Image Decoding

The image decoding class CImageDecoder (CBufferedImageDecoder is described later) provides functions to decode images held in descriptors or files from standard formats for use with devices such as screens, browsers and viewer applications. This class provides these features via a plugin framework. The standard formats supported by the decoder plugins are shown in the table in Overview.

For the purposes of this description, the decoding process has been broken down into the following sections:

• Creation - The creation of the CImageDecoder object and special considerations when doing so.

• Conversion - Covers the basic form of image decoding. More advanced features such as progressive and buffered decoding are described separately.

• Enquiry features - Additional features that enable you to retrieve information stored in certain types of images, for example frame information, comments and so on.

• Streamed and progressive decoding - Reading partial image files and displaying image data before the entire image is read.

• Buffered decoding - The decoding of an image using a buffered input rather than a file or descriptor.

CImageDecoder decodes images stored in files or in descriptors. The decoder object is owned by the client and must be deleted once decoding is finished. CImageDecoder objects cannot be reused to decode other images - each image requires its own instance of the decoder.

CImageDecoder objects are created using the CImageDecoder::DataNewL() and CImageDecoder::FileNewL() factory functions for images held in descriptors or files respectively:

In creating a CImageDecoder object, a suitable plugin has to be associated with the image to be decoded. How the plugin is specified depends on the factory function you use to create the object and what parameters you specify. There are four alternatives:

• MIME type - the plugin is determined by looking up a specified MIME type against a list of known MIME type/plugin implementations. Use the following "File" factory function (or its "Data" equivalent).

  static CImageDecoder* FileNewL(RFs& aFs, const TDesC& aSourceFilename, const TDesC8& aMIMEType, const TOptions aOptions = EOptionNone);

• Image type and sub-type - the plugin is determined by looking up the specified image type and sub-types against a list of known type/sub-type plugin implementations. Use the following "File" factory function (or its "Data" equivalent).

  static CImageDecoder* FileNewL(RFs& aFs, const TDesC& aSourceFilename, const TOptions aOptions = EOptionNone, const TUid aImageType = KNullUid, const TUid aImageSubType = KNullUid, const TUid aDecoderUid = KNullUid);

  Note: For images types that do not have sub-types use KNullUid.

• Implementation UID - the plugin is determined by looking up its specific UID and the image type and sub-type. Use the same FileNewL() factory function shown in "Image type and sub-type", but in addition to the type and sub-type parameters, also specify the plugin UID using aDecoderUid.

  Note: Implementation UID is supported because it is possible to have more than one plugin for a particular image type. This method is recommended if the application or calling DLL needs to exploit features of a particular plugin.

• Automatic detection - no MIME type, format type/sub-type or UIDs are specified. The plugin to use is determined by analysing the header information of the specified image. Use the same FileNewL() factory function shown in "Image type and sub-type", but do not specify any values for aImageType aImageSubType or aDecoderUid.

  If no suitable plugin can be found, the factory function leaves with KErrNotFound. If a suitable plugin is found, but that plugin cannot interpret the image data the factory function leaves with KErrCorrupt.

  Note: Some image formats cannot be automatically detected because they do not contain sufficient header information, for example, WBMP and OTA images are known to fall into this category. Under such circumstances, use one of the other three alternative for opening the image.

It is possible for more than one valid plugin decoder to be available for a specific image format. If a plugin determined by MIME type or type/sub-type fails to open an image due to KErrCorrupt, it will continue to try with subsequent valid plugins if available. This mechanism is carried out behind the scenes and is largely invisible to the application. If you require a specific function you know to be available within a specific plugin, specify that plugin using aDecoderUid.

As part of the CImageDecoder creation basic information about the image is pre-read from the image headers. This information is used to support certain enquiry methods such as CImageDecoder::FrameCount() and CImageDecoder::FrameInfo() and so on (more on the enquiry methods is supplied in Enquiry features).

You are free to decode any subset of individual frames, in any order, and to repeat the decoding if necessary. To decode an image frame use the asynchronous conversion method CImageDecoder::Convert(), as shown below.

void Convert(TRequestStatus* aRequestStatus, CFbsBitmap& aDestination, TInt aFrameNumber = 0);

void ContinueConvert(TRequestStatus* aRequestStatus);

Note: The use of CImageDecoder::ContinueConvert() is described in Streamed and progressive decoding.

Bitmap masks

There is a second variant of the Convert() function intended for use when decoding images with bitmap masks.

void Convert(TRequestStatus* aRequestStatus, CFbsBitmap& aDestination, CFbsBitmap& aDestinationMask, TInt aFrameNumber = 0);

Within the Symbian OS it is normal to store the bitmap mask separately from the main image data. Calls such as BitBltMasked() expect this data to be provided separately. This separation is true for both CFbsBitmap objects and MBM files.

To determine whether an image contains a bitmap mask use FrameInfo(), the presence or absence of a mask will be indicated the ETransparencyPossible flag. Two types of mask are possible.

• 8-bit alpha blend - indicated by the EAlphaChannel flag of CImageDecoder::FrameInfo() iFlags. The destination mask bitmap must be of type EGray256.

• Simple on/off mask - if no EAlphaChannel flag is set. The destination bitmap mask can be either EGray2 or EGray256.

Note: Images that contain bitmap masks do not have to have those masks decoded if they are not required.

Scaling and display modes

Before Convert() can be used the destination CFbsBitmap objects for image data and possibly bitmap masks must be created. The simplest way to do this is to use a 1 to 1 approach, that is, create the CFbsBitmap object with the same properties as the source image (if at all possible). This could be achieved by using something similar to:

iFrameInfo = &imageDecoder->FrameInfo(FrameNum);

iBitmap->Create(iFrameInfo->iOverallSizeInPixels, iFrameInfo->iFrameDisplayMode );

If the ECanDither flag of CImageDecoder::FrameInfo() iFlags is set then the destination display mode can be adjusted. In such cases it is generally recommended that the current window's display mode is used as this can make drawing faster.

Note: The mask bitmap must have the same size in pixels as the main image, even though the display mode is generally different.

As indicated above the size of the image bitmap is available via FrameInfo() (iOverallSizeInPixels). However, in addition to a direct mapping of an image's size, ReducedSize() function can be used to calculate the reduced size of the decoded bitmap based on the input parameters.

If the EFullyScaleable flag of FrameInfo() iFlags is set, you can specify any size for the CFbsBitmap and the image will be resized appropriately.

Animations

Included in the CImageDecoder::FrameInfo() result are flags that enable the support of GIF animation. These flags are iDelay, ELeaveInPlace, ERestoreToBackground and ERestoreToPrevious. ICL provides these flags to enable the application to implement the animation; ICL does not provide functions to display the animation directly.

Processing the information from these flags will inevitably require a state machine, and additional timers, within the application. Requirements will vary depending on the architecture of the client application itself.

Background colours

Some image formats contain background colour information. This is information is presented by ICL via CImageDecoder::FrameInfo() iBackgroundColor. If a non-masked image with background colour is decoded, the colour is included in the image data. If the decoded image also contains a bitmap mask, then the background colour will be a component of the mask CFbsBitmap object. Depending on your application, you may want to choose your own background colour rather than use the one provided by the source image.

Previous sections shown above have made use of CImageDecode::FrameInfo(). There are however, other enquiry functions that also provide useful information. They are:

• CImageDecode::FrameData()

  Provides access to additional chunks of data within the images, for example, palette and similar lookup tables, copyright information and other strings. This information should be used with care, the returned data is merely a reference to the original data within the CImageDecoder object, and is only valid for the lifetime of the object.

• CImageDecode::NumberOfImageComments() and CImageDecode::ImageCommentL()

  Images can have comments embedded in them. These comments are usually either embedded as a single repository which can be retrieved on an image level, or multiple repositories attached to individual frames within the image. The two functions above return the number of comments attached to the entire image and provide a mechanism for retrieving any of those comments.

• CImageDecode::NumberOfFrameComments() and CImageDecode::FrameCommentL()

  Similar functionality to above, but these functions return the number of comments attached to individually specified frames and provide a mechanism for retrieving any of those comments.

• CImageDecode::FrameInfoStringsLC and CImageDecode::FrameInfoStringsL()

  Provide a list of data for a specified frame, including data items such as format, plugin description, size and so on, in a readable format.

  Although the property strings can be accessed by name, it is recommended that they be generally treated as a list of strings.

CImageDecoder includes facilities to support the simultaneous decoding and display of images as they are being loaded. The decoder does not wait for the entire image to be loaded before processing it, rather it begins as soon as possible, stops when it runs out of data and then carries on when more data is available. The decoding rules that CImageDecoder follows are shown below:

• If there is insufficient data to work out the image format (plugin decoder to use) CImageDecoder::FileNewL() (or its "Data" equivalent) leaves with KErrUnderflow. This is only applicable to instances where automatic detection of the plugin decoder is used. For more information see, "Automatic detection" in Creation.

• As soon as the correct plugin decoder has been determined it is opened and whatever addition image data is available is scanned. The plugin decoder continues to decode image data as it arrives, updating FrameCount() whenever it becomes aware of a new frame within the data. The internal flag CImageDecoder::IsHeaderProcessingComplete() is maintained at EFalse until the entire image has been loaded.

• Frames can be decoded before the entire image is loaded, but the frame to be decoded must be at least partially loaded. If IsHeaderProcessingComplete() is set to EFalse and FrameCount() is equal to or less than the frame to decode, the application must wait for the relevant frame to load. In such circumstances a call to CImageDecoder::ContinueProcessingHeadersL() should be made that scans for any further headers. FrameCount() and IsHeaderProcessingComplete() should then be recalled to determine if the frame has arrived.

• Once FrameCount() is greater than the frame the application wants to decode, it is possible to start to decoding the frame using CImageDecoder::Convert(). Frame headers and their associated data do not always follow each other in some image formats. This has the implication that although FrameCount() has indicated that the frame is available, it may not yet be fully loaded. Under such circumstances as much decoding as possible is undertaken, and Convert() then completes with the error code KErrUnderflow. If the EPartialDecodeInvalid flag (from FrameInfo()) is not set, the partially decoded image can be displayed - for some image formats, a partially decoded image is not generally usable, but this facility is supported by most known formats. Obviously, if all of the image is present, Convert() completes with KErrNone as normal.

• Where only a partial conversion has been completed, CImageDecoder::ContinueConvert() should be used to continue converting when new data arrives. ContinueConvert() continues to convert the frame data where the previous call left off. This function should continue to be called until it returns the error code KErrNone rather than KErrUnderflow.

  Note: The CFbsBitmap must never be resized during a conversion session using ContinueConvert(), if resizing does occur, a panic will be raised by the function. If resizing or any other parameter changes need to be made to the CFbsBitmap, frame decoding should be restarted by using Convert() rather than trying to continue an existing conversion session.

There is no explicit decoder call to say "all of the data is now present". This could be done via inference, continuously checking IsHeaderProcessingComplete() until it returns ETrue. However, this is not advisable as it is possible that IsHeaderProcessingComplete() could continue to return EFalse even after the application has finished sending image data. Your state machine should be written so that it takes this possibility into account.

CBufferedImageDecoder is a wrapper that encapsulates not only a CImageDecoder but also the descriptor used to store the image, and in some circumstances can replace the use of CImageDecoder itself. CBufferedImageDecoder can always be created, even if there is no data available at the time of creation. Another key feature of CBufferedImageDecoder is that it can be reused to decode multiple images. The use of this class is outlined below.

• The decoder is created using the CBufferedImageDecoder::NewL() factory function, and the decoding process is started with one of the two CBufferedImageDecoder::OpenL() calls, specifying either MIME-type or image type/sub-type and UID, which are similar to CImageDecoder::DataNewL(). If not enough data is available to create an internal decoder a call to CBufferedImageDecoder::ValidDecoder() returns EFalse. The decoder keeps an internal copy of the data provided so the client can discard its own copy.

  Warning: The decoding and enquiry functions must not be used until an internal decoder has been created, otherwise calls will panic with EDecoderNotCreated.

• Additional data can be supplied to the decoder, when it becomes available, by using CBufferedImageDecoder::AppendDataL() and CBufferedImageDecoder::ContinueOpenL(), which will copy the data and try to find an appropriate plugin. This step should be repeated until enough data is available to create the internal decoder, indicated by ValidDecoder() returning ETrue. OpenL() and ContinueOpenL() will leave with KErrCorrupt if a suitable decoder is found, but that decoder was unable to interpret the supplied image data.

  Note: As with OpenL(), data provided by AppendDataL() is kept in an internal buffer by the decoder so the client can discard its own copy.

• The process for decoding the image header and frames are the same as for CImageDecoder with IsHeaderProcessingComplete(), FrameCount(), Convert() and ContinueConvert() calls except that the data is not appended to a descriptor, but supplied to the decoder with AppendDataL().

• The decoder can be reused by calling a CBufferedImageDecoder::Reset(), which destroys the internal data buffer and decoder. A new decoder plugin can then be started using OpenL().

• ICL client side architecture - brief overview of the architecture of the image converted library client side classes.

• Overview - general overview of ICL, synchronous and asynchronous methods, active objects and concurrency issues.

• Image decoding - frame information and data, still image decoding, animated image decoding, progressive decoding and buffered decoding.

• Image encoding - encoding/decoding feature comparison, image encoding.

• Additional information - dithering, enabling progressive decoding, MIME type and plugin reporting, threading, memory usage, error handling and performance considerations.