class TumblrFile:
"""
This is the base container class for all downloadable resources associated with Tumblr posts.
"""
CATEGORY = 'misc'
def __init__(self, data, container):
"""
Args:
data(dict): API response data
container(TumblrPost): Parent container
"""
self.log = logging.getLogger('tumdlr.containers.file')
self._data = data
self.container = container
self.url = URL(self._data.get('url', self._data.get('post_url')))
def download(self, context, **kwargs):
"""
Args:
context(tumdlr.main.Context): CLI request context
kwargs(dict): Additional arguments to send with the download request
Returns:
str: Path to the saved file
"""
try:
download(self.url.as_string(), str(self.filepath(context, kwargs)), **kwargs)
except Exception as e:
self.log.warn('Post download failed: %r', self, exc_info=e)
raise TumdlrDownloadError(error_message=str(e), download_url=self.url.as_string())
def filepath(self, context, request_data):
"""
Args:
context(tumdlr.main.Context): CLI request context
request_data(Optional[dict]): Additional arguments to send with the download request
Returns:
Path
"""
# Construct the save basedir
basedir = Path(context.config['Tumdlr']['SavePath'])
# Are we categorizing by user?
if context.config['Categorization']['User']:
self.log.debug('Categorizing by user: %s', self.container.blog.name)
basedir = basedir.joinpath(sanitize_filename(self.container.blog.name))
# Are we categorizing by post type?
if context.config['Categorization']['PostType']:
self.log.debug('Categorizing by type: %s', self.CATEGORY)
basedir = basedir.joinpath(self.CATEGORY)
self.log.debug('Basedir constructed: %s', basedir)
return basedir
def test_stress_authority(self):
# Authority is most ambiguous part of url. Invalid host can contatin
# ':' and '@' (path for example can not contain '?'. And query
# can not contain '#'). The host '//no:99:' will be parsed as 'no:99'
# and in next recomposition it can be written as '//no:99'. But parsing
# of '//no:99:' and '//no:99' will be different.
# # case generation:
# from re import sub
# from itertools import permutations
# cases = set(sub('\d+', '7', ''.join(case))
# for case in set(permutations('::@@77777')))
cases = """7:7:7@7@ 7:7:7@7@7 7:7:7@@ 7:7:7@@7 7:7:@7@ 7:7:@7@7 7:7:@@
7:7:@@7 7:7@7:7@ 7:7@7:7@7 7:7@7:@ 7:7@7:@7 7:7@7@7: 7:7@7@7:7
7:7@7@: 7:7@7@:7 7:7@:7@ 7:7@:7@7 7:7@:@ 7:7@:@7 7:7@@7: 7:7@@7:7
7:7@@: 7:7@@:7 7::7@7@ 7::7@7@7 7::7@@ 7::7@@7 7::@7@ 7::@7@7 7::@@
7::@@7 7:@7:7@ 7:@7:7@7 7:@7:@ 7:@7:@7 7:@7@7: 7:@7@7:7 7:@7@:
7:@7@:7 7:@:7@ 7:@:7@7 7:@:@ 7:@:@7 7:@@7: 7:@@7:7 7:@@: 7:@@:7
7@7:7:7@ 7@7:7:7@7 7@7:7:@ 7@7:7:@7 7@7:7@7: 7@7:7@7:7 7@7:7@:
7@7:7@:7 7@7::7@ 7@7::7@7 7@7::@ 7@7::@7 7@7:@7: 7@7:@7:7 7@7:@:
7@7:@:7 7@7@7:7: 7@7@7:7:7 7@7@7:: 7@7@7::7 7@7@:7: 7@7@:7:7 7@7@::
7@7@::7 7@:7:7@ 7@:7:7@7 7@:7:@ 7@:7:@7 7@:7@7: 7@:7@7:7 7@:7@:
7@:7@:7 7@::7@ 7@::7@7 7@::@ 7@::@7 7@:@7: 7@:@7:7 7@:@: 7@:@:7
7@@7:7: 7@@7:7:7 7@@7:: 7@@7::7 7@@:7: 7@@:7:7 7@@:: 7@@::7 :7:7@7@
:7:7@7@7 :7:7@@ :7:7@@7 :7:@7@ :7:@7@7 :7:@@ :7:@@7 :7@7:7@
:7@7:7@7 :7@7:@ :7@7:@7 :7@7@7: :7@7@7:7 :7@7@: :7@7@:7 :7@:7@
:7@:7@7 :7@:@ :7@:@7 :7@@7: :7@@7:7 :7@@: :7@@:7 ::7@7@ ::7@7@7
::7@@ ::7@@7 ::@7@ ::@7@7 ::@@7 :@7:7@ :@7:7@7 :@7:@ :@7:@7 :@7@7:
:@7@7:7 :@7@: :@7@:7 :@:7@ :@:7@7 :@:@7 :@@7: :@@7:7 :@@:7 @7:7:7@
@7:7:7@7 @7:7:@ @7:7:@7 @7:7@7: @7:7@7:7 @7:7@: @7:7@:7 @7::7@
@7::7@7 @7::@ @7::@7 @7:@7: @7:@7:7 @7:@: @7:@:7 @7@7:7: @7@7:7:7
@7@7:: @7@7::7 @7@:7: @7@:7:7 @7@:: @7@::7 @:7:7@ @:7:7@7 @:7:@
@:7:@7 @:7@7: @:7@7:7 @:7@: @:7@:7 @::7@ @::7@7 @::@7 @:@7: @:@7:7
@:@:7 @@7:7: @@7:7:7 @@7:: @@7::7 @@:7: @@:7:7 @@::7""".split()
for case in cases:
url = URL('//' + case)
# check is all parts defined in original url is defined in parsed
self.assertEqual(url, URL(url.as_string()))
self.assertEqual(url, URL('//' + url.authority))
class TumblrPost:
"""
This is the base container class for all Tumblr post types. It contains data that is always available with any
type of post.
Additional supported post types may extend this class to provide additional metadata parsing
"""
def __init__(self, post, blog):
"""
Args:
post(dict): API response
blog(tumdlr.api.TumblrBlog): Parent blog
"""
self._post = post
self.blog = blog
self.log = logging.getLogger('tumdlr.containers.post')
self.id = None # type: int
self.type = None # type: str
self.url = None # type: URL
self.tags = set()
self.post_date = None # type: str
self.note_count = None # type: int
self.files = []
try:
self._parse_post()
except Exception as e:
self.log.warn('Failed to parse post data: %r', self, exc_info=e)
raise TumdlrParserError(post_data=post)
@property
def is_text(self):
"""
Returns:
bool
"""
return self.type == 'text'
@property
def is_photo(self):
"""
Returns:
bool
"""
return self.type in ['photo', 'link']
@property
def is_video(self):
"""
Returns:
bool
"""
return self.type == 'video'
def _parse_post(self):
self.id = self._post['id']
self.type = self._post['type']
self.url = URL(self._post['post_url']) if 'post_url' in self._post else None
self.tags = set(self._post.get('tags', []))
self.note_count = self._post.get('note_count')
self.post_date = self._post['date']
def __repr__(self):
return "<TumblrPost id='{id}' type='{type}' url='{url}'>"\
.format(id=self.id, type=self.type, url=self.url)
def __str__(self):
return self.url.as_string() if self.url else ''