logger.warning('This feature is deprecated, please use --cookie to set your cookie.')
+    csrf_token = _get_csrf_token(request('get', url=constant.LOGIN_URL).text)
+    if os.getenv('DEBUG'):
+        logger.info('Getting CSRF token ...')
+
+    if os.getenv('DEBUG'):
+        logger.info('CSRF token is {}'.format(csrf_token))
 
     login_dict = {
         'csrfmiddlewaretoken': csrf_token,
         'username_or_email': username,
         'password': password,
     }
-    resp = s.post(constant.LOGIN_URL, data=login_dict)
+    resp = request('post', url=constant.LOGIN_URL, data=login_dict)
+
+    if 'You\'re loading pages way too quickly.' in resp.text or 'Really, slow down' in resp.text:
+        csrf_token = _get_csrf_token(resp.text)
+        resp = request('post', url=resp.url, data={'csrfmiddlewaretoken': csrf_token, 'next': '/'})
+
     if 'Invalid username/email or password' in resp.text:
         logger.error('Login failed, please check your username and password')
         exit(1)
 
-    html = BeautifulSoup(s.get(constant.FAV_URL).content, 'html.parser')
+    if 'You\'re loading pages way too quickly.' in resp.text or 'Really, slow down' in resp.text:
+        logger.error('Using nhentai --cookie \'YOUR_COOKIE_HERE\' to save your Cookie.')
+        exit(2)
+
+
+def _get_title_and_id(response):
+    result = []
+    html = BeautifulSoup(response, 'html.parser')
+    doujinshi_search_result = html.find_all('div', attrs={'class': 'gallery'})
+    for doujinshi in doujinshi_search_result:
+        doujinshi_container = doujinshi.find('div', attrs={'class': 'caption'})
+        title = doujinshi_container.text.strip()
+        title = title if len(title) < 85 else title[:82] + '...'
+        id_ = re.search('/g/(\d+)/', doujinshi.a['href']).group(1)
+        result.append({'id': id_, 'title': title})
+
+    return result
+
+
+def favorites_parser(page=None):
+    result = []
+    html = BeautifulSoup(request('get', constant.FAV_URL).content, 'html.parser')
     count = html.find('span', attrs={'class': 'count'})
     if not count:
         logger.error("Can't get your number of favorited doujins. Did the login failed?")
+        return []
 
     count = int(count.text.strip('(').strip(')').replace(',', ''))
     if count == 0:
@@ -56,36 +76,31 @@ def login_parser(username, password):
         return []
     pages = int(count / 25)
 
-    if pages:
-        pages += 1 if count % (25 * pages) else 0
+    if page:
+        page_range_list = page
     else:
-        pages = 1
-
-    logger.info('You have %d favorites in %d pages.' % (count, pages))
-
-    if os.getenv('DEBUG'):
-        pages = 1
+        if pages:
+            pages += 1 if count % (25 * pages) else 0
+        else:
+            pages = 1
 
-    ret = []
-    doujinshi_id = re.compile('data-id="([\d]+)"')
+        logger.info('You have %d favorites in %d pages.' % (count, pages))
 
-    def _callback(request, result):
-        ret.append(result)
+        if os.getenv('DEBUG'):
+            pages = 1
 
-    thread_pool = threadpool.ThreadPool(5)
+        page_range_list = range(1, pages + 1)
 
-    for page in range(1, pages+1):
+    for page in page_range_list:
         try:
             logger.info('Getting doujinshi ids of page %d' % page)
-            resp = s.get(constant.FAV_URL + '?page=%d' % page).text
-            ids = doujinshi_id.findall(resp)
-            requests_ = threadpool.makeRequests(doujinshi_parser, ids, _callback)
-            [thread_pool.putRequest(req) for req in requests_]
-            thread_pool.wait()
+            resp = request('get', constant.FAV_URL + '?page=%d' % page).content
+
+            result.extend(_get_title_and_id(resp))
         except Exception as e:
             logger.error('Error: %s, continue', str(e))
 
-    return ret
+    return result
 
 
 def doujinshi_parser(id_):
@@ -93,74 +108,79 @@ def doujinshi_parser(id_):
         raise Exception('Doujinshi id({0}) is not valid'.format(id_))
 
     id_ = int(id_)
-    logger.log(15, 'Fetching information of doujinshi id {0}'.format(id_))
+    logger.log(15, 'Fetching doujinshi information of id {0}'.format(id_))
     doujinshi = dict()
     doujinshi['id'] = id_
-    url = '{0}/{1}'.format(constant.DETAIL_URL, id_)
-    i=0
-    while i<5:
-        try:
-            response = request('get', url).json()
-        except Exception as e:
-            i+=1
-            if not i<5:
-                logger.critical(str(e))
-                exit(1)
-            continue
-        break
+    url = '{0}/{1}/'.format(constant.DETAIL_URL, id_)
+
+    try:
+        response = request('get', url)
+        if response.status_code in (200, ):
+            response = response.content
+        elif response.status_code in (404,):
+            logger.error("Doujinshi with id {0} cannot be found".format(id_))
+            return []
+        else:
+            logger.debug('Slow down and retry ({}) ...'.format(id_))
+            time.sleep(1)
+            return doujinshi_parser(str(id_))
+
+    except Exception as e:
+        logger.warn('Error: {}, ignored'.format(str(e)))
+        return None
 
-    doujinshi['name'] = response['title']['english']
-    doujinshi['subtitle'] = response['title']['japanese']
-    doujinshi['img_id'] = response['media_id']
-    doujinshi['ext'] = ''.join(map(lambda s: s['t'], response['images']['pages']))
-    doujinshi['pages'] = len(response['images']['pages'])
+    html = BeautifulSoup(response, 'html.parser')
+    doujinshi_info = html.find('div', attrs={'id': 'info'})
 
-    # gain information of the doujinshi
-    needed_fields = ['character', 'artist', 'language', 'tag']
-    for tag in response['tags']:
-        tag_type = tag['type']
-        if tag_type in needed_fields:
-            if tag_type == 'tag':
-                if tag_type not in doujinshi:
-                    doujinshi[tag_type] = {}
+    title = doujinshi_info.find('h1').text
+    subtitle = doujinshi_info.find('h2')
 
-                tag['name'] = tag['name'].replace(' ', '-')
-                tag['name'] = tag['name'].lower()
-                doujinshi[tag_type][tag['name']] = tag['id']
-            elif tag_type not in doujinshi:
-                doujinshi[tag_type] = tag['name']
-            else:
-                doujinshi[tag_type] += ', ' + tag['name']
+    doujinshi['name'] = title
+    doujinshi['subtitle'] = subtitle.text if subtitle else ''
 
-    return doujinshi
+    doujinshi_cover = html.find('div', attrs={'id': 'cover'})
+    img_id = re.search('/galleries/([\d]+)/cover\.(jpg|png|gif)$', doujinshi_cover.a.img.attrs['data-src'])
 
+    ext = []
+    for i in html.find_all('div', attrs={'class': 'thumb-container'}):
+        _, ext_name = os.path.basename(i.img.attrs['data-src']).rsplit('.', 1)
+        ext.append(ext_name)
 
-def search_parser(keyword, page):
-    logger.debug('Searching doujinshis using keywords {0}'.format(keyword))
-    result = []
-    i=0
-    while i<5:
-        try:
-            response = request('get', url=constant.SEARCH_URL, params={'query': keyword, 'page': page}).json()
-        except Exception as e:
-            i+=1
-            if not i<5:
-                logger.critical(str(e))
-                logger.warn('If you are in China, please configure the proxy to fu*k GFW.')
-                exit(1)
-            continue
-        break
+    if not img_id:
+        logger.critical('Tried yo get image id failed')
+        exit(1)
+
+    doujinshi['img_id'] = img_id.group(1)
+    doujinshi['ext'] = ext
 
-    if 'result' not in response:
-        raise Exception('No result in response')
+    for _ in doujinshi_info.find_all('div', class_='tag-container field-name'):
+        if re.search('Pages:', _.text):
+            pages = _.find('span', class_='name').string
+    doujinshi['pages'] = int(pages)
 
-    for row in response['result']:
-        title = row['title']['english']
-        title = title[:85] + '..' if len(title) > 85 else title
-        result.append({'id': row['id'], 'title': title})
+    # gain information of the doujinshi
+    information_fields = doujinshi_info.find_all('div', attrs={'class': 'field-name'})
+    needed_fields = ['Characters', 'Artists', 'Languages', 'Tags', 'Parodies', 'Groups', 'Categories']
+    for field in information_fields:
+        field_name = field.contents[0].strip().strip(':')
+        if field_name in needed_fields:
+            data = [sub_field.find('span', attrs={'class': 'name'}).contents[0].strip() for sub_field in
+                    field.find_all('a', attrs={'class': 'tag'})]
+            doujinshi[field_name.lower()] = ', '.join(data)
+
+    time_field = doujinshi_info.find('time')
+    if time_field.has_attr('datetime'):
+        doujinshi['date'] = time_field['datetime']
+    return doujinshi
 
+
+def old_search_parser(keyword, sorting='date', page=1):
+    logger.debug('Searching doujinshis of keyword {0}'.format(keyword))
+    response = request('get', url=constant.SEARCH_URL, params={'q': keyword, 'page': page, 'sort': sorting}).content
+
+    result = _get_title_and_id(response)
     if not result:
-        logger.warn('No results for keywords {}'.format(keyword))
+        logger.warn('Not found anything of keyword {}'.format(keyword))
 
     return result
 
@@ -170,88 +190,96 @@ def print_doujinshi(doujinshi_list):
         return
     doujinshi_list = [(i['id'], i['title']) for i in doujinshi_list]
     headers = ['id', 'doujinshi']
-    logger.info('Search Result\n' +
+    logger.info('Search Result || Found %i doujinshis \n' % doujinshi_list.__len__() +
                 tabulate(tabular_data=doujinshi_list, headers=headers, tablefmt='rst'))
 
 
-def tag_parser(tag_id, max_page=1):
-    logger.info('Searching for doujinshi with tag id {0}'.format(tag_id))
+def search_parser(keyword, sorting, page, is_page_all=False):
+    # keyword = '+'.join([i.strip().replace(' ', '-').lower() for i in keyword.split(',')])
     result = []
-    i=0
-    while i<5:
-        try:
-            response = request('get', url=constant.TAG_API_URL, params={'sort': 'popular', 'tag_id': tag_id}).json()
-        except Exception as e:
-            i+=1
-            if not i<5:
+    if not page:
+        page = [1]
+
+    if is_page_all:
+        url = request('get', url=constant.SEARCH_URL, params={'query': keyword}).url
+        init_response = request('get', url.replace('%2B', '+')).json()
+        page = range(1, init_response['num_pages']+1)
+
+    total = '/{0}'.format(page[-1]) if is_page_all else ''
+    for p in page:
+        i = 0
+
+        logger.info('Searching doujinshis using keywords "{0}" on page {1}{2}'.format(keyword, p, total))
+        while i < 3:
+            try:
+                url = request('get', url=constant.SEARCH_URL, params={'query': keyword,
+                                                                      'page': p, 'sort': sorting}).url
+                response = request('get', url.replace('%2B', '+')).json()
+            except Exception as e:
                 logger.critical(str(e))
-                exit(1)
-            continue
-        break
-    page = max_page if max_page <= response['num_pages'] else int(response['num_pages'])
-
-    for i in range(1, page+1):
-        logger.info('Getting page {} ...'.format(i))
-
-        if page != 1:
-            i=0
-            while i<5:
-                try:
-                    response = request('get', url=constant.TAG_API_URL, params={'sort': 'popular', 'tag_id': tag_id}).json()
-                except Exception as e:
-                    i+=1
-                    if not i<5:
-                        logger.critical(str(e))
-                        exit(1)
-                    continue
-                break
-    for row in response['result']:
-        title = row['title']['english']
-        title = title[:85] + '..' if len(title) > 85 else title
-        result.append({'id': row['id'], 'title': title})
 
-    if not result:
-        logger.warn('No results for tag id {}'.format(tag_id))
-    
+            break
+
+        if 'result' not in response:
+            logger.warn('No result in response in page {}'.format(p))
+            break
+
+        for row in response['result']:
+            title = row['title']['english']
+            title = title[:85] + '..' if len(title) > 85 else title
+            result.append({'id': row['id'], 'title': title})
+
+        if not result:
+            logger.warn('No results for keywords {}'.format(keyword))
+
     return result
 
 
-def tag_guessing(tag_name):
-    tag_name = tag_name.lower()
-    tag_name = tag_name.replace(' ', '-')
-    logger.info('Trying to get tag_id of tag \'{0}\''.format(tag_name))
-    i=0
-    while i<5:
+def __api_suspended_doujinshi_parser(id_):
+    if not isinstance(id_, (int,)) and (isinstance(id_, (str,)) and not id_.isdigit()):
+        raise Exception('Doujinshi id({0}) is not valid'.format(id_))
+
+    id_ = int(id_)
+    logger.log(15, 'Fetching information of doujinshi id {0}'.format(id_))
+    doujinshi = dict()
+    doujinshi['id'] = id_
+    url = '{0}/{1}'.format(constant.DETAIL_URL, id_)
+    i = 0
+    while 5 > i:
         try:
-            response = request('get', url='%s/%s' % (constant.TAG_URL, tag_name)).content
+            response = request('get', url).json()
         except Exception as e:
-            i+=1
-            if not i<5:
+            i += 1
+            if not i < 5:
                 logger.critical(str(e))
                 exit(1)
             continue
         break
 
-    html = BeautifulSoup(response, 'html.parser')
-    first_item = html.find('div', attrs={'class': 'gallery'})
-    if not first_item:
-        logger.error('Cannot find doujinshi id of tag \'{0}\''.format(tag_name))
-        return
+    doujinshi['name'] = response['title']['english']
+    doujinshi['subtitle'] = response['title']['japanese']
+    doujinshi['img_id'] = response['media_id']
+    doujinshi['ext'] = ''.join([i['t'] for i in response['images']['pages']])
+    doujinshi['pages'] = len(response['images']['pages'])
 
-    doujinshi_id = re.findall('(\d+)', first_item.a.attrs['href'])
-    if not doujinshi_id:
-        logger.error('Cannot find doujinshi id of tag \'{0}\''.format(tag_name))
-        return
+    # gain information of the doujinshi
+    needed_fields = ['character', 'artist', 'language', 'tag', 'parody', 'group', 'category']
+    for tag in response['tags']:
+        tag_type = tag['type']
+        if tag_type in needed_fields:
+            if tag_type == 'tag':
+                if tag_type not in doujinshi:
+                    doujinshi[tag_type] = {}
 
-    ret = doujinshi_parser(doujinshi_id[0])
-    if 'tag' in ret and tag_name in ret['tag']:
-        tag_id = ret['tag'][tag_name]
-        logger.info('Tag id of tag \'{0}\' is {1}'.format(tag_name, tag_id))
-    else:
-        logger.error('Cannot find doujinshi id of tag \'{0}\''.format(tag_name))
-        return
+                tag['name'] = tag['name'].replace(' ', '-')
+                tag['name'] = tag['name'].lower()
+                doujinshi[tag_type][tag['name']] = tag['id']
+            elif tag_type not in doujinshi:
+                doujinshi[tag_type] = tag['name']
+            else:
+                doujinshi[tag_type] += ', ' + tag['name']
 
-    return tag_id
+    return doujinshi
 
 
 if __name__ == '__main__':