[gfycat] add 'user' and 'search' extractors

gallery_dl/extractor/gfycat.py

@@ -6,9 +6,11 @@
 # it under the terms of the GNU General Public License version 2 as
 # published by the Free Software Foundation.
 
-"""Extract images from https://gfycat.com/"""
+"""Extractors for https://gfycat.com/"""
 
 from .common import Extractor, Message
+from .. import text
+from ..cache import cache
 
 
 class GfycatExtractor(Extractor):
@@ -47,11 +49,45 @@ class GfycatExtractor(Extractor):
         return ()
 
 
+class GfycatUserExtractor(GfycatExtractor):
+    """Extractor for gfycat user profiles"""
+    subcategory = "user"
+    directory_fmt = ("{category}", "{userName}")
+    pattern = r"(?:https?://)?gfycat\.com/@([^/?&#]+)"
+    test = ("https://gfycat.com/@gretta", {
+        "pattern": r"https://giant\.gfycat\.com/[A-Za-z]+\.mp4",
+        "count": ">= 100",
+    })
+
+    def gfycats(self):
+        return GfycatAPI(self).user(self.key)
+
+
+class GfycatSearchExtractor(GfycatExtractor):
+    """Extractor for gfycat search results"""
+    subcategory = "search"
+    directory_fmt = ("{category}", "Search", "{search}")
+    pattern = r"(?:https?://)?gfycat\.com/gifs/search/([^/?&#]+)"
+    test = ("https://gfycat.com/gifs/search/funny+animals", {
+        "pattern": r"https://\w+\.gfycat\.com/[A-Za-z]+\.mp4",
+        "archive": False,
+        "range": "100-300",
+        "count": "> 200",
+    })
+
+    def metadata(self):
+        self.key = text.unquote(self.key).replace("+", " ")
+        return {"search": self.key}
+
+    def gfycats(self):
+        return GfycatAPI(self).search(self.key)
+
+
 class GfycatImageExtractor(GfycatExtractor):
     """Extractor for individual images from gfycat.com"""
     subcategory = "image"
     pattern = (r"(?:https?://)?(?:\w+\.)?gfycat\.com"
-               r"/(?:gifs/detail/|\w+/)?([A-Za-z]+)")
+               r"/(?:gifs/detail/|\w+/)?([A-Za-z]{8,})")
     test = (
         ("https://gfycat.com/GrayGenerousCowrie", {
             "url": "e0b5e1d7223108249b15c3c7898dd358dbfae045",
@@ -86,3 +122,54 @@ class GfycatImageExtractor(GfycatExtractor):
     def gfycats(self):
         url = "https://api.gfycat.com/v1/gfycats/" + self.key
         return (self.request(url).json()["gfyItem"],)
+
+
+class GfycatAPI():
+    API_ROOT = "https://api.gfycat.com"
+    ACCESS_KEY = "Anr96uuqt9EdamSCwK4txKPjMsf2M95Rfa5FLLhPFucu8H5HTzeutyAa"
+
+    def __init__(self, extractor):
+        self.extractor = extractor
+        self.headers = {}
+
+    def gfycat(self, gfycat_id):
+        endpoint = "/v1/gfycats/" + gfycat_id
+        return self._call(endpoint)["gfyItem"]
+
+    def user(self, user):
+        endpoint = "/v1/users/{}/gfycats".format(user.lower())
+        params = {"count": 100}
+        return self._pagination(endpoint, params)
+
+    def search(self, query):
+        endpoint = "/v1/gfycats/search"
+        params = {"search_text": query, "count": 150}
+        return self._pagination(endpoint, params)
+
+    @cache(keyarg=1, maxage=3600)
+    def _authenticate_impl(self, category):
+        url = "https://weblogin." + category + ".com/oauth/webtoken"
+        data = {"access_key": self.ACCESS_KEY}
+        headers = {"Referer": self.extractor.root + "/",
+                   "Origin" : self.extractor.root}
+        response = self.extractor.request(
+            url, method="POST", headers=headers, json=data)
+        return "Bearer " + response.json()["access_token"]
+
+    def _call(self, endpoint, params=None):
+        url = self.API_ROOT + endpoint
+        self.headers["Authorization"] = self._authenticate_impl(
+            self.extractor.category)
+        return self.extractor.request(
+            url, params=params, headers=self.headers).json()
+
+    def _pagination(self, endpoint, params):
+        while True:
+            data = self._call(endpoint, params)
+            gfycats = data["gfycats"]
+            yield from gfycats
+
+            if "found" not in data and len(gfycats) < params["count"] or \
+                    not data["gfycats"]:
+                return
+            params["cursor"] = data["cursor"]