From fc2bf6d58b664f4413af21ec2b1d6d45059bada8 Mon Sep 17 00:00:00 2001
From: Ilya Samartsev <ignz@yandex-team.ru>
Date: Fri, 4 Sep 2020 17:40:40 +0300
Subject: [PATCH] add tests and small fixes

---
 aiogram/utils/text_decorations.py         | 16 +++++++++++----
 tests/test_utils/test_text_decorations.py | 25 +++++++++++++++++++++++
 2 files changed, 37 insertions(+), 4 deletions(-)
 create mode 100644 tests/test_utils/test_text_decorations.py

diff --git a/aiogram/utils/text_decorations.py b/aiogram/utils/text_decorations.py
index 58812d57..81592465 100644
--- a/aiogram/utils/text_decorations.py
+++ b/aiogram/utils/text_decorations.py
@@ -57,7 +57,7 @@ class TextDecoration(ABC):
         """
         result = "".join(
             self._unparse_entities(
-                text.encode('utf-16-le'), sorted(entities, key=lambda item: item.offset) if entities else []
+                self._add_surrogates(text), sorted(entities, key=lambda item: item.offset) if entities else []
             )
         )
         return result
@@ -76,8 +76,8 @@ class TextDecoration(ABC):
         for index, entity in enumerate(entities):
             if entity.offset * 2 < offset:
                 continue
-            if entity.offset * 2> offset:
-                yield self.quote(text[offset : entity.offset * 2].decode('utf-16-le'))
+            if entity.offset * 2 > offset:
+                yield self.quote(self._remove_surrogates(text[offset : entity.offset * 2]))
             start = entity.offset * 2
             offset = entity.offset * 2 + entity.length * 2
 
@@ -94,7 +94,15 @@ class TextDecoration(ABC):
             )
 
         if offset < length:
-            yield self.quote(text[offset:length].decode('utf-16-le'))
+            yield self.quote(self._remove_surrogates(text[offset:length]))
+
+    @staticmethod
+    def _add_surrogates(text: str):
+        return text.encode('utf-16-le')
+
+    @staticmethod
+    def _remove_surrogates(text: bytes):
+        return text.decode('utf-16-le')
 
     @abstractmethod
     def link(self, value: str, link: str) -> str:  # pragma: no cover
diff --git a/tests/test_utils/test_text_decorations.py b/tests/test_utils/test_text_decorations.py
new file mode 100644
index 00000000..dd0e595d
--- /dev/null
+++ b/tests/test_utils/test_text_decorations.py
@@ -0,0 +1,25 @@
+from aiogram.types import MessageEntity, MessageEntityType
+from aiogram.utils import text_decorations
+
+
+class TestTextDecorations:
+  def test_unparse_entities_normal_text(self):
+    assert text_decorations.markdown_decoration.unparse(
+      "hi i'm bold and italic and still bold",
+      entities=[
+        MessageEntity(offset=3, length=34, type=MessageEntityType.BOLD),
+        MessageEntity(offset=12, length=10, type=MessageEntityType.ITALIC),
+      ]
+    ) == "hi *i'm bold _and italic_\r and still bold*"
+
+  def test_unparse_entities_emoji_text(self):
+    """
+    emoji is encoded as two chars in json
+    """
+    assert text_decorations.markdown_decoration.unparse(
+      "🚀 i'm bold and italic and still bold",
+      entities=[
+        MessageEntity(offset=3, length=34, type=MessageEntityType.BOLD),
+        MessageEntity(offset=12, length=10, type=MessageEntityType.ITALIC),
+      ]
+    ) == "🚀 *i'm bold _and italic_\r and still bold*"