Deduplicate emotes

matrix_puppeteer_line/db/__init__.py

@@ -5,11 +5,12 @@ from .user import User
 from .puppet import Puppet
 from .portal import Portal
 from .message import Message
+from .media import Media
 
 
 def init(db: Database) -> None:
-    for table in (User, Puppet, Portal, Message):
+    for table in (User, Puppet, Portal, Message, Media):
         table.db = db
 
 
-__all__ = ["upgrade_table", "User", "Puppet", "Portal", "Message"]
+__all__ = ["upgrade_table", "User", "Puppet", "Portal", "Message", "Media"]

matrix_puppeteer_line/db/media.py

@@ -0,0 +1,51 @@
+# matrix-puppeteer-line - A very hacky Matrix-LINE bridge based on running LINE's Chrome extension in Puppeteer
+# Copyright (C) 2020-2021 Tulir Asokan, Andrew Ferrazzutti
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+from typing import Optional, ClassVar, TYPE_CHECKING
+
+from attr import dataclass
+
+from mautrix.types import ContentURI
+from mautrix.util.async_db import Database
+
+fake_db = Database("") if TYPE_CHECKING else None
+
+
+@dataclass
+class Media:
+    db: ClassVar[Database] = fake_db
+
+    media_id: str
+    mxc: ContentURI
+    # TODO Consider whether mime_type, file_name, and size are needed.
+
+    async def insert(self) -> None:
+        q = ("INSERT INTO media (media_id, mxc) "
+             "VALUES ($1, $2)")
+        await self.db.execute(q, self.media_id, self.mxc)
+
+    async def update(self) -> None:
+        q = ("UPDATE media SET mxc=$2 "
+             "WHERE media_id=$1")
+        await self.db.execute(q, self.media_id, self.mxc)
+
+    @classmethod
+    async def get_by_id(cls, media_id: str) -> Optional[ContentURI]:
+        q = ("SELECT media_id, mxc "
+             "FROM media WHERE media_id=$1")
+        row = await cls.db.fetchrow(q, media_id)
+        if not row:
+            return None
+        return cls(**row)

matrix_puppeteer_line/db/upgrade.py

@@ -60,3 +60,11 @@ async def upgrade_avatars(conn: Connection) -> None:
        ADD COLUMN IF NOT EXISTS icon_path TEXT,
        ADD COLUMN IF NOT EXISTS icon_mxc TEXT
    """)
+
+
+@upgrade_table.register(description="Deduplicated media")
+async def upgrade_media(conn: Connection) -> None:
+    await conn.execute("""CREATE TABLE IF NOT EXISTS media (
+        media_id   TEXT PRIMARY KEY,
+        mxc        TEXT NOT NULL
+    )""")

matrix_puppeteer_line/portal.py

@@ -33,7 +33,7 @@ from mautrix.errors import MatrixError
 from mautrix.util.simple_lock import SimpleLock
 from mautrix.util.network_retry import call_with_net_retry
 
-from .db import Portal as DBPortal, Message as DBMessage
+from .db import Portal as DBPortal, Message as DBMessage, Media as DBMedia
 from .config import Config
 from .rpc import ChatInfo, Participant, Message, Client, PathImage
 from . import user as u, puppet as p, matrix as m
@@ -212,6 +212,7 @@ class Portal(DBPortal, BasePortal):
 
         event_id = None
         if evt.image_url:
+            # TODO Deduplicate stickers, but only if encryption is disabled
             content = await self._handle_remote_photo(source, intent, evt)
             event_id = await self._send_message(intent, content, timestamp=evt.timestamp)
         elif evt.html and not evt.html.isspace():
@@ -244,21 +245,22 @@ class Portal(DBPortal, BasePortal):
                     if msg_html:
                         msg_html += chunk["data"]
                 elif ctype == "img":
-                    if not msg_html:
-                        msg_html = msg_text
-
                     cclass = chunk["class"]
                     if cclass == "emojione":
                         alt = chunk["alt"]
+                        media_id = None
                     else:
-                        alt = f':{"?" if "alt" not in chunk else "".join(filter(lambda char: char.isprintable(), chunk["alt"]))}:'
+                        alt = "".join(filter(lambda char: char.isprintable(), chunk["alt"])).strip()
+                        alt = f':{alt if alt else "n/a"}:'
+                        media_id = f'{chunk.get("data-stickon-pkg-cd", 0)}/{chunk.get("data-stickon-stk-cd", 0)}'
 
+                    # NOTE Not encrypting content linked to by HTML tags
+                    if not self.encrypted:
+                        media_mxc = await self._get_mxc_for_remote_media(source, intent, chunk["src"], media_id)
+                        if not msg_html:
+                            msg_html = msg_text
+                        msg_html += f'<img data-mx-emoticon src="{media_mxc}" alt="{alt}" title="{alt}" height="32">'
                     msg_text += alt
-                    # TODO Make a standalone function for this, and cache mxc in DB
-                    #      ID is some combination of data-stickon-pkg-cd, data-stickon-stk-cd, src
-                    resp = await source.client.read_image(chunk["src"])
-                    media_info = await self._reupload_remote_media(resp.data, intent, resp.mime)
-                    msg_html += f'<img data-mx-emoticon src="{media_info.mxc}" alt="{alt}" title="{alt}" height="32">'
 
             content = TextMessageEventContent(
                 msgtype=MessageType.TEXT,
@@ -279,9 +281,25 @@ class Portal(DBPortal, BasePortal):
                                         msgtype=MessageType.IMAGE, body=media_info.file_name,
                                         info=ImageInfo(mimetype=media_info.mime_type, size=media_info.size))
 
+    async def _get_mxc_for_remote_media(self, source: 'u.User', intent: IntentAPI,
+                                        media_url: str, media_id: Optional[str] = None
+                                        ) -> ContentURI:
+        if not media_id:
+            media_id = media_url
+        media_info = await DBMedia.get_by_id(media_id)
+        if not media_info:
+            self.log.debug(f"Did not find existing mxc URL for {media_id}, uploading media now")
+            resp = await source.client.read_image(media_url)
+            media_info = await self._reupload_remote_media(resp.data, intent, resp.mime, disable_encryption=True)
+            await DBMedia(media_id=media_id, mxc=media_info.mxc).insert()
+            self.log.debug(f"Uploaded media as {media_info.mxc}")
+        else:
+            self.log.debug(f"Found existing mxc URL for {media_id}: {media_info.mxc}")
+        return media_info.mxc
+
     async def _reupload_remote_media(self, data: bytes, intent: IntentAPI,
-                                     mime_type: str = None, file_name: str = None
-                                     ) -> ReuploadedMediaInfo:
+                                     mime_type: str = None, file_name: str = None,
+                                     disable_encryption: bool = True) -> ReuploadedMediaInfo:
         if not mime_type:
             mime_type = magic.from_buffer(data, mime=True)
         upload_mime_type = mime_type
@@ -290,7 +308,7 @@ class Portal(DBPortal, BasePortal):
         upload_file_name = file_name
 
         decryption_info = None
-        if self.encrypted and encrypt_attachment:
+        if self.encrypted and encrypt_attachment and not disable_encryption:
             data, decryption_info = encrypt_attachment(data)
             upload_mime_type = "application/octet-stream"
             upload_file_name = None