diff --git a/gallery_dl/extractor/3dbooru.py b/gallery_dl/extractor/3dbooru.py
index ee6fd5c0..d0e59ad6 100644
--- a/gallery_dl/extractor/3dbooru.py
+++ b/gallery_dl/extractor/3dbooru.py
@@ -71,7 +71,7 @@ class ThreedeebooruPopularExtractor(booru.MoebooruPopularMixin,
                r"/post/popular_(?P<scale>by_(?:day|week|month)|recent)"
                r"(?:\?(?P<query>[^#]*))?")
     test = ("http://behoimi.org/post/popular_by_month?month=2&year=2013", {
-        "url": "a447e115fdab60c25ab71c4fdb1b9f509bc23f99",
+        "url": "c70268dce441a9ccc3383c244ec15edb059f494f",
         "count": 20,
     })
 
diff --git a/gallery_dl/extractor/dynastyscans.py b/gallery_dl/extractor/dynastyscans.py
index 251b2205..f532cac8 100644
--- a/gallery_dl/extractor/dynastyscans.py
+++ b/gallery_dl/extractor/dynastyscans.py
@@ -24,21 +24,21 @@ class DynastyscansBase():
 
     def _parse_image_page(self, image_id):
         url = "{}/images/{}".format(self.root, image_id)
-        page = self.request(url).text
+        extr = text.extract_from(self.request(url).text)
 
-        date, pos = text.extract(page, "class='create_at'>", "</span>")
-        tags, pos = text.extract(page, "class='tags'>", "</span>", pos)
-        src , pos = text.extract(page, "class='btn-group'>", "</div>", pos)
-        url , pos = text.extract(page, ' src="', '"', pos)
+        date = extr("class='create_at'>", "</span>")
+        tags = extr("class='tags'>", "</span>")
+        src = extr("class='btn-group'>", "</div>")
+        url = extr(' src="', '"')
 
         src = text.extract(src, 'href="', '"')[0] if "Source<" in src else ""
 
         return {
-            "url": self.root + url,
+            "url"     : self.root + url,
             "image_id": text.parse_int(image_id),
-            "tags": text.split_html(text.unescape(tags)),
-            "date": text.remove_html(date),
-            "source": text.unescape(src),
+            "tags"    : text.split_html(text.unescape(tags)),
+            "date"    : text.remove_html(date),
+            "source"  : text.unescape(src),
         }
 
 
@@ -59,28 +59,26 @@ class DynastyscansChapterExtractor(DynastyscansBase, ChapterExtractor):
     )
 
     def metadata(self, page):
-        info  , pos = text.extract(page, "<h3 id='chapter-title'><b>", "</b>")
-        author, pos = text.extract(page, " by ", "</a>", pos)
-        group , pos = text.extract(page, '"icon-print"></i> ', '</span>', pos)
-        date  , pos = text.extract(page, '"icon-calendar"></i> ', '<', pos)
-
+        extr = text.extract_from(page)
         match = re.match(
             (r"(?:<a[^>]*>)?([^<]+)(?:</a>)?"  # manga name
              r"(?: ch(\d+)([^:<]*))?"  # chapter info
              r"(?:: (.+))?"),  # title
-            info
+            extr("<h3 id='chapter-title'><b>", "</b>"),
         )
+        author = extr(" by ", "</a>")
+        group = extr('"icon-print"></i> ', '</span>')
 
         return {
-            "manga": text.unescape(match.group(1)),
-            "chapter": text.parse_int(match.group(2)),
+            "manga"   : text.unescape(match.group(1)),
+            "chapter" : text.parse_int(match.group(2)),
             "chapter_minor": match.group(3) or "",
-            "title": text.unescape(match.group(4) or ""),
-            "author": text.remove_html(author),
-            "group": (text.remove_html(group) or
-                      text.extract(group, ' alt="', '"')[0] or ""),
-            "date": date,
-            "lang": "en",
+            "title"   : text.unescape(match.group(4) or ""),
+            "author"  : text.remove_html(author),
+            "group"   : (text.remove_html(group) or
+                         text.extract(group, ' alt="', '"')[0] or ""),
+            "date"    : extr('"icon-calendar"></i> ', '<'),
+            "lang"    : "en",
             "language": "English",
         }
 
diff --git a/gallery_dl/extractor/fallenangels.py b/gallery_dl/extractor/fallenangels.py
index 896b7754..cb131612 100644
--- a/gallery_dl/extractor/fallenangels.py
+++ b/gallery_dl/extractor/fallenangels.py
@@ -39,17 +39,16 @@ class FallenangelsChapterExtractor(ChapterExtractor):
         ChapterExtractor.__init__(self, match, url)
 
     def metadata(self, page):
+        extr = text.extract_from(page)
         lang = "vi" if self.version == "truyen" else "en"
-        data = {
-            "chapter": self.chapter,
+        return {
+            "manga"   : extr('name="description" content="', ' Chapter '),
+            "title"   : extr(':  ', ' - Page 1'),
+            "chapter" : self.chapter,
             "chapter_minor": self.minor or "",
-            "lang": lang,
+            "lang"    : lang,
             "language": util.code_to_language(lang),
         }
-        return text.extract_all(page, (
-            ("manga", 'name="description" content="', ' Chapter '),
-            ("title", ':  ', ' - Page 1'),
-        ), values=data)[0]
 
     @staticmethod
     def images(page):
@@ -83,26 +82,24 @@ class FallenangelsMangaExtractor(MangaExtractor):
         MangaExtractor.__init__(self, match, url)
 
     def chapters(self, page):
-        language = util.code_to_language(self.lang)
+        extr = text.extract_from(page)
         results = []
-        pos = 0
-        while True:
-            test, pos = text.extract(page, '<li style="', '', pos)
-            if test is None:
-                return results
-            volume , pos = text.extract(page, 'class="volume-', '"', pos)
-            url    , pos = text.extract(page, 'href="', '"', pos)
-            chapter, pos = text.extract(page, '>', '<', pos)
-            title  , pos = text.extract(page, '<em>', '</em>', pos)
+        language = util.code_to_language(self.lang)
+        while extr('<li style="', '"'):
+            vol = extr('class="volume-', '"')
+            url = extr('href="', '"')
+            cha = extr('>', '<')
+            title = extr('<em>', '</em>')
 
-            manga, _, chapter = chapter.rpartition(" ")
+            manga, _, chapter = cha.rpartition(" ")
             chapter, dot, minor = chapter.partition(".")
             results.append((url, {
-                "manga": manga,
-                "title": text.unescape(title),
-                "volume": text.parse_int(volume),
-                "chapter": text.parse_int(chapter),
+                "manga"   : manga,
+                "title"   : text.unescape(title),
+                "volume"  : text.parse_int(vol),
+                "chapter" : text.parse_int(chapter),
                 "chapter_minor": dot + minor,
-                "lang": self.lang,
+                "lang"    : self.lang,
                 "language": language,
             }))
+        return results
diff --git a/gallery_dl/extractor/foolslide.py b/gallery_dl/extractor/foolslide.py
index 92de190a..14baa366 100644
--- a/gallery_dl/extractor/foolslide.py
+++ b/gallery_dl/extractor/foolslide.py
@@ -38,6 +38,7 @@ class FoolslideBase(SharedConfigMixin):
         data["volume"] = text.parse_int(info[2])
         data["chapter"] = text.parse_int(info[3])
         data["chapter_minor"] = "." + info[4] if len(info) >= 5 else ""
+        data["title"] = data["chapter_string"].partition(":")[2].strip()
         return data
 
 
@@ -74,14 +75,11 @@ class FoolslideChapterExtractor(FoolslideBase, ChapterExtractor):
             yield Message.Url, url, data
 
     def metadata(self, page):
-        _      , pos = text.extract(page, '<h1 class="tbtitle dnone">', '')
-        manga  , pos = text.extract(page, 'title="', '"', pos)
-        chapter, pos = text.extract(page, 'title="', '"', pos)
-        chapter = text.unescape(chapter)
+        extr = text.extract_from(page)
+        extr('<h1 class="tbtitle dnone">', '')
         return self.parse_chapter_url(self.chapter_url, {
-            "manga": text.unescape(manga).strip(),
-            "title": chapter.partition(":")[2].strip(),
-            "chapter_string": chapter,
+            "manga"         : text.unescape(extr('title="', '"')).strip(),
+            "chapter_string": text.unescape(extr('title="', '"')),
         })
 
     def images(self, page):
@@ -101,25 +99,20 @@ class FoolslideMangaExtractor(FoolslideBase, MangaExtractor):
     pattern_fmt = r"(/series/[^/?&#]+)"
 
     def chapters(self, page):
-        manga , pos = text.extract(page, '<h1 class="title">', '</h1>')
-        author, pos = text.extract(page, '<b>Author</b>: ', '<br', pos)
-        artist, pos = text.extract(page, '<b>Artist</b>: ', '<br', pos)
-        manga = text.unescape(manga).strip()
+        extr = text.extract_from(page)
+        manga = text.unescape(extr('<h1 class="title">', '</h1>')).strip()
+        author = extr('<b>Author</b>: ', '<br')
+        artist = extr('<b>Artist</b>: ', '<br')
 
         results = []
         while True:
-            url, pos = text.extract(
-                page, '<div class="title"><a href="', '"', pos)
+            url = extr('<div class="title"><a href="', '"')
             if not url:
                 return results
-
-            chapter, pos = text.extract(page, 'title="', '"', pos)
-            group  , pos = text.extract(page, 'title="', '"', pos)
-
             results.append((url, self.parse_chapter_url(url, {
                 "manga": manga, "author": author, "artist": artist,
-                "group": group, "chapter_string": chapter,
-                "title": chapter.partition(": ")[2] or "",
+                "chapter_string": extr('title="', '"'),
+                "group"         : extr('title="', '"'),
             })))
 
 
@@ -166,7 +159,7 @@ EXTRACTORS = {
         "test-manga":
             ("https://reader.kireicake.com/series/wonderland/", {
                 "url": "d067b649af1cc88fa8c8b698fde04a10909fd169",
-                "keyword": "99caa336a9d48e27e3b8e56a0a1e6faf9fc13a51",
+                "keyword": "268f43772fb239888ca5c5f6a4f65f99ffb3eefb",
             }),
     },
     "powermanga": {
@@ -214,7 +207,7 @@ EXTRACTORS = {
         "test-manga":
             ("http://sensescans.com/reader/series/hakkenden/", {
                 "url": "2360ccb0ead0ff2f5e27b7aef7eb17b9329de2f2",
-                "keyword": "122cf92c32e6428c50f56ffaf29d06b96750ed71",
+                "keyword": "4919f2bfed38e3a34dc984ec8d1dbd7a03044e23",
             }),
     },
     "worldthree": {
diff --git a/gallery_dl/extractor/hentaicafe.py b/gallery_dl/extractor/hentaicafe.py
index ba095d98..7141affb 100644
--- a/gallery_dl/extractor/hentaicafe.py
+++ b/gallery_dl/extractor/hentaicafe.py
@@ -21,7 +21,7 @@ class HentaicafeChapterExtractor(foolslide.FoolslideChapterExtractor):
                r"(/manga/read/[^/?&#]+/[a-z-]+/\d+/\d+(?:/\d+)?)")
     test = ("https://hentai.cafe/manga/read/saitom-box/en/0/1/", {
         "url": "8c6a8c56875ba3ed7ab0a74a64f9960077767fc2",
-        "keyword": "3688ddd3f3077c93eaa8021477ef66d18dc6c159",
+        "keyword": "7182d262810faa692827c947d2f360dfcb8d5e43",
     })
     root = "https://hentai.cafe"
 
@@ -51,7 +51,7 @@ class HentaicafeMangaExtractor(foolslide.FoolslideMangaExtractor):
         # foolslide URL
         ("https://hentai.cafe/manga/series/saitom-box/", {
             "url": "ca3e8a91531fd6acd863d93ac3afbd8ead06a076",
-            "keyword": "46012b857eb1a1394bc55c0efe7aa4e7f704d10d",
+            "keyword": "f0ece32d958f889d8229ed4052716d398a0a875c",
         }),
     )
     root = "https://hentai.cafe"
diff --git a/gallery_dl/extractor/hentaifox.py b/gallery_dl/extractor/hentaifox.py
index 417c095b..cf4871f7 100644
--- a/gallery_dl/extractor/hentaifox.py
+++ b/gallery_dl/extractor/hentaifox.py
@@ -31,25 +31,21 @@ class HentaifoxGalleryExtractor(HentaifoxBase, GalleryExtractor):
         GalleryExtractor.__init__(self, match)
         self.gallery_id = match.group(2)
 
-    def metadata(self, page):
-        title, pos = text.extract(page, "<h1>", "</h1>")
-        data = text.extract_all(page, (
-            ("parody"    , ">Parodies:"  , "</a></span>"),
-            ("characters", ">Characters:", "</a></span>"),
-            ("tags"      , ">Tags:"      , "</a></span>"),
-            ("artist"    , ">Artists:"   , "</a></span>"),
-            ("group"     , ">Groups:"    , "</a></span>"),
-            ("type"      , ">Category:"  , "</a></span>"),
-        ), pos)[0]
+    def metadata(self, page, split=text.split_html):
+        extr = text.extract_from(page)
 
-        for key, value in data.items():
-            data[key] = text.split_html(value)[::2]
-        data["gallery_id"] = text.parse_int(self.gallery_id)
-        data["title"] = text.unescape(title)
-        data["type"] = data["type"][0] if data["type"] else ""
-        data["language"] = "English"
-        data["lang"] = "en"
-        return data
+        return {
+            "gallery_id": text.parse_int(self.gallery_id),
+            "title"     : text.unescape(extr("<h1>", "</h1>")),
+            "parody"    : split(extr(">Parodies:"  , "</a></span>"))[::2],
+            "characters": split(extr(">Characters:", "</a></span>"))[::2],
+            "tags"      : split(extr(">Tags:"      , "</a></span>"))[::2],
+            "artist"    : split(extr(">Artists:"   , "</a></span>"))[::2],
+            "group"     : split(extr(">Groups:"    , "</a></span>"))[::2],
+            "type"      : text.remove_html(extr(">Category:", "</a></span>")),
+            "language"  : "English",
+            "lang"      : "en",
+        }
 
     def images(self, page):
         return [
diff --git a/gallery_dl/extractor/livedoor.py b/gallery_dl/extractor/livedoor.py
index f0ecfd47..64af9d86 100644
--- a/gallery_dl/extractor/livedoor.py
+++ b/gallery_dl/extractor/livedoor.py
@@ -38,18 +38,14 @@ class LivedoorExtractor(Extractor):
         """Return an iterable with post objects"""
 
     def _load(self, data, body):
-        pid  , pos = text.extract(data, "id : '"   , "'")
-        title, pos = text.extract(data, "title : '", "'", pos)
-        cat1 , pos = text.extract(data, "name:'"   , "'", pos)
-        cat2 , pos = text.extract(data, "name:'"   , "'", pos)
-        date , pos = text.extract(data, "date : '" , "'", pos)
-        tags , pos = text.extract(body, '</dt><dd>', '</dl>')
+        extr = text.extract_from(data)
+        tags = text.extract(body, '</dt><dd>', '</dl>')[0]
 
         return {
-            "id"        : text.parse_int(pid),
-            "title"     : title,
-            "date"      : date,
-            "categories": [cat1, cat2],
+            "id"        : text.parse_int(extr("id : '", "'")),
+            "title"     : extr("title : '", "'"),
+            "categories": [extr("name:'", "'"), extr("name:'", "'")],
+            "date"      : extr("date : '", "'"),
             "tags"      : text.split_html(tags),
             "user"      : self.user,
             "body"      : body,
@@ -108,22 +104,15 @@ class LivedoorBlogExtractor(LivedoorExtractor):
         url = "{}/{}".format(self.root, self.user)
 
         while url:
-            page = self.request(url).text
-            pos = 0
-
+            extr = text.extract_from(self.request(url).text)
             while True:
-                data, pos = text.extract(page, '.articles.push(', ');', pos)
+                data = extr('.articles.push(', ');')
                 if not data:
                     break
-                body, pos = text.extract(
-                    page,
-                    '<div class="article-body-inner">',
-                    '<!-- articleBody End -->',
-                    pos,
-                )
+                body = extr('<div class="article-body-inner">',
+                            '<!-- articleBody End -->')
                 yield self._load(data, body)
-
-            url = text.extract(page, '<a rel="next" href="', '"', pos)[0]
+            url = extr('<a rel="next" href="', '"')
 
 
 class LivedoorPostExtractor(LivedoorExtractor):
@@ -148,13 +137,8 @@ class LivedoorPostExtractor(LivedoorExtractor):
     def posts(self):
         url = "{}/{}/archives/{}.html".format(
             self.root, self.user, self.post_id)
-        page = self.request(url).text
-
-        data, pos = text.extract(page, 'articles :', '</script>')
-        body, pos = text.extract(
-            page,
-            '<div class="article-body-inner">',
-            '<!-- articleBody End -->',
-            pos,
-        )
+        extr = text.extract_from(self.request(url).text)
+        data = extr('articles :', '</script>')
+        body = extr('<div class="article-body-inner">',
+                    '<!-- articleBody End -->')
         return (self._load(data, body),)
diff --git a/gallery_dl/extractor/luscious.py b/gallery_dl/extractor/luscious.py
index 71c32ef5..668a8c09 100644
--- a/gallery_dl/extractor/luscious.py
+++ b/gallery_dl/extractor/luscious.py
@@ -62,7 +62,7 @@ class LusciousAlbumExtractor(LusciousBase, GalleryExtractor):
     test = (
         ("https://luscious.net/albums/okinami-no-koigokoro_277031/", {
             "url": "7e4984a271a1072ac6483e4228a045895aff86f3",
-            "keyword": "1b1e8981afa250a0181e31b15ce3cbaa37c00856",
+            "keyword": "f9c34e1a5b0c1f119e9f644c99933ecf7d7dbfd2",
             "content": "b3a747a6464509440bd0ff6d1267e6959f8d6ff3",
         }),
         ("https://luscious.net/albums/virgin-killer-sweater_282582/", {
diff --git a/gallery_dl/extractor/tsumino.py b/gallery_dl/extractor/tsumino.py
index 834ce47f..62a91732 100644
--- a/gallery_dl/extractor/tsumino.py
+++ b/gallery_dl/extractor/tsumino.py
@@ -77,42 +77,31 @@ class TsuminoGalleryExtractor(TsuminoBase, GalleryExtractor):
         GalleryExtractor.__init__(self, match, url)
 
     def metadata(self, page):
-        extr = text.extract
-        title, pos = extr(page, '"og:title" content="', '"')
-        thumb, pos = extr(page, '"og:image" content="', '"', pos)
+        extr = text.extract_from(page)
+        title = extr('"og:title" content="', '"')
         title_en, _, title_jp = text.unescape(title).partition("/")
         title_en = title_en.strip()
         title_jp = title_jp.strip()
 
-        uploader  , pos = extr(page, 'id="Uploader">'  , '</div>', pos)
-        date      , pos = extr(page, 'id="Uploaded">'  , '</div>', pos)
-        rating    , pos = extr(page, 'id="Rating">'    , '</div>', pos)
-        gtype     , pos = extr(page, 'id="Category">'  , '</div>', pos)
-        collection, pos = extr(page, 'id="Collection">', '</div>', pos)
-        group     , pos = extr(page, 'id="Group">'     , '</div>', pos)
-        artist    , pos = extr(page, 'id="Artist">'    , '</div>', pos)
-        parody    , pos = extr(page, 'id="Parody">'    , '</div>', pos)
-        character , pos = extr(page, 'id="Character">' , '</div>', pos)
-        tags      , pos = extr(page, 'id="Tag">'       , '</div>', pos)
-
         return {
             "gallery_id": text.parse_int(self.gallery_id),
-            "title": title_en or title_jp,
-            "title_en": title_en,
-            "title_jp": title_jp,
-            "thumbnail": thumb,
-            "uploader": text.remove_html(uploader),
-            "date": date.strip(),
-            "rating": text.parse_float(rating.partition(" ")[0]),
-            "type": text.remove_html(gtype),
-            "collection": text.remove_html(collection),
-            "group": text.split_html(group),
-            "artist": text.split_html(artist),
-            "parody": text.split_html(parody),
-            "characters": text.split_html(character),
-            "tags": text.split_html(tags),
-            "language": "English",
-            "lang": "en",
+            "title"     : title_en or title_jp,
+            "title_en"  : title_en,
+            "title_jp"  : title_jp,
+            "thumbnail" : extr('"og:image" content="', '"'),
+            "uploader"  : text.remove_html(extr('id="Uploader">', '</div>')),
+            "date"      : extr('id="Uploaded">', '</div>').strip(),
+            "rating"    : text.parse_float(extr(
+                'id="Rating">', '</div>').partition(" ")[0]),
+            "type"      : text.remove_html(extr('id="Category">'  , '</div>')),
+            "collection": text.remove_html(extr('id="Collection">', '</div>')),
+            "group"     : text.split_html(extr('id="Group">'      , '</div>')),
+            "artist"    : text.split_html(extr('id="Artist">'     , '</div>')),
+            "parody"    : text.split_html(extr('id="Parody">'     , '</div>')),
+            "characters": text.split_html(extr('id="Character">'  , '</div>')),
+            "tags"      : text.split_html(extr('id="Tag">'        , '</div>')),
+            "language"  : "English",
+            "lang"      : "en",
         }
 
     def images(self, page):
diff --git a/gallery_dl/extractor/twitter.py b/gallery_dl/extractor/twitter.py
index 0d496aff..aa63feaf 100644
--- a/gallery_dl/extractor/twitter.py
+++ b/gallery_dl/extractor/twitter.py
@@ -87,18 +87,15 @@ class TwitterExtractor(Extractor):
 
     @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="'       , '"'),
-            ("user_id"   , 'data-user-id="'    , '"'),
-        ))[0]
-        for key in ("tweet_id", "retweet_id", "user_id"):
-            data[key] = text.parse_int(data[key])
-        data["retweeter"] = data["retweeter"] or ""
-        return data
+        extr = text.extract_from(tweet)
+        return {
+            "tweet_id"  : text.parse_int(extr('data-tweet-id="'  , '"')),
+            "retweet_id": text.parse_int(extr('data-retweet-id="', '"')),
+            "retweeter" : extr('data-retweeter="'  , '"'),
+            "user"      : extr('data-screen-name="', '"'),
+            "username"  : extr('data-name="'       , '"'),
+            "user_id"   : text.parse_int(extr('data-user-id="'   , '"')),
+        }
 
     def _tweets_from_api(self, url):
         params = {