moment/src/backend/matrix_client.py

1388 lines
47 KiB
Python
Raw Normal View History

2019-12-19 22:46:16 +11:00
# SPDX-License-Identifier: LGPL-3.0-or-later
2019-12-19 04:16:15 +11:00
"""Matrix client and related classes."""
import asyncio
import html
import io
import logging as log
import platform
2019-11-10 00:52:16 +11:00
import re
import sys
2019-10-31 02:10:40 +11:00
import traceback
from contextlib import suppress
from copy import deepcopy
from datetime import datetime, timedelta
from functools import partial
2019-07-16 06:14:08 +10:00
from pathlib import Path
2019-11-05 06:18:01 +11:00
from typing import (
TYPE_CHECKING, Any, ClassVar, Dict, List, NamedTuple, Optional, Set, Tuple,
Type, Union,
2019-11-05 06:18:01 +11:00
)
2019-11-10 00:52:16 +11:00
from urllib.parse import urlparse
2019-12-02 20:06:21 +11:00
from uuid import UUID, uuid4
import cairosvg
from PIL import Image as PILImage
from pymediainfo import MediaInfo
2019-11-05 06:18:01 +11:00
import nio
from nio.crypto import AsyncDataT as UploadData
from nio.crypto import async_generator_from_data
2019-11-05 06:18:01 +11:00
from . import __app_name__, __display_name__, utils
from .errors import (
2019-12-05 00:59:14 +11:00
BadMimeType, InvalidUserId, InvalidUserInContext, MatrixError,
MatrixNotFound, MatrixTooLarge, UneededThumbnail,
UserFromOtherServerDisallowed,
)
from .html_markdown import HTML_PROCESSOR as HTML
from .media_cache import Media, Thumbnail
from .models.items import Event, Member, Room, Upload, UploadStatus, ZeroDate
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
from .models.model_store import ModelStore
from .nio_callbacks import NioCallbacks
from .pyotherside_events import AlertRequested, LoopException
if TYPE_CHECKING:
from .backend import Backend
if sys.version_info >= (3, 7):
current_task = asyncio.current_task
else:
current_task = asyncio.Task.current_task
CryptDict = Dict[str, Any]
class UploadReturn(NamedTuple):
2019-12-19 04:16:15 +11:00
"""Details for an uploaded file."""
mxc: str
mime: str
decryption_dict: Dict[str, Any]
class MatrixImageInfo(NamedTuple):
2019-12-19 04:16:15 +11:00
"""Image informations to be passed for Matrix file events."""
width: int
height: int
mime: str
size: int
def as_dict(self) -> Dict[str, Union[int, str]]:
"""Return a dict ready to be included in a Matrix file events."""
return {
"w": self.width,
"h": self.height,
"mimetype": self.mime,
"size": self.size,
}
2019-11-05 06:18:01 +11:00
class MatrixClient(nio.AsyncClient):
2019-12-19 04:16:15 +11:00
"""A client for an account to interact with a matrix homeserver."""
2019-12-12 03:42:59 +11:00
user_id_regex = re.compile(r"^@.+:.+")
room_id_or_alias_regex = re.compile(r"^[#!].+:.+")
http_s_url_regex = re.compile(r"^https?://")
2019-12-12 03:42:59 +11:00
lazy_load_filter: ClassVar[Dict[str, Any]] = {
"room": {
"ephemeral": {"lazy_load_members": True},
"state": {"lazy_load_members": True},
"timeline": {"lazy_load_members": True},
"account_data": {"lazy_load_members": True},
},
}
low_limit_filter: ClassVar[Dict[str, Any]] = {
"presence": {"limit": 1},
"room": {
"ephemeral": {"limit": 1},
"timeline": {
"limit": 5,
# This kind says another event was redacted, but we wouldn't
# have it in our model, so nothing would be shown
"not_types": ["m.room.redaction"],
},
},
}
no_profile_events_filter: ClassVar[Dict[str, Any]] = {
"room": {
"timeline": {"not_types": ["m.room.member"]},
},
}
no_unknown_events_filter: ClassVar[Dict[str, Any]] = {
"room": {
"timeline": {
"not_types": [
"m.room.message.feedback",
"m.room.pinned_events",
"m.call.*",
"m.room.third_party_invite",
"m.room.tombstone",
"m.reaction",
],
},
},
}
2019-12-19 04:16:15 +11:00
def __init__(self,
backend,
user: str,
homeserver: str = "https://matrix.org",
device_id: Optional[str] = None) -> None:
if not urlparse(homeserver).scheme:
raise ValueError(
f"homeserver is missing scheme (e.g. https://): {homeserver}",
)
store = Path(backend.appdirs.user_data_dir) / "encryption"
store.mkdir(parents=True, exist_ok=True)
2019-07-09 03:08:46 +10:00
super().__init__(
homeserver = homeserver,
user = user,
device_id = device_id,
store_path = store,
config = nio.AsyncClientConfig(
max_timeout_retry_wait_time = 10,
# TODO: pass a custom encryption DB pickle key?
),
2019-07-09 03:08:46 +10:00
)
self.backend: "Backend" = backend
self.models: ModelStore = self.backend.models
self.open_room: Optional[str] = None
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
self.profile_task: Optional[asyncio.Future] = None
self.server_config_task: Optional[asyncio.Future] = None
self.sync_task: Optional[asyncio.Future] = None
2020-03-08 20:24:07 +11:00
self.upload_monitors: Dict[UUID, nio.TransferMonitor] = {}
self.upload_tasks: Dict[UUID, asyncio.Task] = {}
self.send_message_tasks: Dict[UUID, asyncio.Task] = {}
2020-03-08 20:24:07 +11:00
self.first_sync_done: asyncio.Event = asyncio.Event()
self.first_sync_date: Optional[datetime] = None
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
self.past_tokens: Dict[str, str] = {} # {room_id: token}
self.fully_loaded_rooms: Set[str] = set() # {room_id}
self.loaded_once_rooms: Set[str] = set() # {room_id}
self.cleared_events_rooms: Set[str] = set() # {room_id}
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
self.nio_callbacks = NioCallbacks(self)
def __repr__(self) -> str:
return "%s(user_id=%r, homeserver=%r, device_id=%r)" % (
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
type(self).__name__, self.user_id, self.homeserver, self.device_id,
)
async def _send(self, *args, **kwargs) -> nio.Response:
"""Raise a `MatrixError` subclass for any `nio.ErrorResponse`.
This function is called by `nio.AsyncClient`'s methods to send
requests to the server. Return normal responses, but catch any
`ErrorResponse` to turn them into `MatrixError` exceptions we raise.
"""
response = await super()._send(*args, **kwargs)
if isinstance(response, nio.ErrorResponse):
raise MatrixError.from_nio(response)
return response
@staticmethod
def default_device_name() -> str:
2019-12-19 04:16:15 +11:00
"""Device name to set at login if the user hasn't set a custom one."""
os_name = platform.system()
if not os_name: # unknown OS
return __display_name__
# On Linux, the kernel version is returned, so for a one-time-set
# device name it would quickly be outdated.
os_ver = platform.release() if os_name == "Windows" else ""
return f"{__display_name__} on {os_name} {os_ver}".rstrip()
async def login(self, password: str, device_name: str = "") -> None:
2019-12-19 04:16:15 +11:00
"""Login to the server using the account's password."""
await super().login(
password, device_name or self.default_device_name(),
)
2019-12-19 04:16:15 +11:00
asyncio.ensure_future(self._start())
async def resume(self, user_id: str, token: str, device_id: str) -> None:
2019-12-19 04:16:15 +11:00
"""Login to the server using an existing access token."""
response = nio.LoginResponse(user_id, device_id, token)
await self.receive_response(response)
2019-12-19 04:16:15 +11:00
asyncio.ensure_future(self._start())
2019-11-22 19:27:20 +11:00
async def logout(self) -> None:
2019-12-19 04:16:15 +11:00
"""Logout from the server. This will delete the device."""
tasks = (self.profile_task, self.sync_task, self.server_config_task)
for task in tasks:
2019-11-22 19:27:20 +11:00
if task:
task.cancel()
with suppress(asyncio.CancelledError):
await task
await super().logout()
await self.close()
@property
def syncing(self) -> bool:
2019-12-19 04:16:15 +11:00
"""Return whether this client is currently syncing with the server."""
if not self.sync_task:
return False
return not self.sync_task.done()
2019-12-19 04:16:15 +11:00
async def _start(self) -> None:
"""Fetch our user profile, server config and enter the sync loop."""
2019-12-19 04:16:15 +11:00
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
def on_profile_response(future) -> None:
2019-12-19 04:16:15 +11:00
"""Update our model `Account` with the received profile details."""
if future.cancelled(): # Account logged out
return
exception = future.exception()
if exception:
log.warn("On %s client startup: %r", self.user_id, exception)
self.profile_task = asyncio.ensure_future(
self.backend.get_profile(self.user_id),
)
self.profile_task.add_done_callback(on_profile_response)
2019-11-12 23:47:03 +11:00
return
resp = future.result()
account = self.models["accounts"][self.user_id]
2019-11-12 23:47:03 +11:00
account.profile_updated = datetime.now()
account.display_name = resp.displayname or ""
account.avatar_url = resp.avatar_url or ""
def on_server_config_response(future) -> None:
"""Update our model `Account` with the received config details."""
if future.cancelled(): # Account logged out
return
exception = future.exception()
if exception:
log.warn("On %s client startup: %r", self.user_id, exception)
self.server_config_task = asyncio.ensure_future(
self.get_server_config(),
)
self.server_config_task.add_done_callback(
on_server_config_response,
)
return
account = self.models["accounts"][self.user_id]
account.max_upload_size = future.result()
2019-11-22 19:27:20 +11:00
self.profile_task = asyncio.ensure_future(
self.backend.get_profile(self.user_id),
)
self.profile_task.add_done_callback(on_profile_response)
self.server_config_task = asyncio.ensure_future(
self.get_server_config(),
)
self.server_config_task.add_done_callback(on_server_config_response)
filter1 = deepcopy(self.lazy_load_filter)
utils.dict_update_recursive(filter1, self.low_limit_filter)
cfg = self.backend.ui_settings
if cfg["hideProfileChangeEvents"] or cfg["hideMembershipEvents"]:
utils.dict_update_recursive(filter1, self.no_profile_events_filter)
if cfg["hideUnknownEvents"]:
filter1["room"]["timeline"]["not_types"].extend(
self.no_unknown_events_filter["room"]["timeline"]["not_types"],
)
2019-10-31 02:10:40 +11:00
while True:
try:
self.sync_task = asyncio.ensure_future(self.sync_forever(
timeout = 10_000,
sync_filter = self.lazy_load_filter,
first_sync_filter = filter1,
))
2019-11-22 19:27:20 +11:00
await self.sync_task
break # task cancelled
except Exception as err:
2019-10-31 02:10:40 +11:00
trace = traceback.format_exc().rstrip()
if isinstance(err, MatrixError) and err.http_code >= 500:
log.warning(
"Server failure during sync for %s:\n%s",
self.user_id,
trace,
)
else:
LoopException(str(err), err, trace)
await asyncio.sleep(5)
async def get_server_config(self) -> int:
"""Return the maximum upload size on this server"""
return (await self.content_repository_config()).upload_size
async def can_kick(self, room_id: str, target_user_id: str) -> bool:
"""Return whether we can kick a certain user in a room."""
levels = self.all_rooms[room_id].power_levels
return levels.can_user_kick(self.user_id, target_user_id)
async def can_ban(self, room_id: str, target_user_id: str) -> bool:
"""Return whether we can ban/unbun a certain user in a room."""
levels = self.all_rooms[room_id].power_levels
return levels.can_user_ban(self.user_id, target_user_id)
2019-07-05 16:45:30 +10:00
@property
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
def all_rooms(self) -> Dict[str, nio.MatrixRoom]:
2019-12-19 04:16:15 +11:00
"""Return dict containing both our joined and invited rooms."""
2019-07-05 16:45:30 +10:00
return {**self.invited_rooms, **self.rooms}
async def send_text(self, room_id: str, text: str) -> None:
2019-12-19 04:16:15 +11:00
"""Send a markdown `m.text` or `m.notice` (with `/me`) message ."""
from_md = partial(HTML.from_markdown, room_id=room_id)
2020-03-23 07:21:29 +11:00
2019-07-20 10:55:52 +10:00
escape = False
if text.startswith("//") or text.startswith(r"\/"):
escape = True
text = text[1:]
if text.startswith("/me ") and not escape:
2019-10-29 04:49:55 +11:00
event_type = nio.RoomMessageEmote
2019-07-20 10:55:52 +10:00
text = text[len("/me "): ]
content = {"body": text, "msgtype": "m.emote"}
2020-03-23 07:21:29 +11:00
to_html = from_md(text, inline=True, outgoing=True)
echo_body = from_md(text, inline=True)
2019-07-20 10:55:52 +10:00
else:
2019-10-29 04:49:55 +11:00
event_type = nio.RoomMessageText
2019-07-20 10:55:52 +10:00
content = {"body": text, "msgtype": "m.text"}
2020-03-23 07:21:29 +11:00
to_html = from_md(text, outgoing=True)
echo_body = from_md(text)
2019-07-22 07:41:43 +10:00
if to_html not in (html.escape(text), f"<p>{html.escape(text)}</p>"):
content["format"] = "org.matrix.custom.html"
content["formatted_body"] = to_html
2019-07-04 11:20:49 +10:00
# Can't use the standard Matrix transaction IDs; they're only visible
# to the sender so our other accounts wouldn't be able to replace
# local echoes by real messages.
tx_id = uuid4()
content[f"{__app_name__}.transaction_id"] = str(tx_id)
2019-10-29 04:49:55 +11:00
mentions = HTML.mentions_in_html(echo_body)
await self._local_echo(
room_id, tx_id, event_type, content=echo_body, mentions=mentions,
)
await self._send_message(room_id, content, tx_id)
2019-10-29 06:27:36 +11:00
2020-03-08 20:24:07 +11:00
async def toggle_pause_upload(
self, room_id: str, uuid: Union[str, UUID],
) -> None:
if isinstance(uuid, str):
uuid = UUID(uuid)
pause = not self.upload_monitors[uuid].pause
self.upload_monitors[uuid].pause = pause
self.models[room_id, "uploads"][str(uuid)].paused = pause
async def cancel_upload(self, uuid: Union[str, UUID]) -> None:
if isinstance(uuid, str):
uuid = UUID(uuid)
self.upload_tasks[uuid].cancel()
2020-03-08 20:24:07 +11:00
2019-10-29 06:27:36 +11:00
async def send_file(self, room_id: str, path: Union[Path, str]) -> None:
2019-12-19 04:16:15 +11:00
"""Send a `m.file`, `m.image`, `m.audio` or `m.video` message."""
2019-12-02 20:06:21 +11:00
item_uuid = uuid4()
try:
await self._send_file(item_uuid, room_id, path)
2019-12-06 01:00:23 +11:00
except (nio.TransferCancelledError, asyncio.CancelledError):
2019-12-06 23:44:45 +11:00
log.info("Deleting item for cancelled upload %s", item_uuid)
2020-03-08 20:24:07 +11:00
del self.upload_monitors[item_uuid]
del self.upload_tasks[item_uuid]
del self.models[room_id, "uploads"][str(item_uuid)]
2019-12-02 20:06:21 +11:00
async def _send_file(
self, item_uuid: UUID, room_id: str, path: Union[Path, str],
) -> None:
2020-03-13 05:41:00 +11:00
"""Upload and monitor a file + thumbnail and send the built event."""
2019-12-19 04:16:15 +11:00
# TODO: this function is way too complex, and most of it should be
2019-12-19 04:16:15 +11:00
# refactored into nio.
transaction_id = uuid4()
path = Path(path)
encrypt = room_id in self.encrypted_rooms
try:
size = path.resolve().stat().st_size
except (PermissionError, FileNotFoundError):
# This error will be caught again by the try block later below
size = 0
2019-12-06 23:44:45 +11:00
monitor = nio.TransferMonitor(size)
2020-03-08 20:24:07 +11:00
upload_item = Upload(item_uuid, path, total_size=size)
self.models[room_id, "uploads"][str(item_uuid)] = upload_item
2019-12-06 01:00:23 +11:00
self.upload_monitors[item_uuid] = monitor
self.upload_tasks[item_uuid] = current_task() # type: ignore
def on_transferred(transferred: int) -> None:
upload_item.uploaded = transferred
2019-12-06 01:00:23 +11:00
def on_speed_changed(speed: float) -> None:
upload_item.speed = speed
upload_item.time_left = monitor.remaining_time or timedelta(0)
2019-12-06 01:00:23 +11:00
monitor.on_transferred = on_transferred
monitor.on_speed_changed = on_speed_changed
2019-12-02 17:57:47 +11:00
try:
url, mime, crypt_dict = await self.upload(
2019-12-15 08:40:31 +11:00
lambda *_: path,
filename = path.name,
filesize = size,
2020-03-08 20:24:07 +11:00
encrypt = encrypt,
monitor = monitor,
2019-12-02 17:57:47 +11:00
)
2020-03-08 20:24:07 +11:00
# FIXME: nio might not catch the cancel in time
if monitor.cancel:
raise nio.TransferCancelledError()
except (MatrixError, OSError) as err:
2019-12-02 17:57:47 +11:00
upload_item.status = UploadStatus.Error
upload_item.error = type(err)
upload_item.error_args = err.args
2019-12-02 20:06:21 +11:00
2019-12-06 01:00:23 +11:00
# Wait for cancellation from UI, see parent send_file() method
2019-12-02 20:06:21 +11:00
while True:
await asyncio.sleep(0.1)
upload_item.status = UploadStatus.Caching
await Media.from_existing_file(self.backend.media_cache, url, path)
kind = (mime or "").split("/")[0]
thumb_url: str = ""
thumb_info: Optional[MatrixImageInfo] = None
2019-10-29 06:27:36 +11:00
content: dict = {
f"{__app_name__}.transaction_id": str(transaction_id),
2019-10-29 06:27:36 +11:00
"body": path.name,
"info": {
"mimetype": mime,
"size": upload_item.total_size,
2019-10-29 06:27:36 +11:00
},
}
if encrypt:
content["file"] = {"url": url, **crypt_dict}
else:
content["url"] = url
2019-10-29 06:27:36 +11:00
if kind == "image":
is_svg = mime == "image/svg+xml"
event_type = \
nio.RoomEncryptedImage if encrypt else nio.RoomMessageImage
2019-10-29 06:27:36 +11:00
content["msgtype"] = "m.image"
content["info"]["w"], content["info"]["h"] = (
await utils.svg_dimensions(path) if is_svg else
2019-10-29 06:27:36 +11:00
PILImage.open(path).size
)
2019-10-29 06:27:36 +11:00
try:
thumb_data, thumb_info = await self.generate_thumbnail(
path, is_svg=is_svg,
)
except UneededThumbnail:
pass
except Exception:
trace = traceback.format_exc().rstrip()
log.warning("Failed thumbnailing %s:\n%s", path, trace)
else:
thumb_ext = "png" if thumb_info.mime == "image/png" else "jpg"
thumb_name = f"{path.stem}_thumbnail.{thumb_ext}"
2019-12-07 07:44:25 +11:00
upload_item.status = UploadStatus.Uploading
upload_item.filepath = Path(thumb_name)
upload_item.total_size = len(thumb_data)
2019-12-02 17:57:47 +11:00
try:
upload_item.total_size = thumb_info.size
monitor = nio.TransferMonitor(thumb_info.size)
monitor.on_transferred = on_transferred
monitor.on_speed_changed = on_speed_changed
self.upload_monitors[item_uuid] = monitor
2020-03-08 20:24:07 +11:00
2019-12-02 17:57:47 +11:00
thumb_url, _, thumb_crypt_dict = await self.upload(
2019-12-15 08:40:31 +11:00
lambda *_: thumb_data,
2020-03-10 08:33:07 +11:00
filename = f"{path.stem}_sample{path.suffix}",
filesize = thumb_info.size,
2019-12-02 17:57:47 +11:00
encrypt = encrypt,
2020-03-08 20:24:07 +11:00
monitor = monitor,
2019-12-02 17:57:47 +11:00
)
2020-03-08 20:24:07 +11:00
# FIXME: nio might not catch the cancel in time
if monitor.cancel:
raise nio.TransferCancelledError()
2019-12-02 17:57:47 +11:00
except MatrixError as err:
log.warning(f"Failed uploading thumbnail {path}: {err}")
else:
2019-12-07 07:44:25 +11:00
upload_item.status = UploadStatus.Caching
2019-12-02 17:57:47 +11:00
await Thumbnail.from_bytes(
self.backend.media_cache,
thumb_url,
2020-03-10 08:33:07 +11:00
path.name,
2019-12-02 17:57:47 +11:00
thumb_data,
wanted_size = (content["info"]["w"],
content["info"]["h"]),
)
if encrypt:
content["info"]["thumbnail_file"] = {
"url": thumb_url,
**thumb_crypt_dict,
}
else:
content["info"]["thumbnail_url"] = thumb_url
2019-12-19 04:16:15 +11:00
content["info"]["thumbnail_info"] = thumb_info.as_dict()
2019-10-29 06:27:36 +11:00
elif kind == "audio":
event_type = \
nio.RoomEncryptedAudio if encrypt else nio.RoomMessageAudio
2019-10-29 06:27:36 +11:00
content["msgtype"] = "m.audio"
content["info"]["duration"] = getattr(
MediaInfo.parse(path).tracks[0], "duration", 0,
) or 0
elif kind == "video":
event_type = \
nio.RoomEncryptedVideo if encrypt else nio.RoomMessageVideo
2019-10-29 06:27:36 +11:00
content["msgtype"] = "m.video"
tracks = MediaInfo.parse(path).tracks
content["info"]["duration"] = \
getattr(tracks[0], "duration", 0) or 0
content["info"]["w"] = max(
getattr(t, "width", 0) or 0 for t in tracks
)
content["info"]["h"] = max(
getattr(t, "height", 0) or 0 for t in tracks
)
else:
event_type = \
nio.RoomEncryptedFile if encrypt else nio.RoomMessageFile
2019-10-29 06:27:36 +11:00
content["msgtype"] = "m.file"
content["filename"] = path.name
2020-03-08 20:24:07 +11:00
del self.upload_monitors[item_uuid]
del self.upload_tasks[item_uuid]
del self.models[room_id, "uploads"][str(upload_item.id)]
2019-10-29 06:27:36 +11:00
await self._local_echo(
room_id,
transaction_id,
event_type,
inline_content = path.name,
media_url = url,
media_title = path.name,
media_width = content["info"].get("w", 0),
media_height = content["info"].get("h", 0),
media_duration = content["info"].get("duration", 0),
media_size = content["info"]["size"],
media_mime = content["info"]["mimetype"],
thumbnail_url = thumb_url,
thumbnail_width =
content["info"].get("thumbnail_info", {}).get("w", 0),
thumbnail_height =
content["info"].get("thumbnail_info", {}).get("h", 0),
2020-03-10 03:06:58 +11:00
thumbnail_mime =
content["info"].get("thumbnail_info", {}).get("mimetype", ""),
2019-10-29 06:27:36 +11:00
)
await self._send_message(room_id, content, transaction_id)
2019-10-29 04:49:55 +11:00
async def _local_echo(
2019-12-19 04:16:15 +11:00
self,
room_id: str,
transaction_id: UUID,
event_type: Type[nio.Event],
**event_fields,
2019-10-29 04:49:55 +11:00
) -> None:
2019-12-19 04:16:15 +11:00
"""Register a local model `Event` while waiting for the server.
When the user sends a message, we want to show instant feedback in
the UI timeline without waiting for the servers to receive our message
and retransmit it to us.
The event will be locally echoed for all our accounts that are members
of the `room_id` room.
This allows sending messages from other accounts within the same
composer without having to go to another page in the UI,
and getting direct feedback for these accounts in the timeline.
When we do get the real event retransmited by the server, it will
replace the local one we registered.
"""
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
our_info = self.models[self.user_id, room_id, "members"][self.user_id]
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
content = event_fields.get("content", "").strip()
if content and "inline_content" not in event_fields:
event_fields["inline_content"] = HTML.filter(
content, inline=True, room_id=room_id,
)
2019-10-29 04:49:55 +11:00
event = Event(
id = f"echo-{transaction_id}",
event_id = "",
event_type = event_type,
date = datetime.now(),
sender_id = self.user_id,
sender_name = our_info.display_name,
sender_avatar = our_info.avatar_url,
is_local_echo = True,
links = Event.parse_links(content),
2019-10-29 06:27:36 +11:00
**event_fields,
2019-07-04 11:20:49 +10:00
)
2019-10-29 04:49:55 +11:00
for user_id in self.models["accounts"]:
if user_id in self.models[self.user_id, room_id, "members"]:
key = f"echo-{transaction_id}"
self.models[user_id, room_id, "events"][key] = deepcopy(event)
2019-07-04 11:20:49 +10:00
2019-10-29 04:49:55 +11:00
await self.set_room_last_event(room_id, event)
async def _send_message(
self, room_id: str, content: dict, transaction_id: UUID,
) -> None:
2019-12-19 04:16:15 +11:00
"""Send a message event with `content` dict to a room."""
2019-08-17 02:23:34 +10:00
self.send_message_tasks[transaction_id] = \
current_task() # type: ignore
async with self.backend.send_locks[room_id]:
await self.room_send(
2019-07-09 03:08:46 +10:00
room_id = room_id,
message_type = "m.room.message",
content = content,
ignore_unverified_devices = True,
)
2019-07-05 12:25:06 +10:00
2019-07-04 11:20:49 +10:00
async def load_all_room_members(self, room_id: str) -> None:
"""Request a room's full member list if it hasn't already been loaded.
Member lazy-loading is used to accelerate the initial sync with the
server. This method will be called from QML to load a room's entire
member list when the user is currently viewing the room.
"""
room = self.all_rooms[room_id]
if not room.members_synced:
await super().joined_members(room_id)
await self.register_nio_room(self.all_rooms[room_id])
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
async def load_past_events(self, room_id: str) -> bool:
"""Ask the server for 100 previous events of the room.
2019-12-19 04:16:15 +11:00
Events from before the client was started will be requested and
registered into our models.
Returns whether there are any messages left to load.
"""
if room_id in self.fully_loaded_rooms or \
room_id in self.invited_rooms or \
room_id in self.cleared_events_rooms or \
self.models[self.user_id, "rooms"][room_id].left:
2019-07-05 16:45:30 +10:00
return False
2019-08-28 05:00:50 +10:00
await self.first_sync_done.wait()
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
2019-12-15 04:50:21 +11:00
while not self.past_tokens.get(room_id):
# If a new room was added, wait for onSyncResponse to set the token
await asyncio.sleep(0.1)
2019-07-05 16:45:30 +10:00
response = await self.room_messages(
room_id = room_id,
start = self.past_tokens[room_id],
limit = 100 if room_id in self.loaded_once_rooms else 20,
message_filter = self.lazy_load_filter,
2019-07-05 16:45:30 +10:00
)
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
self.loaded_once_rooms.add(room_id)
2019-07-05 16:45:30 +10:00
more_to_load = True
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
self.past_tokens[room_id] = response.end
2019-07-05 16:45:30 +10:00
for event in response.chunk:
if isinstance(event, nio.RoomCreateEvent):
self.fully_loaded_rooms.add(room_id)
more_to_load = False
2019-07-05 16:45:30 +10:00
for cb in self.event_callbacks:
if (cb.filter is None or isinstance(event, cb.filter)):
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
await cb.func(self.all_rooms[room_id], event)
2019-07-05 16:45:30 +10:00
return more_to_load
2019-11-10 05:20:53 +11:00
async def new_direct_chat(self, invite: str, encrypt: bool = False) -> str:
2019-12-19 04:16:15 +11:00
"""Create a room and invite a single user in it for a direct chat."""
2019-11-10 05:20:53 +11:00
if invite == self.user_id:
raise InvalidUserInContext(invite)
2019-12-12 03:42:59 +11:00
if not self.user_id_regex.match(invite):
2019-12-05 00:59:14 +11:00
raise InvalidUserId(invite)
# Raise MatrixNotFound if profile doesn't exist
await self.get_profile(invite)
2019-11-10 05:20:53 +11:00
2019-12-30 05:30:15 +11:00
response = await super().room_create(
2019-11-10 05:20:53 +11:00
invite = [invite],
is_direct = True,
visibility = nio.RoomVisibility.private,
initial_state =
[nio.EnableEncryptionBuilder().as_dict()] if encrypt else [],
2019-12-30 05:30:15 +11:00
)
return response.room_id
2019-11-10 05:20:53 +11:00
async def new_group_chat(
2019-11-09 06:32:12 +11:00
self,
name: Optional[str] = None,
topic: Optional[str] = None,
public: bool = False,
encrypt: bool = False,
2019-11-09 06:32:12 +11:00
federate: bool = True,
) -> str:
2019-12-19 04:16:15 +11:00
"""Create a new matrix room with the purpose of being a group chat."""
2019-11-09 06:32:12 +11:00
2019-12-30 05:30:15 +11:00
response = await super().room_create(
name = name or None,
topic = topic or None,
2019-11-09 06:32:12 +11:00
federate = federate,
visibility =
nio.RoomVisibility.public if public else
nio.RoomVisibility.private,
initial_state =
[nio.EnableEncryptionBuilder().as_dict()] if encrypt else [],
2019-12-30 05:30:15 +11:00
)
return response.room_id
2019-11-09 06:32:12 +11:00
2019-11-10 00:52:16 +11:00
async def room_join(self, alias_or_id_or_url: str) -> str:
2019-12-19 04:16:15 +11:00
"""Join an existing matrix room."""
2019-11-10 00:52:16 +11:00
string = alias_or_id_or_url.strip()
if self.http_s_url_regex.match(string):
2019-11-10 00:52:16 +11:00
for part in urlparse(string).fragment.split("/"):
2019-12-12 03:42:59 +11:00
if self.room_id_or_alias_regex.match(part):
2019-11-10 00:52:16 +11:00
string = part
break
else:
raise ValueError(f"No alias or room id found in url {string}")
2019-12-12 03:42:59 +11:00
if not self.room_id_or_alias_regex.match(string):
raise ValueError("Not an alias or room id")
2019-12-30 05:30:15 +11:00
response = await super().join(string)
return response.room_id
2019-11-10 00:52:16 +11:00
2019-07-08 12:19:17 +10:00
async def room_forget(self, room_id: str) -> None:
2019-12-19 04:16:15 +11:00
"""Leave a joined room (or decline an invite) and forget its history.
If all the members of a room leave and forget it, that room
will be marked as suitable for destruction by the server.
"""
self.models[self.user_id, "rooms"].pop(room_id, None)
self.models["every_room"].pop((self.user_id, room_id), None)
self.models.pop((self.user_id, room_id, "events"), None)
self.models.pop((self.user_id, room_id, "members"), None)
try:
await super().room_leave(room_id)
except MatrixNotFound: # already left
pass
2019-07-08 12:19:17 +10:00
await super().room_forget(room_id)
2019-12-12 03:42:59 +11:00
async def room_mass_invite(
self, room_id: str, *user_ids: str,
) -> Tuple[List[str], List[Tuple[str, Exception]]]:
2019-12-19 04:16:15 +11:00
"""Invite users to a room in parallel.
Returns a tuple with:
- A list of users we successfully invited
- A list of `(user_id, Exception)` tuples for those failed to invite.
"""
2019-12-12 03:42:59 +11:00
user_ids = tuple(
uid for uid in user_ids
# Server would return a 403 forbidden for users already in the room
if uid not in self.all_rooms[room_id].users
)
async def invite(user_id: str):
if not self.user_id_regex.match(user_id):
return InvalidUserId(user_id)
if not self.rooms[room_id].federate:
_, user_server = user_id.split(":", maxsplit=1)
_, room_server = room_id.split(":", maxsplit=1)
user_server = re.sub(r":443$", "", user_server)
room_server = re.sub(r":443$", "", room_server)
if user_server != room_server:
return UserFromOtherServerDisallowed(user_id)
2019-12-12 03:42:59 +11:00
try:
await self.get_profile(user_id)
except MatrixNotFound as err:
return err
2019-12-12 03:42:59 +11:00
return await self.room_invite(room_id, user_id)
2019-12-12 03:42:59 +11:00
coros = [invite(uid) for uid in user_ids]
successes = []
errors: list = []
responses = await asyncio.gather(*coros)
for user_id, response in zip(user_ids, responses):
if isinstance(response, nio.RoomInviteError):
errors.append((user_id, MatrixError.from_nio(response)))
elif isinstance(response, Exception):
errors.append((user_id, response))
else:
successes.append(user_id)
return (successes, errors)
async def get_redacted_event_content(
self,
nio_type: Type[nio.Event],
redacter: str,
sender: str,
reason: str = "",
) -> str:
"""Get content to be displayed in place of a redacted event."""
kind = (
"message" if issubclass(nio_type, nio.RoomMessage) else
"media" if issubclass(nio_type, nio.RoomMessageMedia) else
"event"
)
content = f"%1 removed this {kind}" if redacter == sender else \
f"%1's {kind} was removed by %2"
if reason:
content = f"{content}, reason: {reason}"
return content
2020-03-27 07:31:57 +11:00
async def room_mass_redact(
self, room_id: str, reason: str, *event_client_ids: str,
) -> List[nio.RoomRedactResponse]:
"""Redact events from a room in parallel."""
2020-03-27 07:31:57 +11:00
tasks = []
for user_id in self.backend.clients:
for client_id in event_client_ids:
event = self.models[user_id, room_id, "events"].get(client_id)
if not event:
continue
if event.is_local_echo:
if user_id == self.user_id:
uuid = UUID(event.id.replace("echo-", ""))
self.send_message_tasks[uuid].cancel()
event.is_local_echo = False
else:
if user_id == self.user_id:
tasks.append(
self.room_redact(room_id, event.event_id, reason),
)
event.is_local_echo = True
event.content = await self.get_redacted_event_content(
event.event_type, self.user_id, event.sender_id, reason,
)
event.event_type = nio.RedactedEvent
return await asyncio.gather(*tasks)
2019-12-12 03:42:59 +11:00
async def generate_thumbnail(
self, data: UploadData, is_svg: bool = False,
) -> Tuple[bytes, MatrixImageInfo]:
2019-12-19 04:16:15 +11:00
"""Create a thumbnail from an image, return the bytes and info."""
png_modes = ("1", "L", "P", "RGBA")
data = b"".join([c async for c in async_generator_from_data(data)])
is_svg = await utils.guess_mime(data) == "image/svg+xml"
if is_svg:
svg_width, svg_height = await utils.svg_dimensions(data)
data = cairosvg.svg2png(
bytestring = data,
parent_width = svg_width,
parent_height = svg_height,
)
thumb = PILImage.open(io.BytesIO(data))
small = thumb.width <= 800 and thumb.height <= 600
is_jpg_png = thumb.format in ("JPEG", "PNG")
jpgable_png = thumb.format == "PNG" and thumb.mode not in png_modes
2019-07-16 06:14:08 +10:00
if small and is_jpg_png and not jpgable_png and not is_svg:
raise UneededThumbnail()
if not small:
thumb.thumbnail((800, 600))
with io.BytesIO() as out:
if thumb.mode in png_modes:
thumb.save(out, "PNG", optimize=True)
mime = "image/png"
else:
thumb.convert("RGB").save(out, "JPEG", optimize=True)
mime = "image/jpeg"
thumb_data = out.getvalue()
thumb_size = len(thumb_data)
2020-03-10 01:03:22 +11:00
if thumb_size >= len(data) and is_jpg_png and not is_svg:
raise UneededThumbnail()
info = MatrixImageInfo(thumb.width, thumb.height, mime, thumb_size)
return (thumb_data, info)
async def upload(
self,
2019-12-15 08:40:31 +11:00
data_provider: nio.DataProvider,
filename: Optional[str] = None,
filesize: Optional[int] = None,
mime: Optional[str] = None,
2019-12-15 08:40:31 +11:00
encrypt: bool = False,
monitor: Optional[nio.TransferMonitor] = None,
) -> UploadReturn:
2019-12-19 04:16:15 +11:00
"""Upload a file to the matrix homeserver."""
if filesize > self.models["accounts"][self.user_id].max_upload_size:
raise MatrixTooLarge()
2019-12-15 08:40:31 +11:00
mime = mime or await utils.guess_mime(data_provider(0, 0))
response, decryption_dict = await super().upload(
2019-12-15 08:40:31 +11:00
data_provider,
"application/octet-stream" if encrypt else mime,
filename,
encrypt,
2019-12-06 01:00:23 +11:00
monitor,
filesize,
)
2019-07-16 06:14:08 +10:00
return UploadReturn(response.content_uri, mime, decryption_dict)
2019-07-16 06:14:08 +10:00
async def set_avatar_from_file(self, path: Union[Path, str]) -> None:
2019-12-19 04:16:15 +11:00
"""Upload an image to the homeserver and set it as our avatar."""
path = Path(path)
mime = await utils.guess_mime(path)
if mime.split("/")[0] != "image":
raise BadMimeType(wanted="image/*", got=mime)
mxc, *_ = await self.upload(
data_provider = lambda *_: path,
filename = path.name,
filesize = path.resolve().stat().st_size,
mime = mime,
)
await self.set_avatar(mxc)
2019-07-16 06:14:08 +10:00
2019-08-29 01:42:52 +10:00
async def import_keys(self, infile: str, passphrase: str) -> None:
2019-12-19 04:16:15 +11:00
"""Import decryption keys from a file, then retry decrypting events."""
2019-11-24 02:14:14 +11:00
await super().import_keys(infile, passphrase)
await self.retry_decrypting_events()
2019-08-28 17:54:53 +10:00
2019-09-08 09:17:32 +10:00
async def export_keys(self, outfile: str, passphrase: str) -> None:
2019-12-19 04:16:15 +11:00
"""Export our decryption keys to a file."""
2019-09-08 09:17:32 +10:00
path = Path(outfile)
path.parent.mkdir(parents=True, exist_ok=True)
# The QML dialog asks the user if he wants to overwrite before this
if path.exists():
path.unlink()
2019-09-08 09:17:32 +10:00
await super().export_keys(outfile, passphrase)
async def retry_decrypting_events(self) -> None:
2019-12-19 04:16:15 +11:00
"""Retry decrypting room `Event`s in our model we failed to decrypt."""
for sync_id, model in self.models.items():
if not (isinstance(sync_id, tuple) and
2020-03-16 07:00:29 +11:00
len(sync_id) == 3 and
sync_id[0] == self.user_id and
sync_id[2] == "events"):
continue
2020-03-16 07:00:29 +11:00
_, room_id, _ = sync_id
for ev in deepcopy(model).values():
room = self.all_rooms[room_id]
if isinstance(ev.source, nio.MegolmEvent):
try:
decrypted = self.decrypt_event(ev.source)
if not decrypted:
raise nio.EncryptionError()
except nio.EncryptionError:
continue
for cb in self.event_callbacks:
if not cb.filter or isinstance(decrypted, cb.filter):
await asyncio.coroutine(cb.func)(room, decrypted)
async def clear_events(self, room_id: str) -> None:
2020-03-13 05:41:00 +11:00
"""Remove every `Event` of a room we registered in our model.
2019-12-19 04:16:15 +11:00
The events will be gone from the UI, until the client is restarted.
"""
self.cleared_events_rooms.add(room_id)
model = self.models[self.user_id, room_id, "events"]
2019-09-09 01:49:47 +10:00
if model:
model.clear()
self.models[self.user_id, "rooms"][room_id].last_event_date = \
ZeroDate
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
# Functions to register data into models
async def event_is_past(self, ev: Union[nio.Event, Event]) -> bool:
2019-12-19 04:16:15 +11:00
"""Return whether an event was created before this client started."""
if not self.first_sync_date:
return True
if isinstance(ev, Event):
return ev.date < self.first_sync_date
date = datetime.fromtimestamp(ev.server_timestamp / 1000)
return date < self.first_sync_date
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
async def set_room_last_event(self, room_id: str, item: Event) -> None:
2019-12-19 04:16:15 +11:00
"""Set the `last_event` for a `Room` using data in our `Event` model.
The `last_event` is notably displayed in the UI room subtitles.
"""
room = self.models[self.user_id, "rooms"][room_id]
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
if item.date > room.last_event_date:
room.last_event_date = item.date
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
2019-12-19 04:16:15 +11:00
async def register_nio_room(
self, room: nio.MatrixRoom, left: bool = False,
) -> None:
"""Register a `nio.MatrixRoom` as a `Room` object in our model."""
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
# Add room
inviter = getattr(room, "inviter", "") or ""
levels = room.power_levels
can_send_state = partial(levels.can_user_send_state, self.user_id)
can_send_msg = partial(levels.can_user_send_message, self.user_id)
try:
registered = self.models[self.user_id, "rooms"][room.room_id]
last_event_date = registered.last_event_date
typing_members = registered.typing_members
2020-03-23 14:55:48 +11:00
mentions = registered.mentions
unreads = registered.unreads
except KeyError:
last_event_date = datetime.fromtimestamp(0)
typing_members = []
2020-03-23 14:55:48 +11:00
mentions = 0
unreads = 0
room_item = Room(
id = room.room_id,
for_account = self.user_id,
given_name = room.name or "",
2019-11-28 01:03:49 +11:00
display_name = room.display_name or "",
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
avatar_url = room.gen_avatar_url or "",
plain_topic = room.topic or "",
topic = HTML.filter(
room.topic or "", inline=True, room_id=room.room_id,
),
inviter_id = inviter,
inviter_name = room.user_name(inviter) if inviter else "",
inviter_avatar =
(room.avatar_url(inviter) or "") if inviter else "",
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
left = left,
2019-12-12 03:42:59 +11:00
typing_members = typing_members,
encrypted = room.encrypted,
invite_required = room.join_rule == "invite",
guests_allowed = room.guest_access == "can_join",
can_invite = levels.can_user_invite(self.user),
2020-04-20 01:12:35 +10:00
can_kick = levels.can_user_kick(self.user),
can_redact_all = levels.can_user_redact(self.user),
can_send_messages = can_send_msg(),
can_set_name = can_send_state("m.room.name"),
can_set_topic = can_send_state("m.room.topic"),
can_set_avatar = can_send_state("m.room.avatar"),
can_set_encryption = can_send_state("m.room.encryption"),
can_set_join_rules = can_send_state("m.room.join_rules"),
can_set_guest_access = can_send_state("m.room.guest_access"),
2019-12-12 03:42:59 +11:00
last_event_date = last_event_date,
2020-03-23 14:55:48 +11:00
mentions = mentions,
unreads = unreads,
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
)
self.models[self.user_id, "rooms"][room.room_id] = room_item
self.models["every_room"][self.user_id, room.room_id] = room_item
# List members that left the room, then remove them from our model
left_the_room = [
user_id
for user_id in self.models[self.user_id, room.room_id, "members"]
if user_id not in room.users
]
for user_id in left_the_room:
del self.models[self.user_id, room.room_id, "members"][user_id]
HTML.rooms_user_id_names[room.room_id].pop(user_id, None)
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
# Add the room members to the added room
new_dict = {
user_id: Member(
id = user_id,
2019-12-13 01:03:39 +11:00
display_name = room.user_name(user_id) # disambiguated
if member.display_name else "",
avatar_url = member.avatar_url or "",
typing = user_id in room.typing_users,
power_level = member.power_level,
invited = member.invited,
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
) for user_id, member in room.users.items()
}
self.models[self.user_id, room.room_id, "members"].update(new_dict)
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
for user_id, member in room.users.items():
if member.display_name:
HTML.rooms_user_id_names[room.room_id][user_id] = \
member.display_name
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
2019-12-19 04:16:15 +11:00
async def get_member_name_avatar(
self, room_id: str, user_id: str,
) -> Tuple[str, str]:
"""Return a room member's display name and avatar.
If the member isn't found in the room (e.g. they left), their
profile is retrieved using `MatrixClient.backend.get_profile()`.
"""
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
try:
item = self.models[self.user_id, room_id, "members"][user_id]
2019-12-19 04:16:15 +11:00
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
except KeyError: # e.g. user is not anymore in the room
2019-11-12 23:47:03 +11:00
try:
info = await self.backend.get_profile(user_id)
return (info.displayname or "", info.avatar_url or "")
2019-12-19 04:16:15 +11:00
2019-11-12 23:47:03 +11:00
except MatrixError:
return ("", "")
2019-12-19 04:16:15 +11:00
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
else:
return (item.display_name, item.avatar_url)
async def register_nio_event(
self,
room: nio.MatrixRoom,
ev: nio.Event,
event_id: str = "",
**fields,
) -> None:
"""Register a `nio.Event` as a `Event` object in our model."""
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
await self.register_nio_room(room)
sender_name, sender_avatar = \
await self.get_member_name_avatar(room.room_id, ev.sender)
target_id = getattr(ev, "state_key", "") or ""
target_name, target_avatar = \
await self.get_member_name_avatar(room.room_id, target_id) \
if target_id else ("", "")
content = fields.get("content", "").strip()
if content and "inline_content" not in fields:
fields["inline_content"] = HTML.filter(
content, inline=True, room_id=room.room_id,
)
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
# Create Event ModelItem
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
item = Event(
id = event_id or ev.event_id,
event_id = ev.event_id,
event_type = type(ev),
source = ev,
date = datetime.fromtimestamp(ev.server_timestamp / 1000),
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
sender_id = ev.sender,
sender_name = sender_name,
sender_avatar = sender_avatar,
target_id = target_id,
target_name = target_name,
target_avatar = target_avatar,
links = Event.parse_links(content),
2019-09-14 12:34:20 +10:00
**fields,
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
)
# Add the Event to model
model = self.models[self.user_id, room.room_id, "events"]
tx_id = ev.source.get("content", {}).get(
f"{__app_name__}.transaction_id",
)
from_us = ev.sender in self.backend.clients
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
if from_us and tx_id and f"echo-{tx_id}" in model:
item.id = f"echo-{tx_id}"
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
model[item.id] = item
Big performance refactoring & various improvements Instead of passing all sorts of events for the JS to handle and manually add to different data models, we now handle everything we can in Python. For any change, the python models send a sync event with their contents (no more than 4 times per second) to JS, and the QSyncable library's JsonListModel takes care of converting it to a QML ListModel and sending the appropriate signals. The SortFilterProxyModel library is not used anymore, the only case where we need to filter/sort something now is when the user interacts with the "Filter rooms" or "Filter members" fields. These cases are handled by a simple JS function. We now keep separated room and timeline models for different accounts, the previous approach of sharing all the data we could between accounts created a lot of complications (local echoes, decrypted messages replacing others, etc). The users's own account profile changes are now hidden in the timeline. On startup, if all events for a room were only own profile changes, more events will be loaded. Any kind of image format supported by Qt is now handled by the pyotherside image provider, instead of just PNG/JPG. SVGs which previously caused errors are supported as well. The typing members bar paddings/margins are fixed. The behavior of the avatar/"upload a profile picture" overlay is fixed. Config files read from disk are now cached (TODO: make them reloadable again). Pylint is not used anymore because of all its annoying false warnings and lack of understanding for dataclasses, it is replaced by flake8 with a custom config and various plugins. Debug mode is now considered on if the program was compiled with the right option, instead of taking an argument from CLI. When on, C++ will set a flag in the Window QML component. The loading screen is now unloaded after the UI is ready, where previously it just stayed in the background invisible and wasted CPU. The overall refactoring and improvements make us now able to handle rooms with thousand of members and no lazy-loading, where previously everything would freeze and simply scrolling up to load past events in any room would block the UI for a few seconds.
2019-08-11 22:01:22 +10:00
await self.set_room_last_event(room.room_id, item)
# Alerts
if from_us or await self.event_is_past(ev):
return
AlertRequested()
if self.open_room != room.room_id:
account = self.models["accounts"][self.user_id]
room = self.models[self.user_id, "rooms"][room.room_id]
account.total_unread += 1
room.unreads += 1
content = fields.get("content", "")
if HTML.user_id_link_in_html(content, self.user_id):
account.total_mentions += 1
room.mentions += 1