libcamera  v0.4.0+139-39419ce4
Supporting cameras in Linux since 2019
Public Types | Public Member Functions | Public Attributes | Protected Attributes | List of all members
libcamera::Converter Class Referenceabstract

Abstract Base Class for converter. More...

Inheritance diagram for libcamera::Converter:
Inheritance graph
[legend]
Collaboration diagram for libcamera::Converter:
Collaboration graph
[legend]

Public Types

enum  Feature { None = 0, InputCrop = (1 << 0) }
 Specify the features supported by the converter.
 
enum  Alignment { Down = 0, Up }
 The alignment mode specified when adjusting the converter input or output sizes.
 
using Features = Flags< Feature >
 A bitwise combination of features supported by the converter.
 

Public Member Functions

 Converter (MediaDevice *media, Features features=Feature::None)
 Construct a Converter instance. More...
 
virtual int loadConfiguration (const std::string &filename)=0
 Load converter configuration from file. More...
 
virtual bool isValid () const =0
 Check if the converter configuration is valid. More...
 
virtual std::vector< PixelFormatformats (PixelFormat input)=0
 Retrieve the list of supported pixel formats for an input pixel format. More...
 
virtual SizeRange sizes (const Size &input)=0
 Retrieve the range of minimum and maximum output sizes for an input size. More...
 
virtual Size adjustInputSize (const PixelFormat &pixFmt, const Size &size, Alignment align=Alignment::Down)=0
 Adjust the converter input size to a valid value. More...
 
virtual Size adjustOutputSize (const PixelFormat &pixFmt, const Size &size, Alignment align=Alignment::Down)=0
 Adjust the converter output size to a valid value. More...
 
virtual std::tuple< unsigned int, unsigned int > strideAndFrameSize (const PixelFormat &pixelFormat, const Size &size)=0
 Retrieve the output stride and frame size for an input configutation. More...
 
virtual int validateOutput (StreamConfiguration *cfg, bool *adjusted, Alignment align=Alignment::Down)=0
 Validate and possibily adjust cfg to a valid converter output. More...
 
virtual int configure (const StreamConfiguration &inputCfg, const std::vector< std::reference_wrapper< StreamConfiguration >> &outputCfgs)=0
 Configure a set of output stream conversion from an input stream. More...
 
virtual bool isConfigured (const Stream *stream) const =0
 Check if a given stream is configured. More...
 
virtual int exportBuffers (const Stream *stream, unsigned int count, std::vector< std::unique_ptr< FrameBuffer >> *buffers)=0
 Export buffers from the converter device. More...
 
virtual int start ()=0
 Start the converter streaming operation. More...
 
virtual void stop ()=0
 Stop the converter streaming operation.
 
virtual int queueBuffers (FrameBuffer *input, const std::map< const Stream *, FrameBuffer *> &outputs)=0
 Queue buffers to converter device. More...
 
virtual int setInputCrop (const Stream *stream, Rectangle *rect)=0
 Set the crop rectangle rect for stream. More...
 
virtual std::pair< Rectangle, RectangleinputCropBounds ()=0
 Retrieve the crop bounds of the converter. More...
 
virtual std::pair< Rectangle, RectangleinputCropBounds (const Stream *stream)=0
 Retrieve the crop bounds for stream. More...
 
const std::string & deviceNode () const
 The converter device node attribute accessor. More...
 
Features features () const
 Retrieve the features supported by the converter. More...
 

Public Attributes

Signal< FrameBuffer * > inputBufferReady
 A signal emitted when the input frame buffer completes.
 
Signal< FrameBuffer * > outputBufferReady
 A signal emitted on each frame buffer completion of the output queue.
 

Protected Attributes

Features features_
 Stores the features supported by the converter.
 

Detailed Description

Abstract Base Class for converter.

The Converter class is an Abstract Base Class defining the interfaces of converter implementations.

Converters offer scaling and pixel format conversion services on an input stream. The converter can output multiple streams with individual conversion parameters from the same input stream.

Constructor & Destructor Documentation

◆ Converter()

libcamera::Converter::Converter ( MediaDevice media,
Features  features = Feature::None 
)

Construct a Converter instance.

Parameters
[in]mediaThe media device implementing the converter
[in]featuresFeatures flags representing supported features

This searches for the entity implementing the data streaming function in the media graph entities and use its device node as the converter device node.

Member Function Documentation

◆ adjustInputSize()

libcamera::Converter::adjustInputSize ( const PixelFormat pixFmt,
const Size size,
Alignment  align = Alignment::Down 
)
pure virtual

Adjust the converter input size to a valid value.

Parameters
[in]pixFmtThe pixel format of the converter input stream
[in]sizeThe converter input size to adjust to a valid value
[in]alignThe desired alignment
Returns
The adjusted converter input size or a null Size if size cannot be adjusted

Implemented in libcamera::V4L2M2MConverter.

◆ adjustOutputSize()

libcamera::Converter::adjustOutputSize ( const PixelFormat pixFmt,
const Size size,
Alignment  align = Alignment::Down 
)
pure virtual

Adjust the converter output size to a valid value.

Parameters
[in]pixFmtThe pixel format of the converter output stream
[in]sizeThe converter output size to adjust to a valid value
[in]alignThe desired alignment
Returns
The adjusted converter output size or a null Size if size cannot be adjusted

Implemented in libcamera::V4L2M2MConverter.

◆ configure()

libcamera::Converter::configure ( const StreamConfiguration inputCfg,
const std::vector< std::reference_wrapper< StreamConfiguration >> &  outputCfgs 
)
pure virtual

Configure a set of output stream conversion from an input stream.

Parameters
[in]inputCfgInput stream configuration
[out]outputCfgsA list of output stream configurations
Returns
0 on success or a negative error code otherwise

Implemented in libcamera::V4L2M2MConverter.

◆ deviceNode()

libcamera::Converter::deviceNode ( ) const
inline

The converter device node attribute accessor.

Returns
The converter device node string

◆ exportBuffers()

libcamera::Converter::exportBuffers ( const Stream stream,
unsigned int  count,
std::vector< std::unique_ptr< FrameBuffer >> *  buffers 
)
pure virtual

Export buffers from the converter device.

Parameters
[in]streamOutput stream pointer exporting the buffers
[in]countNumber of buffers to allocate
[out]buffersVector to store allocated buffers

This function operates similarly to V4L2VideoDevice::exportBuffers() on the output stream indicated by the output.

Returns
The number of allocated buffers on success or a negative error code otherwise

Implemented in libcamera::V4L2M2MConverter.

◆ features()

libcamera::Converter::features ( ) const
inline

Retrieve the features supported by the converter.

Returns
The converter Features flags

◆ formats()

libcamera::Converter::formats ( PixelFormat  input)
pure virtual

Retrieve the list of supported pixel formats for an input pixel format.

Parameters
[in]inputInput pixel format to retrieve output pixel format list for
Returns
The list of supported output pixel formats

Implemented in libcamera::V4L2M2MConverter.

◆ inputCropBounds() [1/2]

libcamera::Converter::inputCropBounds ( )
pure virtual

Retrieve the crop bounds of the converter.

Retrieve the minimum and maximum crop bounds of the converter. This can be used to query the crop bounds before configuring a stream.

Returns
A pair containing the minimum and maximum crop bound in that order

Implemented in libcamera::V4L2M2MConverter.

◆ inputCropBounds() [2/2]

libcamera::Converter::inputCropBounds ( const Stream stream)
pure virtual

Retrieve the crop bounds for stream.

Parameters
[in]streamThe output stream

Retrieve the minimum and maximum crop bounds for stream. The converter should support cropping (Feature::InputCrop).

The crop bounds depend on the configuration of the output stream and hence this function should be called after the stream has been configured using configure().

When called with an unconfigured stream, this function returns a pair of null rectangles.

Returns
A pair containing the minimum and maximum crop bound in that order

Implemented in libcamera::V4L2M2MConverter.

◆ isConfigured()

libcamera::Converter::isConfigured ( const Stream stream) const
pure virtual

Check if a given stream is configured.

Parameters
[in]streamThe output stream
Returns
True if the stream is configured or false otherwise

Implemented in libcamera::V4L2M2MConverter.

◆ isValid()

libcamera::Converter::isValid ( ) const
pure virtual

Check if the converter configuration is valid.

Returns
True is the converter is valid, false otherwise

Implemented in libcamera::V4L2M2MConverter.

◆ loadConfiguration()

libcamera::Converter::loadConfiguration ( const std::string &  filename)
pure virtual

Load converter configuration from file.

Parameters
[in]filenameThe file name path

Load converter dependent configuration parameters to apply on the hardware.

Returns
0 on success or a negative error code otherwise

◆ queueBuffers()

libcamera::Converter::queueBuffers ( FrameBuffer input,
const std::map< const Stream *, FrameBuffer *> &  outputs 
)
pure virtual

Queue buffers to converter device.

Parameters
[in]inputThe frame buffer to apply the conversion
[out]outputsThe container holding the output stream pointers and their respective frame buffer outputs.

This function queues the input frame buffer on the output streams of the outputs map key and retrieve the output frame buffer indicated by the buffer map value.

Returns
0 on success or a negative error code otherwise

Implemented in libcamera::V4L2M2MConverter.

◆ setInputCrop()

libcamera::Converter::setInputCrop ( const Stream stream,
Rectangle rect 
)
pure virtual

Set the crop rectangle rect for stream.

Parameters
[in]streamThe output stream
[in,out]rectThe crop rectangle to apply and return the rectangle that is actually applied

Set the crop rectangle rect for stream provided the converter supports cropping. The converter has the Feature::InputCrop flag in this case.

The underlying hardware can adjust the rectangle supplied by the user due to hardware constraints. The caller can inspect rect to determine the actual rectangle that has been applied by the converter, after this function returns.

Returns
0 on success or a negative error code otherwise

Implemented in libcamera::V4L2M2MConverter.

◆ sizes()

libcamera::Converter::sizes ( const Size input)
pure virtual

Retrieve the range of minimum and maximum output sizes for an input size.

Parameters
[in]inputInput stream size to retrieve range for
Returns
A range of output image sizes

Implemented in libcamera::V4L2M2MConverter.

◆ start()

libcamera::Converter::start ( )
pure virtual

Start the converter streaming operation.

Returns
0 on success or a negative error code otherwise

Implemented in libcamera::V4L2M2MConverter.

◆ strideAndFrameSize()

libcamera::Converter::strideAndFrameSize ( const PixelFormat pixelFormat,
const Size size 
)
pure virtual

Retrieve the output stride and frame size for an input configutation.

Parameters
[in]pixelFormatInput stream pixel format
[in]sizeInput stream size
Returns
A tuple indicating the stride and frame size or an empty tuple on error

Implemented in libcamera::V4L2M2MConverter.

◆ validateOutput()

libcamera::Converter::validateOutput ( StreamConfiguration cfg,
bool *  adjusted,
Alignment  align = Alignment::Down 
)
pure virtual

Validate and possibily adjust cfg to a valid converter output.

Parameters
[in,out]cfgThe StreamConfiguration to validate and adjust
[out]adjustedSet to true if cfg has been adjusted
[in]alignThe desired alignment
Returns
0 if cfg is valid or has been adjusted, a negative error code otherwise if cfg cannot be adjusted

Implemented in libcamera::V4L2M2MConverter.


The documentation for this class was generated from the following files: