Lonami Exo
df1dfdf8ea
Remove some redundant except
2018-06-14 16:13:46 +02:00
Lonami Exo
c9ea1bafc0
Apply @andr-04 asyncio commits to TcpClient
2018-06-14 16:08:23 +02:00
Lonami Exo
64dd957189
Fix None first_query and TcpClient.disconnect()
2018-06-11 19:51:01 +02:00
Lonami Exo
7e68274f26
Keep consistent structure and remove done TODO
2018-06-09 15:42:10 +02:00
Lonami Exo
be279ce3f5
Make TLMessage always have a valid TLObject
...
This simplifies the flow instead of having separate request/body
attributes, and also means that BinaryReader.tgread_object() can
be used without so many special cases.
2018-06-09 13:48:27 +02:00
Lonami Exo
f7e8907c6f
Create RpcResult class and generalise core special cases
...
This results in a cleaner MTProtoSender, which now can always
read a TLObject with a guaranteed item, if the message is OK.
2018-06-09 13:13:55 +02:00
Lonami Exo
9477c75fce
Fix basic requests sending and receiving
2018-06-06 21:42:48 +02:00
Lonami Exo
e469258ab9
Create a new MTProtoSender structure and its foundation
...
This means that the TcpClient and the Connection (currently only
ConnectionTcpFull) will no longer be concerned about handling
errors, but the MTProtoSender will.
The foundation of the library will now be based on asyncio.
2018-06-06 20:41:01 +02:00
Lonami Exo
2fb5215f5f
Fix parsers misbehaving with None text
2018-06-03 13:48:43 +02:00
Lonami Exo
8d7c7a19c0
Add some setters for custom.Message
2018-06-03 11:53:18 +02:00
Lonami Exo
e2a0de1913
Don't retry forever on TcpClient.connect()
2018-04-25 10:06:11 +02:00
Lonami Exo
eabaa3854a
Replace offset with match.start() to allow custom regex
2018-04-03 13:47:40 +02:00
Lonami Exo
33e908de42
Fix markdown regex not supporting [] inside URLs
2018-03-22 19:02:08 +01:00
Lonami Exo
987cf41ec6
Higher timeout and log them as warning if any data was received
...
This might be the cause for "number of retries reached 0" so
more specific logging calls might be useful. If while reading
a response it times out but some data had already been read,
said data will be lost.
The sequence of events that triggered reaching 0 retries was:
- Sending requests with IDs XYZ
- socket.timeout while reading
- Items timed out. Retrying
- Processing RPC result
- Received response for XYZ
- Lost request with ID XYZ
2018-03-18 20:08:00 +01:00
Lonami Exo
a7f98fd3cc
Ignore bad file descriptor while closing
2018-03-02 17:26:42 +01:00
Lonami Exo
229cd78df0
Fix markdown's URL regex not acceping newlines
2018-02-27 14:10:02 +01:00
Lonami Exo
6f16aeb553
Add logging calls on the TcpClient
2018-02-26 13:41:07 +01:00
Dmitry Bukhta
0731a1d698
Raise ProxyConnectionError instead looping forever ( #621 )
...
We shouldn't try reconnecting when using a proxy if what's
unavailable is the proxy server (and not Telegram servers).
2018-02-20 17:05:02 +01:00
Lonami Exo
83d9d1d78e
Fix markdown parser not inverting delimiters dict
2018-02-16 20:30:19 +01:00
Lonami Exo
75d99fbb53
Fix HTML entity parsing failing when needing surrogates
2018-02-15 11:52:46 +01:00
Lonami Exo
2ffe2b71dc
Except OSError with errno.WSAEACCES when connecting
...
"OSError: [WinError 10013] An attempt was made to access a
socket in a way forbidden by its access permissions."
2018-02-03 11:44:05 +01:00
frizzlywitch
3b8365f871
Remove square braces from IPv6 addresses ( #561 )
2018-01-26 14:38:13 +01:00
Tulir Asokan
5f2f04c6c2
Add HTML parse mode ( #554 )
2018-01-22 10:06:11 +01:00
Lonami Exo
33e50aaee1
Reuse .on_response/.__str__/.stringify, override iff necessary
2018-01-19 12:12:52 +01:00
Lonami Exo
46b088d44c
Also handle ECONNREFUSED on .connect() (report on #392 )
2018-01-08 12:26:32 +01:00
Lonami Exo
59a1a6aef2
Stop working with bytes on the markdown parser
2018-01-07 16:19:41 +01:00
Lonami Exo
3eafe18d0b
Implement MtProto 2.0 ( closes #484 , thanks @delivrance!)
...
Huge shoutout to @delivrance's pyrogram, specially this commit:
pyrogram/pyrogram/commit/42f9a2d6994baaf9ecad590d1ff4d175a8c56454
2018-01-06 01:55:11 +01:00
Lonami Exo
ec4ca5dbfc
More consistent with asyncio branch (style/small fixes)
...
Like passing an extra (invalid) dt parameter when serializing
a datetime, and handling more errors in the TcpClient class.
2018-01-05 18:31:48 +01:00
Dmitry D. Chernov
6ec6967ff9
Make exception types correspond to Python docs
2017-12-28 09:48:02 +10:00
Dmitry D. Chernov
4a2a64ce2f
TcpClient: Catch ConnectionError instead of its particular cases
...
That can be more reliable, especially in the case of using PySocks.
2017-12-23 05:45:23 +10:00
Lonami Exo
992017ddf8
Except ConnectionAbortedError on TcpClient
2017-12-22 11:27:57 +01:00
Lonami Exo
605c103f29
Add unparse markdown method
2017-11-26 17:16:59 +01:00
Lonami Exo
57a70d0d47
Document the extensions/ module
2017-11-26 17:14:28 +01:00
Dmitry D. Chernov
f99d14558f
binary_reader: Parse TL 'date' to UTC datetime instead of local
2017-11-23 02:06:43 +10:00
Lonami Exo
9767774147
Fix import in markdown parser not being relative
2017-11-17 15:57:48 +01:00
Lonami Exo
346c5bb303
Add method to md parser to extract text surrounded by entities
2017-11-16 19:13:13 +01:00
Lonami Exo
e5deaf5db8
Fix c4e07cf, md parsing adding unfinished entity at wrong offset
2017-11-16 19:07:53 +01:00
Lonami Exo
4ddbc78699
Ensure IPv6 addresses are surrounded by '[]' ( #425 )
2017-11-16 13:47:15 +01:00
Lonami
48e96ca15f
Fix ._get_connected failing when .fileno() == 0 ( #427 )
2017-11-14 12:01:33 +01:00
Andrey Egorov
84d48ef7bf
Safer check to determine whether sockets are connected ( #427 )
2017-11-12 16:51:32 +01:00
Lonami Exo
c4e07cff57
Fix unfinished markdown delimiters being stripped away
2017-11-10 11:44:27 +01:00
Lonami Exo
cb3f20db65
Clean up markdown parsing since tuples aren't used anymore
2017-11-10 11:41:49 +01:00
Lonami Exo
7d75eebdab
Make markdown parser use only Telegram's MessageEntity's
2017-11-10 11:07:36 +01:00
Lonami Exo
83af705cc8
Add more comments to the markdown parser
2017-11-06 11:32:40 +01:00
Lonami Exo
3a2c3a9497
Fix URL regex for markdown was greedy (fix-up)
2017-11-06 11:22:58 +01:00
Lonami Exo
07ece83aba
Fix overlapping markdown entities being skipped
2017-11-06 10:37:22 +01:00
Lonami Exo
4f80429215
Work on byte level when parsing markdown
...
Reasoning: instead encoding every character one by one as we
encounter them to use half their length as the correct offset,
we can simply encode the whole string at once as utf-16le and
work with that directly.
2017-11-06 10:29:32 +01:00
Viktor Oreshkin
49eb281251
Proper offset calculation for markdown ( #407 )
...
Dan suca
If Dan shared it with Traitor I'll not have to spend my time on this
Not a, sorry for not letting you sleep
k thx bye
Will this stay in history?
2017-11-06 00:17:22 +01:00
Lonami Exo
82cac4836c
Fix markdown URL parsing using character index instead offset
2017-10-30 11:15:53 +01:00
Lonami Exo
0a14aa1bc6
Remove additional check when calculating emojies length
...
This special check treated some emojies as 3 characters long but
this shouldn't have actually been done, likely due to the old
regex matching more things as emoji than it should (which would
have count as 2 too, making up for 1+3 from the new is_emoji()).
2017-10-30 10:56:39 +01:00
