Add StreamReader.readinto()

[tontinton]

BPO	41305
Nosy	@njsmith, @asvetlov, @1st1, @aeros, @tontinton
PRs	bpo-41305: Add StreamReader.readinto() #21491

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = None
created_at = <Date 2020-07-15.16:16:43.987>
labels = ['type-feature', '3.10', 'expert-asyncio']
title = 'Add StreamReader.readinto()'
updated_at = <Date 2020-07-16.21:34:43.883>
user = 'https://github.com/tontinton'

bugs.python.org fields:

activity = <Date 2020-07-16.21:34:43.883>
actor = 'yselivanov'
assignee = 'none'
closed = False
closed_date = None
closer = None
components = ['asyncio']
creation = <Date 2020-07-15.16:16:43.987>
creator = 'tontinton'
dependencies = []
files = []
hgrepos = []
issue_num = 41305
keywords = ['patch']
message_count = 10.0
messages = ['373702', '373714', '373719', '373720', '373732', '373750', '373765', '373770', '373772', '373773']
nosy_count = 5.0
nosy_names = ['njs', 'asvetlov', 'yselivanov', 'aeros', 'tontinton']
pr_nums = ['21491']
priority = 'normal'
resolution = None
stage = 'patch review'
status = 'open'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue41305'
versions = ['Python 3.10']

[tontinton]

Add a StreamReader.readinto(buf) function.

Exactly like StreamReader.read() with *n* being equal to the length of buf.

Instead of allocating a new buffer, copy the read buffer into buf.

[1st1]

We don't want to extend StreamReader with new APIs as ultimately we plan to deprecate it. A new streams API is needed, perhaps inspired by Trio. Sadly, I'm -1 on this one.

[tontinton]

ok.

Im interested in learning about the new api.
Is it documented somewhere?

[tontinton]

Ah it's trio...

[1st1]

Im interested in learning about the new api.

There are two problems with the current API:

1. Reader and writer are separate objects, while they should be one.
2. Writer.write is not a coroutine, although it should be one.

There are other minor nits, but that's the crux of the problem. So we need a new streams design + a new set of APIs to work with it (and streams are in many places, like in the subprocess APIs).

Trio was going to stabilize their own streaming API and we thought it would be great if our new API was compatible with it (not sure if they did stabilize it or not).

I was going to lead the project myself (and still am) but dropped the ball and we missed 3.9 to do this. If you want to start working on this I'd be glad to assist with discussions & reviews.

[tontinton]

Ok actually that sounds really important, I am interested.

But to begin doing something like this I need to know what's the general design.

Is it simply combining stream reader and stream writer into a single object and changing the write() function to always wait the write (thus deprecating drain) and that's it?

If there is not more to it I can probably do this pretty quickly, I mean it seems easy on the surface.

If there is more to it then I would like a more thorough explanation. Maybe we should chat about this.

[1st1]

Is it simply combining stream reader and stream writer into a single object and changing the write() function to always wait the write (thus deprecating drain) and that's it?

Pretty much. We might also rename a few APIs here and there, like "close()" should become an "async def aclose()"

We also will likely want to define a few ABCs.

Which brings me to the most important point: what we need it not coding it (yet), but rather drafting the actual proposal and posting it to https://discuss.python.org/c/async-sig/20. Once a formal proposal is there we can proceed with the implementation.

It's a bit of a process to follow, but we have to do it this way.

[tontinton]

Which brings me to the most important point: what we need it not coding it (yet), but rather drafting the actual proposal and posting it to https://discuss.python.org/c/async-sig/20. Once a formal proposal is there we can proceed with the implementation.

Posted: https://discuss.python.org/t/discussion-on-a-new-api-for-asyncio/4725

By the way I know it's unrelated but I want a CR on #21446 I think it's also very important.

[tontinton]

By the way if we will eventually combine StreamReader and StreamWriter won't this function (readinto) be useful then?

Maybe we should consider adding it right now.

Tell me your thoughts on this.

[1st1]

By the way if we will eventually combine StreamReader and StreamWriter won't this function (readinto) be useful then?

Yes. But StreamReader and StreamWriter will stay for the foreseeable future for backwards compatibility pretty much frozen in time.

So I'd like to start treating them as legacy APIs now.

[gvanrossum]

Okay, let's close this issue -- we're not going to add readinto(). I also left a note on the 2-year-old Discourse thread about the proposed new streams API, trying to revive that.