Feature request: automatic string decoding for writable streams #7315
Labels
feature request
Issues that request new features to be added to Node.js.
stream
Issues and PRs related to the stream subsystem.
Comments
mscdex
added
stream
4 tasks
addaleax
added a commit
to addaleax/node
that referenced
this issue
Jan 6, 2017
Support decoding the input of writable streams to a specific decoding before passing it to `_write()`. By default, all data written to a writable stream is encoded into Buffers. This change enables the reverse situation, i.e. when it is desired by the stream implementer to process all input as strings, whether it was passed to `write()` as a string or not. This makes sense for multi-byte character encodings where the buffers that are written using `write()` may contain partial characters, so calling `chunk.toString()` is not universally applicable. Fixes: nodejs#7315
|
@nodejs/streams |
|
This would be very nice to have, if someone would like to take it. Happy to coach, but it is tricky to implement. #7425 has an original implementation which we need to be rechecked for performance regressions. If there are none, we are 👍 . A very important thing to do would be to document how to do the boilerplate. See https://github.com/mcollina/split2/blob/master/index.js as an example. |
|
See #15369. |
|
Put into https://github.com/nodejs/node/projects/13 backlog |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
It would be nice if
Writablestreams had a configuration option to decode buffers to strings (using aStringDecoderinstance). This is handy for example when writing aTransformstream that takes utf8 input.Currently you have to set up your own
StringDecoderinstance and write the data through that first, which is kind of annoying. For non-multibyte encodings this isn't necessary though since you can justchunk.toString()inside your_write()/_transform()handler.The text was updated successfully, but these errors were encountered: