From 10365394d7ad2577ca1ec76de3fbeedd31aaeeb3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mike=20F=C3=A4hrmann?= <mike_faehrmann@web.de>
Date: Fri, 17 Aug 2018 20:04:11 +0200
Subject: [PATCH] [twitter] add support for user-timelines (closes #96)

also adds a 'retweets' option to filter retweeted content
---
 CHANGELOG.md                    |   2 +
 docs/configuration.rst          |   9 +++
 docs/gallery-dl.conf            |   4 +
 docs/supportedsites.rst         |   2 +-
 gallery_dl/extractor/twitter.py | 133 ++++++++++++++++++++++++--------
 gallery_dl/version.py           |   2 +-
 6 files changed, 118 insertions(+), 34 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index bf7232e2..144a09e7 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,7 @@
 # Changelog
 
+## Unreleased
+
 ## 1.5.1 - 2018-08-17
 - Added support for:
   - `piczel` - https://piczel.tv/
diff --git a/docs/configuration.rst b/docs/configuration.rst
index 4bfe9201..88289cc0 100644
--- a/docs/configuration.rst
+++ b/docs/configuration.rst
@@ -642,6 +642,15 @@ Description A (comma-separated) list of post types to extract images, etc. from.
 =========== =====
 
 
+extractor.twitter.retweets
+--------------------------
+=========== =====
+Type        ``bool``
+Default     ``true``
+Description Extract images from retweets.
+=========== =====
+
+
 extractor.[booru].tags
 ----------------------
 =========== =====
diff --git a/docs/gallery-dl.conf b/docs/gallery-dl.conf
index 8674e171..66cc0c45 100644
--- a/docs/gallery-dl.conf
+++ b/docs/gallery-dl.conf
@@ -107,6 +107,10 @@
             "posts": "photo",
             "reblogs": true
         },
+        "twitter":
+        {
+            "retweets": true
+        },
         "yandere":
         {
             "tags": false
diff --git a/docs/supportedsites.rst b/docs/supportedsites.rst
index 1edbdeb0..d87949a6 100644
--- a/docs/supportedsites.rst
+++ b/docs/supportedsites.rst
@@ -76,7 +76,7 @@ SmugMug              https://www.smugmug.com/            |Albums, individ-5|
 Subapics             https://subapics.com/               Chapters, Manga
 The /b/ Archive      https://thebarchive.com/            Threads
 Tumblr               https://www.tumblr.com/             Images from Users, Likes, Posts, Tag-Searches      Optional (OAuth)
-Twitter              https://twitter.com/                Tweets
+Twitter              https://twitter.com/                Images from Users, Tweets
 Warosu               https://warosu.org/                 Threads
 World Three          http://www.slide.world-three.org/   Chapters, Manga
 XVideos              https://www.xvideos.com/            Images from Users, Galleries
diff --git a/gallery_dl/extractor/twitter.py b/gallery_dl/extractor/twitter.py
index 36779000..01f205f2 100644
--- a/gallery_dl/extractor/twitter.py
+++ b/gallery_dl/extractor/twitter.py
@@ -12,52 +12,121 @@ from .common import Extractor, Message
 from .. import text
 
 
-class TwitterTweetExtractor(Extractor):
-    """Extractor for images from tweets on twitter.com"""
+class TwitterExtractor(Extractor):
+    """Base class for twitter extractors"""
     category = "twitter"
-    subcategory = "tweet"
     directory_fmt = ["{category}", "{user}"]
     filename_fmt = "{tweet_id}_{num}.{extension}"
     archive_fmt = "{tweet_id}_{num}"
-    pattern = [r"(?:https?://)?(?:www\.|mobile\.)?twitter\.com/"
-               r"(([^/]+)/status/(\d+))"]
+    root = "https://twitter.com"
+
+    def __init__(self):
+        Extractor.__init__(self)
+        self.user = None
+        self.retweets = self.config("retweets", True)
+
+    def items(self):
+        yield Message.Version, 1
+        yield Message.Directory, self.metadata()
+
+        for tweet in self.tweets():
+            data = self._data_from_tweet(tweet)
+            if not self.retweets and data["retweet_id"]:
+                continue
+
+            images = text.extract_iter(
+                tweet, 'data-image-url="', '"')
+            for data["num"], url in enumerate(images, 1):
+                text.nameext_from_url(url, data)
+                yield Message.Url, url + ":orig", data
+
+    def metadata(self):
+        """Return general metadata"""
+
+    def tweets(self):
+        """Yield HTML content of all relevant tweets"""
+
+    @staticmethod
+    def _data_from_tweet(tweet):
+        data = text.extract_all(tweet, (
+            ("tweet_id"  , 'data-tweet-id="'   , '"'),
+            ("retweet_id", 'data-retweet-id="' , '"'),
+            ("retweeter" , 'data-retweeter="'  , '"'),
+            ("user"      , 'data-screen-name="', '"'),
+            ("username"  , 'data-name="'       , '"'),
+            ("userid"    , 'data-user-id="'    , '"'),
+        ))[0]
+        for key in ("tweet_id", "retweet_id", "userid"):
+            data[key] = text.parse_int(data[key])
+        data["retweeter"] = data["retweeter"] or ""
+        return data
+
+
+class TwitterUserExtractor(TwitterExtractor):
+    """Extractor for all tweeted images of a user"""
+    subcategory = "user"
+    archive_fmt = "{tweet_id}_{num}"
+    pattern = [r"(?:https?://)?(?:www\.|mobile\.)?twitter\.com"
+               r"/([^/?&#]+)/?$"]
+    test = [("https://twitter.com/PicturesEarth", {
+        "range": (1, 40),
+        "url": "2f4d51cbba81e56c1c755677b3ad58fc167c9771",
+        "keyword": "611066e488c233e0b1bd2ab45d5f7fca1335f691",
+    })]
+
+    def __init__(self, match):
+        TwitterExtractor.__init__(self)
+        self.user = match.group(1)
+
+    def metadata(self):
+        return {"user": self.user}
+
+    def tweets(self):
+        url = "{}/i/profiles/show/{}/timeline/tweets".format(
+            self.root, self.user)
+        params = {}
+        tweet = None
+
+        while True:
+            data = self.request(url, params=params).json()
+            html = data["items_html"]
+
+            for tweet in text.extract_iter(
+                    html, '<div class="tweet ', '\n</li>'):
+                yield tweet
+
+            if not tweet or not data["has_more_items"]:
+                return
+            params["max_position"] = text.extract(
+                tweet, 'data-tweet-id="', '"')[0]
+
+
+class TwitterTweetExtractor(TwitterExtractor):
+    """Extractor for images from tweets on twitter.com"""
+    subcategory = "tweet"
+    pattern = [r"(?:https?://)?(?:www\.|mobile\.)?twitter\.com"
+               r"/([^/?&#]+)/status/(\d+)"]
     test = [
         ("https://twitter.com/PicturesEarth/status/672897688871018500", {
             "url": "d9e68d41301d2fe382eb27711dea28366be03b1a",
-            "keyword": "7a6eac2bc88bbf16d0671ebb38e31f708d940ee8",
+            "keyword": "6ea5cdc97e4e8b2b133f7dfb2048b85a608ef041",
             "content": "a1f2f04cb2d8df24b1afa7a39910afda23484342",
         }),
         ("https://twitter.com/perrypumas/status/894001459754180609", {
             "url": "c8a262a9698cb733fb27870f5a8f75faf77d79f6",
-            "keyword": "334cd0c1f85c3e66923b44740f17407ce444931e",
+            "keyword": "9d0960bc4e1b7407319f8826a43bccf2df60abcf",
         }),
     ]
 
     def __init__(self, match):
-        Extractor.__init__(self)
-        self.path, self.user, self.tid = match.groups()
+        TwitterExtractor.__init__(self)
+        self.user, self.tweet_id = match.groups()
 
-    def items(self):
-        page = self.request("https://twitter.com/" + self.path).text
-        data = self.get_job_metadata()
-        imgs = self.get_image_urls(page)
-        data["count"] = len(imgs)
-        yield Message.Version, 1
-        yield Message.Directory, data
-        for data["num"], url in enumerate(imgs, 1):
-            yield Message.Url, url + ":orig", text.nameext_from_url(url, data)
+    def metadata(self):
+        return {"user": self.user, "tweet_id": self.tweet_id}
 
-    def get_job_metadata(self):
-        """Collect metadata for extractor-job"""
-        return {
-            "user": self.user,
-            "tweet_id": self.tid,
-        }
-
-    @staticmethod
-    def get_image_urls(page):
-        """Extract and return a list of all image-urls"""
-        tweet = text.extract(
-            page, '<div class="follow-bar">', '<ul class="stats" ')[0]
-        return list(text.extract_iter(
-            tweet, '<img data-aria-label-part src="', '"'))
+    def tweets(self):
+        url = "{}/{}/status/{}".format(self.root, self.user, self.tweet_id)
+        page = self.request(url).text
+        return (text.extract(
+            page, '<div class="tweet ', '<ul class="stats')[0],)
diff --git a/gallery_dl/version.py b/gallery_dl/version.py
index da76da93..a55aa19e 100644
--- a/gallery_dl/version.py
+++ b/gallery_dl/version.py
@@ -6,4 +6,4 @@
 # it under the terms of the GNU General Public License version 2 as
 # published by the Free Software Foundation.
 
-__version__ = "1.5.1"
+__version__ = "1.5.2-dev"