From 0e3d08ae7581a84749b4d2cb3d0a0858c43f5601 Mon Sep 17 00:00:00 2001
From: Dan <14043624+delivrance@users.noreply.github.com>
Date: Mon, 22 Jan 2018 00:26:43 +0100
Subject: [PATCH] Add HTML style parse mode

---
 pyrogram/client/client.py       |   3 +-
 pyrogram/extensions/__init__.py |   1 +
 pyrogram/extensions/html.py     | 103 ++++++++++++++++++++++++++++++++
 3 files changed, 106 insertions(+), 1 deletion(-)
 create mode 100644 pyrogram/extensions/html.py

diff --git a/pyrogram/client/client.py b/pyrogram/client/client.py
index 131ce037..d325cc01 100644
--- a/pyrogram/client/client.py
+++ b/pyrogram/client/client.py
@@ -47,7 +47,7 @@ from pyrogram.api.types import (
     InputPeerUser, InputPeerChat, InputPeerChannel
 )
 from pyrogram.crypto import CTR
-from pyrogram.extensions import Markdown
+from pyrogram.extensions import Markdown, HTML
 from pyrogram.session import Auth, Session
 
 log = logging.getLogger(__name__)
@@ -89,6 +89,7 @@ class Client:
         self.peers_by_username = {}
 
         self.markdown = Markdown(self.peers_by_id)
+        self.html = HTML(self.peers_by_id)
 
         self.config = None
         self.proxy = None
diff --git a/pyrogram/extensions/__init__.py b/pyrogram/extensions/__init__.py
index cd918a09..e60b4da1 100644
--- a/pyrogram/extensions/__init__.py
+++ b/pyrogram/extensions/__init__.py
@@ -16,4 +16,5 @@
 # You should have received a copy of the GNU Lesser General Public License
 # along with Pyrogram.  If not, see <http://www.gnu.org/licenses/>.
 
+from .html import HTML
 from .markdown import Markdown
diff --git a/pyrogram/extensions/html.py b/pyrogram/extensions/html.py
new file mode 100644
index 00000000..99aa36ad
--- /dev/null
+++ b/pyrogram/extensions/html.py
@@ -0,0 +1,103 @@
+import re
+from struct import unpack
+
+from pyrogram.api.types import (
+    MessageEntityBold as Bold,
+    MessageEntityItalic as Italic,
+    MessageEntityCode as Code,
+    MessageEntityTextUrl as Url,
+    MessageEntityPre as Pre,
+    InputMessageEntityMentionName as Mention
+)
+
+
+class HTML:
+    SMP_RE = re.compile(r"[\U00010000-\U0010FFFF]")
+
+    BOLD_RE = r"(?P<b><b>(?P<b_body>.*?)</b>)"
+    STRONG_RE = r"(?P<strong><strong>(?P<strong_body>.*?)</strong>)"
+    ITALIC_RE = r"(?P<i><i>(?P<i_body>.*?)</i>)"
+    EMPATHIZE_RE = r"(?P<em><em>(?P<em_body>.*?)</em>)"
+    CODE_RE = r"(?P<code><code>(?P<code_body>.*?)</code>)"
+    PRE_RE = r"(?P<pre><pre>(?P<pre_body>.*?)</pre>)"
+    MENTION_RE = r"(?P<mention><a href=\"tg://user\?id=(?P<user_id>\d+?)\">(?P<mention_text>.*?)</a>)"
+    URL_RE = r"(?P<url><a href=\"(?P<url_path>.*?)\">(?P<url_text>.*?)</a>)"
+
+    HTML_RE = re.compile("|".join([BOLD_RE, STRONG_RE, ITALIC_RE, EMPATHIZE_RE, CODE_RE, PRE_RE, MENTION_RE, URL_RE]))
+
+    @classmethod
+    def add_surrogates(cls, text):
+        return cls.SMP_RE.sub(
+            lambda match:  # Split SMP in two surrogates
+            "".join(chr(i) for i in unpack("<HH", match.group().encode("utf-16le"))),
+            text
+        )
+
+    @staticmethod
+    def remove_surrogates(text):
+        return text.encode("utf-16", "surrogatepass").decode("utf-16")
+
+    def __init__(self, peers_by_id):
+        self.peers_by_id = peers_by_id
+
+    def parse(self, text):
+        entities = []
+        text = self.add_surrogates(text)
+        offset = 0
+
+        # TODO: Beautify ifs
+        for match in self.HTML_RE.finditer(text):
+            start = match.start() - offset
+
+            if match.group("b"):
+                pattern = match.group("b")
+                body = match.group("b_body")
+                entity = Bold(start, len(body))
+                offset += 7
+            elif match.group("strong"):
+                pattern = match.group("strong")
+                body = match.group("strong_body")
+                entity = Bold(start, len(body))
+                offset += 17
+            elif match.group("i"):
+                pattern = match.group("i")
+                body = match.group("i_body")
+                entity = Italic(start, len(body))
+                offset += 7
+            elif match.group("em"):
+                pattern = match.group("em")
+                body = match.group("em_body")
+                entity = Italic(start, len(body))
+                offset += 9
+            elif match.group("code"):
+                pattern = match.group("code")
+                body = match.group("code_body")
+                entity = Code(start, len(body))
+                offset += 13
+            elif match.group("pre"):
+                pattern = match.group("pre")
+                body = match.group("pre_body")
+                entity = Pre(start, len(body), "")
+                offset += 11
+            elif match.group("mention"):
+                pattern = match.group("mention")
+                body = match.group("mention_text")
+                user_id = match.group("user_id")
+                entity = Mention(start, len(body), self.peers_by_id[int(user_id)])
+                offset += len(user_id) + 28
+            elif match.group("url"):
+                pattern = match.group("url")
+                body = match.group("url_text")
+                path = match.group("url_path")
+                entity = Url(start, len(body), path)
+                offset += len(path) + 15
+            else:
+                continue
+
+            entities.append(entity)
+            text = text.replace(pattern, body)
+
+        return dict(
+            message=self.remove_surrogates(text),
+            entities=entities
+        )