First Alpha Release for Version 4.5

Resume a previously-aborted post download loop (#732 )
With this change, Instaloader is capable of resuming a previously-aborted download loop. To do so, it creates a JSON file within the target directory when interrupted, that contains all the necessary information to later resume that operation. Resuming an interrupted download is supported for most, but not all targets. It is supported for: - Regular profile posts, - IGTV posts - Saved posts, - Tagged posts, - Explore posts.
2026-10-17 06:02:18 -07:00 · 2020-07-21 17:37:45 +02:00 · 2020-07-21 17:28:55 +02:00 · 2020-07-18 15:32:05 +02:00 · 2020-07-18 15:24:31 +02:00 · 2020-07-15 10:26:22 +02:00
36 changed files with 1538 additions and 583 deletions
@@ -24,4 +24,4 @@ If applicable, add error messages and tracebacks to help explain your problem.
 **Additional context**
 Add any other context about the problem here.

-(please also see https://instaloader.github.io/contributing.html for how to report a bug)
+<!-- please also see https://instaloader.github.io/contributing.html for how to report a bug -->
@@ -0,0 +1,11 @@
+blank_issues_enabled: false
+contact_links:
+  - name: Contributing to Instaloader
+    url: https://instaloader.github.io/contributing.html
+    about: Guidelines for how to report an issue.
+  - name: Instaloader Troubleshooting
+    url: https://instaloader.github.io/troubleshooting.html
+    about: Many common problems and their workarounds are described here.
+  - name: Instaloader Documentation
+    url: https://instaloader.github.io/
+    about: Many questions might be already answered here.
@@ -15,10 +15,10 @@ A clear and concise description of what you want to happen.
 A clear and concise description of any alternative solutions or features you've considered.

 **If the feature request is accepted, would you be willing to submit a PR?**
-
-Yes / No _(Help can be provided if you need assistance submitting a PR)_
+Yes / No
+<!-- Help can be provided if you need assistance submitting a PR -->

 **Additional context**
 Add any other context about the feature request here.

-(please also see https://instaloader.github.io/contributing.html for how to suggest a feature)
+<!-- please also see https://instaloader.github.io/contributing.html for how to suggest a feature -->
@@ -6,3 +6,7 @@ labels: question
 ---

 Your question here...
+
+<!-- Please see our documentation: https://instaloader.github.io/ -->
+
+<!-- Basic questions regarding Instaloader's usage might be more appropriate to ask on Stack Overflow -->
@@ -0,0 +1,17 @@
+
+
+<!--
+Please describe:
+
+- A motivation for this change, e.g.
+  - Fixes # .
+  - More general: What problem does the pull request solve?
+
+- The changes proposed in this pull request
+
+- The completeness of this change
+  - Is it just a proof of concept?
+  - Is the documentation updated (if appropriate)?
+  - Do you consider it ready to be merged or is it a draft?
+  - Can we help you at some point?
+-->
@@ -0,0 +1,30 @@
+name: PyLint and MyPy
+
+on: [push, pull_request]
+
+jobs:
+  lint:
+    name: PyLint and MyPy
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: [3.8]
+    steps:
+      - name: Checkout Instaloader Repository
+        uses: actions/checkout@v2
+        with:
+          fetch-depth: 0  # needed for building docs
+      - name: Setup Python
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install Dependencies
+        run: |
+          python -m pip install pipenv==2020.6.2
+          pipenv sync --dev
+      - name: PyLint
+        run: pipenv run pylint instaloader
+      - name: MyPy
+        run: pipenv run mypy -m instaloader
+      - name: Build Documentation
+        run: pipenv run make -C docs html SPHINXOPTS="-W -n"
@@ -2,14 +2,25 @@ name: Mark stale issues and pull requests

 on:
  schedule:
-  - cron: "15 16 * * *"
+  - cron: "10 14 * * *"

 jobs:
  stale:

+    name: Mark stale issues and pull requests
    runs-on: ubuntu-latest
    
    steps:
+    - uses: actions/stale@v3
+      with:
+        repo-token: ${{ secrets.GITHUB_TOKEN }}
+        only-labels: 'question'
+        stale-issue-message: 'There has been no activity on this question for an extended period of time. This issue will be closed after further 14 days of inactivity.'
+        stale-issue-label: 'stale'
+        exempt-issue-labels: 'leave open'
+        days-before-stale: 15
+        days-before-close: -1
+        remove-stale-when-updated: false
    - uses: actions/stale@v1
      with:
        repo-token: ${{ secrets.GITHUB_TOKEN }}
@@ -21,4 +32,3 @@ jobs:
        exempt-pr-label: 'leave open'
        days-before-stale: 90
        days-before-close: 14
-        operations-per-run: 7
@@ -4,6 +4,7 @@ on: [push, pull_request]

 jobs:
  build:
+    name: Windows EXE build
    runs-on: windows-latest
    steps:
      - name: Checkout Instaloader repository
@@ -27,7 +27,7 @@ deploy:
    repo: instaloader/instaloader.github.io
    target-branch: master
    on:
-      tags: true
+      branch: master
      python: 3.6
  - provider: script
    script: deploy/arch/deploy.sh $TRAVIS_TAG
@@ -2,7 +2,7 @@

 .. badges-start

-|travis| |pypi| |pyversion| |license| |aur| |contributors| |downloads| |saythanks|
+|travis| |pypi| |pyversion| |license| |aur| |contributors| |downloads|

 .. |travis| image:: https://img.shields.io/travis/instaloader/instaloader/master.svg
   :alt: Travis-CI Build Status
@@ -23,10 +23,6 @@
   :alt: Contributor Count
   :target: https://github.com/instaloader/instaloader/graphs/contributors

-.. |saythanks| image:: https://img.shields.io/badge/Say%20Thanks-!-ff7efd.svg
-   :alt: Say Thanks!
-   :target: https://saythanks.io/to/aandergr
-
 .. |aur| image:: https://img.shields.io/aur/version/instaloader.svg
   :alt: Arch User Repository Package
   :target: https://aur.archlinux.org/packages/instaloader/
@@ -53,11 +49,14 @@
  directory accordingly,

 - allows **fine-grained customization** of filters and where to store
-  downloaded media.
+  downloaded media,
+
+- automatically **resumes previously-interrupted** download iterations.

 ::

-    instaloader [--comments] [--geotags] [--stories] [--highlights] [--tagged]
+    instaloader [--comments] [--geotags]
+                [--stories] [--highlights] [--tagged] [--igtv]
                [--login YOUR-USERNAME] [--fast-update]
                profile | "#hashtag" | :stories | :feed | :saved

@@ -8,33 +8,18 @@ import sys
 # copy the required files into repo root
 shutil.copy('docs/favicon.ico', '.')
 shutil.copy('deploy/windows/instaloader.spec', '.')
-shutil.unpack_archive('deploy/windows/ps', '.', 'xztar')

 code = """
+import contextlib
 import psutil
 import subprocess

 def __main():
-    grandparentpid = psutil.Process(os.getppid()).ppid()
-    grandparentpidsearchstring = ' ' + str(grandparentpid) + ' '
-    if hasattr(sys, "_MEIPASS"):
-        ps = os.path.join(sys._MEIPASS, 'tasklist.exe')
-    else:
-        ps = 'tasklist'
-    popen = subprocess.Popen(ps, stdout=subprocess.PIPE, universal_newlines=True)
-    for examine in iter(popen.stdout.readline, ""):
-        if grandparentpidsearchstring in examine:
-            pname = examine
-            break
-    popen.stdout.close()
-    return_code = popen.wait()
-    if return_code:
-        raise subprocess.CalledProcessError(return_code, ps)
-    if pname[0:12] == 'explorer.exe':
-        subprocess.Popen("cmd /K \\\"{0}\\\"".format(os.path.splitext(os.path.basename(sys.argv[0]))[0]))
-    else:
-        main()
-
+    with contextlib.suppress(AttributeError, psutil.Error):
+        if psutil.Process().parent().parent().name() == "explorer.exe":
+            subprocess.Popen("powershell -NoExit -Command \\\"& {0}\\\"".format(sys.argv[0]))
+            return
+    main()

 if __name__ == "__main__":
    __main()
@@ -5,7 +5,7 @@ block_cipher = None

 a = Analysis(['__main__.py'],
             binaries=[],
-             datas=[('tasklist.exe', '.'), ('framedyn.dll', '.')],
+             datas=[],
             hiddenimports=[],
             hookspath=[],
             runtime_hooks=[],
@@ -18,14 +18,17 @@ code {
    padding: 0.7em;
    color: #fff; }
  .highlight .c1 {
-    color: #008d06; }
+    color: #666; }
  .highlight .k, .highlight .kn, .highlight .ow {
-    color: #f48400;
-    font-weight: bold; }
-  .highlight .nb {
+    color: #008d06; }
+  .highlight .nb, .highlight .ne, .highlight .nf, .highlight .vm {
    color: #f48400; }
-  .highlight .s1, .highlight .s2, .highlight .mi {
+  .highlight .s1, .highlight .s2, .highlight .sa, .highlight .se, .highlight .si, .highlight .mi {
    color: #ff7efd; }
+  .highlight .c1, .highlight .nf, .highlight .sa, .highlight .se, .highlight .si {
+    font-style: italic; }
+  .highlight .k, .highlight .kn, .highlight .ow {
+    font-weight: bold; }

 .doc-content {
  padding-top: 0.5em;
@@ -37,21 +37,28 @@ code {
  }

  .c1 {
-    color: $color_instaloader_main
+    color: #666;
  }

  .k, .kn, .ow {
-    color: $color_instaloader_accent;
-    font-weight: bold
+    color: $color_instaloader_main
  }

-  .nb {
+  .nb, .ne, .nf, .vm {
    color: $color_instaloader_accent
  }

-  .s1, .s2, .mi {
+  .s1, .s2, .sa, .se, .si, .mi {
    color: $color_instaloader_hyper
  }
+
+  .c1, .nf, .sa, .se, .si {
+    font-style: italic;
+  }
+
+  .k, .kn, .ow {
+    font-weight: bold
+  }
 }

 .doc-content {
@@ -51,9 +51,6 @@
        <li class="nav-item">
          <a target="_blank" class="nav-link" href="https://github.com/instaloader/instaloader">GitHub</a>
        </li>
-        <li class="nav-item d-sm-none d-lg-block">
-          <a target="_blank" class="nav-link" href="https://saythanks.io/to/aandergr">Say Thanks!</a>
-        </li>
      </ul>
      <form class="form-inline" action="search.html" method="get">
        <div class="input-group my-2 my-sm-0 ml-2">
@@ -37,7 +37,7 @@ picture, video or sidecar (set of multiple pictures/videos) posted in a user's
 profile. :class:`Instaloader` provides methods to iterate over Posts from a
 certain source::

-    for post in L.get_hashtag_posts('cat'):
+    for post in instaloader.Hashtag.from_name(L.context, 'cat').get_posts():
        # post is an instance of instaloader.Post
        L.download_post(post, target='#cat')

@@ -63,7 +63,7 @@ certain source::
 - :meth:`Instaloader.get_explore_posts`
   Media that is suggested by Instagram to explore.

- :meth:`Instaloader.get_hashtag_posts`
+- :meth:`Hashtag.get_posts`
   Media associated with given hashtag.

 With the :class:`Profile` class, Instaloader also makes it easy to access
@@ -155,6 +155,14 @@ Profiles
 .. autoclass:: Profile
   :no-show-inheritance:

+Hashtags
+""""""""
+
+.. autoclass:: Hashtag
+   :no-show-inheritance:
+
+   .. versionadded:: 4.4
+
 TopSearchResults
 """"""""""""""""

@@ -166,7 +174,7 @@ TopSearchResults
 Loading and Saving
 """"""""""""""""""

-:class:`Post`, :class:`StoryItem` and :class:`Profile` can be saved and loaded
+:class:`Post`, :class:`StoryItem`, :class:`Profile` and :class:`Hashtag` can be saved and loaded
 to/from JSON files.

 .. autofunction:: load_structure_from_file
@@ -218,8 +226,32 @@ Exceptions

 .. autoexception:: TooManyRequestsException

+Resumable Iterations
+^^^^^^^^^^^^^^^^^^^^
+
+.. versionadded:: 4.5
+
+.. autoclass:: NodeIterator
+   :no-show-inheritance:
+
+.. autoclass:: FrozenNodeIterator
+   :no-show-inheritance:
+
+.. autofunction:: resumable_iteration
+
 ``InstaloaderContext`` (Low-level functions)
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

+``InstaloaderContext``
+""""""""""""""""""""""
+
 .. autoclass:: InstaloaderContext
   :no-show-inheritance:
+
+``RateController``
+""""""""""""""""""
+
+.. autoclass:: RateController
+   :no-show-inheritance:
+
+   .. versionadded:: 4.5
@@ -51,7 +51,7 @@ invoke it with
    instaloader --login=your_username profile [profile ...]

 When logging in, Instaloader **stores the session cookies** in a file in your
-temporary directory, which will be reused later the next time :option:`--login`
+home directory, which will be reused later the next time :option:`--login`
 is given.  So you can download private profiles **non-interactively** when you
 already have a valid session cookie file.

@@ -77,10 +77,13 @@ Instaloader supports the following targets:
      instructs Instaloader to also **download the user's stories**,

   - :option:`--highlights`
-      to **download highlights of each profile that is downloaded**, and
+      to **download highlights of each profile that is downloaded**,

   - :option:`--tagged`
-      to **download posts where the user is tagged**.
+      to **download posts where the user is tagged**, and
+
+   - :option:`--igtv`
+      to **download IGTV videos**.

 - ``"#hashtag"``
   Posts with a certain **hashtag** (the quotes are usually necessary),
@@ -108,8 +111,12 @@ Instaloader supports the following targets:
   ``profile`` (requires :option:`--login`).

 - ``-post``
-   The single **post** with the given shortcode. Must be preceeded by ``--`` in
-   the argument list to not be mistaken as an option flag.
+   Replace **post** with the post's shortcode to download single post. Must be preceeded by ``--`` in
+   the argument list to not be mistaken as an option flag::
+
+    instaloader -- -B_K4CykAOtf
+
+

   .. versionadded:: 4.1

@@ -288,17 +295,18 @@ Instaloader as Cronjob
 Instaloader is suitable for running as a cronjob to periodically update your
 personal Instagram archive. The :option:`--quiet` option disables user
 interactions and logging of non-error messages. To non-interactively use
-Instaloader logged-in, create a session file, e.g. in your home directory::
+Instaloader logged-in, create a session file::

-   instaloader --login=your_username --sessionfile=~/.instaloader-session
+   instaloader --login=your_username

-Then use the same parameters in your cronjob to load the session and download
+Then use the same parameter in your cronjob to load the session and download
 the given targets::

-   instaloader --login=your_username --sessionfile=~/.instaloader-session --quiet <target> [...]
+   instaloader --login=your_username --quiet <target> [...]

-Without :option:`--sessionfile` option, Instaloader saves the session file in
-a path within your temporary directory.
+Instaloader saves the session file to
+``~/.config/instaloader/session-YOUR-USERNAME``. See
+:option:`--sessionfile` option for how to override this path.

 Programming Instaloader
 ^^^^^^^^^^^^^^^^^^^^^^^
@@ -129,6 +129,12 @@ What to Download of each Profile

   .. versionadded:: 4.1

+.. option:: --igtv
+
+   Also download IGTV videos.
+
+   .. versionadded:: 4.3
+
 .. option:: --stories-only

   .. deprecated:: 4.1
@@ -187,9 +193,8 @@ Instaloader to login.

 .. option:: --sessionfile SESSIONFILE, -f SESSIONFILE

-   Path for loading and storing session key file.  Defaults to a path within
-   your temporary directory, encoding your local username and your Instagram
-   profile name.
+   Path for loading and storing session key file.  Defaults to
+   ``~/.config/instaloader/session-YOUR-USERNAME``.

 .. option:: --password YOUR-PASSWORD, -p YOUR-PASSWORD

@@ -218,6 +223,29 @@ How to Download
   ``#hashtag`` or the profile name. Defaults to ``{date_utc}_UTC``.
   See :ref:`filename-specification` for a list of supported tokens.

+.. option:: --resume-prefix prefix
+
+   For many targets, Instaloader is capable of resuming a previously-aborted
+   download loop.  To do so, it creates a JSON file within the target directory
+   when interrupted.  This option controls the prefix for filenames that are
+   used to save the information to resume an interrupted download.  The default
+   prefix is ``iterator``.
+
+   Resuming an interrupted download is supported for most, but not all targets.
+   JSON files with resume information are always compressed, regardless of
+   :option:`--no-compress-json`.
+
+   This feature is turned off entirely with :option:`--no-resume`.
+
+   .. versionadded:: 4.5
+
+.. option:: --no-resume
+
+   Do not resume a previously-aborted download iteration, and do not save such
+   information when interrupted.
+
+   .. versionadded:: 4.5
+
 .. option:: --user-agent USER_AGENT

   User Agent to use for HTTP requests. Per default, Instaloader pretends being
@@ -237,6 +265,17 @@ How to Download

   .. versionadded:: 4.2

+   .. warning::
+      We discourage from using the "commit mode" due to its bugs when being used
+      with other command line options (e.g. :issue:`257` with
+      :option:`--fast-update` and :issue:`483` with :option:`--tagged`).
+
+.. option:: --request-timeout N
+
+   Seconds to wait before timing out a connection request.
+
+   .. versionadded:: 4.3
+
 Miscellaneous Options
 ^^^^^^^^^^^^^^^^^^^^^

@@ -246,12 +285,6 @@ Miscellaneous Options
   if login credentials are needed but not given.
   This is handy for running :ref:`instaloader-as-cronjob`.

-.. option:: --request-timeout N
-
-   Seconds to wait before timing out a connection request.
-
-   .. versionadded:: 4.3
-
 .. option:: +args.txt

   Read arguments from file `args.txt`, a shortcut to provide arguments from
@@ -28,8 +28,9 @@ Download Posts in a Specific Period
 -----------------------------------

 To only download Instagram pictures (and metadata) that are within a specific
-period, you can play around with :func:`~itertools.dropwhile` and
-:func:`~itertools.takewhile` from :mod:`itertools` like in this snippet.
+period, you can simply use :func:`~itertools.dropwhile` and
+:func:`~itertools.takewhile` from :mod:`itertools` on a generator that returns
+Posts in **exact chronological order**, such as :meth:`Profile.get_posts`.

 .. literalinclude:: codesnippets/121_since_until.py

@@ -37,6 +38,14 @@ See also :class:`Post`, :meth:`Instaloader.download_post`.

 Discussed in :issue:`121`.

+The code example with :func:`~itertools.dropwhile` and
+:func:`~itertools.takewhile` makes the assumption that the post iterator returns
+posts in exact chronological order.  As discussed in :issue:`666`, the following
+approach fits for an **almost chronological order**, where up to *k* older posts
+are inserted into an otherwise chronological order, such as an Hashtag feed.
+
+.. literalinclude:: codesnippets/666_historical_hashtag_data.py
+
 Likes of a Profile / Ghost Followers
 ------------------------------------

@@ -2,7 +2,7 @@ import instaloader

 L = instaloader.Instaloader()

-posts = L.get_hashtag_posts('urbanphotography')
+posts = instaloader.Hashtag.from_name(L.context, 'urbanphotography').get_posts()

 users = set()

@@ -5,13 +5,11 @@ import instaloader

 L = instaloader.Instaloader()

-posts = L.get_hashtag_posts('urbanphotography')
-# or
-# posts = instaloader.Profile.from_username(L.context, PROFILE).get_posts()
+posts = instaloader.Profile.from_username(L.context, "instagram").get_posts()

 SINCE = datetime(2015, 5, 1)
 UNTIL = datetime(2015, 3, 1)

 for post in takewhile(lambda p: p.date > UNTIL, dropwhile(lambda p: p.date > SINCE, posts)):
    print(post.date)
-    L.download_post(post, '#urbanphotography')
+    L.download_post(post, "instagram")
@@ -0,0 +1,53 @@
+from argparse import ArgumentParser
+from glob import glob
+from os.path import expanduser
+from platform import system
+from sqlite3 import OperationalError, connect
+
+try:
+    from instaloader import ConnectionException, Instaloader
+except ModuleNotFoundError:
+    raise SystemExit("Instaloader not found.\n  pip install [--user] instaloader")
+
+
+def get_cookiefile():
+    default_cookiefile = {
+        "Windows": "~/AppData/Roaming/Mozilla/Firefox/Profiles/*/cookies.sqlite",
+        "Darwin": "~/Library/Application Support/Firefox/Profiles/*/cookies.sqlite",
+    }.get(system(), "~/.mozilla/firefox/*/cookies.sqlite")
+    cookiefiles = glob(expanduser(default_cookiefile))
+    if not cookiefiles:
+        raise SystemExit("No Firefox cookies.sqlite file found. Use -c COOKIEFILE.")
+    return cookiefiles[0]
+
+
+def import_session(cookiefile, sessionfile):
+    print("Using cookies from {}.".format(cookiefile))
+    conn = connect(cookiefile)
+    try:
+        cookie_data = conn.execute(
+            "SELECT name, value FROM moz_cookies WHERE baseDomain='instagram.com'"
+        )
+    except OperationalError:
+        cookie_data = conn.execute(
+            "SELECT name, value FROM moz_cookies WHERE host LIKE '%instagram.com'"
+        )
+    instaloader = Instaloader(max_connection_attempts=1)
+    instaloader.context._session.cookies.update(cookie_data)
+    username = instaloader.test_login()
+    if not username:
+        raise SystemExit("Not logged in. Are you logged in successfully in Firefox?")
+    print("Imported session cookie for {}.".format(username))
+    instaloader.context.username = username
+    instaloader.save_session_to_file(sessionfile)
+
+
+if __name__ == "__main__":
+    p = ArgumentParser()
+    p.add_argument("-c", "--cookiefile")
+    p.add_argument("-f", "--sessionfile")
+    args = p.parse_args()
+    try:
+        import_session(args.cookiefile or get_cookiefile(), args.sessionfile)
+    except (ConnectionException, OperationalError) as e:
+        raise SystemExit("Cookie import failed: {}".format(e))
@@ -0,0 +1,30 @@
+from datetime import datetime
+import instaloader
+
+L = instaloader.Instaloader()
+
+posts = instaloader.Hashtag.from_name(L.context, "urbanphotography").get_posts()
+
+SINCE = datetime(2020, 5, 10)  # further from today, inclusive
+UNTIL = datetime(2020, 5, 11)  # closer to today, not inclusive
+
+k = 0  # initiate k
+k_list = []  # uncomment this to tune k
+
+for post in posts:
+    postdate = post.date
+
+    if postdate > UNTIL:
+        continue
+    elif postdate <= SINCE:
+        k += 1
+        if k == 50:
+            break
+        else:
+            continue
+    else:
+        L.download_post(post, "#urbanphotography")
+        k = 0  # set k to 0
+        # if you want to tune k, uncomment below to get your k max
+        #k_list.append(k)
+#max(k_list)
@@ -1,23 +0,0 @@
-from glob import glob
-from os.path import expanduser
-from sqlite3 import connect
-
-from instaloader import ConnectionException, Instaloader
-
-# FIREFOXCOOKIEFILE = "/home/alex/.mozilla/firefox/l96w6b90.default/cookies.sqlite"
-FIREFOXCOOKIEFILE = glob(expanduser("~/.mozilla/firefox/*.default/cookies.sqlite"))[0]
-
-instaloader = Instaloader(max_connection_attempts=1)
-instaloader.context._session.cookies.update(connect(FIREFOXCOOKIEFILE)
-                                            .execute("SELECT name, value FROM moz_cookies "
-                                                     "WHERE baseDomain='instagram.com'"))
-
-try:
-    username = instaloader.test_login()
-    if not username:
-        raise ConnectionException()
-except ConnectionException:
-    raise SystemExit("Cookie import failed. Are you logged in successfully in Firefox?")
-
-instaloader.context.username = username
-instaloader.save_session_to_file()
@@ -18,6 +18,7 @@
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 #
 import os
+import platform
 import subprocess
 import sys
 import typing
@@ -49,7 +50,25 @@ autodoc_member_order = 'bysource'
 intersphinx_mapping = {'python': ('https://docs.python.org/3', None),
                       'requests': ('https://requests.kennethreitz.org/en/master/', None)}

-extlinks = {'issue': ('https://github.com/instaloader/instaloader/issues/%s', 'issue #')}
+nitpick_ignore = [('py:class', 'typing.Tuple')]
+
+current_release = subprocess.check_output(["git", "describe", "--abbrev=0"]).decode("ascii")[1:-1]
+date_format = "%e %b %Y" if platform.system() != "Windows" else "%d %b %Y"
+current_release_date = subprocess.check_output(
+    ["git", "log", "-1", "--tags", "--format=%ad", "--date=format:"+date_format]).decode("ascii")[:-1]
+
+html_context = {'current_release': current_release, 'current_release_date': current_release_date}
+
+extlinks = {
+    'issue': (
+        'https://github.com/instaloader/instaloader/issues/%s',
+        'Issue #'
+    ),
+    'example': (
+        'https://raw.githubusercontent.com/instaloader/instaloader/v' + current_release + '/docs/codesnippets/%s',
+        'Example '
+    ),
+}

 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']
@@ -69,7 +88,7 @@ master_doc = 'index'

 # General information about the project.
 project = 'Instaloader'
-copyright = '2017, Alexander Graf and Andre Koch-Kramer'
+copyright = '2017-2020, Alexander Graf and Andre Koch-Kramer'
 author = 'Alexander Graf and Andre Koch-Kramer'

 # The version info for the project you're documenting, acts as replacement for
@@ -353,12 +372,6 @@ texinfo_documents = [
 # texinfo_no_detailmenu = False


-current_release = subprocess.check_output(["git", "describe", "--abbrev=0"]).decode("ascii")[1:-1]
-current_release_date = subprocess.check_output(["git", "log", "-1", "--tags", "--format=%ad", "--date=format:%e %b %Y"]).decode("ascii")[:-1]
-
-html_context = {'current_release': current_release, 'current_release_date': current_release_date}
-
-
 def setup(app):
    typing.TYPE_CHECKING = True
    app.add_stylesheet("instaloader.css")
@@ -35,8 +35,8 @@ reporting a problem, please keep the following in mind:
 - If not obvious, describe **which behavior you expected**
  instead of what actually happened.

- If you think an issue has been closed accidentally or inappropriately, feel
-  free to reopen it.
+- If we have closed an issue apparently inadvertently or inappropriately, please
+  let us know.

 Writing Code or Improving the Documentation
 -------------------------------------------
@@ -45,28 +45,25 @@ Changes of the Instaloader source can be proposed as a
 `Pull Request <https://github.com/instaloader/instaloader/pulls>`__. There are only
 few things to consider:

- Sometimes, the most current code is not in the ``master`` branch. Check that
-  you forked from the most recent branch.
+- Base your Pull Request on the ``master`` branch if it fixes a bug,
+  or the ``upcoming/v4.X`` branch (if it exists at the moment of submitting the PR)
+  otherwise.

 - We use `Pylint <https://www.pylint.org/>`__ for error and syntax checking of
-  the source. The file ``.travis.yml`` in the project's root directory
-  shows how it is invoked. Note that sometimes it might be better to disable a
-  warning rather than adapting the code to Pylint's desires.
+  the source and `MyPy <https://github.com/python/mypy>`__ for type checking.
+  Beware that sometimes it might be better to disable a warning rather than
+  adapting the code to a tool's desires.

 - The documentation source is located in the ``docs`` folder. The file
  ``cli-options.rst`` is merely an RST-formatted copy of ``instaloader --help``
  output, of which the source is in ``instaloader/__main__.py``.

- Feel free to contact us, even if you "only" have Proof-of-Concepts or
-  not-fully integrated changes. They already might be an advance for the
-  project.
-
-Suggesting Features
-------------------
+Proposing Features
+------------------

 .. goal-start

-Instaloader's goal is to mimick the browser's behavior to access the data that
+Instaloader's goal is to mimic the browser's behavior to access the data that
 is available through the Instagram web interface, pack this data into complete
 and easily-(re)usable python objects, and provide a user interface for the most
 common downloading and metadata collection tasks, without changing any of the
@@ -74,10 +71,9 @@ online data.

 .. goal-end

-If you have an idea of how Instaloader should be enhanced, but do not want to
-implement the feature yourself, feel free to open a ticket in our 
-`Issue Tracker <https://github.com/instaloader/instaloader/issues>`__.
-Please consider the following:
+Prior spending effort on implementing a new feature, it might be appropriate to
+clarify how it could fit into the project's scope or discuss implementation
+details. If you feel the need to do so, please create a "feature suggestion".

 - Instaloader already has plenty of features. **Check the documentation**
  beforehand to ensure your desired suggestion is not already implemented.
@@ -86,16 +82,21 @@ Please consider the following:
  an issue suggesting the same or a similar feature, share your thoughts in a
  comment there, instead of opening a new issue.

- If possible, provide us a **use case of the feature**: How could the user
-  invoke the new function? Which problem would it solve? If new information is
-  obtained, how would it be further processed?
+- **Motivate the feature**, i.e.

- If not obvious, briefly motivate how your suggested feature **conforms with
-  Instaloader's project goal**.
+  - Provide us a **use case of the feature**: How could the user
+    invoke the new function? Which problem would it solve? If new information is
+    obtained, how would it be further processed?

- **Be patient**. Naturally, bugs and pull requests have a higher priority than
-  feature suggestions. Keep in mind that this is a free software project, and
-  unfortunately we only have limited time to work on it.
+  - Describe already-working **alternatives of the feature** and how they
+    compare to your proposed feature.
+
+  - Briefly describe how your suggested feature **conforms with Instaloader's
+    project goal**.
+
+- Explain your **solution ideas**. Describe your ideas on how the feature could
+  be implemented and the underlying problem could be solved. Also **describe
+  alternatives** that you have considered.

 Donations
 ---------
@@ -40,12 +40,15 @@ See :ref:`install` for more options on how to install Instaloader.
 - allows **fine-grained customization** of filters and where to store
  downloaded media,

+- automatically **resumes previously-interrupted** download iterations,
+
 - is free `open source <https://github.com/instaloader/instaloader>`__
  software written in Python.

 ::

-    instaloader [--comments] [--geotags] [--stories] [--highlights] [--tagged]
+    instaloader [--comments] [--geotags]
+                [--stories] [--highlights] [--tagged] [--igtv]
                [--login YOUR-USERNAME] [--fast-update]
                profile | "#hashtag" | %location_id |
                :stories | :feed | :saved
@@ -27,12 +27,26 @@ expired. In case the retry continuously fails for some reason, which should not
 happen in normal conditions, consider adjusting the
 :option:`--max-connection-attempts` option.

+There have been observations that services, that in their nature offer
+promiscious IP addresses, such as cloud, VPN and public proxy services, might be
+subject to significantly stricter limits for anonymous access. However,
+logged-in accesses (see :option:`--login`) do not seem to be affected.
+
+Too many queries in the last time
+---------------------------------
+
 **"Too many queries in the last time"** is not an error. It is a notice that the
 rate limit has almost been reached, according to Instaloader's own rate
 accounting mechanism. We regularly adjust this mechanism to match Instagram's
 current rate limiting.

-Login error
+Private but not followed
+------------------------
+
+You have to follow a private account to access most of its associated
+information.
+
+Login Error
 -----------

 Instaloader's login *should* work fine, both with and without
@@ -40,21 +54,31 @@ Two-Factor-Authentication. It also supports handling the *checkpoint challenge*,
 issued when Instagram suspects authentication activity on your account, by
 pointing the user to an URL to be opened in a browser.

-Nevertheless, in :issue:`92` users report problems with logging in. To still use
-Instaloader's logged-in functionality, you may use the following script to
-workaround login problems by importing the session cookies from Firefox and
-bypassing Instaloader's login.
+Nevertheless, in :issue:`92` and :issue:`615` users reported problems with
+logging in. We recommend to always keep the sessionfile which Instaloader
+creates when using :option:`--login`. If a sessionfile is present,
+:option:`--login` does not make make use of the failure-prone login procedure.
+Also, sessionfiles usually do not expire and can be copied between different
+computers without any problems.

-.. literalinclude:: codesnippets/92_import_firefox_session.py
+If you do not have a sessionfile present, you may use the following script
+(:example:`615_import_firefox_session.py`) to workaround login problems by
+importing the session cookies from Firefox and bypassing Instaloader's login and
+so still use Instaloader's logged-in functionality.

-To use this,
+.. literalinclude:: codesnippets/615_import_firefox_session.py

-#. login to Instagram in Firefox, 
+To use this script,

-#. execute the snippet,
+#. Download the script: :example:`615_import_firefox_session.py`,

-#. then, ``instaloader -l USERNAME`` should work fine.
+#. Login to Instagram in Firefox,

-If you do not use your default firefox profile, or your operating system has the
-paths differently set up, you may have to alter the ``FIREFOXCOOKIEFILE``
-variable first.
+#. Execute the snippet, e.g. with ``python 615_import_firefox_session.py``,
+
+#. Then, ``instaloader -l USERNAME`` should work fine.
+
+This script also supports specifying a cookiefile path, which may be useful if
+you use multiple Firefox profiles or if your operating system has the directory
+structure differently set up. Also, you can specify an alternative sessionfile
+path.
@@ -1,7 +1,7 @@
 """Download pictures (or videos) along with their captions and other metadata from Instagram."""


-__version__ = '4.3a1'
+__version__ = '4.5a1'


 try:
@@ -14,6 +14,7 @@ else:

 from .exceptions import *
 from .instaloader import Instaloader
-from .instaloadercontext import InstaloaderContext
-from .structures import (Highlight, Post, PostSidecarNode, PostComment, PostCommentAnswer, PostLocation, Profile, Story,
-                         StoryItem, TopSearchResults, load_structure_from_file, save_structure_to_file)
+from .instaloadercontext import InstaloaderContext, RateController
+from .nodeiterator import NodeIterator, FrozenNodeIterator, resumable_iteration
+from .structures import (Hashtag, Highlight, Post, PostSidecarNode, PostComment, PostCommentAnswer, PostLocation,
+                         Profile, Story, StoryItem, TopSearchResults, load_structure_from_file, save_structure_to_file)
@@ -3,6 +3,7 @@
 import ast
 import datetime
 import os
+import re
 import sys
 from argparse import ArgumentParser, SUPPRESS
 from typing import List, Optional
@@ -19,7 +20,8 @@ def usage_string():
    argv0 = os.path.basename(sys.argv[0])
    argv0 = "instaloader" if argv0 == "__main__.py" else argv0
    return """
-{0} [--comments] [--geotags] [--stories] [--highlights] [--tagged]
+{0} [--comments] [--geotags]
+{2:{1}} [--stories] [--highlights] [--tagged] [--igtv]
 {2:{1}} [--login YOUR-USERNAME] [--fast-update]
 {2:{1}} profile | "#hashtag" | %%location_id | :stories | :feed | :saved
 {0} --help""".format(argv0, len(argv0), '')
@@ -61,7 +63,10 @@ def _main(instaloader: Instaloader, targetlist: List[str],
          username: Optional[str] = None, password: Optional[str] = None,
          sessionfile: Optional[str] = None,
          download_profile_pic: bool = True, download_posts=True,
-          download_stories: bool = False, download_highlights: bool = False, download_tagged: bool = False,
+          download_stories: bool = False,
+          download_highlights: bool = False,
+          download_tagged: bool = False,
+          download_igtv: bool = False,
          fast_update: bool = False,
          max_count: Optional[int] = None, post_filter_str: Optional[str] = None,
          storyitem_filter_str: Optional[str] = None) -> None:
@@ -77,6 +82,8 @@ def _main(instaloader: Instaloader, targetlist: List[str],
        instaloader.context.log('Only download storyitems with property "{}".'.format(storyitem_filter_str))
    # Login, if desired
    if username is not None:
+        if not re.match(r"^[A-Za-z0-9._]+$", username):
+            instaloader.context.error("Warning: Parameter \"{}\" for --login is not a valid username.".format(username))
        try:
            instaloader.load_session_from_file(username, sessionfile)
        except FileNotFoundError as err:
@@ -132,18 +139,19 @@ def _main(instaloader: Instaloader, targetlist: List[str],
            # strip '/' characters to be more shell-autocompletion-friendly
            target = target.rstrip('/')
            with instaloader.context.error_catcher(target):
-                if target[0] == '@':
+                if re.match(r"^@[A-Za-z0-9._]+$", target):
                    instaloader.context.log("Retrieving followees of %s..." % target[1:])
                    profile = Profile.from_username(instaloader.context, target[1:])
                    for followee in profile.get_followees():
                        instaloader.save_profile_id(followee)
                        profiles.add(followee)
-                elif target[0] == '#':
+                elif re.match(r"^#\w+$", target):
                    instaloader.download_hashtag(hashtag=target[1:], max_count=max_count, fast_update=fast_update,
-                                                 post_filter=post_filter)
-                elif target[0] == '-':
+                                                 post_filter=post_filter,
+                                                 profile_pic=download_profile_pic, posts=download_posts)
+                elif re.match(r"^-[A-Za-z0-9-_]+$", target):
                    instaloader.download_post(Post.from_shortcode(instaloader.context, target[1:]), target)
-                elif target[0] == "%":
+                elif re.match(r"^%[0-9]+$", target):
                    instaloader.download_location(location=target[1:], max_count=max_count, fast_update=fast_update,
                                                  post_filter=post_filter)
                elif target == ":feed":
@@ -154,11 +162,12 @@ def _main(instaloader: Instaloader, targetlist: List[str],
                elif target == ":saved":
                    instaloader.download_saved_posts(fast_update=fast_update, max_count=max_count,
                                                     post_filter=post_filter)
-                else:
+                elif re.match(r"^[A-Za-z0-9._]+$", target):
                    try:
                        profile = instaloader.check_profile_id(target)
                        if instaloader.context.is_logged_in and profile.has_blocked_viewer:
-                            if download_profile_pic or ((download_posts or download_tagged) and not profile.is_private):
+                            if download_profile_pic or ((download_posts or download_tagged or download_igtv)
+                                                        and not profile.is_private):
                                raise ProfileNotExistsException("{} blocked you; But we download her anonymously."
                                                                .format(target))
                            else:
@@ -169,7 +178,7 @@ def _main(instaloader: Instaloader, targetlist: List[str],
                        # Not only our profile.has_blocked_viewer condition raises ProfileNotExistsException,
                        # check_profile_id() also does, since access to blocked profile may be responded with 404.
                        if instaloader.context.is_logged_in and (download_profile_pic or download_posts or
-                                                                 download_tagged):
+                                                                 download_tagged or download_igtv):
                            instaloader.context.log(err)
                            instaloader.context.log("Trying again anonymously, helps in case you are just blocked.")
                            with instaloader.anonymous_copy() as anonymous_loader:
@@ -179,20 +188,28 @@ def _main(instaloader: Instaloader, targetlist: List[str],
                                                              .format(target, err))
                        else:
                            raise
+                else:
+                    target_type = {
+                        '#': 'hashtag',
+                        '%': 'location',
+                        '-': 'shortcode',
+                    }.get(target[0], 'username')
+                    raise ProfileNotExistsException('Invalid {} {}'.format(target_type, target))
        if len(profiles) > 1:
            instaloader.context.log("Downloading {} profiles: {}".format(len(profiles),
                                                                         ' '.join([p.username for p in profiles])))
-        if profiles and download_profile_pic and not instaloader.context.is_logged_in:
-            instaloader.context.error("Warning: Use --login to download HD version of profile pictures.")
+        if profiles and (download_profile_pic or download_posts) and not instaloader.context.is_logged_in:
+            instaloader.context.error("Warning: Use --login to download higher-quality versions of pictures.")
        instaloader.download_profiles(profiles,
-                                      download_profile_pic, download_posts, download_tagged, download_highlights,
-                                      download_stories, fast_update, post_filter, storyitem_filter)
+                                      download_profile_pic, download_posts, download_tagged, download_igtv,
+                                      download_highlights, download_stories,
+                                      fast_update, post_filter, storyitem_filter)
        if anonymous_retry_profiles:
            instaloader.context.log("Downloading anonymously: {}"
                                    .format(' '.join([p.username for p in anonymous_retry_profiles])))
            with instaloader.anonymous_copy() as anonymous_loader:
                anonymous_loader.download_profiles(anonymous_retry_profiles,
-                                                   download_profile_pic, download_posts, download_tagged,
+                                                   download_profile_pic, download_posts, download_tagged, download_igtv,
                                                   fast_update=fast_update, post_filter=post_filter)
    except KeyboardInterrupt:
        print("\nInterrupted by user.", file=sys.stderr)
@@ -287,6 +304,8 @@ def main():
                        help='Also download highlights of each profile that is downloaded. Requires --login.')
    g_prof.add_argument('--tagged', action='store_true',
                        help='Also download posts where each profile is tagged.')
+    g_prof.add_argument('--igtv', action='store_true',
+                        help='Also download IGTV videos.')

    g_cond = parser.add_argument_group("Which Posts to Download")

@@ -332,6 +351,13 @@ def main():
                            '--dirname-pattern. {profile} is replaced by the profile name,'
                            '{target} is replaced by the target you specified, i.e. either :feed'
                            '#hashtag or the profile name. Defaults to \'{date_utc}_UTC\'')
+    g_how.add_argument('--resume-prefix', metavar='PREFIX',
+                       help='Prefix for filenames that are used to save the information to resume an interrupted '
+                            'download.')
+    g_how.add_argument('--no-resume', action='store_true',
+                       help='Do not resume a previously-aborted download iteration, and do not save such information '
+                            'when interrupted.')
+    g_how.add_argument('--use-aged-resume-files', action='store_true', help=SUPPRESS)
    g_how.add_argument('--user-agent',
                       help='User Agent to use for HTTP requests. Defaults to \'{}\'.'.format(default_user_agent()))
    g_how.add_argument('-S', '--no-sleep', action='store_true', help=SUPPRESS)
@@ -339,10 +365,7 @@ def main():
                       help='Maximum number of connection attempts until a request is aborted. Defaults to 3. If a '
                            'connection fails, it can be manually skipped by hitting CTRL+C. Set this to 0 to retry '
                            'infinitely.')
-    g_how.add_argument('--commit-mode', action='store_true',
-                       help='Tries to ensure downloaded images avoid corruption in case of unexpected interruption. '
-                       'If the last picture is corrupted, Instaloader will fix the picture the next time it is run. '
-                       'Requires the JSON metadata to be saved.')
+    g_how.add_argument('--commit-mode', action='store_true', help=SUPPRESS)
    g_how.add_argument('--request-timeout', metavar='N', type=float,
                       help='seconds to wait before timing out a connection request')

@@ -378,6 +401,10 @@ def main():
                raise SystemExit("--no-captions and --post-metadata-txt or --storyitem-metadata-txt given; "
                                 "That contradicts.")

+        if args.no_resume and args.resume_prefix:
+            raise SystemExit("--no-resume and --resume-prefix given; That contradicts.")
+        resume_prefix = (args.resume_prefix if args.resume_prefix else 'iterator') if not args.no_resume else None
+
        if args.no_pictures and args.fast_update:
            raise SystemExit('--no-pictures and --fast-update cannot be used together.')

@@ -386,9 +413,6 @@ def main():
        download_posts = not (args.no_posts or args.stories_only or args.profile_pic_only)
        download_stories = args.stories or args.stories_only

-        if args.commit_mode and args.no_metadata_json:
-            raise SystemExit('--commit-mode requires JSON metadata to be saved.')
-
        loader = Instaloader(sleep=not args.no_sleep, quiet=args.quiet, user_agent=args.user_agent,
                             dirname_pattern=args.dirname_pattern, filename_pattern=args.filename_pattern,
                             download_pictures=not args.no_pictures,
@@ -400,7 +424,8 @@ def main():
                             storyitem_metadata_txt_pattern=storyitem_metadata_txt_pattern,
                             max_connection_attempts=args.max_connection_attempts,
                             request_timeout=args.request_timeout,
-                             commit_mode=args.commit_mode)
+                             resume_prefix=resume_prefix,
+                             check_resume_bbd=not args.use_aged_resume_files)
        _main(loader,
              args.profile,
              username=args.login.lower() if args.login is not None else None,
@@ -411,6 +436,7 @@ def main():
              download_stories=download_stories,
              download_highlights=args.highlights,
              download_tagged=args.tagged,
+              download_igtv=args.igtv,
              fast_update=args.fast_update,
              max_count=int(args.count) if args.count is not None else None,
              post_filter_str=args.post_filter,
@@ -1,6 +1,5 @@
 import getpass
 import json
-import lzma
 import os
 import platform
 import re
@@ -20,13 +19,28 @@ import requests
 import urllib3  # type: ignore

 from .exceptions import *
-from .instaloadercontext import InstaloaderContext
-from .structures import (Highlight, JsonExportable, Post, PostLocation, Profile, Story, StoryItem,
-                         save_structure_to_file, load_structure_from_file)
+from .instaloadercontext import InstaloaderContext, RateController
+from .nodeiterator import NodeIterator, resumable_iteration
+from .structures import (Hashtag, Highlight, JsonExportable, Post, PostLocation, Profile, Story, StoryItem,
+                         load_structure_from_file, save_structure_to_file)


 def get_default_session_filename(username: str) -> str:
    """Returns default session filename for given username."""
+    sessionfilename = "session-{}".format(username)
+    if platform.system() == "Windows":
+        # on Windows, use %LOCALAPPDATA%\Instaloader\session-USERNAME
+        localappdata = os.getenv("LOCALAPPDATA")
+        if localappdata is not None:
+            return os.path.join(localappdata, "Instaloader", sessionfilename)
+        # legacy fallback - store in temp dir if %LOCALAPPDATA% is not set
+        return os.path.join(tempfile.gettempdir(), ".instaloader-" + getpass.getuser(), sessionfilename)
+    # on Unix, use ~/.config/instaloader/session-USERNAME
+    return os.path.join(os.getenv("XDG_CONFIG_HOME", os.path.expanduser("~/.config")), "instaloader", sessionfilename)
+
+
+def get_legacy_session_filename(username: str) -> str:
+    """Returns legacy (until v4.4.3) default session filename for given username."""
    dirname = tempfile.gettempdir() + "/" + ".instaloader-" + getpass.getuser()
    filename = dirname + "/" + "session-" + username
    return filename.lower()
@@ -103,22 +117,20 @@ class _ArbitraryItemFormatter(string.Formatter):


 class _PostPathFormatter(_ArbitraryItemFormatter):
-    def vformat(self, format_string, args, kwargs):
-        """Override :meth:`string.Formatter.vformat` for character substitution in paths for Windows, see issue #84."""
-        ret = super().vformat(format_string, args, kwargs)
-        if platform.system() == 'Windows':
-            ret = ret.replace(':', '\ua789').replace('<', '\ufe64').replace('>', '\ufe65').replace('\"', '\uff02')
-            ret = ret.replace('\\', '\ufe68').replace('|', '\uff5c').replace('?', '\ufe16').replace('*', '\uff0a')
-        return ret
-
    def get_value(self, key, args, kwargs):
-        """Replaces '/' with similar looking Division Slash and on windows newline with space"""
        ret = super().get_value(key, args, kwargs)
        if not isinstance(ret, str):
            return ret
+        return self.sanitize_path(ret)
+
+    @staticmethod
+    def sanitize_path(ret: str) -> str:
+        """Replaces '/' with similar looking Division Slash and some other illegal filename characters on Windows."""
        ret = ret.replace('/', '\u2215')
        if platform.system() == 'Windows':
-            ret = ret.replace('\n', ' ')
+            ret = ret.replace(':', '\uff1a').replace('<', '\ufe64').replace('>', '\ufe65').replace('\"', '\uff02')
+            ret = ret.replace('\\', '\ufe68').replace('|', '\uff5c').replace('?', '\ufe16').replace('*', '\uff0a')
+            ret = ret.replace('\n', ' ').replace('\r', ' ')
        return ret


@@ -141,8 +153,10 @@ class Instaloader:
       txt file.
    :param storyitem_metadata_txt_pattern: :option:`--storyitem-metadata-txt`, default is empty (=none)
    :param max_connection_attempts: :option:`--max-connection-attempts`
-    :param commit_mode: :option:`--commit-mode`
-    :param request_timeout: "option:`--request-timeout`, set per-request timeout (seconds)
+    :param request_timeout: :option:`--request-timeout`, set per-request timeout (seconds)
+    :param rate_controller: Generator for a :class:`RateController` to override rate controlling behavior
+    :param resume_prefix: :option:`--resume-prefix`, or None for :option:`--no-resume`.
+    :param check_resume_bbd: Whether to check the date of expiry of resume files and reject them if expired.

    .. attribute:: context

@@ -166,9 +180,12 @@ class Instaloader:
                 storyitem_metadata_txt_pattern: str = None,
                 max_connection_attempts: int = 3,
                 request_timeout: Optional[float] = None,
-                 commit_mode: bool = False):
+                 rate_controller: Optional[Callable[[InstaloaderContext], RateController]] = None,
+                 resume_prefix: Optional[str] = "iterator",
+                 check_resume_bbd: bool = True):

-        self.context = InstaloaderContext(sleep, quiet, user_agent, max_connection_attempts, request_timeout)
+        self.context = InstaloaderContext(sleep, quiet, user_agent, max_connection_attempts,
+                                          request_timeout, rate_controller)

        # configuration parameters
        self.dirname_pattern = dirname_pattern or "{target}"
@@ -184,12 +201,8 @@ class Instaloader:
            else post_metadata_txt_pattern
        self.storyitem_metadata_txt_pattern = '' if storyitem_metadata_txt_pattern is None \
            else storyitem_metadata_txt_pattern
-        self.commit_mode = commit_mode
-        if self.commit_mode and not self.save_metadata:
-            raise InvalidArgumentException("Commit mode requires JSON metadata to be saved.")
-
-        # Used to keep state in commit mode
-        self._committed = None  # type: Optional[bool]
+        self.resume_prefix = resume_prefix
+        self.check_resume_bbd = check_resume_bbd

    @contextmanager
    def anonymous_copy(self):
@@ -211,7 +224,8 @@ class Instaloader:
            storyitem_metadata_txt_pattern=self.storyitem_metadata_txt_pattern,
            max_connection_attempts=self.context.max_connection_attempts,
            request_timeout=self.context.request_timeout,
-            commit_mode=self.commit_mode)
+            resume_prefix=self.resume_prefix,
+            check_resume_bbd=self.check_resume_bbd)
        yield new_loader
        self.context.error_log.extend(new_loader.context.error_log)
        new_loader.context.error_log = []  # avoid double-printing of errors
@@ -237,15 +251,9 @@ class Instaloader:
        if filename_suffix is not None:
            filename += '_' + filename_suffix
        filename += '.' + file_extension
-        # A post is considered "commited" if the json file exists and is not malformed.
-        if self.commit_mode:
-            if self._committed and os.path.isfile(filename):
-                self.context.log(filename + ' exists', end=' ', flush=True)
-                return False
-        else:
-            if os.path.isfile(filename):
-                self.context.log(filename + ' exists', end=' ', flush=True)
-                return False
+        if os.path.isfile(filename):
+            self.context.log(filename + ' exists', end=' ', flush=True)
+            return False
        self.context.get_and_write_raw(url, filename)
        os.utime(filename, (datetime.now().timestamp(), mtime.timestamp()))
        return True
@@ -357,43 +365,73 @@ class Instaloader:
        os.utime(filename, (datetime.now().timestamp(), mtime.timestamp()))
        self.context.log('geo', end=' ', flush=True)

+    def format_filename_within_target_path(self,
+                                           target: Union[str, Path],
+                                           owner_profile: Optional[Profile],
+                                           identifier: str,
+                                           name_suffix: str,
+                                           extension: str):
+        """Returns a filename within the target path.
+
+        .. versionadded:: 4.5"""
+        if ((format_string_contains_key(self.dirname_pattern, 'profile') or
+             format_string_contains_key(self.dirname_pattern, 'target'))):
+            profile_str = owner_profile.username.lower() if owner_profile is not None else target
+            return os.path.join(self.dirname_pattern.format(profile=profile_str, target=target),
+                                '{0}_{1}.{2}'.format(identifier, name_suffix, extension))
+        else:
+            return os.path.join(self.dirname_pattern.format(),
+                                '{0}_{1}_{2}.{3}'.format(target, identifier, name_suffix, extension))
+
    @_retry_on_connection_error
-    def download_profilepic(self, profile: Profile, _attempt: int = 1) -> None:
-        """Downloads and saves profile pic."""
+    def download_title_pic(self, url: str, target: Union[str, Path], name_suffix: str, owner_profile: Optional[Profile],
+                           _attempt: int = 1) -> None:
+        """Downloads and saves a picture that does not have an association with a Post or StoryItem, such as a
+        Profile picture or a Highlight cover picture. Modification time is taken from the HTTP response headers.
+
+        .. versionadded:: 4.3"""

        def _epoch_to_string(epoch: datetime) -> str:
            return epoch.strftime('%Y-%m-%d_%H-%M-%S_UTC')

-        profile_pic_response = self.context.get_raw(profile.profile_pic_url)
+        http_response = self.context.get_raw(url)
        date_object = None  # type: Optional[datetime]
-        if 'Last-Modified' in profile_pic_response.headers:
-            date_object = datetime.strptime(profile_pic_response.headers["Last-Modified"], '%a, %d %b %Y %H:%M:%S GMT')
-            profile_pic_bytes = None
-            profile_pic_identifier = _epoch_to_string(date_object)
+        if 'Last-Modified' in http_response.headers:
+            date_object = datetime.strptime(http_response.headers["Last-Modified"], '%a, %d %b %Y %H:%M:%S GMT')
+            pic_bytes = None
+            pic_identifier = _epoch_to_string(date_object)
        else:
-            profile_pic_bytes = profile_pic_response.content
-            profile_pic_identifier = md5(profile_pic_bytes).hexdigest()[:16]
-        profile_pic_extension = 'jpg'
-        if ((format_string_contains_key(self.dirname_pattern, 'profile') or
-             format_string_contains_key(self.dirname_pattern, 'target'))):
-            filename = '{0}/{1}_profile_pic.{2}'.format(self.dirname_pattern.format(profile=profile.username.lower(),
-                                                                                    target=profile.username.lower()),
-                                                        profile_pic_identifier, profile_pic_extension)
-        else:
-            filename = '{0}/{1}_{2}_profile_pic.{3}'.format(self.dirname_pattern.format(), profile.username.lower(),
-                                                            profile_pic_identifier, profile_pic_extension)
-        content_length = profile_pic_response.headers.get('Content-Length', None)
+            pic_bytes = http_response.content
+            pic_identifier = md5(pic_bytes).hexdigest()[:16]
+        filename = self.format_filename_within_target_path(target, owner_profile, pic_identifier, name_suffix, 'jpg')
+        content_length = http_response.headers.get('Content-Length', None)
        if os.path.isfile(filename) and (not self.context.is_logged_in or
                                         (content_length is not None and
                                          os.path.getsize(filename) >= int(content_length))):
            self.context.log(filename + ' already exists')
            return None
        os.makedirs(os.path.dirname(filename), exist_ok=True)
-        self.context.write_raw(profile_pic_bytes if profile_pic_bytes else profile_pic_response, filename)
+        self.context.write_raw(pic_bytes if pic_bytes else http_response, filename)
        if date_object:
            os.utime(filename, (datetime.now().timestamp(), date_object.timestamp()))
        self.context.log('')  # log output of _get_and_write_raw() does not produce \n

+    def download_profilepic(self, profile: Profile) -> None:
+        """Downloads and saves profile pic."""
+        self.download_title_pic(profile.profile_pic_url, profile.username.lower(), 'profile_pic', profile)
+
+    def download_highlight_cover(self, highlight: Highlight, target: Union[str, Path]) -> None:
+        """Downloads and saves Highlight cover picture.
+
+        .. versionadded:: 4.3"""
+        self.download_title_pic(highlight.cover_url, target, 'cover', highlight.owner_profile)
+
+    def download_hashtag_profilepic(self, hashtag: Hashtag) -> None:
+        """Downloads and saves the profile picture of a Hashtag.
+
+        .. versionadded:: 4.4"""
+        self.download_title_pic(hashtag.profile_pic_url, '#' + hashtag.name, 'profile_pic', None)
+
    @_requires_login
    def save_session_to_file(self, filename: Optional[str] = None) -> None:
        """Saves internally stored :class:`requests.Session` object.
@@ -422,6 +460,8 @@ class Instaloader:
        """
        if filename is None:
            filename = get_default_session_filename(username)
+            if not os.path.exists(filename):
+                filename = get_legacy_session_filename(username)
        with open(filename, 'rb') as sessionfile:
            self.context.load_session_from_file(username, sessionfile)
            self.context.log("Loaded session from %s." % filename)
@@ -466,12 +506,11 @@ class Instaloader:
        """

        dirname = _PostPathFormatter(post).format(self.dirname_pattern, target=target)
-        filename = dirname + '/' + self.format_filename(post, target=target)
+        filename = os.path.join(dirname, self.format_filename(post, target=target))
        os.makedirs(os.path.dirname(filename), exist_ok=True)

        # Download the image(s) / video thumbnail and videos within sidecars if desired
        downloaded = True
-        self._committed = self.check_if_committed(filename)
        if self.download_pictures:
            if post.typename == 'GraphSidecar':
                edge_number = 1
@@ -595,7 +634,7 @@ class Instaloader:

        date_local = item.date_local
        dirname = _PostPathFormatter(item).format(self.dirname_pattern, target=target)
-        filename = dirname + '/' + self.format_filename(item, target=target)
+        filename = os.path.join(dirname, self.format_filename(item, target=target))
        os.makedirs(os.path.dirname(filename), exist_ok=True)
        downloaded = False
        if not item.is_video or self.download_video_thumbnails is True:
@@ -646,6 +685,9 @@ class Instaloader:

        .. versionadded:: 4.1

+        .. versionchanged:: 4.3
+           Also downloads and saves the Highlight's cover pictures.
+
        :param user: ID or Profile of the user whose highlights should get downloaded.
        :param fast_update: If true, abort when first already-downloaded picture is encountered
        :param filename_target: Replacement for {target} in dirname_pattern and filename_pattern
@@ -655,7 +697,12 @@ class Instaloader:
        """
        for user_highlight in self.get_highlights(user):
            name = user_highlight.owner_username
+            highlight_target = (filename_target
+                                if filename_target
+                                else (Path(_PostPathFormatter.sanitize_path(name)) /
+                                      _PostPathFormatter.sanitize_path(user_highlight.title)))  # type: Union[str, Path]
            self.context.log("Retrieving highlights \"{}\" from profile {}".format(user_highlight.title, name))
+            self.download_highlight_cover(user_highlight, highlight_target)
            totalcount = user_highlight.itemcount
            count = 1
            for item in user_highlight.get_items():
@@ -666,12 +713,86 @@ class Instaloader:
                count += 1
                with self.context.error_catcher('Download highlights \"{}\" from user {}'.format(user_highlight.title,
                                                                                                 name)):
-                    downloaded = self.download_storyitem(item, filename_target
-                                                         if filename_target
-                                                         else Path(name) / Path(user_highlight.title))
+                    downloaded = self.download_storyitem(item, highlight_target)
                    if fast_update and not downloaded:
                        break

+    def posts_download_loop(self,
+                            posts: Iterator[Post],
+                            target: Union[str, Path],
+                            fast_update: bool = False,
+                            post_filter: Optional[Callable[[Post], bool]] = None,
+                            max_count: Optional[int] = None,
+                            total_count: Optional[int] = None,
+                            owner_profile: Optional[Profile] = None) -> None:
+        """
+        Download the Posts returned by given Post Iterator.
+
+        ..versionadded:: 4.4
+
+        ..versionchanged:: 4.5
+           Transparently resume an aborted operation if `posts` is a :class:`NodeIterator`.
+
+        :param posts: Post Iterator to loop through.
+        :param target: Target name.
+        :param fast_update: :option:`--fast-update`.
+        :param post_filter: :option:`--post-filter`.
+        :param max_count: Maximum count of Posts to download (:option:`--count`).
+        :param total_count: Total number of posts returned by given iterator.
+        :param owner_profile: Associated profile, if any.
+        """
+        displayed_count = (max_count if total_count is None or max_count is not None and max_count < total_count
+                           else total_count)
+        with resumable_iteration(
+                context=self.context,
+                iterator=posts,
+                load=load_structure_from_file,
+                save=save_structure_to_file,
+                format_path=lambda magic: self.format_filename_within_target_path(
+                    target, owner_profile, self.resume_prefix or '', magic, 'json.xz'
+                ),
+                check_bbd=self.check_resume_bbd,
+                enabled=self.resume_prefix is not None
+        ) as resume_info:
+            is_resuming, start_index = resume_info
+            for number, post in enumerate(posts):
+                if max_count is not None and number + start_index >= max_count:
+                    break
+                if displayed_count is not None:
+                    self.context.log("[{0:{w}d}/{1:{w}d}] ".format(number + start_index + 1, displayed_count,
+                                                                   w=len(str(displayed_count))),
+                                     end="", flush=True)
+                else:
+                    self.context.log("[{:3d}] ".format(number + start_index + 1), end="", flush=True)
+                if post_filter is not None:
+                    try:
+                        if not post_filter(post):
+                            self.context.log("{} skipped".format(post))
+                            continue
+                    except (InstaloaderException, KeyError, TypeError) as err:
+                        self.context.error("{} skipped. Filter evaluation failed: {}".format(post, err))
+                        continue
+                with self.context.error_catcher("Download {} of {}".format(post, target)):
+                    # The PostChangedException gets raised if the Post's id/shortcode changed while obtaining
+                    # additional metadata. This is most likely the case if a HTTP redirect takes place while
+                    # resolving the shortcode URL.
+                    # The `post_changed` variable keeps the fast-update functionality alive: A Post which is
+                    # obained after a redirect has probably already been downloaded as a previous Post of the
+                    # same Profile.
+                    # Observed in issue #225: https://github.com/instaloader/instaloader/issues/225
+                    post_changed = False
+                    while True:
+                        try:
+                            downloaded = self.download_post(post, target=target)
+                            break
+                        except PostChangedException:
+                            post_changed = True
+                            continue
+                    if fast_update and not downloaded and not post_changed:
+                        # disengage fast_update for first post when resuming
+                        if not is_resuming or number > 0:
+                            break
+
    @_requires_login
    def get_feed_posts(self) -> Iterator[Post]:
        """Get Posts of the user's feed.
@@ -684,8 +805,10 @@ class Instaloader:

        while True:
            feed = data["user"]["edge_web_feed_timeline"]
-            yield from (Post(self.context, edge["node"]) for edge in feed["edges"]
-                        if not edge["node"]["__typename"] == "GraphSuggestedUserFeedUnit")
+            for edge in feed["edges"]:
+                node = edge["node"]
+                if node.get("__typename") in Post.supported_graphql_types() and node.get("shortcode") is not None:
+                    yield Post(self.context, node)
            if not feed["page_info"]["has_next_page"]:
                break
            data = self.context.graphql_query("d6f4427fbe92d846298cf93df0b937d3",
@@ -714,20 +837,7 @@ class Instaloader:
        :raises LoginRequiredException: If called without being logged in.
        """
        self.context.log("Retrieving pictures from your feed...")
-        count = 1
-        for post in self.get_feed_posts():
-            if max_count is not None and count > max_count:
-                break
-            name = post.owner_username
-            if post_filter is not None and not post_filter(post):
-                self.context.log("<pic by %s skipped>" % name, flush=True)
-                continue
-            self.context.log("[%3i] %s " % (count, name), end="", flush=True)
-            count += 1
-            with self.context.error_catcher('Download feed'):
-                downloaded = self.download_post(post, target=':feed')
-                if fast_update and not downloaded:
-                    break
+        self.posts_download_loop(self.get_feed_posts(), ":feed", fast_update, post_filter, max_count=max_count)

    @_requires_login
    def download_saved_posts(self, max_count: int = None, fast_update: bool = False,
@@ -740,20 +850,11 @@ class Instaloader:
        :raises LoginRequiredException: If called without being logged in.
        """
        self.context.log("Retrieving saved posts...")
-        count = 1
-        assert self.context.username is not None
-        for post in Profile.from_username(self.context, self.context.username).get_saved_posts():
-            if max_count is not None and count > max_count:
-                break
-            if post_filter is not None and not post_filter(post):
-                self.context.log("<{} skipped>".format(post), flush=True)
-                continue
-            self.context.log("[{:>3}] ".format(count), end=str(), flush=True)
-            count += 1
-            with self.context.error_catcher('Download saved posts'):
-                downloaded = self.download_post(post, target=':saved')
-                if fast_update and not downloaded:
-                    break
+        assert self.context.username is not None  # safe due to @_requires_login; required by typechecker
+        node_iterator = Profile.from_username(self.context, self.context.username).get_saved_posts()
+        self.posts_download_loop(node_iterator, ":saved",
+                                 fast_update, post_filter,
+                                 max_count=max_count, total_count=node_iterator.count)

    @_requires_login
    def get_location_posts(self, location: str) -> Iterator[Post]:
@@ -804,53 +905,38 @@ class Instaloader:
           Require being logged in (as required by Instagram)
        """
        self.context.log("Retrieving pictures for location {}...".format(location))
-        count = 1
-        for post in self.get_location_posts(location):
-            if max_count is not None and count > max_count:
-                break
-            self.context.log('[{0:3d}] %{1} '.format(count, location), end='', flush=True)
-            if post_filter is not None and not post_filter(post):
-                self.context.log('<skipped>')
-                continue
-            count += 1
-            with self.context.error_catcher('Download location {}'.format(location)):
-                downloaded = self.download_post(post, target='%' + location)
-                if fast_update and not downloaded:
-                    break
+        self.posts_download_loop(self.get_location_posts(location), "%" + location, fast_update, post_filter,
+                                 max_count=max_count)

    @_requires_login
-    def get_explore_posts(self) -> Iterator[Post]:
+    def get_explore_posts(self) -> NodeIterator[Post]:
        """Get Posts which are worthy of exploring suggested by Instagram.

        :return: Iterator over Posts of the user's suggested posts.
+        :rtype: NodeIterator[Post]
        :raises LoginRequiredException: If called without being logged in.
        """
-        data = self.context.get_json('explore/', {})
-        yield from (Post(self.context, node)
-                    for node in self.context.graphql_node_list("df0dcc250c2b18d9fd27c5581ef33c7c",
-                                                               {}, 'https://www.instagram.com/explore/',
-                                                               lambda d: d['data']['user']['edge_web_discover_media'],
-                                                               data.get('rhx_gis')))
+        return NodeIterator(
+            self.context,
+            'df0dcc250c2b18d9fd27c5581ef33c7c',
+            lambda d: d['data']['user']['edge_web_discover_media'],
+            lambda n: Post(self.context, n),
+            query_referer='https://www.instagram.com/explore/',
+        )

    def get_hashtag_posts(self, hashtag: str) -> Iterator[Post]:
-        """Get Posts associated with a #hashtag."""
-        has_next_page = True
-        end_cursor = None
-        while has_next_page:
-            if end_cursor:
-                params = {'__a': 1, 'max_id': end_cursor}
-            else:
-                params = {'__a': 1}
-            hashtag_data = self.context.get_json('explore/tags/{0}/'.format(hashtag),
-                                                 params)['graphql']['hashtag']['edge_hashtag_to_media']
-            yield from (Post(self.context, edge['node']) for edge in hashtag_data['edges'])
-            has_next_page = hashtag_data['page_info']['has_next_page']
-            end_cursor = hashtag_data['page_info']['end_cursor']
+        """Get Posts associated with a #hashtag.

-    def download_hashtag(self, hashtag: str,
+        .. deprecated:: 4.4
+           Use :meth:`Hashtag.get_posts`."""
+        return Hashtag.from_name(self.context, hashtag).get_posts()
+
+    def download_hashtag(self, hashtag: Union[Hashtag, str],
                         max_count: Optional[int] = None,
                         post_filter: Optional[Callable[[Post], bool]] = None,
-                         fast_update: bool = False) -> None:
+                         fast_update: bool = False,
+                         profile_pic: bool = True,
+                         posts: bool = True) -> None:
        """Download pictures of one hashtag.

        To download the last 30 pictures with hashtag #cat, do::
@@ -858,26 +944,34 @@ class Instaloader:
            loader = Instaloader()
            loader.download_hashtag('cat', max_count=30)

-        :param hashtag: Hashtag to download, without leading '#'
+        :param hashtag: Hashtag to download, as instance of :class:`Hashtag`, or string without leading '#'
        :param max_count: Maximum count of pictures to download
        :param post_filter: function(post), which returns True if given picture should be downloaded
        :param fast_update: If true, abort when first already-downloaded picture is encountered
+        :param profile_pic: not :option:`--no-profile-pic`.
+        :param posts: not :option:`--no-posts`.
+
+        .. versionchanged:: 4.4
+           Add parameters `profile_pic` and `posts`.
        """
-        hashtag = hashtag.lower()
-        self.context.log("Retrieving pictures with hashtag {}...".format(hashtag))
-        count = 1
-        for post in self.get_hashtag_posts(hashtag):
-            if max_count is not None and count > max_count:
-                break
-            self.context.log('[{0:3d}] #{1} '.format(count, hashtag), end='', flush=True)
-            if post_filter is not None and not post_filter(post):
-                self.context.log('<skipped>')
-                continue
-            count += 1
-            with self.context.error_catcher('Download hashtag #{}'.format(hashtag)):
-                downloaded = self.download_post(post, target='#' + hashtag)
-                if fast_update and not downloaded:
-                    break
+        if isinstance(hashtag, str):
+            with self.context.error_catcher("Get hashtag #{}".format(hashtag)):
+                hashtag = Hashtag.from_name(self.context, hashtag)
+        if not isinstance(hashtag, Hashtag):
+            return
+        target = "#" + hashtag.name
+        if profile_pic:
+            with self.context.error_catcher("Download profile picture of {}".format(target)):
+                self.download_hashtag_profilepic(hashtag)
+        if posts:
+            self.context.log("Retrieving pictures with hashtag #{}...".format(hashtag.name))
+            self.posts_download_loop(hashtag.get_all_posts(), target, fast_update, post_filter,
+                                     max_count=max_count)
+        if self.save_metadata:
+            json_filename = '{0}/{1}'.format(self.dirname_pattern.format(profile=target,
+                                                                         target=target),
+                                             target)
+            self.save_metadata_json(json_filename, hashtag)

    def download_tagged(self, profile: Profile, fast_update: bool = False,
                        target: Optional[str] = None,
@@ -886,24 +980,30 @@ class Instaloader:

        .. versionadded:: 4.1"""
        self.context.log("Retrieving tagged posts for profile {}.".format(profile.username))
-        count = 1
-        for post in profile.get_tagged_posts():
-            self.context.log("[%3i/???] " % (count), end="", flush=True)
-            count += 1
-            if post_filter is not None and not post_filter(post):
-                self.context.log('<{} skipped>'.format(post))
-            with self.context.error_catcher('Download tagged {}'.format(profile.username)):
-                downloaded = self.download_post(post, target if target else Path(profile.username) / Path(':tagged'))
-                if fast_update and not downloaded:
-                    break
+        self.posts_download_loop(profile.get_tagged_posts(),
+                                 target if target
+                                 else (Path(_PostPathFormatter.sanitize_path(profile.username)) /
+                                       _PostPathFormatter.sanitize_path(':tagged')),
+                                 fast_update, post_filter)
+
+    def download_igtv(self, profile: Profile, fast_update: bool = False,
+                      post_filter: Optional[Callable[[Post], bool]] = None) -> None:
+        """Download IGTV videos of a profile.
+
+        .. versionadded:: 4.3"""
+        self.context.log("Retrieving IGTV videos for profile {}.".format(profile.username))
+        self.posts_download_loop(profile.get_igtv_posts(), profile.username, fast_update, post_filter,
+                                 total_count=profile.igtvcount, owner_profile=profile)

    def _get_id_filename(self, profile_name: str) -> str:
        if ((format_string_contains_key(self.dirname_pattern, 'profile') or
             format_string_contains_key(self.dirname_pattern, 'target'))):
-            return '{0}/id'.format(self.dirname_pattern.format(profile=profile_name.lower(),
-                                                               target=profile_name.lower()))
+            return os.path.join(self.dirname_pattern.format(profile=profile_name.lower(),
+                                                            target=profile_name.lower()),
+                                'id')
        else:
-            return '{0}/{1}_id'.format(self.dirname_pattern.format(), profile_name.lower())
+            return os.path.join(self.dirname_pattern.format(),
+                                '{0}_id'.format(profile_name.lower()))

    def save_profile_id(self, profile: Profile):
        """
@@ -969,7 +1069,10 @@ class Instaloader:

    def download_profiles(self, profiles: Set[Profile],
                          profile_pic: bool = True, posts: bool = True,
-                          tagged: bool = False, highlights: bool = False, stories: bool = False,
+                          tagged: bool = False,
+                          igtv: bool = False,
+                          highlights: bool = False,
+                          stories: bool = False,
                          fast_update: bool = False,
                          post_filter: Optional[Callable[[Post], bool]] = None,
                          storyitem_filter: Optional[Callable[[Post], bool]] = None,
@@ -980,6 +1083,7 @@ class Instaloader:
        :param profile_pic: not :option:`--no-profile-pic`.
        :param posts: not :option:`--no-posts`.
        :param tagged: :option:`--tagged`.
+        :param igtv: :option:`--igtv`.
        :param highlights: :option:`--highlights`.
        :param stories: :option:`--stories`.
        :param fast_update: :option:`--fast-update`.
@@ -989,7 +1093,11 @@ class Instaloader:
           Whether :exc:`LoginRequiredException` and :exc:`PrivateProfileNotFollowedException` should be raised or
           catched and printed with :meth:`InstaloaderContext.error_catcher`.

-        .. versionadded:: 4.1"""
+        .. versionadded:: 4.1
+
+        .. versionchanged:: 4.3
+           Add `igtv` parameter.
+        """

        @contextmanager
        def _error_raiser(_str):
@@ -1009,13 +1117,13 @@ class Instaloader:

                # Save metadata as JSON if desired.
                if self.save_metadata:
-                    json_filename = '{0}/{1}_{2}'.format(self.dirname_pattern.format(profile=profile_name,
-                                                                                     target=profile_name),
-                                                         profile_name, profile.userid)
+                    json_filename = os.path.join(self.dirname_pattern.format(profile=profile_name,
+                                                                             target=profile_name),
+                                                 '{0}_{1}'.format(profile_name, profile.userid))
                    self.save_metadata_json(json_filename, profile)

                # Catch some errors
-                if profile.is_private and (tagged or highlights or posts):
+                if profile.is_private and (tagged or igtv or highlights or posts):
                    if not self.context.is_logged_in:
                        raise LoginRequiredException("--login=USERNAME required.")
                    if not profile.followed_by_viewer and self.context.username != profile.username:
@@ -1026,6 +1134,11 @@ class Instaloader:
                    with self.context.error_catcher('Download tagged of {}'.format(profile_name)):
                        self.download_tagged(profile, fast_update=fast_update, post_filter=post_filter)

+                # Download IGTV, if requested
+                if igtv:
+                    with self.context.error_catcher('Download IGTV of {}'.format(profile_name)):
+                        self.download_igtv(profile, fast_update=fast_update, post_filter=post_filter)
+
                # Download highlights, if requested
                if highlights:
                    with self.context.error_catcher('Download highlights of {}'.format(profile_name)):
@@ -1034,32 +1147,8 @@ class Instaloader:
                # Iterate over pictures and download them
                if posts:
                    self.context.log("Retrieving posts from profile {}.".format(profile_name))
-                    totalcount = profile.mediacount
-                    count = 1
-                    for post in profile.get_posts():
-                        self.context.log("[%3i/%3i] " % (count, totalcount), end="", flush=True)
-                        count += 1
-                        if post_filter is not None and not post_filter(post):
-                            self.context.log('<skipped>')
-                            continue
-                        with self.context.error_catcher("Download {} of {}".format(post, profile_name)):
-                            # The PostChangedException gets raised if the Post's id/shortcode changed while obtaining
-                            # additional metadata. This is most likely the case if a HTTP redirect takes place while
-                            # resolving the shortcode URL.
-                            # The `post_changed` variable keeps the fast-update functionality alive: A Post which is
-                            # obained after a redirect has probably already been downloaded as a previous Post of the
-                            # same Profile.
-                            # Observed in issue #225: https://github.com/instaloader/instaloader/issues/225
-                            post_changed = False
-                            while True:
-                                try:
-                                    downloaded = self.download_post(post, target=profile_name)
-                                    break
-                                except PostChangedException:
-                                    post_changed = True
-                                    continue
-                            if fast_update and not downloaded and not post_changed:
-                                break
+                    self.posts_download_loop(profile.get_posts(), profile_name, fast_update, post_filter,
+                                             total_count=profile.mediacount, owner_profile=profile)

        if stories and profiles:
            with self.context.error_catcher("Download stories"):
@@ -1138,37 +1227,8 @@ class Instaloader:

        # Iterate over pictures and download them
        self.context.log("Retrieving posts from profile {}.".format(profile_name))
-        totalcount = profile.mediacount
-        count = 1
-        for post in profile.get_posts():
-            self.context.log("[%3i/%3i] " % (count, totalcount), end="", flush=True)
-            count += 1
-            if post_filter is not None and not post_filter(post):
-                self.context.log('<skipped>')
-                continue
-            with self.context.error_catcher('Download profile {}'.format(profile_name)):
-                downloaded = self.download_post(post, target=profile_name)
-                if fast_update and not downloaded:
-                    break
-
-    def check_if_committed(self, filename: str) -> bool:
-        """Checks to see if the current post has been committed.
-
-        A post is considered committed if its json metadata file exists and is not malformed.
-
-        .. versionadded:: 4.2
-        """
-        if os.path.isfile(filename + '.json.xz'):
-            filename += '.json.xz'
-        elif os.path.isfile(filename + '.json'):
-            filename += '.json'
-        else:
-            return False
-        try:
-            load_structure_from_file(self.context, filename)
-            return True
-        except (FileNotFoundError, lzma.LZMAError, json.decoder.JSONDecodeError):
-            return False
+        self.posts_download_loop(profile.get_posts(), profile_name, fast_update, post_filter,
+                                 total_count=profile.mediacount, owner_profile=profile)

    def interactive_login(self, username: str) -> None:
        """Logs in and internally stores session, asking user for password interactively.
@@ -1,5 +1,6 @@
 import hashlib
 import json
+import os
 import pickle
 import random
 import re
@@ -52,7 +53,8 @@ class InstaloaderContext:
    """

    def __init__(self, sleep: bool = True, quiet: bool = False, user_agent: Optional[str] = None,
-                 max_connection_attempts: int = 3, request_timeout: Optional[float] = None):
+                 max_connection_attempts: int = 3, request_timeout: Optional[float] = None,
+                 rate_controller: Optional[Callable[["InstaloaderContext"], "RateController"]] = None):

        self.user_agent = user_agent if user_agent is not None else default_user_agent()
        self.request_timeout = request_timeout
@@ -68,9 +70,7 @@ class InstaloaderContext:
        # error log, filled with error() and printed at the end of Instaloader.main()
        self.error_log = []                      # type: List[str]

-        # For the adaption of sleep intervals (rate control)
-        self._graphql_query_timestamps = dict()  # type: Dict[str, List[float]]
-        self._graphql_earliest_next_request_time = 0.0
+        self._rate_controller = rate_controller(self) if rate_controller is not None else RateController(self)

        # Can be set to True for testing, disables supression of InstaloaderContext._error_catcher
        self.raise_all_errors = False
@@ -113,7 +113,7 @@ class InstaloaderContext:
    def close(self):
        """Print error log and close session"""
        if self.error_log and not self.quiet:
-            print("\nErrors occured:", file=sys.stderr)
+            print("\nErrors or warnings occurred:", file=sys.stderr)
            for err in self.error_log:
                print(err, file=sys.stderr)
        self._session.close()
@@ -215,8 +215,11 @@ class InstaloaderContext:
        session.headers.update({'X-CSRFToken': csrf_token})
        # Not using self.get_json() here, because we need to access csrftoken cookie
        self.do_sleep()
+        # Workaround credits to pgrimaud.
+        # See: https://github.com/pgrimaud/instagram-user-feed/commit/96ad4cf54d1ad331b337f325c73e664999a6d066
+        enc_password = '#PWD_INSTAGRAM_BROWSER:0:{}:{}'.format(int(datetime.now().timestamp()), passwd)
        login = session.post('https://www.instagram.com/accounts/login/ajax/',
-                             data={'password': passwd, 'username': user}, allow_redirects=True)
+                             data={'enc_password': enc_password, 'username': user}, allow_redirects=True)
        try:
            resp_json = login.json()
        except json.decoder.JSONDecodeError:
@@ -239,6 +242,12 @@ class InstaloaderContext:
                                                                                               resp_json['message']))
            else:
                raise ConnectionException("Login error: \"{}\" status.".format(resp_json['status']))
+        if 'authenticated' not in resp_json:
+            # Issue #472
+            if 'message' in resp_json:
+                raise ConnectionException("Login error: Unexpected response, \"{}\".".format(resp_json['message']))
+            else:
+                raise ConnectionException("Login error: Unexpected response, this might indicate a blocked IP.")
        if not resp_json['authenticated']:
            if resp_json['user']:
                # '{"authenticated": false, "user": true, "status": "ok"}'
@@ -285,77 +294,6 @@ class InstaloaderContext:
        if self.sleep:
            time.sleep(min(random.expovariate(0.7), 5.0))

-    def _dump_query_timestamps(self, current_time: float):
-        """Output the number of GraphQL queries grouped by their query_hash within the last time."""
-        windows = [10, 11, 15, 20, 30, 60]
-        print("GraphQL requests:", file=sys.stderr)
-        for query_hash, times in self._graphql_query_timestamps.items():
-            print("  {}".format(query_hash), file=sys.stderr)
-            for window in windows:
-                reqs_in_sliding_window = sum(t > current_time - window * 60 for t in times)
-                print("    last {} minutes: {} requests".format(window, reqs_in_sliding_window), file=sys.stderr)
-
-    def _graphql_request_count_per_sliding_window(self, query_hash: str) -> int:
-        """Return how many GraphQL requests can be done within the sliding window."""
-        if self.is_logged_in:
-            max_reqs = {'1cb6ec562846122743b61e492c85999f': 20,
-                        '33ba35852cb50da46f5b5e889df7d159': 20,
-                        'iphone': 100,
-                        'other': 100}
-        else:
-            max_reqs = {'1cb6ec562846122743b61e492c85999f': 200,
-                        '33ba35852cb50da46f5b5e889df7d159': 200,
-                        'other': 200}
-        return max_reqs.get(query_hash) or min(max_reqs.values())
-
-    def _graphql_query_waittime(self, query_hash: str, current_time: float, untracked_queries: bool = False) -> float:
-        """Calculate time needed to wait before GraphQL query can be executed."""
-        sliding_window = 660
-        if query_hash not in self._graphql_query_timestamps:
-            self._graphql_query_timestamps[query_hash] = []
-        self._graphql_query_timestamps[query_hash] = list(filter(lambda t: t > current_time - 60 * 60,
-                                                                 self._graphql_query_timestamps[query_hash]))
-        reqs_in_sliding_window = list(filter(lambda t: t > current_time - sliding_window,
-                                             self._graphql_query_timestamps[query_hash]))
-        count_per_sliding_window = self._graphql_request_count_per_sliding_window(query_hash)
-        if len(reqs_in_sliding_window) < count_per_sliding_window and not untracked_queries:
-            return max(0, self._graphql_earliest_next_request_time - current_time)
-        next_request_time = min(reqs_in_sliding_window) + sliding_window + 6
-        if untracked_queries:
-            self._graphql_earliest_next_request_time = next_request_time
-        return round(max(next_request_time, self._graphql_earliest_next_request_time) - current_time)
-
-    def _ratecontrol_graphql_query(self, query_hash: str, untracked_queries: bool = False):
-        """Called before a GraphQL query is made in order to stay within Instagram's rate limits.
-
-        :param query_hash: The query_hash parameter of the query.
-        :param untracked_queries: True, if 429 has been returned to apply 429 logic.
-        """
-        if not untracked_queries:
-            waittime = self._graphql_query_waittime(query_hash, time.monotonic(), untracked_queries)
-            assert waittime >= 0
-            if waittime > 10:
-                self.log('\nToo many queries in the last time. Need to wait {} seconds, until {:%H:%M}.'
-                         .format(waittime, datetime.now() + timedelta(seconds=waittime)))
-            time.sleep(waittime)
-            if query_hash not in self._graphql_query_timestamps:
-                self._graphql_query_timestamps[query_hash] = [time.monotonic()]
-            else:
-                self._graphql_query_timestamps[query_hash].append(time.monotonic())
-        else:
-            text_for_429 = ("HTTP error code 429 was returned because too many queries occured in the last time. "
-                            "Please do not use Instagram in your browser or run multiple instances of Instaloader "
-                            "in parallel.")
-            print(textwrap.fill(text_for_429), file=sys.stderr)
-            current_time = time.monotonic()
-            waittime = self._graphql_query_waittime(query_hash, current_time, untracked_queries)
-            assert waittime >= 0
-            if waittime > 10:
-                self.log('The request will be retried in {} seconds, at {:%H:%M}.'
-                         .format(waittime, datetime.now() + timedelta(seconds=waittime)))
-            self._dump_query_timestamps(current_time)
-            time.sleep(waittime)
-
    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.
@@ -376,11 +314,11 @@ class InstaloaderContext:
        try:
            self.do_sleep()
            if is_graphql_query:
-                self._ratecontrol_graphql_query(params['query_hash'])
+                self._rate_controller.wait_before_query(params['query_hash'])
            if is_iphone_query:
-                self._ratecontrol_graphql_query('iphone')
+                self._rate_controller.wait_before_query('iphone')
            if is_other_query:
-                self._ratecontrol_graphql_query('other')
+                self._rate_controller.wait_before_query('other')
            resp = sess.get('https://{0}/{1}'.format(host, path), params=params, allow_redirects=False)
            while resp.is_redirect:
                redirect_url = resp.headers['location']
@@ -436,12 +374,13 @@ class InstaloaderContext:
                    raise ConnectionException(error_string) from err
            self.error(error_string + " [retrying; skip with ^C]", repeat_at_end=False)
            try:
-                if is_graphql_query and isinstance(err, TooManyRequestsException):
-                    self._ratecontrol_graphql_query(params['query_hash'], untracked_queries=True)
-                if is_iphone_query and isinstance(err, TooManyRequestsException):
-                    self._ratecontrol_graphql_query('iphone', untracked_queries=True)
-                if is_other_query and isinstance(err, TooManyRequestsException):
-                    self._ratecontrol_graphql_query('other', untracked_queries=True)
+                if isinstance(err, TooManyRequestsException):
+                    if is_graphql_query:
+                        self._rate_controller.handle_429(params['query_hash'])
+                    if is_iphone_query:
+                        self._rate_controller.handle_429('iphone')
+                    if is_other_query:
+                        self._rate_controller.handle_429('other')
                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)
@@ -489,7 +428,12 @@ class InstaloaderContext:
                          edge_extractor: Callable[[Dict[str, Any]], Dict[str, Any]],
                          rhx_gis: Optional[str] = None,
                          first_data: Optional[Dict[str, Any]] = None) -> Iterator[Dict[str, Any]]:
-        """Retrieve a list of GraphQL nodes."""
+        """
+        Retrieve a list of GraphQL nodes.
+
+        ..deprecated:: 4.5
+           Use :class:`NodeIterator` instead, which provides more functionality.
+        """

        def _query():
            query_variables['first'] = self._graphql_page_length
@@ -527,8 +471,8 @@ class InstaloaderContext:

        .. versionadded:: 4.2.1"""
        with copy_session(self._session, self.request_timeout) as tempsession:
-            tempsession.headers['User-Agent'] = 'Instagram 10.3.2 (iPhone7,2; iPhone OS 9_3_3; en_US; en-US; ' \
-                                                'scale=2.00; 750x1334) AppleWebKit/420+'
+            tempsession.headers['User-Agent'] = 'Instagram 123.1.0.26.115 (iPhone12,1; iOS 13_3; en_US; en-US; ' \
+                                                'scale=2.00; 1656x3584; 190542906)'
            for header in ['Host', 'Origin', 'X-Instagram-AJAX', 'X-Requested-With']:
                tempsession.headers.pop(header, None)
            return self.get_json(path, params, 'i.instagram.com', tempsession)
@@ -538,11 +482,12 @@ class InstaloaderContext:

        .. versionadded:: 4.2.1"""
        self.log(filename, end=' ', flush=True)
-        with open(filename, 'wb') as file:
+        with open(filename + '.temp', 'wb') as file:
            if isinstance(resp, requests.Response):
                shutil.copyfileobj(resp.raw, file)
            else:
                file.write(resp)
+        os.rename(filename + '.temp', filename)

    def get_raw(self, url: str, _attempt=1) -> requests.Response:
        """Downloads a file anonymously.
@@ -584,3 +529,97 @@ class InstaloaderContext:
        if self._root_rhx_gis is None:
            self._root_rhx_gis = self.get_json('', {}).get('rhx_gis', '')
        return self._root_rhx_gis or None
+
+
+class RateController:
+    """
+    Class providing request tracking and rate controlling to stay within rate limits.
+
+    It can be overridden to change Instaloader's behavior regarding rate limits, for example to raise a custom
+    exception when the rate limit is hit::
+
+       import instaloader
+
+       class MyRateController(instaloader.RateController):
+           def sleep(self, secs):
+               raise MyCustomException()
+
+       L = instaloader.Instaloader(rate_controller=lambda ctx: MyRateController(ctx))
+    """
+
+    def __init__(self, context: InstaloaderContext):
+        self._context = context
+        self._graphql_query_timestamps = dict()  # type: Dict[str, List[float]]
+        self._graphql_earliest_next_request_time = 0.0
+
+    def sleep(self, secs: float):
+        """Wait given number of seconds."""
+        # Not static, to allow for the behavior of this method to depend on context-inherent properties, such as
+        # whether we are logged in.
+        # pylint:disable=no-self-use
+        time.sleep(secs)
+
+    def _dump_query_timestamps(self, current_time: float):
+        windows = [10, 11, 15, 20, 30, 60]
+        print("GraphQL requests:", file=sys.stderr)
+        for query_hash, times in self._graphql_query_timestamps.items():
+            print("  {}".format(query_hash), file=sys.stderr)
+            for window in windows:
+                reqs_in_sliding_window = sum(t > current_time - window * 60 for t in times)
+                print("    last {} minutes: {} requests".format(window, reqs_in_sliding_window), file=sys.stderr)
+
+    def count_per_sliding_window(self, query_type: str) -> int:
+        """Return how many GraphQL requests can be done within the sliding window."""
+        # Not static, to allow for the count_per_sliding_window to depend on context-inherent properties, such as
+        # whether we are logged in.
+        # pylint:disable=no-self-use,unused-argument
+        return 200
+
+    def query_waittime(self, query_type: str, current_time: float, untracked_queries: bool = False) -> float:
+        """Calculate time needed to wait before GraphQL query can be executed."""
+        sliding_window = 660
+        if query_type not in self._graphql_query_timestamps:
+            self._graphql_query_timestamps[query_type] = []
+        self._graphql_query_timestamps[query_type] = list(filter(lambda t: t > current_time - 60 * 60,
+                                                                 self._graphql_query_timestamps[query_type]))
+        reqs_in_sliding_window = list(filter(lambda t: t > current_time - sliding_window,
+                                             self._graphql_query_timestamps[query_type]))
+        count_per_sliding_window = self.count_per_sliding_window(query_type)
+        if len(reqs_in_sliding_window) < count_per_sliding_window and not untracked_queries:
+            return max(0.0, self._graphql_earliest_next_request_time - current_time)
+        next_request_time = min(reqs_in_sliding_window) + sliding_window + 6
+        if untracked_queries:
+            self._graphql_earliest_next_request_time = next_request_time
+        return max(next_request_time, self._graphql_earliest_next_request_time) - current_time
+
+    def wait_before_query(self, query_type: str) -> None:
+        """This method is called before a query to Instagram. It calls :meth:`RateController.sleep` to wait
+        until the request can be made."""
+        waittime = self.query_waittime(query_type, time.monotonic(), False)
+        assert waittime >= 0
+        if waittime > 15:
+            self._context.log("\nToo many queries in the last time. Need to wait {} seconds, until {:%H:%M}."
+                              .format(round(waittime), datetime.now() + timedelta(seconds=waittime)))
+        if waittime > 0:
+            self.sleep(waittime)
+        if query_type not in self._graphql_query_timestamps:
+            self._graphql_query_timestamps[query_type] = [time.monotonic()]
+        else:
+            self._graphql_query_timestamps[query_type].append(time.monotonic())
+
+    def handle_429(self, query_type: str) -> None:
+        """This method is called to handle a 429 Too Many Requests response. It calls :meth:`RateController.sleep` to
+         wait until we can repeat the same request."""
+        text_for_429 = ("HTTP error code 429 was returned because too many queries occurred in the last time. "
+                        "Please do not use Instagram in your browser or run multiple instances of Instaloader "
+                        "in parallel.")
+        print(textwrap.fill(text_for_429), file=sys.stderr)
+        current_time = time.monotonic()
+        waittime = self.query_waittime(query_type, current_time, True)
+        assert waittime >= 0
+        if waittime > 15:
+            self._context.log("The request will be retried in {} seconds, at {:%H:%M}."
+                              .format(round(waittime), datetime.now() + timedelta(seconds=waittime)))
+        self._dump_query_timestamps(current_time)
+        if waittime > 0:
+            self.sleep(waittime)
@@ -0,0 +1,261 @@
+import base64
+import hashlib
+import json
+import os
+from contextlib import contextmanager
+from datetime import datetime, timedelta
+from lzma import LZMAError
+from typing import Any, Callable, Dict, Iterator, NamedTuple, Optional, Tuple, TypeVar
+
+from .exceptions import InvalidArgumentException, QueryReturnedBadRequestException
+from .instaloadercontext import InstaloaderContext
+
+FrozenNodeIterator = NamedTuple('FrozenNodeIterator',
+                                [('query_hash', str),
+                                 ('query_variables', Dict),
+                                 ('query_referer', Optional[str]),
+                                 ('context_username', Optional[str]),
+                                 ('total_index', int),
+                                 ('best_before', Optional[float]),
+                                 ('remaining_data', Optional[Dict])])
+FrozenNodeIterator.__doc__ = \
+    """A serializable representation of a :class:`NodeIterator` instance, saving its iteration state."""
+FrozenNodeIterator.query_hash.__doc__ = """The GraphQL ``query_hash`` parameter."""
+FrozenNodeIterator.query_variables.__doc__ = """The GraphQL ``query_variables`` parameter."""
+FrozenNodeIterator.query_referer.__doc__ = """The HTTP referer used for the GraphQL query."""
+FrozenNodeIterator.context_username.__doc__ = """The username who created the iterator, or ``None``."""
+FrozenNodeIterator.total_index.__doc__ = """Number of items that have already been returned."""
+FrozenNodeIterator.best_before.__doc__ = """Date when parts of the stored nodes might have expired."""
+FrozenNodeIterator.remaining_data.__doc__ = \
+    """The already-retrieved, yet-unprocessed ``edges`` and the ``page_info`` at time of freezing."""
+
+
+T = TypeVar('T')
+
+
+class NodeIterator(Iterator[T]):
+    """
+    Iterate the nodes within edges in a GraphQL pagination. Instances of this class are returned by many (but not all)
+    of Instaloader's :class:`Post`-returning functions (such as :meth:`Profile.get_posts` etc.).
+
+    What makes this iterator special is its ability to freeze/store its current state, e.g. to interrupt an iteration,
+    and later thaw/resume from where it left off.
+
+    You can freeze a NodeIterator with :meth:`NodeIterator.freeze`::
+
+       post_iterator = profile.get_posts()
+       try:
+           for post in post_iterator:
+               do_something_with(post)
+       except KeyboardInterrupt:
+           save("resume_information.json", post_iterator.freeze())
+
+    and later reuse it with :meth:`NodeIterator.thaw` on an equally-constructed NodeIterator::
+
+       post_iterator = profile.get_posts()
+       post_iterator.thaw(load("resume_information.json"))
+
+    A :class:`FrozenNodeIterator` can only be thawn with a matching NodeIterator, i.e. a NodeIterator instance that has
+    been constructed with the same parameters as the instance that is represented by the :class:`FrozenNodeIterator` in
+    question. This is to ensure that an iteration cannot be resumed in a wrong, unmatching loop. As a quick way to
+    distinguish iterators that are saved e.g. in files, there is the :attr:`NodeIterator.magic` string: Two
+    NodeIterators are matching if they have the same magic.
+
+    See also :func:`resumable_iteration` for a high-level context manager that handles a resumable iteration.
+    """
+
+    _graphql_page_length = 50
+    shelf_life = timedelta(days=29)
+
+    def __init__(self,
+                 context: InstaloaderContext,
+                 query_hash: str,
+                 edge_extractor: Callable[[Dict[str, Any]], Dict[str, Any]],
+                 node_wrapper: Callable[[Dict], T],
+                 query_variables: Optional[Dict[str, Any]] = None,
+                 query_referer: Optional[str] = None,
+                 first_data: Optional[Dict[str, Any]] = None):
+        self._context = context
+        self._query_hash = query_hash
+        self._edge_extractor = edge_extractor
+        self._node_wrapper = node_wrapper
+        self._query_variables = query_variables if query_variables is not None else {}
+        self._query_referer = query_referer
+        self._data = first_data
+        self._page_index = 0
+        self._total_index = 0
+        self._best_before = (None if first_data is None else
+                             datetime.now() + NodeIterator.shelf_life)
+
+    def _query(self, after: Optional[str] = None) -> Dict:
+        pagination_variables = {'first': NodeIterator._graphql_page_length}  # type: Dict[str, Any]
+        if after is not None:
+            pagination_variables['after'] = after
+        try:
+            data = self._edge_extractor(
+                self._context.graphql_query(
+                    self._query_hash, {**self._query_variables, **pagination_variables}, self._query_referer
+                )
+            )
+            self._best_before = datetime.now() + NodeIterator.shelf_life
+            return data
+        except QueryReturnedBadRequestException:
+            new_page_length = int(NodeIterator._graphql_page_length / 2)
+            if new_page_length >= 12:
+                NodeIterator._graphql_page_length = new_page_length
+                self._context.error("HTTP Error 400 (Bad Request) on GraphQL Query. Retrying with shorter page length.",
+                                    repeat_at_end=False)
+                return self._query(after)
+            else:
+                raise
+
+    def __iter__(self):
+        return self
+
+    def __next__(self):
+        if self._data is None:
+            self._data = self._query()
+        if self._page_index < len(self._data['edges']):
+            node = self._data['edges'][self._page_index]['node']
+            page_index, total_index = self._page_index, self._total_index
+            try:
+                self._page_index += 1
+                self._total_index += 1
+            except KeyboardInterrupt:
+                self._page_index, self._total_index = page_index, total_index
+                raise
+            return self._node_wrapper(node)
+        if self._data['page_info']['has_next_page']:
+            query_response = self._query(self._data['page_info']['end_cursor'])
+            page_index, data = self._page_index, self._data
+            try:
+                self._page_index = 0
+                self._data = query_response
+            except KeyboardInterrupt:
+                self._page_index, self._data = page_index, data
+                raise
+            return self.__next__()
+        raise StopIteration()
+
+    @property
+    def count(self) -> Optional[int]:
+        """The ``count`` as returned by Instagram. This is not always the total count this iterator will yield."""
+        return self._data.get('count') if self._data is not None else None
+
+    @property
+    def total_index(self) -> int:
+        """Number of items that have already been returned."""
+        return self._total_index
+
+    @property
+    def magic(self) -> str:
+        """Magic string for easily identifying a matching iterator file for resuming (hash of some parameters)."""
+        if 'blake2b' not in hashlib.algorithms_available:
+            magic_hash = hashlib.new('sha224')
+        else:
+            # Use blake2b when possible, i.e. on Python >= 3.6.
+            magic_hash = hashlib.blake2b(digest_size=6)  # type:ignore  # pylint: disable=no-member
+        magic_hash.update(json.dumps(
+            [self._query_hash, self._query_variables, self._query_referer, self._context.username]
+        ).encode())
+        return base64.urlsafe_b64encode(magic_hash.digest()).decode()
+
+    def freeze(self) -> FrozenNodeIterator:
+        """Freeze the iterator for later resuming."""
+        remaining_data = None
+        if self._data is not None:
+            remaining_data = {**self._data,
+                              'edges': (self._data['edges'][(max(self._page_index - 1, 0)):])}
+        return FrozenNodeIterator(
+            query_hash=self._query_hash,
+            query_variables=self._query_variables,
+            query_referer=self._query_referer,
+            context_username=self._context.username,
+            total_index=max(self.total_index - 1, 0),
+            best_before=self._best_before.timestamp() if self._best_before else None,
+            remaining_data=remaining_data,
+        )
+
+    def thaw(self, frozen: FrozenNodeIterator) -> None:
+        """Use this iterator for resuming from earlier iteration."""
+        if self._total_index or self._page_index:
+            raise InvalidArgumentException("thaw() called on already-used iterator.")
+        if (self._query_hash != frozen.query_hash or
+                self._query_variables != frozen.query_variables or
+                self._query_referer != frozen.query_referer or
+                self._context.username != frozen.context_username):
+            raise InvalidArgumentException("Mismatching resume information.")
+        self._total_index = frozen.total_index
+        self._best_before = datetime.fromtimestamp(frozen.best_before) if frozen.best_before else None
+        self._data = frozen.remaining_data
+
+
+@contextmanager
+def resumable_iteration(context: InstaloaderContext,
+                        iterator: Iterator,
+                        load: Callable[[InstaloaderContext, str], Any],
+                        save: Callable[[FrozenNodeIterator, str], None],
+                        format_path: Callable[[str], str],
+                        check_bbd: bool = True,
+                        enabled: bool = True) -> Iterator[Tuple[bool, int]]:
+    """
+    High-level context manager to handle a resumable iteration that can be interrupted with a KeyboardInterrupt.
+
+    It can be used as follows to automatically load a previously-saved state into the iterator, save the iterator's
+    state when interrupted, and delete the resume file upon completion::
+
+       post_iterator = profile.get_posts()
+       with resumable_iteration(
+               context=L.context,
+               iterator=post_iterator,
+               load=lambda _, path: FrozenNodeIterator(**json.load(open(path))),
+               save=lambda fni, path: json.dump(fni._asdict(), open(path, 'w')),
+               format_path=lambda magic: "resume_info_{}.json".format(magic)
+       ) as resume_info:
+           is_resuming, start_index = resume_info
+           for post in post_iterator:
+               do_something_with(post)
+
+    It yields a tuple (is_resuming, start_index).
+
+    When the passed iterator is not a :class:`NodeIterator`, it behaves as if ``resumable_iteration`` was not used,
+    just executing the inner body.
+
+    :param context: The :class:`InstaloaderContext`.
+    :param iterator: The fresh :class:`NodeIterator`.
+    :param load: Loads a FrozenNodeIterator from given path. The object is ignored if it has a different type.
+    :param save: Saves the given FrozenNodeIterator to the given path.
+    :param format_path: Returns the path to the resume file for the given magic.
+    :param check_bbd: Whether to check the best before date and reject an expired FrozenNodeIterator.
+    :param enabled: Set to False to disable all functionality and simply execute the inner body.
+    """
+    if not enabled or not isinstance(iterator, NodeIterator):
+        yield False, 0
+        return
+    is_resuming = False
+    start_index = 0
+    resume_file_path = format_path(iterator.magic)
+    resume_file_exists = os.path.isfile(resume_file_path)
+    if resume_file_exists:
+        try:
+            fni = load(context, resume_file_path)
+            if not isinstance(fni, FrozenNodeIterator):
+                raise InvalidArgumentException("Invalid type.")
+            if check_bbd and fni.best_before and datetime.fromtimestamp(fni.best_before) < datetime.now():
+                raise InvalidArgumentException("\"Best before\" date exceeded.")
+            iterator.thaw(fni)
+            is_resuming = True
+            start_index = iterator.total_index
+            context.log("Resuming from {}.".format(resume_file_path))
+        except (InvalidArgumentException, LZMAError, json.decoder.JSONDecodeError) as exc:
+            context.error("Warning: Not resuming from {}: {}".format(resume_file_path, exc))
+    try:
+        yield is_resuming, start_index
+    except KeyboardInterrupt:
+        os.makedirs(os.path.dirname(resume_file_path), exist_ok=True)
+        save(iterator.freeze(), resume_file_path)
+        context.log("\nSaved resume information to {}.".format(resume_file_path))
+        raise
+    if resume_file_exists:
+        os.unlink(resume_file_path)
+        context.log("Iteration complete, deleted resume information file {}.".format(resume_file_path))
@@ -9,7 +9,7 @@ from typing import Any, Dict, Iterator, List, Optional, Union
 from . import __version__
 from .exceptions import *
 from .instaloadercontext import InstaloaderContext
-
+from .nodeiterator import FrozenNodeIterator, NodeIterator

 PostSidecarNode = namedtuple('PostSidecarNode', ['is_video', 'display_url', 'video_url'])
 PostSidecarNode.__doc__ = "Item of a Sidecar Post."
@@ -69,8 +69,11 @@ class Post:
        self._node = node
        self._owner_profile = owner_profile
        self._full_metadata_dict = None  # type: Optional[Dict[str, Any]]
-        self._rhx_gis_str = None         # type: Optional[str]
        self._location = None            # type: Optional[PostLocation]
+        self._iphone_struct_ = None
+        if 'iphone_struct' in node:
+            # if loaded from JSON with load_structure_from_file()
+            self._iphone_struct_ = node['iphone_struct']

    @classmethod
    def from_shortcode(cls, context: InstaloaderContext, shortcode: str):
@@ -98,6 +101,11 @@ class Post:
            raise InvalidArgumentException("Wrong mediaid {0}, unable to convert to shortcode".format(str(mediaid)))
        return b64encode(mediaid.to_bytes(9, 'big'), b'-_').decode().replace('A', ' ').lstrip().replace(' ', 'A')

+    @staticmethod
+    def supported_graphql_types() -> List[str]:
+        """The values of __typename fields that the :class:`Post` class can handle."""
+        return ["GraphImage", "GraphVideo", "GraphSidecar"]
+
    def _asdict(self):
        node = self._node
        if self._full_metadata_dict:
@@ -106,6 +114,8 @@ class Post:
            node['owner'] = self.owner_profile._asdict()
        if self._location:
            node['location'] = self._location._asdict()
+        if self._iphone_struct_:
+            node['iphone_struct'] = self._iphone_struct_
        return node

    @property
@@ -131,9 +141,17 @@ class Post:

    def _obtain_metadata(self):
        if not self._full_metadata_dict:
-            pic_json = self._context.get_json("p/{0}/".format(self.shortcode), params={})
-            self._full_metadata_dict = pic_json['entry_data']['PostPage'][0]['graphql']['shortcode_media']
-            self._rhx_gis_str = pic_json.get('rhx_gis')
+            pic_json = self._context.graphql_query(
+                '2b0673e0dc4580674a88d426fe00ea90',
+                {'shortcode': self.shortcode}
+            )
+            self._full_metadata_dict = pic_json['data']['shortcode_media']
+            if self._full_metadata_dict is None:
+                # issue #449
+                self._context.error("Fetching Post metadata failed (issue #449). "
+                                    "The following data has been returned:\n"
+                                    + json.dumps(pic_json['entry_data'], indent=2))
+                raise BadResponseException("Fetching Post metadata failed.")
            if self.shortcode != self._full_metadata_dict['shortcode']:
                self._node.update(self._full_metadata_dict)
                raise PostChangedException
@@ -145,9 +163,13 @@ class Post:
        return self._full_metadata_dict

    @property
-    def _rhx_gis(self) -> Optional[str]:
-        self._obtain_metadata()
-        return self._rhx_gis_str
+    def _iphone_struct(self) -> Dict[str, Any]:
+        if not self._context.is_logged_in:
+            raise LoginRequiredException("--login required to access iPhone media info endpoint.")
+        if not self._iphone_struct_:
+            data = self._context.get_iphone_json(path='api/v1/media/{}/info/'.format(self.mediaid), params={})
+            self._iphone_struct_ = data['items'][0]
+        return self._iphone_struct_

    def _field(self, *keys) -> Any:
        """Lookups given fields in _node, and if not found in _full_metadata. Raises KeyError if not found anywhere."""
@@ -214,6 +236,13 @@ class Post:
    @property
    def url(self) -> str:
        """URL of the picture / video thumbnail of the post"""
+        if self.typename == "GraphImage" and self._context.is_logged_in:
+            try:
+                orig_url = self._iphone_struct['image_versions2']['candidates'][0]['url']
+                url = re.sub(r'&se=\d+(&?)', r'\1', orig_url)
+                return url
+            except (InstaloaderException, KeyError, IndexError) as err:
+                self._context.error('{} Unable to fetch high quality image version of {}.'.format(err, self))
        return self._node["display_url"] if "display_url" in self._node else self._node["display_src"]

    @property
@@ -224,10 +253,22 @@ class Post:
    def get_sidecar_nodes(self) -> Iterator[PostSidecarNode]:
        """Sidecar nodes of a Post with typename==GraphSidecar."""
        if self.typename == 'GraphSidecar':
-            for edge in self._field('edge_sidecar_to_children', 'edges'):
+            edges = self._field('edge_sidecar_to_children', 'edges')
+            if any(edge['node']['is_video'] for edge in edges):
+                # video_url is only present in full metadata, issue #558.
+                edges = self._full_metadata['edge_sidecar_to_children']['edges']
+            for idx, edge in enumerate(edges):
                node = edge['node']
                is_video = node['is_video']
-                yield PostSidecarNode(is_video=is_video, display_url=node['display_url'],
+                display_url = node['display_url']
+                if not is_video and self._context.is_logged_in:
+                    try:
+                        carousel_media = self._iphone_struct['carousel_media']
+                        orig_url = carousel_media[idx]['image_versions2']['candidates'][0]['url']
+                        display_url = re.sub(r'&se=\d+(&?)', r'\1', orig_url)
+                    except (InstaloaderException, KeyError, IndexError) as err:
+                        self._context.error('{} Unable to fetch high quality image version of {}.'.format(err, self))
+                yield PostSidecarNode(is_video=is_video, display_url=display_url,
                                      video_url=node['video_url'] if is_video else None)

    @property
@@ -347,7 +388,7 @@ class Post:
                                     created_at_utc=datetime.utcfromtimestamp(node['created_at']),
                                     text=node['text'],
                                     owner=Profile(self._context, node['owner']),
-                                     likes_count=node['edge_liked_by']['count'])
+                                     likes_count=node.get('edge_liked_by', {}).get('count', 0))

        def _postcommentanswers(node):
            if 'edge_threaded_comments' not in node:
@@ -361,11 +402,14 @@ class Post:
                # If the answer's metadata already contains all comments, don't do GraphQL requests to obtain them
                yield from (_postcommentanswer(comment['node']) for comment in answer_edges)
                return
-            yield from (_postcommentanswer(answer_node) for answer_node in
-                        self._context.graphql_node_list("51fdd02b67508306ad4484ff574a0b62",
-                                                        {'comment_id': node['id']},
-                                                        'https://www.instagram.com/p/' + self.shortcode + '/',
-                                                        lambda d: d['data']['comment']['edge_threaded_comments']))
+            yield from NodeIterator(
+                self._context,
+                '51fdd02b67508306ad4484ff574a0b62',
+                lambda d: d['data']['comment']['edge_threaded_comments'],
+                _postcommentanswer,
+                {'comment_id': node['id']},
+                'https://www.instagram.com/p/{0}/'.format(self.shortcode),
+            )

        def _postcomment(node):
            return PostComment(*_postcommentanswer(node),
@@ -373,29 +417,22 @@ class Post:
        if self.comments == 0:
            # Avoid doing additional requests if there are no comments
            return
-        try:
-            comment_edges = self._field('edge_media_to_parent_comment', 'edges')
-            answers_count = sum([edge['node']['edge_threaded_comments']['count'] for edge in comment_edges])
-            threaded_comments_available = True
-        except KeyError:
-            comment_edges = self._field('edge_media_to_comment', 'edges')
-            answers_count = 0
-            threaded_comments_available = False
+
+        comment_edges = self._field('edge_media_to_comment', 'edges')
+        answers_count = sum([edge['node'].get('edge_threaded_comments', {}).get('count', 0) for edge in comment_edges])

        if self.comments == len(comment_edges) + answers_count:
            # If the Post's metadata already contains all parent comments, don't do GraphQL requests to obtain them
            yield from (_postcomment(comment['node']) for comment in comment_edges)
            return
-        yield from (_postcomment(node) for node in
-                    self._context.graphql_node_list(
-                        "97b41c52301f77ce508f55e66d17620e" if threaded_comments_available
-                        else "f0986789a5c5d17c2400faebf16efd0d",
-                        {'shortcode': self.shortcode},
-                        'https://www.instagram.com/p/' + self.shortcode + '/',
-                        lambda d:
-                        d['data']['shortcode_media'][
-                            'edge_media_to_parent_comment' if threaded_comments_available else 'edge_media_to_comment'],
-                        self._rhx_gis))
+        yield from NodeIterator(
+            self._context,
+            '97b41c52301f77ce508f55e66d17620e',
+            lambda d: d['data']['shortcode_media']['edge_media_to_parent_comment'],
+            _postcomment,
+            {'shortcode': self.shortcode},
+            'https://www.instagram.com/p/{0}/'.format(self.shortcode),
+        )

    def get_likes(self) -> Iterator['Profile']:
        """Iterate over all likes of the post. A :class:`Profile` instance of each likee is yielded."""
@@ -407,11 +444,38 @@ class Post:
            # If the Post's metadata already contains all likes, don't do GraphQL requests to obtain them
            yield from (Profile(self._context, like['node']) for like in likes_edges)
            return
-        yield from (Profile(self._context, node) for node in
-                    self._context.graphql_node_list("1cb6ec562846122743b61e492c85999f", {'shortcode': self.shortcode},
-                                                    'https://www.instagram.com/p/' + self.shortcode + '/',
-                                                    lambda d: d['data']['shortcode_media']['edge_liked_by'],
-                                                    self._rhx_gis))
+        yield from NodeIterator(
+            self._context,
+            '1cb6ec562846122743b61e492c85999f',
+            lambda d: d['data']['shortcode_media']['edge_liked_by'],
+            lambda n: Profile(self._context, n),
+            {'shortcode': self.shortcode},
+            'https://www.instagram.com/p/{0}/'.format(self.shortcode),
+        )
+
+    @property
+    def is_sponsored(self) -> bool:
+        """
+        Whether Post is a sponsored post, equivalent to non-empty :meth:`Post.sponsor_users`.
+
+        .. versionadded:: 4.4
+        """
+        try:
+            sponsor_edges = self._field('edge_media_to_sponsor_user', 'edges')
+        except KeyError:
+            return False
+        return bool(sponsor_edges)
+
+    @property
+    def sponsor_users(self) -> List['Profile']:
+        """
+        The Post's sponsors.
+
+        .. versionadded:: 4.4
+        """
+        return ([] if not self.is_sponsored else
+                [Profile(self._context, edge['node']['sponsor']) for edge in
+                 self._field('edge_media_to_sponsor_user', 'edges')])

    @property
    def location(self) -> Optional[PostLocation]:
@@ -468,7 +532,6 @@ class Profile:
        self._has_public_story = None  # type: Optional[bool]
        self._node = node
        self._has_full_metadata = False
-        self._rhx_gis = None
        self._iphone_struct_ = None
        if 'iphone_struct' in node:
            # if loaded from JSON with load_structure_from_file()
@@ -518,10 +581,11 @@ class Profile:

    def _asdict(self):
        json_node = self._node.copy()
-        # remove posts
+        # remove posts to avoid "Circular reference detected" exception
        json_node.pop('edge_media_collections', None)
        json_node.pop('edge_owner_to_timeline_media', None)
        json_node.pop('edge_saved_media', None)
+        json_node.pop('edge_felix_video_timeline', None)
        if self._iphone_struct_:
            json_node['iphone_struct'] = self._iphone_struct_
        return json_node
@@ -529,10 +593,9 @@ class Profile:
    def _obtain_metadata(self):
        try:
            if not self._has_full_metadata:
-                metadata = self._context.get_json('{}/'.format(self.username), params={})
+                metadata = self._context.get_json('{}/feed/'.format(self.username), params={})
                self._node = metadata['entry_data']['ProfilePage'][0]['graphql']['user']
                self._has_full_metadata = True
-                self._rhx_gis = metadata.get('rhx_gis')
        except (QueryReturnedNotFoundException, KeyError) as err:
            top_search_results = TopSearchResults(self._context, self.username)
            similar_profiles = [profile.username for profile in top_search_results.get_profiles()]
@@ -598,6 +661,10 @@ class Profile:
    def mediacount(self) -> int:
        return self._metadata('edge_owner_to_timeline_media', 'count')

+    @property
+    def igtvcount(self) -> int:
+        return self._metadata('edge_felix_video_timeline', 'count')
+
    @property
    def followers(self) -> int:
        return self._metadata('edge_followed_by', 'count')
@@ -610,6 +677,16 @@ class Profile:
    def external_url(self) -> Optional[str]:
        return self._metadata('external_url')

+    @property
+    def is_business_account(self) -> bool:
+        """.. versionadded:: 4.4"""
+        return self._metadata('is_business_account')
+
+    @property
+    def business_category_name(self) -> str:
+        """.. versionadded:: 4.4"""
+        return self._metadata('business_category_name')
+
    @property
    def biography(self) -> str:
        return self._metadata('biography')
@@ -651,8 +728,7 @@ class Profile:
                                                        'include_reel': False, 'include_suggested_users': False,
                                                        'include_logged_out_extras': True,
                                                        'include_highlight_reels': False},
-                                                       'https://www.instagram.com/{}/'.format(self.username),
-                                                       self._rhx_gis)
+                                                       'https://www.instagram.com/{}/'.format(self.username))
            self._has_public_story = data['data']['user']['has_public_story']
        assert self._has_public_story is not None
        return self._has_public_story
@@ -703,73 +779,126 @@ class Profile:
 	   Use :attr:`profile_pic_url`."""
        return self.profile_pic_url

-    def get_posts(self) -> Iterator[Post]:
-        """Retrieve all posts from a profile."""
-        self._obtain_metadata()
-        yield from (Post(self._context, node, self) for node in
-                    self._context.graphql_node_list("472f257a40c653c64c666ce877d59d2b",
-                                                    {'id': self.userid},
-                                                    'https://www.instagram.com/{0}/'.format(self.username),
-                                                    lambda d: d['data']['user']['edge_owner_to_timeline_media'],
-                                                    self._rhx_gis,
-                                                    self._metadata('edge_owner_to_timeline_media')))
+    def get_posts(self) -> NodeIterator[Post]:
+        """Retrieve all posts from a profile.

-    def get_saved_posts(self) -> Iterator[Post]:
-        """Get Posts that are marked as saved by the user."""
+        :rtype:NodeIterator[Post]"""
+        self._obtain_metadata()
+        return NodeIterator(
+            self._context,
+            '472f257a40c653c64c666ce877d59d2b',
+            lambda d: d['data']['user']['edge_owner_to_timeline_media'],
+            lambda n: Post(self._context, n, self),
+            {'id': self.userid},
+            'https://www.instagram.com/{0}/'.format(self.username),
+            self._metadata('edge_owner_to_timeline_media'),
+        )
+
+    def get_saved_posts(self) -> NodeIterator[Post]:
+        """Get Posts that are marked as saved by the user.
+
+        :rtype:NodeIterator[Post]"""

        if self.username != self._context.username:
            raise LoginRequiredException("--login={} required to get that profile's saved posts.".format(self.username))

        self._obtain_metadata()
-        yield from (Post(self._context, node) for node in
-                    self._context.graphql_node_list("f883d95537fbcd400f466f63d42bd8a1",
-                                                    {'id': self.userid},
-                                                    'https://www.instagram.com/{0}/'.format(self.username),
-                                                    lambda d: d['data']['user']['edge_saved_media'],
-                                                    self._rhx_gis,
-                                                    self._metadata('edge_saved_media')))
+        return NodeIterator(
+            self._context,
+            'f883d95537fbcd400f466f63d42bd8a1',
+            lambda d: d['data']['user']['edge_saved_media'],
+            lambda n: Post(self._context, n),
+            {'id': self.userid},
+            'https://www.instagram.com/{0}/'.format(self.username),
+            self._metadata('edge_saved_media'),
+        )

-    def get_tagged_posts(self) -> Iterator[Post]:
+    def get_tagged_posts(self) -> NodeIterator[Post]:
        """Retrieve all posts where a profile is tagged.

+        :rtype: NodeIterator[Post]
+
        .. versionadded:: 4.0.7"""
        self._obtain_metadata()
-        yield from (Post(self._context, node, self if int(node['owner']['id']) == self.userid else None) for node in
-                    self._context.graphql_node_list("e31a871f7301132ceaab56507a66bbb7",
-                                                    {'id': self.userid},
-                                                    'https://www.instagram.com/{0}/'.format(self.username),
-                                                    lambda d: d['data']['user']['edge_user_to_photos_of_you'],
-                                                    self._rhx_gis))
+        return NodeIterator(
+            self._context,
+            'e31a871f7301132ceaab56507a66bbb7',
+            lambda d: d['data']['user']['edge_user_to_photos_of_you'],
+            lambda n: Post(self._context, n, self if int(n['owner']['id']) == self.userid else None),
+            {'id': self.userid},
+            'https://www.instagram.com/{0}/'.format(self.username),
+        )

-    def get_followers(self) -> Iterator['Profile']:
+    def get_igtv_posts(self) -> NodeIterator[Post]:
+        """Retrieve all IGTV posts.
+
+        :rtype: NodeIterator[Post]
+
+        .. versionadded:: 4.3"""
+        self._obtain_metadata()
+        return NodeIterator(
+            self._context,
+            'bc78b344a68ed16dd5d7f264681c4c76',
+            lambda d: d['data']['user']['edge_felix_video_timeline'],
+            lambda n: Post(self._context, n, self),
+            {'id': self.userid},
+            'https://www.instagram.com/{0}/channel/'.format(self.username),
+            self._metadata('edge_felix_video_timeline'),
+        )
+
+    def get_followers(self) -> NodeIterator['Profile']:
        """
        Retrieve list of followers of given profile.
        To use this, one needs to be logged in and private profiles has to be followed.
+
+        :rtype:NodeIterator[Profile]
        """
        if not self._context.is_logged_in:
            raise LoginRequiredException("--login required to get a profile's followers.")
        self._obtain_metadata()
-        yield from (Profile(self._context, node) for node in
-                    self._context.graphql_node_list("37479f2b8209594dde7facb0d904896a",
-                                                    {'id': str(self.userid)},
-                                                    'https://www.instagram.com/' + self.username + '/',
-                                                    lambda d: d['data']['user']['edge_followed_by'],
-                                                    self._rhx_gis))
+        return NodeIterator(
+            self._context,
+            '37479f2b8209594dde7facb0d904896a',
+            lambda d: d['data']['user']['edge_followed_by'],
+            lambda n: Profile(self._context, n),
+            {'id': str(self.userid)},
+            'https://www.instagram.com/{0}/'.format(self.username),
+        )

-    def get_followees(self) -> Iterator['Profile']:
+    def get_followees(self) -> NodeIterator['Profile']:
        """
        Retrieve list of followees (followings) of given profile.
        To use this, one needs to be logged in and private profiles has to be followed.
+
+        :rtype:NodeIterator[Profile]
        """
        if not self._context.is_logged_in:
            raise LoginRequiredException("--login required to get a profile's followees.")
        self._obtain_metadata()
-        yield from (Profile(self._context, node) for node in
-                    self._context.graphql_node_list("58712303d941c6855d4e888c5f0cd22f",
-                                                    {'id': str(self.userid)},
-                                                    'https://www.instagram.com/' + self.username + '/',
-                                                    lambda d: d['data']['user']['edge_follow'],
-                                                    self._rhx_gis))
+        return NodeIterator(
+            self._context,
+            '58712303d941c6855d4e888c5f0cd22f',
+            lambda d: d['data']['user']['edge_follow'],
+            lambda n: Profile(self._context, n),
+            {'id': str(self.userid)},
+            'https://www.instagram.com/{0}/'.format(self.username),
+        )
+
+    def get_similar_accounts(self) -> Iterator['Profile']:
+        """
+        Retrieve list of suggested / similar accounts for this profile.
+        To use this, one needs to be logged in.
+
+        .. versionadded:: 4.4
+        """
+        if not self._context.is_logged_in:
+            raise LoginRequiredException("--login required to get a profile's similar accounts.")
+        self._obtain_metadata()
+        yield from (Profile(self._context, edge["node"]) for edge in
+                    self._context.graphql_query("ad99dd9d3646cc3c0dda65debcd266a7",
+                                                {"user_id": str(self.userid), "include_chaining": True},
+                                                "https://www.instagram.com/{0}/"
+                                                .format(self.username))["data"]["user"]["edge_chaining"]["edges"])


 class StoryItem:
@@ -1067,6 +1196,169 @@ class Highlight(Story):
        yield from (StoryItem(self._context, item, self.owner_profile) for item in self._items)


+class Hashtag:
+    """
+    An Hashtag.
+
+    Analogous to :class:`Profile`, get an instance with::
+
+       L = Instaloader()
+       hashtag = Hashtag.from_name(L.context, HASHTAG)
+
+    To then download the Hashtag's Posts, do::
+
+       for post in hashtag.get_posts():
+          L.download_post(post, target="#"+hashtag.name)
+
+    Also, this class implements == and is hashable.
+    """
+    def __init__(self, context: InstaloaderContext, node: Dict[str, Any]):
+        assert "name" in node
+        self._context = context
+        self._node = node
+        self._has_full_metadata = False
+
+    @classmethod
+    def from_name(cls, context: InstaloaderContext, name: str):
+        """
+        Create a Hashtag instance from a given hashtag name, without preceeding '#'. Raises an Exception if there is no
+        hashtag with the given name.
+
+        :param context: :attr:`Instaloader.context`
+        :param name: Hashtag, without preceeding '#'
+        :raises: :class:`QueryReturnedNotFoundException`
+        """
+        # pylint:disable=protected-access
+        hashtag = cls(context, {'name': name.lower()})
+        hashtag._obtain_metadata()
+        return hashtag
+
+    @property
+    def name(self):
+        """Hashtag name lowercased, without preceeding '#'"""
+        return self._node["name"].lower()
+
+    def _query(self, params):
+        return self._context.get_json("explore/tags/{0}/".format(self.name),
+                                      params)["graphql"]["hashtag"]
+
+    def _obtain_metadata(self):
+        if not self._has_full_metadata:
+            self._node = self._query({"__a": 1})
+            self._has_full_metadata = True
+
+    def _asdict(self):
+        json_node = self._node.copy()
+        # remove posts
+        json_node.pop("edge_hashtag_to_top_posts", None)
+        json_node.pop("edge_hashtag_to_media", None)
+        return json_node
+
+    def __repr__(self):
+        return "<Hashtag #{}>".format(self.name)
+
+    def __eq__(self, other: object) -> bool:
+        if isinstance(other, Hashtag):
+            return self.name == other.name
+        return NotImplemented
+
+    def __hash__(self) -> int:
+        return hash(self.name)
+
+    def _metadata(self, *keys) -> Any:
+        try:
+            d = self._node
+            for key in keys:
+                d = d[key]
+            return d
+        except KeyError:
+            self._obtain_metadata()
+            d = self._node
+            for key in keys:
+                d = d[key]
+            return d
+
+    @property
+    def hashtagid(self) -> int:
+        return int(self._metadata("id"))
+
+    @property
+    def profile_pic_url(self) -> str:
+        return self._metadata("profile_pic_url")
+
+    @property
+    def description(self) -> str:
+        return self._metadata("description")
+
+    @property
+    def allow_following(self) -> bool:
+        return self._metadata("allow_following")
+
+    @property
+    def is_following(self) -> bool:
+        return self._metadata("is_following")
+
+    @property
+    def is_top_media_only(self) -> bool:
+        return self._metadata("is_top_media_only")
+
+    def get_related_tags(self) -> Iterator["Hashtag"]:
+        """Yields similar hashtags."""
+        yield from (Hashtag(self._context, edge["node"])
+                    for edge in self._metadata("edge_hashtag_to_related_tags", "edges"))
+
+    def get_top_posts(self) -> Iterator[Post]:
+        """Yields the top posts of the hashtag."""
+        yield from (Post(self._context, edge["node"])
+                    for edge in self._metadata("edge_hashtag_to_top_posts", "edges"))
+
+    @property
+    def mediacount(self) -> int:
+        """
+        The count of all media associated with this hashtag.
+
+        The number of posts with a certain hashtag may differ from the number of posts that can actually be accessed, as
+        the hashtag count might include private posts
+        """
+        return self._metadata("edge_hashtag_to_media", "count")
+
+    def get_posts(self) -> Iterator[Post]:
+        """Yields the posts associated with this hashtag."""
+        self._metadata("edge_hashtag_to_media", "edges")
+        self._metadata("edge_hashtag_to_media", "page_info")
+        conn = self._metadata("edge_hashtag_to_media")
+        yield from (Post(self._context, edge["node"]) for edge in conn["edges"])
+        while conn["page_info"]["has_next_page"]:
+            data = self._query({'__a': 1, 'max_id': conn["page_info"]["end_cursor"]})
+            conn = data["edge_hashtag_to_media"]
+            yield from (Post(self._context, edge["node"]) for edge in conn["edges"])
+
+    def get_all_posts(self) -> Iterator[Post]:
+        """Yields all posts, i.e. all most recent posts and the top posts, in chronological order."""
+        sorted_top_posts = iter(sorted(self.get_top_posts(), key=lambda p: p.date_utc, reverse=True))
+        other_posts = self.get_posts()
+        next_top = next(sorted_top_posts, None)
+        next_other = next(other_posts, None)
+        while next_top is not None or next_other is not None:
+            if next_other is None:
+                yield from sorted_top_posts
+                break
+            if next_top is None:
+                yield from other_posts
+                break
+            if next_top == next_other:
+                yield next_top
+                next_top = next(sorted_top_posts, None)
+                next_other = next(other_posts, None)
+                continue
+            if next_top.date_utc > next_other.date_utc:
+                yield next_top
+                next_top = next(sorted_top_posts, None)
+            else:
+                yield next_other
+                next_other = next(other_posts, None)
+
+
 class TopSearchResults:
    """
    An invocation of this class triggers a search on Instagram for the provided search string.
@@ -1126,6 +1418,17 @@ class TopSearchResults:
            if name:
                yield name

+    def get_hashtags(self) -> Iterator[Hashtag]:
+        """
+        Provides the hashtags from the search result.
+
+        .. versionadded:: 4.4
+        """
+        for hashtag in self._node.get('hashtags', []):
+            node = hashtag.get('hashtag', {})
+            if 'name' in node:
+                yield Hashtag(self._context, node)
+
    @property
    def searchstring(self) -> str:
        """
@@ -1134,17 +1437,17 @@ class TopSearchResults:
        return self._searchstring


-JsonExportable = Union[Post, Profile, StoryItem]
+JsonExportable = Union[Post, Profile, StoryItem, Hashtag, FrozenNodeIterator]


 def save_structure_to_file(structure: JsonExportable, filename: str) -> None:
-    """Saves a :class:`Post`, :class:`Profile` or :class:`StoryItem` to a '.json' or '.json.xz' file such that it can
-    later be loaded by :func:`load_structure_from_file`.
+    """Saves a :class:`Post`, :class:`Profile`, :class:`StoryItem` or :class:`Hashtag` to a '.json' or '.json.xz' file
+    such that it can later be loaded by :func:`load_structure_from_file`.

    If the specified filename ends in '.xz', the file will be LZMA compressed. Otherwise, a pretty-printed JSON file
    will be created.

-    :param structure: :class:`Post`, :class:`Profile` or :class:`StoryItem`
+    :param structure: :class:`Post`, :class:`Profile`, :class:`StoryItem` or :class:`Hashtag`
    :param filename: Filename, ends in '.json' or '.json.xz'
    """
    json_structure = {'node': structure._asdict(),
@@ -1159,8 +1462,8 @@ def save_structure_to_file(structure: JsonExportable, filename: str) -> None:


 def load_structure_from_file(context: InstaloaderContext, filename: str) -> JsonExportable:
-    """Loads a :class:`Post`, :class:`Profile` or :class:`StoryItem` from a '.json' or '.json.xz' file that
-    has been saved by :func:`save_structure_to_file`.
+    """Loads a :class:`Post`, :class:`Profile`, :class:`StoryItem` or :class:`Hashtag` from a '.json' or '.json.xz' file
+    that has been saved by :func:`save_structure_to_file`.

    :param context: :attr:`Instaloader.context` linked to the new object, used for additional queries if neccessary.
    :param filename: Filename, ends in '.json' or '.json.xz'
@@ -1181,6 +1484,10 @@ def load_structure_from_file(context: InstaloaderContext, filename: str) -> Json
            return Profile(context, json_structure['node'])
        elif node_type == "StoryItem":
            return StoryItem(context, json_structure['node'])
+        elif node_type == "Hashtag":
+            return Hashtag(context, json_structure['node'])
+        elif node_type == "FrozenNodeIterator":
+            return FrozenNodeIterator(**json_structure['node'])
        else:
            raise InvalidArgumentException("{}: Not an Instaloader JSON.".format(filename))
    elif 'shortcode' in json_structure:
@@ -5,13 +5,14 @@ import shutil
 import tempfile
 import unittest
 from itertools import islice
-from typing import Dict, List
+from typing import Optional

 import instaloader

 PROFILE_WITH_HIGHLIGHTS = 325732271
 PUBLIC_PROFILE = "selenagomez"
 PUBLIC_PROFILE_ID = 460563723
+PUBLIC_PROFILE_WITH_IGTV = "natgeo"
 HASHTAG = "kitten"
 LOCATION = "362629379"
 OWN_USERNAME = "aandergr"
@@ -22,8 +23,7 @@ PRIVATE_PROFILE_ID = 1706625676
 EMPTY_PROFILE = "not_public"
 EMPTY_PROFILE_ID = 1928659031

-# Preserve query timestamps (rate control) between tests to not get rate limited
-instaloadercontext_query_timestamps = dict()  # type: Dict[str, List[float]]
+ratecontroller = None  # type: Optional[instaloader.RateController]


 class TestInstaloaderAnonymously(unittest.TestCase):
@@ -36,13 +36,15 @@ class TestInstaloaderAnonymously(unittest.TestCase):
                                         download_comments=True,
                                         save_metadata=True)
        self.L.context.raise_all_errors = True
-        # pylint:disable=protected-access
-        self.L.context._graphql_query_timestamps = instaloadercontext_query_timestamps.copy()
+        if ratecontroller is not None:
+            # pylint:disable=protected-access
+            ratecontroller._context = self.L.context
+            self.L.context._rate_controller = ratecontroller

    def tearDown(self):
        # pylint:disable=global-statement,protected-access
-        global instaloadercontext_query_timestamps
-        instaloadercontext_query_timestamps = self.L.context._graphql_query_timestamps.copy()
+        global ratecontroller
+        ratecontroller = self.L.context._rate_controller
        self.L.close()
        os.chdir('/')
        print("Removing {}".format(self.dir))
@@ -72,7 +74,7 @@ class TestInstaloaderAnonymously(unittest.TestCase):
        self.L.download_hashtag(HASHTAG, NORMAL_MAX_COUNT)

    def test_hashtag_paging(self):
-        for count, post in enumerate(self.L.get_hashtag_posts(HASHTAG)):
+        for count, post in enumerate(instaloader.Hashtag.from_name(self.L.context, HASHTAG).get_all_posts()):
            print(post)
            if count == PAGING_MAX_COUNT:
                break
@@ -93,6 +95,10 @@ class TestInstaloaderAnonymously(unittest.TestCase):
        self.assertEqual(EMPTY_PROFILE.lower(),
                         instaloader.Profile.from_id(self.L.context, EMPTY_PROFILE_ID).username)

+    def test_get_username_by_name_empty(self):
+        self.assertEqual(EMPTY_PROFILE_ID,
+                         instaloader.Profile.from_username(self.L.context, EMPTY_PROFILE).userid)
+
    def test_post_from_mediaid(self):
        for post in instaloader.Profile.from_username(self.L.context, PUBLIC_PROFILE).get_posts():
            post2 = instaloader.Post.from_mediaid(self.L.context, post.mediaid)
@@ -104,6 +110,11 @@ class TestInstaloaderAnonymously(unittest.TestCase):
                           PAGING_MAX_COUNT):
            print(post)

+    def test_public_profile_igtv(self):
+        for post in islice(instaloader.Profile.from_username(self.L.context, PUBLIC_PROFILE_WITH_IGTV).get_igtv_posts(),
+                           PAGING_MAX_COUNT):
+            print(post)
+

 class TestInstaloaderLoggedIn(TestInstaloaderAnonymously):

@@ -172,6 +183,11 @@ class TestInstaloaderLoggedIn(TestInstaloaderAnonymously):
        for f in profile.get_followers():
            print(f.username)

+    def test_get_similar_accounts(self):
+        profile = instaloader.Profile.from_username(self.L.context, OWN_USERNAME)
+        for f in profile.get_similar_accounts():
+            print(f.username)
+
    def test_get_likes(self):
        for post in instaloader.Profile.from_username(self.L.context, OWN_USERNAME).get_posts():
            for like in post.get_likes():
Author	SHA1	Message	Date
Alexander Graf	f71ea562af	First Alpha Release for Version 4.5	2020-07-21 17:37:45 +02:00
Alexander Graf	bc40b82f94	Resume a previously-aborted post download loop (#732 ) With this change, Instaloader is capable of resuming a previously-aborted download loop. To do so, it creates a JSON file within the target directory when interrupted, that contains all the necessary information to later resume that operation. Resuming an interrupted download is supported for most, but not all targets. It is supported for: - Regular profile posts, - IGTV posts - Saved posts, - Tagged posts, - Explore posts.	2020-07-21 17:28:55 +02:00
Alexander Graf	c817d1901a	Merge branch 'master' into upcoming/v4.5	2020-07-18 15:32:05 +02:00
Alexander Graf	e232c82d5e	Release of Version 4.4.5	2020-07-18 15:24:31 +02:00
Alexander Graf	601dd0b106	First Release Candidate for Version 4.4.5	2020-07-15 10:26:22 +02:00
Alexander Graf	4c20b573e2	Fix KeyError on Posts with very few comments	2020-07-15 10:17:20 +02:00
Cadence EmberandAlexander Graf	db63e5c7e5	Bypass IP-based redirects to /accounts/login (#727 ) * Bypass IP-based redirects to /accounts/login There are two main changes made: For users, we request /{username}/feed/ instead of /{username}/. For some reason, this completely bypasses the login redirect. This page doesn't work in browser while blocked, but fortunately all the data we need is just present in the HTML page. For posts, we change from using the /p/ page to using the graphql endpoint for the same data, which is still subject to graphql rate limits, but is not subject to login redirects. The data is identical between the two pages, apart from the object keys being sorted differently and rhx_gis being missing on graphql. Yes, this now unblocks access from VPNs, Tor, cloud servers, etc. * Apply requested patch to fix comments * Remove rhx_gis from Post and Profile Co-authored-by: Alexander Graf <17130992+aandergr@users.noreply.github.com>	2020-07-15 09:46:39 +02:00
Alexander Graf	fbe05a1add	More flexible Rate Controller (#716 ) Move InstaloaderContext's rate controlling logic into a class RateController with fine-grained methods to enable easily changing Instaloader's behavior regarding rate limits.	2020-07-11 13:54:31 +02:00
Alexander Graf	e21d34124d	Add codesnippet for almost chronological order Such as for downloading hashtag feeds, as discussed in #666 and contributed by @e2tovar. Also change comment color to grey in codesnippets in documentation.	2020-06-21 19:38:16 +02:00
Lars Lindqvist	fa62025ea3	Deprecate commit-mode (#697 ) Rather than checking the json file to make sure posts have been successfully downloaded, data is stored in a temporary file which is renamed when downloading has finished, as suggested in #257.	2020-06-20 16:23:31 +02:00
michalens	66b3ec6928	Add explanation on downloading single post (#694 )	2020-06-19 19:39:08 +02:00
Alexander Graf	f7d82495bd	Consistently use native '\' path separator on Windows Use os.path.join() for joining paths rather than +'/'+. os.path.join() uses '/' on Unix-ish and '\' on Windows. On Windows, this makes the following improvements: - Visual consistency with / and \ not being mixed-up in one path (os.path.join() has already been used in many places). - Allow using the \\?\ bypass method, e.g. with --dirname-pattern=\\?\D:\{target} The \\?\ bypass method fails when '/' is used as path separator. With the \\?\ bypass method, we can create directories whose name would otherwise be considered invalid, such as directories starting with "aux.".	2020-06-19 19:33:46 +02:00
Alexander Graf	b0a430051e	Restrict GH Action lint workflow on Python 3.8 Now that Travis-CI status reporting works again, there is no need to run the tests twice.	2020-06-18 08:33:25 +02:00
Alexander Graf	8b56b90e72	Less intrusive PR template	2020-06-18 08:32:46 +02:00
Alexander Graf	0c93e63615	Update troubleshooting.rst, Minor fixes to contributing.rst	2020-06-16 21:44:22 +02:00
Alexander Graf	65025e61d9	Update issue/PR templates, contributing.rst, CI - Keep documentation in sync with master branch - Run PyLint and MyPy on Github Actions - Update Issue templates, enforce usage of issue templates, add links to resources in Instaloader Documentation - Update contributing.rst	2020-06-16 20:26:58 +02:00
Alexander Graf	11baf493db	Release of Version 4.4.4	2020-06-11 20:10:52 +02:00
Alexander Graf	2345d90e0e	Fix docs building on Windows	2020-06-09 13:03:48 +02:00
Alexander Graf	71b09aed0f	Warn if --login parameter is not a username Closes #624.	2020-06-09 11:44:06 +02:00
Alexander Graf	fd6d8542c9	Add a unit test for an empty private profile	2020-06-09 11:44:06 +02:00
Alexander Graf	df6fbbee76	Third Release Candidate for Version 4.4.4	2020-06-08 18:51:46 +02:00
Alexander Graf	1661bf7722	Remove unused import	2020-06-08 18:44:43 +02:00
Alexander Graf	8f9723cf62	Fix "Use --login ..." warning at startup	2020-06-08 18:40:13 +02:00
Alexander Graf	e1567a0342	Second Release Candidate for Version 4.4.4	2020-06-08 18:22:06 +02:00
Alexander Graf	610b6242a0	Get higher image quality if logged-in Addresses a suggestion in #630.	2020-06-08 17:53:39 +02:00
Alexander Graf	d90f90988a	Fix ValueError when executing from other drive on Windows Fixes #671.	2020-06-08 17:01:42 +02:00
Alexander Graf	d3c747a306	First Release Candidate for Version 4.4.4	2020-06-06 10:52:42 +02:00
Alexander Graf	225353dc21	Download highest-resolution images if logged-in (#662 ) If logged-in, Post.url and Post.get_sidecar_nodes() now use the iPhone API endpoint to obtain the highest-resolution image version. Resolves #630. A notice "Warning: Use --login to download higher-quality versions of pictures" is issued if the user tries to download posts without being logged-in. Further, the message "Errors occurred:", which is displayed when the InstaloaderContext is closed, has been changed to "Errors or warnings occurred:".	2020-06-06 10:51:03 +02:00
Alexander Graf	ed499cb49c	Let --sessionfile default to a more persistent path (#659 ) The presence of a sessionfile avoids the usage of the failure-prone login mechanism. This commit addresses a suggestion in #615 to store the sessionfile in a persistent path rather than a path within a temporary directory if no --sessionfile parameter is given. The default path is now: $XDG_CONFIG_HOME/instaloader/session-USERNAME or ~/.config/instaloader/session-USERNAME on Unix, %LOCALAPPDATA%\Instaloader\session-USERNAME on Windows. If no file exists in the new path, Instaloader tries loading from the path where the sessionfile was stored before this commit, hence it automatically migrates to the new sessionfile path.	2020-06-06 10:49:41 +02:00
Alexander Graf	1c13d5a30b	Windows EXE: Start Powershell instead of cmd ... if started by file explorer.	2020-06-06 10:37:20 +02:00
Alexander Graf	2fd1422619	Windows EXE: Use psutil for startup terminal detection	2020-06-05 20:12:23 +00:00
Alexander Graf	c818165db4	Update 615_import_firefox_session.py This fixes finding the cookies.sqlite file on MacOS as per https://github.com/instaloader/instaloader/issues/92#issuecomment-638361344.	2020-06-03 20:07:52 +02:00
Alexander Graf	26a78463f1	doc: Discourage using --commit-mode --commit-mode has many bugs, especially #257, #483. For now we discourage using it until it is reimplemented.	2020-06-03 18:03:27 +02:00
Alexander Graf	689dd3613b	doc: Minor style-related changes to code examples	2020-06-03 18:03:01 +02:00
Alexander Graf	27b7c98be2	Update doc section Troubleshooting -> Login Error New Firefox cookie import script, based on a comment in #615: https://github.com/instaloader/instaloader/issues/615#issuecomment-630127064	2020-06-03 17:14:08 +02:00
Alexander Graf	72d86533f2	Downgrade 'old' stale action to more stable v1	2020-05-26 16:03:52 +02:00
Alexander Graf	7bb08a220a	Update issue stale action	2020-05-24 11:25:37 +02:00
Alexander Graf	03aabc52bb	Release of Version 4.4.3	2020-05-22 18:32:05 +02:00
fireattack	9d4df0d5fb	Only substitute characters from auto generated fields (#631 ) Only substitute characters from auto generated fields in _PostPathFormatter This fixes issues of path problem on Windows such as #304, #529, etc. Also made the following changes: 1. Changes substitution of colon to full-width colon (U+FF1A) instead of MODIFIER LETTER COLON (U+A789) to be consistent with others (we use full-width backslash, pipe, etc. already). 2. Also replaces `\r` on Windows with space, just like `\n`.	2020-05-22 18:30:55 +02:00
Alexander Graf	bf2f4f16ed	Release of Version 4.4.2	2020-05-19 09:27:03 +02:00
InfinitayandAlexander Graf	1a3712d112	Workaround for Instagram's encrypted login. (#623 ) * Workaround for Instagram's encrypted login. Credit: https://github.com/pgrimaud/ See: https://github.com/pgrimaud/instagram-user-feed/commit/96ad4cf54d1ad331b337f325c73e664999a6d066 Closes #615. Co-authored-by: Alexander Graf <17130992+aandergr@users.noreply.github.com>	2020-05-19 09:25:08 +02:00
Alexander Graf	65462b01ec	Release of Version 4.4.1	2020-05-14 16:22:42 +02:00
Alexander Graf	9857c5fa24	Regex target matching: Fix hashtag regex Unicode characters are allowed in hashtags, so instead of just [A-Za-z0-9], we now match all \w. Fixes #614.	2020-05-14 16:17:07 +02:00
Alexander Graf	83f013f6ca	Release of Version 4.4	2020-05-11 20:25:43 +02:00
Alexander Graf	e7437310ea	Add versionadded doc annotation for is_business_account and business_category_name	2020-05-11 20:24:04 +02:00
yolo	dcba25a40e	[Issue #601 ] Add is_business_account and business_category_name properties to Profile structure	2020-05-06 08:45:31 +02:00
Alexander Graf	5e4b5fd877	First Beta Release for Version 4.4	2020-04-21 10:22:03 +02:00
Lars Lindqvist	657e76e2d9	Simplify username regex Without passing re.ASCII, \w matches non-ascii letter, such as ö, π and я, which to my knowledge isn't allowed in usernames. Now it's explicit which character ranges are expected in each target.	2020-04-21 10:15:24 +02:00
Lars Lindqvist	78023b3c98	Regex target matching	2020-04-21 10:15:24 +02:00
Alexander Graf	ab67ca30e5	Merge branch 'master' into v4.4-dev	2020-04-21 09:53:01 +02:00
Alexander Graf	0e13c73b62	Include top posts in download_hashtag()	2020-04-19 12:21:27 +02:00
Alexander Graf	b98e9941ae	Omit media when saving Hashtag JSON	2020-04-19 11:30:19 +02:00
Alexander Graf	c9e4f68597	Download Hashtag profilepic and save metadata JSON	2020-04-19 11:16:39 +02:00
Alexander Graf	1338cf0fcd	Fix 'occured' typo in error messages Mentioned in #573 - thanks!	2020-04-18 11:16:55 +02:00
Alexander Graf	83a0f52c60	Hashtag class	2020-04-17 18:49:38 +02:00
Alexander Graf	5c5c633bee	Catch Exceptions raised during filter evaluation Closes #192.	2020-04-10 14:39:40 +02:00
Alexander Graf	e268f9e625	Release of Version 4.3.2	2020-04-10 12:08:24 +02:00
Alexander Graf	793b7e82ba	Limit feed node processing to Post nodes Fixes #478.	2020-04-10 12:05:15 +02:00
Alexander Graf	7329ee10d8	Catch KeyError: 'authenticated' when Logging In Fixes #472.	2020-04-10 12:05:15 +02:00
Alexander Graf	57d2ce9538	Alleviate #449 by catching bad metadata response Instead of ungracefully failing with a TypeError, the Post is now skipped with an error message containing the invalid JSON.	2020-04-10 11:57:36 +02:00
Alexander Graf	dc458a3f3c	Limit feed node processing to Post nodes Fixes #478.	2020-04-10 11:13:07 +02:00
Alexander Graf	08a359408c	Improve Log Output of Skipped Posts	2020-04-10 11:12:09 +02:00
Alexander Graf	a794089c4f	Deduplicate Post Download Loop	2020-04-07 10:06:53 +02:00
Alexander Graf	c56657855d	Release of Version 4.3.1	2020-04-04 10:54:04 +02:00
Alexander Graf	8c7561c5d7	Fix Downloading of Videos within a Sidecar Post Fixes #558.	2020-04-04 10:51:55 +02:00
Alexander Graf	c76385bde2	Unit Test for Profile.get_similar_accounts()	2020-03-25 19:06:05 +01:00
Alexander Graf	b5bf8a0560	Implement Profile.get_similar_accounts() Fixes #489.	2020-03-25 19:00:03 +01:00
Alexander Graf	2790342a09	Catch KeyError: 'authenticated' when Logging In Fixes #472.	2020-03-25 18:17:02 +01:00
Alexander Graf	e725ca2640	Post properties is_sponsored and sponsor_users This allows to access the Paid Partnership With metadata of a Post. Closes #295.	2020-03-25 17:13:46 +01:00
Alexander Graf	9c2e7c70ac	Fix number output in download_igtv()	2020-03-16 13:38:13 +01:00
Alexander Graf	ffbf5b517c	Release of Version 4.3	2020-03-16 09:13:25 +01:00
Alexander Graf	0843e10187	First Release Candidate for Version 4.3	2020-03-15 14:28:16 +01:00
Alexander Graf	78a93c945e	Revert "Revert "Remove nonworking saythanks.io link"" saythanks.io does not work again and the link is now removed. This reverts commit `b2acbcf0e0`.	2020-03-14 17:56:44 +01:00
Alexander Graf	bb3e309195	First Beta Release for Version 4.3	2020-03-14 17:21:07 +01:00
Alexander Graf	6cc2ed54ea	Download IGTV Videos of a Profile (--igtv)	2020-03-14 16:56:51 +01:00
Alexander Graf	41cf260fdd	Fix post filter for downloading tagged posts	2020-03-14 16:55:37 +01:00
Alexander Graf	f910705c51	docs: Fix formatting of request_timeout param	2020-03-07 11:14:07 +01:00
Alexander Graf	aa3f19dcbe	Second Alpha Release for Version 4.3	2020-03-01 17:44:17 +01:00
Alexander Graf	fe258f1abb	Also Download Highlight Cover Pictures Instaloader.download_highlights(), as well as the Instaloader CLI with --highlight option, now also downloads the cover picture of each highlight. This closes #323, where it was suggested.	2020-03-01 17:41:58 +01:00
Alexander Graf	534497bea5	fix misplacement of --request-timeout in docs	2020-02-28 09:49:29 +01:00