Struct Cursor

pub struct Cursor<'a> {
    buf: &'a [u8],
}

Decodes data within pgwire messages.

The API provided is very similar to bytes::Buf, but operations return errors rather than panicking. This is important for safety, as we don’t want to crash if the user sends us malformed pgwire messages.

There are also some special-purpose methods, like Cursor::read_cstr, that are specific to pgwire messages.

Fields

buf: &'a [u8]

Implementations

impl<'a> Cursor<'a>

pub fn new(buf: &'a [u8]) -> Cursor<'a>

Constructs a new Cursor from a byte slice. The cursor will begin decoding from the beginning of the slice.

pub fn peek_byte(&self) -> Result<u8, Error>

Returns the next byte without advancing the cursor.

pub fn read_byte(&mut self) -> Result<u8, Error>

Returns the next byte, advancing the cursor by one byte.

pub fn read_cstr(&mut self) -> Result<&'a str, Error>

Returns the next null-terminated string. The null character is not included the returned string. The cursor is advanced past the null- terminated string.

If there is no null byte remaining in the string, returns CodecError::StringNoTerminator. If the string is not valid UTF-8, returns an io::Error with an error kind of io::ErrorKind::InvalidInput.

NOTE(benesch): it is possible that returning a string here is wrong, and we should be returning bytes, so that we can support messages that are not UTF-8 encoded. At the moment, we’ve not discovered a need for this, though, and using proper strings is convenient.

pub fn read_i16(&mut self) -> Result<i16, Error>

Reads the next 16-bit signed integer, advancing the cursor by two bytes.

pub fn read_i32(&mut self) -> Result<i32, Error>

Reads the next 32-bit signed integer, advancing the cursor by four bytes.

pub fn read_u32(&mut self) -> Result<u32, Error>

Reads the next 32-bit unsigned integer, advancing the cursor by four bytes.

pub fn read_format(&mut self) -> Result<Format, Error>

Reads the next 16-bit format code, advancing the cursor by two bytes.

pub fn advance(&mut self, n: usize)

Advances the cursor by n bytes.

Trait Implementations

impl<'a> Debug for Cursor<'a>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.