mirror of
https://github.com/instaloader/instaloader.git
synced 2024-10-03 22:07:11 +02:00
Merge branch 'master' into v4-dev
This commit is contained in:
Alexander Graf
commit
235522dfd7
3
docs/_static/style.css
vendored
Normal file
3
docs/_static/style.css
vendored
Normal file
@ -0,0 +1,3 @@
|
||||
code {
|
||||
color: #222;
|
||||
}
|
||||
2
docs/_templates/caption.html
vendored
2
docs/_templates/caption.html
vendored
@ -1,2 +0,0 @@
|
||||
<h2><a href="index.html">Instaloader</a></h2>
|
||||
<p>Download pictures (or videos) along with their captions and other metadata from Instagram.</p>
|
||||
10
docs/_templates/layout.html
vendored
10
docs/_templates/layout.html
vendored
@ -1,5 +1,11 @@
|
||||
{% extends "!layout.html" %}
|
||||
{% block extrahead %}
|
||||
<meta name="google-site-verification" content="FNUsDdkOJM9obm9QEdrNBVd9AtAzHzA-aaSMCUrS6C8" />
|
||||
<script async defer src="https://buttons.github.io/buttons.js"></script>
|
||||
{{ super() }}
|
||||
<meta name="google-site-verification" content="FNUsDdkOJM9obm9QEdrNBVd9AtAzHzA-aaSMCUrS6C8" />
|
||||
{% if pagename == "index" %}
|
||||
<link rel="canonical" href="https://instaloader.github.io/" />
|
||||
{% else %}
|
||||
<link rel="canonical" href="https://instaloader.github.io/{{ pagename }}.html" />
|
||||
{% endif %}
|
||||
<script async defer src="https://buttons.github.io/buttons.js"></script>
|
||||
{% endblock %}
|
||||
|
||||
23
docs/_templates/links.html
vendored
23
docs/_templates/links.html
vendored
@ -1,23 +0,0 @@
|
||||
{% if next %}
|
||||
<h3>Next</h3>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="{{ next.link|e }}">{{ next.title }}</a></li>
|
||||
</ul>
|
||||
{% endif %}
|
||||
<h3>Current Release</h3>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a href="https://github.com/instaloader/instaloader/releases/tag/v{{current_release}}">Version <b>{{current_release}}</b></a></li>
|
||||
<li>Released on <b>{{current_release_date}}</b>.</li>
|
||||
</ul>
|
||||
<h3>Links</h3>
|
||||
<ul>
|
||||
<li><a href="https://github.com/instaloader/instaloader/issues">Issue Tracker</a></li>
|
||||
<li><a href="https://github.com/instaloader/instaloader/releases">Version History</a></li>
|
||||
<li><a href="https://pypi.python.org/pypi/instaloader">Instaloader on PyPI</a></li>
|
||||
<li><a href="https://github.com/instaloader/instaloader">Instaloader on GitHub</a></li>
|
||||
</ul>
|
||||
<ul>
|
||||
<li>
|
||||
<a class="github-button" href="https://github.com/instaloader/Instaloader" data-icon="octicon-star" data-size="large" data-show-count="true" aria-label="Star instaloader/Instaloader on GitHub">Star</a>
|
||||
</li>
|
||||
</ul>
|
||||
58
docs/_templates/navbar.html
vendored
Normal file
58
docs/_templates/navbar.html
vendored
Normal file
@ -0,0 +1,58 @@
|
||||
<div id="navbar" class="{{ theme_navbar_class }} navbar-default {% if theme_navbar_fixed_top|tobool -%} navbar-fixed-top{%- endif -%}">
|
||||
<div class="container">
|
||||
<div class="navbar-header">
|
||||
<!-- .btn-navbar is used as the toggle for collapsed navbar content -->
|
||||
<button type="button" class="navbar-toggle" data-toggle="collapse" data-target=".nav-collapse">
|
||||
<span class="icon-bar"></span>
|
||||
<span class="icon-bar"></span>
|
||||
<span class="icon-bar"></span>
|
||||
</button>
|
||||
<a class="navbar-brand" href="{{ pathto(master_doc) }}">
|
||||
{%- block sidebarlogo %}
|
||||
{%- if logo %}<span><img src="{{ pathto('_static/' + logo, 1) }}"></span>{%- endif %}
|
||||
{%- endblock %}
|
||||
{% if theme_navbar_title -%}{{ theme_navbar_title|e }}{%- else -%}{{ project|e }}{%- endif -%}
|
||||
</a>
|
||||
<span class="navbar-text navbar-version pull-left">
|
||||
<a href="https://github.com/instaloader/instaloader/releases/tag/v{{current_release}}" title="Released on {{current_release_date}}.">v<b>{{current_release}}</b></a>
|
||||
</span>
|
||||
</div>
|
||||
|
||||
<div class="collapse navbar-collapse nav-collapse">
|
||||
<ul class="nav navbar-nav">
|
||||
{% if theme_navbar_links %}
|
||||
{%- for link in theme_navbar_links %}
|
||||
<li><a href="{{ pathto(*link[1:]) }}">{{ link[0] }}</a></li>
|
||||
{%- endfor %}
|
||||
{% endif %}
|
||||
{% block navbartoc %}
|
||||
{% include "globaltoc.html" %}
|
||||
{% if theme_navbar_pagenav %}
|
||||
{% include "navbartoc.html" %}
|
||||
{% endif %}
|
||||
{% endblock %}
|
||||
{% if theme_navbar_sidebarrel %}
|
||||
{% block sidebarrel %}
|
||||
{% include "relations.html" %}
|
||||
{% endblock %}
|
||||
{% endif %}
|
||||
</ul>
|
||||
<ul class="nav navbar-nav navbar-right">
|
||||
<li class="dropdown">
|
||||
<a role="button" id="dLabelUsefulLinks" data-toggle="dropdown" data-target="#" href="#">
|
||||
Useful Links
|
||||
<b class="caret"></b>
|
||||
</a>
|
||||
<ul class="dropdown-menu" role="menu" aria-labelledby="dLabelUsefulLinks">
|
||||
<li><a href="https://github.com/instaloader/instaloader/issues">Issue Tracker</a></li>
|
||||
<li><a href="https://github.com/instaloader/instaloader/releases">Version History</a></li>
|
||||
<li><a href="https://pypi.python.org/pypi/instaloader">Instaloader on PyPI</a></li>
|
||||
<li><a href="https://github.com/instaloader/instaloader">Instaloader on GitHub</a></li>
|
||||
<li role="separator" class="divider"></li>
|
||||
<li style="padding-left: 20px;"><a class="github-button" href="https://github.com/instaloader/instaloader" data-icon="octicon-star" data-size="large" data-show-count="true" aria-label="Star Instaloader on GitHub">Star</a></li>
|
||||
</ul>
|
||||
</li>
|
||||
</ul>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
3
docs/_templates/rtdmessage.html
vendored
3
docs/_templates/rtdmessage.html
vendored
@ -1,3 +0,0 @@
|
||||
<h3>New Documentation URL</h3>
|
||||
<div class="danger"><p style="margin-left: 1em"><b>Instaloader's Documentation has been moved to
|
||||
<a class="reference external" href="https://instaloader.github.io/">https://instaloader.github.io</a>.</b></p></div>
|
||||
22
docs/conf.py
22
docs/conf.py
@ -20,6 +20,9 @@
|
||||
import os
|
||||
import subprocess
|
||||
import sys
|
||||
|
||||
import sphinx_bootstrap_theme
|
||||
|
||||
sys.path.insert(0, os.path.abspath('..'))
|
||||
|
||||
# -- General configuration ------------------------------------------------
|
||||
@ -133,16 +136,20 @@ todo_include_todos = False
|
||||
# The theme to use for HTML and HTML Help pages. See the documentation for
|
||||
# a list of builtin themes.
|
||||
#
|
||||
html_theme = 'alabaster'
|
||||
html_theme = 'bootstrap'
|
||||
html_theme_path = sphinx_bootstrap_theme.get_html_theme_path()
|
||||
|
||||
# Theme options are theme-specific and customize the look and feel of a theme
|
||||
# further. For a list of options available for each theme, see the
|
||||
# documentation.
|
||||
#
|
||||
html_theme_options = {
|
||||
'show_powered_by': False,
|
||||
'sidebar_width': '290px',
|
||||
'page_width': '935px' }
|
||||
'navbar_site_name': 'Site Contents',
|
||||
'navbar_pagenav_name': 'Page Contents',
|
||||
'navbar_pagenav': True,
|
||||
'navbar_sidebarrel': True,
|
||||
'nosidebar': True,
|
||||
}
|
||||
|
||||
# Add any paths that contain custom themes here, relative to this directory.
|
||||
# html_theme_path = []
|
||||
@ -191,10 +198,8 @@ html_static_path = ['_static']
|
||||
|
||||
# Custom sidebar templates, maps document names to template names.
|
||||
#
|
||||
if not os.environ.get("READTHEDOCS"):
|
||||
html_sidebars = {'**': ["caption.html", "globaltoc.html", "relations.html", "links.html"] }
|
||||
else:
|
||||
html_sidebars = {'**': ["caption.html", "rtdmessage.html", "globaltoc.html", "relations.html", "links.html"] }
|
||||
#html_sidebars = {'**': ["relations.html", "links.html"] }
|
||||
html_sidebars = {}
|
||||
|
||||
# Additional templates that should be rendered to pages, maps page names to
|
||||
# template names.
|
||||
@ -373,3 +378,4 @@ def skip(app, what, name, obj, skip, options):
|
||||
|
||||
def setup(app):
|
||||
app.connect('autodoc-skip-member', skip)
|
||||
app.add_stylesheet("style.css")
|
||||
|
||||
@ -28,7 +28,7 @@ will be installed automatically, if it is not already installed.
|
||||
|
||||
- If you do not want to use pip, even though it is highly recommended,
|
||||
and prefer to **install Instaloader manually**,
|
||||
`Download the Source <https://github.com/instaloader/instaloader/releases>`__,
|
||||
`Download the Source <https://github.com/instaloader/instaloader/releases/latest>`__,
|
||||
extract the Zip or Tarball and execute ``instaloader.py`` from there.
|
||||
|
||||
- On **Arch Linux**, you may install Instaloader using the
|
||||
@ -36,3 +36,6 @@ will be installed automatically, if it is not already installed.
|
||||
|
||||
- On **Gentoo Linux**, you may install Instaloader using the
|
||||
`Instaloader Ebuild <https://bugs.gentoo.org/642100>`__.
|
||||
|
||||
- On **Windows 10**, you may download the standalone executable from the
|
||||
`current release page <https://github.com/instaloader/instaloader/releases/latest>`__.
|
||||
|
||||
@ -1,3 +1,4 @@
|
||||
requests
|
||||
sphinx
|
||||
sphinx-autodoc-typehints
|
||||
sphinx-bootstrap-theme
|
||||
|
||||
@ -31,14 +31,14 @@ import requests.utils
|
||||
import urllib3
|
||||
|
||||
|
||||
__version__ = '3.3.3'
|
||||
__version__ = '3.3.4'
|
||||
|
||||
# NOTE: duplicated in README.rst and docs/index.rst
|
||||
USAGE_STRING = """
|
||||
{0} [--comments] [--geotags] [--stories]
|
||||
[--login YOUR-USERNAME] [--fast-update]
|
||||
profile | "#hashtag" | :stories | :feed | :saved
|
||||
{0} --help""".format(sys.argv[0])
|
||||
{2:{1}} [--login YOUR-USERNAME] [--fast-update]
|
||||
{2:{1}} profile | "#hashtag" | :stories | :feed | :saved
|
||||
{0} --help""".format(sys.argv[0], len(sys.argv[0]), '')
|
||||
|
||||
try:
|
||||
# pylint:disable=wrong-import-position
|
||||
@ -413,7 +413,7 @@ class Post:
|
||||
if loc_dict is not None:
|
||||
location_json = self._instaloader.get_json("explore/locations/{0}/".format(loc_dict["id"]),
|
||||
params={'__a': 1})
|
||||
return location_json["location"]
|
||||
return location_json["location"] if "location" in location_json else location_json['graphql']['location']
|
||||
|
||||
@staticmethod
|
||||
def json_encoder(obj) -> Dict[str, Any]:
|
||||
@ -525,10 +525,17 @@ class Profile:
|
||||
def requested_by_viewer(self) -> bool:
|
||||
return self._metadata['user']['requested_by_viewer']
|
||||
|
||||
@property
|
||||
def profile_pic_url(self) -> str:
|
||||
return self._metadata["user"]["profile_pic_url_hd"] if "profile_pic_url_hd" in self._metadata["user"] \
|
||||
else self._metadata["user"]["profile_pic_url"]
|
||||
def get_profile_pic_url(self) -> str:
|
||||
"""Return URL of profile picture"""
|
||||
try:
|
||||
with self._instaloader.get_anonymous_session() as anonymous_session:
|
||||
data = self._instaloader.get_json(path='api/v1/users/{0}/info/'.format(self.userid), params={},
|
||||
host='i.instagram.com', session=anonymous_session)
|
||||
return data["user"]["hd_profile_pic_url_info"]["url"]
|
||||
except (InstaloaderException, KeyError) as err:
|
||||
self._instaloader.error('{} Unable to fetch high quality profile pic.'.format(err))
|
||||
return self._metadata["user"]["profile_pic_url_hd"] if "profile_pic_url_hd" in self._metadata["user"] \
|
||||
else self._metadata["user"]["profile_pic_url"]
|
||||
|
||||
def get_posts(self) -> Iterator[Post]:
|
||||
"""Retrieve all posts from a profile."""
|
||||
@ -641,7 +648,7 @@ class Instaloader:
|
||||
|
||||
# configuration parameters
|
||||
self.user_agent = user_agent if user_agent is not None else default_user_agent()
|
||||
self.session = self._get_anonymous_session()
|
||||
self.session = self.get_anonymous_session()
|
||||
self.username = None
|
||||
self.sleep = sleep
|
||||
self.quiet = quiet
|
||||
@ -736,7 +743,7 @@ class Instaloader:
|
||||
:raises QueryReturnedForbiddenException: When the server responds with a 403.
|
||||
:raises ConnectionException: When download repeatedly failed."""
|
||||
try:
|
||||
with self._get_anonymous_session() as anonymous_session:
|
||||
with self.get_anonymous_session() as anonymous_session:
|
||||
resp = anonymous_session.get(url)
|
||||
if resp.status_code == 200:
|
||||
self._log(filename, end=' ', flush=True)
|
||||
@ -763,12 +770,13 @@ class Instaloader:
|
||||
self.error("[skipped by user]", repeat_at_end=False)
|
||||
raise ConnectionException(error_string)
|
||||
|
||||
def get_json(self, url: str, params: Dict[str, Any],
|
||||
def get_json(self, path: str, params: Dict[str, Any], host: str = 'www.instagram.com',
|
||||
session: Optional[requests.Session] = None, _attempt = 1) -> Dict[str, Any]:
|
||||
"""JSON request to Instagram.
|
||||
|
||||
:param url: URL, relative to www.instagram.com/
|
||||
:param path: URL, relative to the given domain which defaults to www.instagram.com/
|
||||
:param params: GET parameters
|
||||
:param host: Domain part of the URL from where to download the requested JSON; defaults to www.instagram.com
|
||||
:param session: Session to use, or None to use self.session
|
||||
:return: Decoded response dictionary
|
||||
:raises QueryReturnedNotFoundException: When the server responds with a 404.
|
||||
@ -785,7 +793,7 @@ class Instaloader:
|
||||
if len(timestamps) < 100 and not untracked_queries:
|
||||
return 0
|
||||
return round(min(timestamps) + sliding_window - current_time) + 6
|
||||
is_graphql_query = 'query_hash' in params and 'graphql/query' in url
|
||||
is_graphql_query = 'query_hash' in params and 'graphql/query' in path
|
||||
if is_graphql_query:
|
||||
query_hash = params['query_hash']
|
||||
waittime = graphql_query_waittime(query_hash)
|
||||
@ -800,11 +808,11 @@ class Instaloader:
|
||||
sess = session if session else self.session
|
||||
try:
|
||||
self._sleep()
|
||||
resp = sess.get('https://www.instagram.com/' + url, params=params, allow_redirects=False)
|
||||
resp = sess.get('https://{0}/{1}'.format(host, path), params=params, allow_redirects=False)
|
||||
while resp.is_redirect:
|
||||
redirect_url = resp.headers['location']
|
||||
self._log('\nHTTP redirect from {} to {}'.format('https://www.instagram.com/' + url, redirect_url))
|
||||
if redirect_url.index('https://www.instagram.com/') == 0:
|
||||
self._log('\nHTTP redirect from https://{0}/{1} to {2}'.format(host, path, redirect_url))
|
||||
if redirect_url.index('https://{}/'.format(host)) == 0:
|
||||
resp = sess.get(redirect_url if redirect_url.endswith('/') else redirect_url + '/',
|
||||
params=params, allow_redirects=False)
|
||||
else:
|
||||
@ -824,7 +832,7 @@ class Instaloader:
|
||||
raise ConnectionException("Returned \"{}\" status.".format(resp_json['status']))
|
||||
return resp_json
|
||||
except (ConnectionException, json.decoder.JSONDecodeError, requests.exceptions.RequestException) as err:
|
||||
error_string = "JSON Query to {}: {}".format(url, err)
|
||||
error_string = "JSON Query to {}: {}".format(path, err)
|
||||
if _attempt == self.max_connection_attempts:
|
||||
raise ConnectionException(error_string)
|
||||
self.error(error_string + " [retrying; skip with ^C]", repeat_at_end=False)
|
||||
@ -840,7 +848,7 @@ class Instaloader:
|
||||
self._log('The request will be retried in {} seconds.'.format(waittime))
|
||||
time.sleep(waittime)
|
||||
self._sleep()
|
||||
return self.get_json(url, params, sess, _attempt + 1)
|
||||
return self.get_json(path=path, params=params, host=host, session=sess, _attempt=_attempt + 1)
|
||||
except KeyboardInterrupt:
|
||||
self.error("[skipped by user]", repeat_at_end=False)
|
||||
raise ConnectionException(error_string)
|
||||
@ -865,7 +873,7 @@ class Instaloader:
|
||||
del header['X-Requested-With']
|
||||
return header
|
||||
|
||||
def _get_anonymous_session(self) -> requests.Session:
|
||||
def get_anonymous_session(self) -> requests.Session:
|
||||
"""Returns our default anonymous requests.Session object."""
|
||||
session = requests.Session()
|
||||
session.cookies.update({'sessionid': '', 'mid': '', 'ig_pr': '1',
|
||||
@ -1091,27 +1099,22 @@ class Instaloader:
|
||||
def _epoch_to_string(epoch: datetime) -> str:
|
||||
return epoch.strftime('%Y-%m-%d_%H-%M-%S')
|
||||
|
||||
with self._get_anonymous_session() as anonymous_session:
|
||||
date_object = datetime.strptime(anonymous_session.head(profile.profile_pic_url).headers["Last-Modified"],
|
||||
profile_pic_url = profile.get_profile_pic_url()
|
||||
with self.get_anonymous_session() as anonymous_session:
|
||||
date_object = datetime.strptime(anonymous_session.head(profile_pic_url).headers["Last-Modified"],
|
||||
'%a, %d %b %Y %H:%M:%S GMT')
|
||||
if ((format_string_contains_key(self.dirname_pattern, 'profile') or
|
||||
format_string_contains_key(self.dirname_pattern, 'target'))):
|
||||
filename = '{0}/{1}_UTC_profile_pic.{2}'.format(self.dirname_pattern.format(profile=profile.username.lower(),
|
||||
target=profile.username.lower()),
|
||||
_epoch_to_string(date_object), profile.profile_pic_url[-3:])
|
||||
_epoch_to_string(date_object), profile_pic_url[-3:])
|
||||
else:
|
||||
filename = '{0}/{1}_{2}_UTC_profile_pic.{3}'.format(self.dirname_pattern.format(), profile.username.lower(),
|
||||
_epoch_to_string(date_object), profile.profile_pic_url[-3:])
|
||||
_epoch_to_string(date_object), profile_pic_url[-3:])
|
||||
if os.path.isfile(filename):
|
||||
self._log(filename + ' already exists')
|
||||
return None
|
||||
url_best = re.sub(r'/s([1-9][0-9]{2})x\1/', '/s2048x2048/', profile.profile_pic_url)
|
||||
url_best = re.sub(r'/vp/[a-f0-9]{32}/[A-F0-9]{8}/', '/', url_best) # remove signature
|
||||
try:
|
||||
self._get_and_write_raw(url_best, filename)
|
||||
except (QueryReturnedForbiddenException, QueryReturnedNotFoundException) as err:
|
||||
self.error('{} Retrying with lower quality version.'.format(err))
|
||||
self._get_and_write_raw(profile.profile_pic_url, filename)
|
||||
self._get_and_write_raw(profile_pic_url, filename)
|
||||
os.utime(filename, (datetime.now().timestamp(), date_object.timestamp()))
|
||||
self._log('') # log output of _get_and_write_raw() does not produce \n
|
||||
|
||||
|
||||
Loading…
Reference in New Issue
Block a user