https://github.com/isaacs/fs-minipass
fs read and write streams based on minipass
https://github.com/isaacs/fs-minipass
npm-cli
Last synced: 8 months ago
JSON representation
fs read and write streams based on minipass
- Host: GitHub
- URL: https://github.com/isaacs/fs-minipass
- Owner: isaacs
- License: isc
- Created: 2017-09-09T20:09:21.000Z (almost 9 years ago)
- Default Branch: main
- Last Pushed: 2024-07-17T21:51:24.000Z (almost 2 years ago)
- Last Synced: 2025-09-28T22:14:58.094Z (9 months ago)
- Topics: npm-cli
- Language: JavaScript
- Homepage:
- Size: 199 KB
- Stars: 20
- Watchers: 10
- Forks: 5
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Codeowners: .github/CODEOWNERS
- Security: SECURITY.md
Awesome Lists containing this project
README
# fs-minipass
Filesystem streams based on [minipass](http://npm.im/minipass).
4 classes are exported:
- ReadStream
- ReadStreamSync
- WriteStream
- WriteStreamSync
When using `ReadStreamSync`, all of the data is made available
immediately upon consuming the stream. Nothing is buffered in memory
when the stream is constructed. If the stream is piped to a writer,
then it will synchronously `read()` and emit data into the writer as
fast as the writer can consume it. (That is, it will respect
backpressure.) If you call `stream.read()` then it will read the
entire file and return the contents.
When using `WriteStreamSync`, every write is flushed to the file
synchronously. If your writes all come in a single tick, then it'll
write it all out in a single tick. It's as synchronous as you are.
The async versions work much like their node builtin counterparts,
with the exception of introducing significantly less Stream machinery
overhead.
## USAGE
It's just streams, you pipe them or read() them or write() to them.
```js
const fsm = require('fs-minipass')
const readStream = new fsm.ReadStream('file.txt')
const writeStream = new fsm.WriteStream('output.txt')
writeStream.write('some file header or whatever\n')
readStream.pipe(writeStream)
```
## ReadStream(path, options)
Path string is required, but somewhat irrelevant if an open file
descriptor is passed in as an option.
Options:
- `fd` Pass in a numeric file descriptor, if the file is already open.
- `readSize` The size of reads to do, defaults to 16MB
- `size` The size of the file, if known. Prevents zero-byte read()
call at the end.
- `autoClose` Set to `false` to prevent the file descriptor from being
closed when the file is done being read.
## WriteStream(path, options)
Path string is required, but somewhat irrelevant if an open file
descriptor is passed in as an option.
Options:
- `fd` Pass in a numeric file descriptor, if the file is already open.
- `mode` The mode to create the file with. Defaults to `0o666`.
- `start` The position in the file to start reading. If not
specified, then the file will start writing at position zero, and be
truncated by default.
- `autoClose` Set to `false` to prevent the file descriptor from being
closed when the stream is ended.
- `flags` Flags to use when opening the file. Irrelevant if `fd` is
passed in, since file won't be opened in that case. Defaults to
`'a'` if a `pos` is specified, or `'w'` otherwise.