fix bug #13

nhentai/doujinshi.py

@@ -5,6 +5,7 @@ from future.builtins import range
 
 from nhentai.constant import DETAIL_URL, IMAGE_URL
 from nhentai.logger import logger
+from nhentai.utils import format_filename
 
 
 class DoujinshiInfo(dict):
@@ -51,7 +52,7 @@ class Doujinshi(object):
             download_queue = []
             for i in range(1, self.pages + 1):
                 download_queue.append('%s/%d/%d.%s' % (IMAGE_URL, int(self.img_id), i, self.ext))
-            self.downloader.download(download_queue, self.id)
+            self.downloader.download(download_queue, format_filename('%s-%s' % (self.id, self.name[:200])))
         else:
             logger.critical('Downloader has not be loaded')
 

nhentai/downloader.py

@@ -1,67 +1,100 @@
-# coding: utf-8
-from __future__ import print_function, unicode_literals
-from tabulate import tabulate
-from future.builtins import range
+# coding: utf-
+from __future__ import unicode_literals, print_function
+from future.builtins import str as text
+import os
+import requests
+import threadpool
+try:
+    from urllib.parse import urlparse
+except ImportError:
+    from urlparse import urlparse
 
-from nhentai.constant import DETAIL_URL, IMAGE_URL
 from nhentai.logger import logger
-from nhentai.utils import format_filename
+from nhentai.parser import request
+from nhentai.utils import Singleton
 
 
-class DoujinshiInfo(dict):
-    def __init__(self, **kwargs):
-        super(DoujinshiInfo, self).__init__(**kwargs)
+requests.packages.urllib3.disable_warnings()
 
-    def __getattr__(self, item):
+
+class NhentaiImageNotExistException(Exception):
+    pass
+
+
+class Downloader(Singleton):
+
+    def __init__(self, path='', thread=1, timeout=30):
+        if not isinstance(thread, (int, )) or thread < 1 or thread > 15:
+            raise ValueError('Invalid threads count')
+        self.path = str(path)
+        self.thread_count = thread
+        self.threads = []
+        self.timeout = timeout
+
+    def _download(self, url, folder='', filename='', retried=0):
+        logger.info('Start downloading: {0} ...'.format(url))
+        filename = filename if filename else os.path.basename(urlparse(url).path)
+        base_filename, extension = os.path.splitext(filename)
         try:
-            return dict.__getitem__(self, item)
-        except KeyError:
-            return ''
-
-
-class Doujinshi(object):
-    def __init__(self, name=None, id=None, img_id=None, ext='jpg', pages=0, **kwargs):
-        self.name = name
-        self.id = id
-        self.img_id = img_id
-        self.ext = ext
-        self.pages = pages
-        self.downloader = None
-        self.url = '%s/%d' % (DETAIL_URL, self.id)
-        self.info = DoujinshiInfo(**kwargs)
-
-    def __repr__(self):
-        return '<Doujinshi: {0}>'.format(self.name)
-
-    def show(self):
-        table = [
-            ["Doujinshi", self.name],
-            ["Subtitle", self.info.subtitle],
-            ["Characters", self.info.characters],
-            ["Authors", self.info.artists],
-            ["Language", self.info.language],
-            ["Tags", self.info.tags],
-            ["URL", self.url],
-            ["Pages", self.pages],
-        ]
-        logger.info(u'Print doujinshi information of {0}\n{1}'.format(self.id, tabulate(table)))
-
-    def download(self):
-        logger.info('Start download doujinshi: %s' % self.name)
-        if self.downloader:
-            download_queue = []
-            for i in range(1, self.pages + 1):
-                download_queue.append('%s/%d/%d.%s' % (IMAGE_URL, int(self.img_id), i, self.ext))
-            self.downloader.download(download_queue, format_filename(self.name))
+            with open(os.path.join(folder, base_filename.zfill(3) + extension), "wb") as f:
+                response = request('get', url, stream=True, timeout=self.timeout)
+                if response.status_code != 200:
+                    raise NhentaiImageNotExistException
+                length = response.headers.get('content-length')
+                if length is None:
+                    f.write(response.content)
                 else:
-            logger.critical('Downloader has not be loaded')
+                    for chunk in response.iter_content(2048):
+                        f.write(chunk)
 
+        except (requests.HTTPError, requests.Timeout) as e:
+            if retried < 3:
+                logger.warning('Warning: {0}, retrying({1}) ...'.format(str(e), retried))
+                return 0, self._download(url=url, folder=folder, filename=filename, retried=retried+1)
+            else:
+                return 0, None
+
+        except NhentaiImageNotExistException as e:
+            os.remove(os.path.join(folder, base_filename.zfill(3) + extension))
+            return -1, url
 
-if __name__ == '__main__':
-    test = Doujinshi(name='test nhentai doujinshi', id=1)
-    print(test)
-    test.show()
-    try:
-        test.download()
         except Exception as e:
-        print('Exception: %s' % str(e))
+            logger.critical(str(e))
+            return 0, None
+
+        return 1, url
+
+    def _download_callback(self, request, result):
+        result, data = result
+        if result == 0:
+            logger.warning('fatal errors occurred, ignored')
+            # exit(1)
+        elif result == -1:
+            logger.warning('url {} return status code 404'.format(data))
+        else:
+            logger.log(15, '{0} download successfully'.format(data))
+
+    def download(self, queue, folder=''):
+        if not isinstance(folder, (text)):
+            folder = str(folder)
+
+        if self.path:
+            folder = os.path.join(self.path, folder)
+
+        if not os.path.exists(folder):
+            logger.warn('Path \'{0}\' not exist.'.format(folder))
+            try:
+                os.makedirs(folder)
+            except EnvironmentError as e:
+                logger.critical('{0}'.format(str(e)))
+                exit(1)
+        else:
+            logger.warn('Path \'{0}\' already exist.'.format(folder))
+
+        queue = [([url], {'folder': folder}) for url in queue]
+
+        self.thread_pool = threadpool.ThreadPool(self.thread_count)
+        requests_ = threadpool.makeRequests(self._download, queue, self._download_callback)
+        [self.thread_pool.putRequest(req) for req in requests_]
+
+        self.thread_pool.wait()

nhentai/utils.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals, print_function
 
 import os
+import string
 from nhentai.logger import logger