Struct Compress 

pub struct Compress { /* private fields */ }

Create a new JPEG file from pixels

Wrapper for jpeg_compress_struct

Implementations

impl Compress

pub fn new(color_space: ColorSpace) -> Self

Compress image using input in this colorspace.

Panics

You need to wrap all use of this library in std::panic::catch_unwind()

By default errors cause unwind (panic) and unwind through the C code, which strictly speaking is not guaranteed to work in Rust (but seems to work fine, at least on x86-64 and ARM).

pub fn new_err(err: Box<ErrorMgr>, color_space: ColorSpace) -> Self

Use a specific error handler instead of the default unwinding one.

Note that the error handler must either abort the process or unwind, it can’t gracefully return due to the design of libjpeg.

color_space refers to input color space

pub fn start_compress<W: Write>(self, writer: W) -> Result<CompressStarted<W>>

Settings can’t be changed after this call. Returns a CompressStarted struct that will handle the rest of the writing.

Panics

It may panic, like all functions of this library.

impl Compress

pub fn components_mut(&mut self) -> &mut [CompInfo]

Expose components for modification, e.g. to set chroma subsampling

pub fn components(&self) -> &[CompInfo]

Read-only view of component information

impl Compress

pub fn set_color_space(&mut self, color_space: ColorSpace)

Set color space of JPEG being written, different from input color space

See jpeg_set_colorspace in libjpeg docs

pub fn set_size(&mut self, width: usize, height: usize)

Image size of the input

pub fn set_gamma(&mut self, gamma: f64)

👎Deprecated:

it doesn’t do anything

libjpeg’s input_gamma = image gamma of input image

pub fn set_pixel_density(&mut self, density: PixelDensity)

Sets pixel density of an image in the JFIF APP0 segment¹. If this method is not called, the resulting JPEG will have a default pixel aspect ratio of 1x1.

---

1. This method is not related to EXIF-based intrinsic image sizing, and does not affect rendering in browsers. ↩

pub fn set_optimize_scans(&mut self, opt: bool)

If true, it will use MozJPEG’s scan optimization. Makes progressive image files smaller.

pub fn set_smoothing_factor(&mut self, smoothing_factor: u8)

If 1-100 (non-zero), it will use MozJPEG’s smoothing.

pub fn set_optimize_coding(&mut self, opt: bool)

Set to false to make files larger for no reason

pub fn set_use_scans_in_trellis(&mut self, opt: bool)

Specifies whether multiple scans should be considered during trellis quantization.

pub fn set_progressive_mode(&mut self)

You can only turn it on

pub fn set_scan_optimization_mode(&mut self, mode: ScanMode)

One scan for all components looks best. Other options may flash grayscale or green images.

pub fn set_fastest_defaults(&mut self)

Reset to libjpeg v6 settings

It gives files identical with libjpeg-turbo

pub fn set_raw_data_in(&mut self, opt: bool)

Advanced. See raw_data_in in libjpeg docs.

pub fn set_quality(&mut self, quality: f32)

Set image quality. Values 60-80 are recommended.

pub fn set_luma_qtable(&mut self, qtable: &QTable)

Instead of quality setting, use a specific quantization table.

pub fn set_chroma_qtable(&mut self, qtable: &QTable)

Instead of quality setting, use a specific quantization table for color.

pub fn set_chroma_sampling_pixel_sizes(&mut self, cb: (u8, u8), cr: (u8, u8))

Sets chroma subsampling, separately for Cb and Cr channels. Instead of setting samples per pixel, like in cinfo’s x_samp_factor, it sets size of chroma “pixels” per luma pixel.

• (1,1), (1,1) == 4:4:4
• (2,1), (2,1) == 4:2:2
• (2,2), (2,2) == 4:2:0

Trait Implementations

impl Drop for Compress

fn drop(&mut self)

Executes the destructor for this type.