y[      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~      ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 5 6 7 8 9 : ; < = > ? @ A B CDEFGHIJKLMNOPQRSTUVWXYZNone!Possible data held by an Exif tag 5Tag values used for exif fields. Completly incomplete  Short typeShort or long typeShort or long type Rational type Rational typeShort or long type Short or long Short or longShortShort Short"Short?JConvert a value to it's corresponding Exif tag. Will often be written as >@*Convert a tag to it's corresponding value.C  !"#$%&'()*+,-./0123456789:;<=>?@[\A  !"#$%&'()*+,-./0123456789:;<=>?@A  !"#$%&'()*+,-./0123456789:;<=> ?@  3  !"#$%&'()*+,-./0123456789:;<=>?@[\None3457]0^1_2`0a1b2c3d4e5f6g8hBecause having a polymorphic get with endianness is to nice to pass on, introducing this helper type class, which is just a superset of Binary, but formalising a parameter passing into it.iAll the IFD must be written in order according to the tag value of the IFD. To avoid getting to much restriction in the serialization code, just sort it.jiGiven an official offset and a list of IFD, update the offset information of the IFD with extended data.Tk]^_l`abcdefgmnopqrstuvwxyz{|}~hijF@k]^_l`abcdefgmnopqrstuvwxyz{|}~h#k]^_l`abcdefgmnopqrstuvwxyz{| }~hijSafeNone+34579>ILNqAThis typeclass exist for performance reason, it allow to pack a pixel value to a simpler "primitive" data type to allow faster writing to moemory.BVPrimitive type asociated to the current pixel It's Word32 for PixelRGBA8 for instanceC<The packing function, allowing to transform to a primitive.D,Inverse transformation, to speed up readingEEHelper class to help extract a luma plane out of an image or a pixelF%Compute the luminance part of a pixelGjExtract a luma plane out of an image. This method is in the typeclass to help performant implementation. LjpegToGrayScale :: FilePath -> FilePath -> IO () jpegToGrayScale source destH;Traversal type matching the definition in the Lens package.3Helpers to embed a rankNTypes inside an ApplicativeIgThis class abstract colorspace conversion. This conversion can be lossy, which ColorConvertible cannotJGPass a pixel from a colorspace (say RGB) to the second one (say YCbCr)K>Helper function to convert a whole image by taking a copy it.L=Implement upcasting for pixel types. Minimal declaration of MV. It is strongly recommended to overload promoteImage to keep performance acceptableMWConvert a pixel type to another pixel type. This operation should never lose any data.NNChange the underlying pixel type of an image by performing a full copy of it.OtDefinition of pixels used in images. Each pixel has a color space, and a representative component (Word8 or Float).PType of the pixel component, "classical" images would have Word8 type as their PixelBaseComponent, HDR image would have Float for instanceQoCall the function for every component of the pixels. For example for RGB pixels mixWith is declared like this: fmixWith f (PixelRGB8 ra ga ba) (PixelRGB8 rb gb bb) = PixelRGB8 (f 0 ra rb) (f 1 ga gb) (f 2 ba bb)RExtension of the Q which separate the treatment of the color components of the alpha value (transparency component). For pixel without alpha components, it is equivalent to mixWith. mixWithAlpha f fa (PixelRGBA8 ra ga ba aa) (PixelRGB8 rb gb bb ab) = PixelRGBA8 (f 0 ra rb) (f 1 ga gb) (f 2 ba bb) (fa aa ab)SReturn the opacity of a pixel, if the pixel has an alpha layer, return the alpha value. If the pixel doesn't have an alpha value, return a value representing the opaqueness.T,Return the number of components of the pixelUApply a function to each component of a pixel. If the color type possess an alpha (transparency channel), it is treated like the other color components.V1Calculate the index for the begining of the pixelW@Calculate theindex for the begining of the pixel at position x yXExtract a pixel at a given position, (x, y), the origin is assumed to be at the corner top left, positive y to the bottom of the imageY'Same as pixelAt but for mutable images.Z0Write a pixel in a mutable image at position x y[Unsafe version of pixelAt, read a pixel at the given index without bound checking (if possible). The index is expressed in number (PixelBaseComponent a)\Unsafe version of readPixel, read a pixel at the given position without bound checking (if possible). The index is expressed in number (PixelBaseComponent a)]Unsafe version of writePixel, write a pixel at the given position without bound checking. This can be _really_ unsafe. The index is expressed in number (PixelBaseComponent a)^rPixel type storing 16bit red, green, blue and alpha (RGBA) information. Values are stored in the following order:RedGreenBlueAlpha`{Classical pixel type storing 8bit red, green, blue and alpha (RGBA) information. Values are stored in the following order:RedGreenBlueAlphabwPixel type storing 16bit cyan, magenta, yellow and black (CMYK) information. Values are stored in the following order:CyanMagentaYellowBlackdvPixel type storing 8bit cyan, magenta, yellow and black (CMYK) information. Values are stored in the following order:CyanMagentaYellowBlackfPixel type storing 8bit luminance, blue difference and red difference (YCbCr) information. Values are stored in the following order: Y (luminance)CbCrhHDR pixel type storing floating point 32bit red, green and blue (RGB) information. Same value range and comments apply as for t,. Values are stored in the following order:RedGreenBluejjPixel type storing 16bit red, green and blue (RGB) information. Values are stored in the following order:RedGreenBluel2Pixel type storing value for the YCCK color space: Y (Luminance)CbCrBlacknqClassic pixel type storing 8bit red, green and blue (RGB) information. Values are stored in the following order:RedGreenBlueplPixel type storing 16bit Luminance (Y) and alpha (A) information. Values are stored in the following order: LuminanceAlpharkPixel type storing 8bit Luminance (Y) and alpha (A) information. Values are stored in the following order: LuminanceAlphatType alias for 32bit floating point greyscale pixels. The standard bounded value range is mapped to the closed interval [0,1] i.e. Kmap promotePixel [0, 1 .. 255 :: Pixel8] == [0/255, 1/255 .. 1.0 :: PixelF]u&Type alias for 32bit greyscale pixels.v&Type alias for 16bit greyscale pixels.wuType alias for 8bit greyscale pixels. For simplicity, greyscale pixels use plain numbers instead of a separate type.xsImage type enumerating all predefined pixel types. It enables loading and use of images of different pixel types.yA greyscale image.z'A greyscale image with 16bit components{A greyscale HDR image|,An image in greyscale with an alpha channel.}4An image in greyscale with alpha channel on 16 bits.~An image in true color.(An image in true color with 16bit depth.An image with HDR pixels,An image in true color and an alpha channel.)A true color image with alpha on 16 bits./An image in the colorspace used by Jpeg images.An image in the colorspace CMYK5An image in the colorspace CMYK and 16 bits precisionImage or pixel buffer, the coordinates are assumed to start from the upper-left corner of the image, with the horizontal position first, then the vertical one. The image can be transformed in place.Width of the image in pixelsHeight of the image in pixels.YThe real image, to extract pixels at some position you should use the helpers functions.XClass modeling transparent pixel, should provide a method to combine transparent pixels"Just return the opaque pixel valueIaccess the transparency (alpha layer) of a given transparent pixel type.>Define plane for the black component of the CMYK color space.?Define plane for the yellow component of the CMYK color space.@Define plane for the magenta component of the CMYK color space.=Define plane for the cyan component of the CMYK color space.%Define the plane for the Cb component%Define the plane for the Cr component'Define the plane for the luma component7Define the plane for the alpha (transparency) component-Define the plane for the blue color component.Define the plane for the green color component,Define the plane for the red color componentClass used to describle plane present in the pixel type. If a pixel has a plane description associated, you can use the plane name to extract planes independently.=Retrieve the index of the component in the given pixel type.-Type for the palette used in Gif & PNG files.JThe main type of this package, one that most functions work on, is Image.Parameterized by the underlying pixel format it forms a rigid type. If you wish to store images of different or unknown pixel formats use x.Image is essentially a rectangular pixel buffer of specified width and height. The coordinates are assumed to start from the upper-left corner of the image, with the horizontal position first and vertical second.Width of the image in pixelsHeight of the image in pixels.]Image pixel data. To extract pixels at a given position you should use the helper functions.Internally pixel data is stored as consecutively packed lines from top to bottom, scanned from left to right within individual lines, from first to last color component within each pixel.QExtract a color plane from an image given a present plane in the image examples: a extractRedPlane :: Image PixelRGB8 -> Image Pixel8 extractRedPlane = extractComponent PlaneRed YExtract a plane of an image. Returns the requested color component as a greyscale image.-If you ask for a component out of bound, the  function will be called.^For any image with an alpha component (transparency), drop it, returning a pure opaque image.A`O(n)` Yield an immutable copy of an image by making a copy of it?`O(n)` Yield a mutable copy of an image by making a copy of it.`O(1)` Unsafe convert an imutable image to an mutable one without copying. The source image shouldn't be used after this operation.`O(1)` Unsafe convert a mutable image to an immutable one without copying. The mutable image may not be used after this operation.?Create a mutable image, filled with the given background color.KCreate a mutable image with garbage as content. All data is uninitialized.Helper function to help extract information from dynamic image. To get the width of a dynamic image, you can use the following snippet: HdynWidth :: DynamicImage -> Int dynWidth img = dynamicMap imageWidth imgEquivalent of the o function for the dynamic images. You can perform pixel colorspace independant operations with this function.For instance, if you want to extract a square crop of any image, without caring about colorspace, you can use the following snippet. dynSquare :: DynamicImage -> DynamicImage dynSquare = dynamicPixelMap squareImage squareImage :: Pixel a => Image a -> Image a squareImage img = generateImage (\x y -> pixelAt img x y) edge edge where edge = min (imageWidth img) (imageHeight img)Create an image given a function to generate pixels. The function will receive values from 0 to width-1 for the x parameter and 0 to height-1 for the y parameter. The coordinates 0,0 are the upper left corner of the image, and (width-1, height-1) the lower right corner..for example, to create a small gradient image: imageCreator :: String -> IO () imageCreator path = writePng path $ generateImage pixelRenderer 250 300 where pixelRenderer x y = PixelRGB8 (fromIntegral x) (fromIntegral y) 128Create an image using a monadic initializer function. The function will receive values from 0 to width-1 for the x parameter and 0 to height-1 for the y parameter. The coordinates 0,0 are the upper left corner of the image, and (width-1, height-1) the lower right corner.{The function is called for each pixel in the line from left to right (0 to width - 1) and for each line (0 to height - 1).Create an image given a function to generate pixels. The function will receive values from 0 to width-1 for the x parameter and 0 to height-1 for the y parameter. The coordinates 0,0 are the upper left corner of the image, and (width-1, height-1) the lower right corner.(the acc parameter is a user defined one.{The function is called for each pixel in the line from left to right (0 to width - 1) and for each line (0 to height - 1).\Fold over the pixel of an image with a raster scan order: from top to bottom, left to rightrFold over the pixel of an image with a raster scan order: from top to bottom, left to right, carrying out a stateFold over the pixel of an image with a raster scan order: from top to bottom, left to right. This functions is analog to the foldMap from the U typeclass, but due to the Pixel constraint, Image cannot be made an instance of it.n equivalent for an image, working at the pixel level. Little example : a brightness function for an rgb image brightnessRGB8 :: Int -> Image PixelRGB8 -> Image PixelRGB8 brightnessRGB8 add = pixelMap brightFunction where up v = fromIntegral (fromIntegral v + add) brightFunction (PixelRGB8 r g b) = PixelRGB8 (up r) (up g) (up b)rTraversal in "raster" order, from left to right the top to bottom. This traversal is matching pixelMap in spirit. Since 3.2.4Traversal providing the pixel position with it's value. The traversal in raster order, from lef to right, then top to bottom. The traversal match pixelMapXY in spirit. Since 3.2.4 Just like K only the function takes the pixel coordinates as additional parameters.dCombine, pixel by pixel and component by component the values of 3 different images. Usage example: averageBrightNess c1 c2 c3 = clamp $ toInt c1 + toInt c2 + toInt c3 where clamp = fromIntegral . min 0 . max 255 toInt :: a -> Int toInt = fromIntegral ziPixelComponent3 averageBrightNess img1 img2 img38Perform a gamma correction for an image with HDR pixels.@Perform a tone mapping operation on an High dynamic range image.mThis function will fill an image with a simple packeable pixel. It will be faster than any unsafeWritePixel.*Fill a packeable pixel between two bounds.DRead a packeable pixel from an image. Equivalent to unsafeReadPixelFWrite a packeable pixel into an image. equivalent to unsafeWritePixel.&Free promotion for identic pixel typesABCDEFGHIJKLMNOPQRFunction for color componentFunction for alpha componentSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~BThe component index, beginning at 0 ending at (componentCount - 1) Source imageWidthHeightBackground colorWidthHeightGenerating function, with x and y params.Width in pixelsHeight in pixelsGenerating function, with x and y params.Width in pixelsHeight in pixels Image width Image heightGenerating functions"Function taking the state, x and y Initial stateWidth in pixelsHeight in pixelsmonadic mapping function Initial stateImage to fold overPixel building function RGB sampleResulting sample*Gamma value, should be between 0.5 and 3.0Image to treat.Exposure parameterImage to treat.Image to write intoPixel to write#Start index in pixel base componentpixel count of pixel to writeImage to read from&Index in (PixelBaseComponent px) countImage to write into&Index in (PixelBaseComponent px) countPixel to write      !"#$%&'()*+,-./01234567ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~Ɔxyz{|}~Hwvutrspqnojkhi`a^_debcfglmLMNOPQRSTUVWXYZ[\]IJKEFGABCDABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwx yz{|}~      !"#$%&'()*+,-./01234567None 4;>LN8Efficient representation of a left fold that preserves the fold's step function, initial accumulator, and extraction functionThis allows the 9M instance to assemble derived folds that traverse the container only once+To specify how the palette will be created."Algorithm used to find the paletteDo we want to apply the dithering to the image. Enabling it often reduce compression ratio but enhance the perceived quality of the final image./Maximum number of color we want in the palette-Define which palette creation method is used.`MedianMeanCut method, provide the best results (visualy) at the cost of increased calculations.FVery fast algorithm (one pass), doesn't provide good looking results.XDefault palette option, which aim at the best quality and maximum possible colors (256)1Reduces an image to a color palette according to  PaletteOpts and returns the  indices image along with its .:TModified median cut algorithm with optional ordered dithering. Returns an image of w+ that acts as a matrix of indices into the .;#A naive one pass Color Quantiation algorithm - Uniform Quantization. Simply take the most significant bits. The maxCols parameter is rounded down to the nearest power of 2, and the bits are divided among the three color channels with priority order green, red, blue. Returns an image of w+ that acts as a matrix of indices into the .<Apply a strict left 8 to a  containerMuch slower than < on lists because ) operations currently do not trigger  build/foldr fusion=Like >, except with a more general ? return value :@ABCDEFGHIJKL8M:;NOPQR<=STUVWXYZ[\]^_`abcdefghij *@ABCDEFGHIJKL8M:;NOPQR<=STUVWXYZ[\]^_`abcdefghijNone>LNiImport a image from an unsafe pointer The pointer must have a size of width * height * componentCount pxklWidth in pixelsHeight in pixelsPointer to the raw dataklklNoneINmType used to read bitsn(Current bit index, current value, stringo"Bitify a list of things to decode.pzDrop all bit until the bit of indice 0, usefull to parse restart marker, as they are byte aligned, but Huffman might not.qsBitify a list of things to decode. Handle Jpeg escape code (0xFF 0x00), thus should be only used in JPEG decoding.r%Append some data bits to a Put monad.s%Append some data bits to a Put monad.$tuvwxyzmn{|}~opqr<The real data to be stored. Actual data should be in the LSB#Number of bit to write from 1 to 32s<The real data to be stored. Actual data should be in the LSB#Number of bit to write from 1 to 32tmn|}~opqrstuvwxyzmn{|}~opqrsNone?Gif image constraint from spec-gif89a, code size max : 12 bits. None None4F BEnumeration used to search in the tables for different components.Represent a compact array of 8 * 8 values. The size is not guarenteed by type system, but if makeMacroBlock is used, everything should be fine size-wise0Tree storing the code used for huffman encoding.4If bit is 0 take the first subtree, if 1, the right.We should output the valueno value present<Helper function to create pure macro block of the good size._Transform parsed coefficients from the jpeg header to a tree which can be used to decode data.#From the Table K.3 of ITU-81 (p153)#From the Table K.4 of ITU-81 (p153)#From the Table K.5 of ITU-81 (p154)None>LN 0 : DC, 1 : AC, stored on 4 bitsStored on 4 bitsAType introduced only to avoid some typeclass overlapping problemStored on 4 bitsStored on 4 bits (begin, end)Encoded as 4 bitsEncoded as 4 bitsEncoded as 4 bitsEncoded as 4 bitsStored with 4 bitsStored with 4 bitsSize: 1012Value 0value 1value 2#Macroblock that can be transformed.rType only used to make clear what kind of integer we are carrying Might be transformed into newtype in the future4Parse JFIF or JFXX information. Right now only JFIF.0Create a new macroblock with the good array size      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\r      !"#$%&'()*+,-./012345678FGH:        !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\None>L]Reference implementation of the DCT, directly implementing the formula of ITU-81. It's slow as hell, perform to many operations, but is accurate and a good reference point.^Fast DCT extracted from libjpeg]_`abcdefghijklm^]^]_`abcdefghijklm^None4>LnYAlgorithm to call to perform an IDCT, return the same block that the one given as input.o0Perform a Jpeg level shift in a mutable fashion.pqrstuvwxyz{|}~nono p qrstuvwxyz{|}~noNone3/Different known interlace methods for PNG imageFNo interlacing, basic data ordering, line by line from left to right.Use the Adam7 ordering, see adam7ReorderingwThe pixels value should be : +---+---+ | c | b | +---+---+ | a | x | +---+---+ x being the current filtered pixel%Filt(x) = Orig(x), Recon(x) = Filt(x)>Filt(x) = Orig(x) - Orig(a), Recon(x) = Filt(x) + Recon(a)>Filt(x) = Orig(x) - Orig(b), Recon(x) = Filt(x) + Recon(b)jFilt(x) = Orig(x) - floor((Orig(a) + Orig(b)) / 2), Recon(x) = Filt(x) + floor((Recon(a) + Recon(b)) / 2)Filt(x) = Orig(x) - PaethPredictor(Orig(a), Orig(b), Orig(c)), Recon(x) = Filt(x) + PaethPredictor(Recon(a), Recon(b), Recon(c))#Low level access to PNG informationThe real uncompressed image8List of raw chunk where some user data might be present.GPNG chunk representing some extra information found in the parsed file.The raw data inside the chunkThe name of the chunk..Data structure during real png loading/parsing4Palette with indices beginning at 0 to elemcount - 1*Raw parsed image which need to be decoded.GWhat kind of information is encoded in the IDAT section of the PngFileStarting from 0Width of the following frameHeight of the following frameEncoded in a Word8#Overwrite output buffer. has value '0',Alpha blend to the output buffer. Has value '1'Encoded in a Word8}No disposal is done on this frame before rendering the next; the contents of the output buffer are left as is. Has Value 0The frame's region of the output buffer is to be cleared to fully transparent black before rendering the next frame. Has Value 1the frame's region of the output buffer is to be reverted to the previous contents before rendering the next frame. Has Value 20 value1 value"Generic header used in PNG images.Image width in number of pixelImage height in number of pixelNumber of bit per sample5Kind of png image (greyscale, true color, indexed...)Compression method used Must be 06If the image is interlaced (for progressive rendering)9Value used to identify a png chunk, must be 4 bytes long.!Parse a palette from a png chunk.4Parse method for a png chunk, without decompression.aImplementation of the get method for the PngRawImage, unpack raw data, without decompressing it.WSignature signalling that the following data will be a png image in the png bit stream(Helper function to help pack signatures.9Signature for the header chunk of png (must be the first)HSignature for a palette chunk in the pgn file. Must occure before iDAT.3Signature for a data chuck (with image parts in it)>Signature for the last chunk of a png image, telling the end.*From the Annex D of the png specification.SCompute the CRC of a raw buffer, as described in annex D of the PNG specification.jZ. None>L;Same as for DcCoefficient, to provide nicer type signatures+Apply a quantization matrix to a macroblock8Unpack an int of the given size encoded from MSB to LSB.Given a size coefficient (how much a pixel span horizontally and vertically), the position of the macroblock, return a list of indices and value to be stored in an array (like the final image)This is one of the most important function of the decoding, it form the barebone decoding pipeline for macroblock. It's all there is to know for macro block transformation!     Component countWidth coefficientHeight coefficientxyComponent index         None35FN>Iteration from to n in monadic context, without data keeping. None !"*35;CKLNgDependent storage used for metadatas. All metadatas of a given kind are unique within this container.RThe current data structure is based on list, so bad performances can be expected.AElement describing a metadata and it's (typed) associated value.%Encode values for unknown informationyStore various additional information about an image. If something is not recognized, it can be stored in an unknown tag. Dot per inch on this x axis. Dot per inch on this y axis. Image width in pixel. Relying on the metadata for this information can avoid the full decompression of the image. Ignored for image writing. Image height in pixels. Relyiung on the metadata for this information can void the full decomrpession of the image. Ignored for image writing.9 unlikely to be decoded, but usefull for metadata writing Exif tag and associated data.5Type describing the original file format of the ilfe.Right based union!Strict left fold of the metadatas!foldMap equivalent for metadatas.URemove an element of the given keys from the metadatas. If not present does nothing.%Search a metadata with the given key.cInsert an element in the metadatas, if an element with the same key is present, it is overwritten.'Create metadatas with a single element.Empty metadatas. Favor Conversion from dpm to dpiConversion from dpi to dpmConversion dpcm -> dpi=Create metadatas indicating the resolution, with DpiX == DpiY6Create metadatas holding width and height information.3Create simple metadatas with Format, Width & Height?Create simple metadatas with Format, Width, Height, DpiX & DpiY8 !"#$%&'// !"#$%&'None 345>LN SAll the instance of this class can be written as a bitmap file using this library.(Pixels per meter)Pixels per meter*in bytesVTry to decode a bitmap image. Right now this function can output the following image:y~Same as  but also extracts metadata./Write an image in a file use the bitmap format.NEncode an image into a bytestring in .bmp format ready to be written on disk.Equivalent to ) but also store the following metadatas: QWrite a dynamic image in a .bmp image file if possible. The same restriction as  apply.@Encode a dynamic image in BMP if possible, supported images are:y~+8Convert an image to a bytestring ready to be serialized. Equivalent to +) but also store the following metadatas: 5,-./0123456789():;<=>*?@ABCDEFGHIJKLMN+ OPQRS   ,-./01 23456789():;<=>*?@ABCDEFGHIJKLMN+ OPQRSNone4>LNTStored on 3 bitsUSection 20 of spec-gif89aVSection 18 of spec-gif89aWStored on 16 bitsXStored on 16 bitsYStored on 8 bitsZStored on 1 bit[Stored on 3 bits\Stored on 1 bit]Stored on 3 bits 7Help to control the behaviour of GIF animation looping. /The animation will stop once the end is reached 2The animation will restart once the end is reached 0The animation will repeat n times before stoping]Delay to wait before showing the next Gif image. The delay is expressed in 100th of seconds.tTransform a raw gif image to an image, without modifying the pixels. This function can output the following images:~uTransform a raw gif image to an image, without modifying the pixels. This function can output the following images:~-Metadatas include Width & Height information.VTransform a raw gif to a list of images, representing all the images of an animation..Extract a list of frame delays from a raw gif.,Default palette to produce greyscale images.'Encode a gif animation to a bytestring.#Every image must have the same size3Every palette must have between one and 256 colors.)Encode a greyscale image to a bytestring.VEncode an image with a given palette. Can return errors if the palette is ill-formed.,A palette must have between 1 and 256 colors,Write a greyscale in a gif file on the disk.4Write a list of images as a gif animation in a file.#Every image must have the same size3Every palette must have between one and 256 colors.+Write a gif image with a palette to a file.,A palette must have between 1 and 256 colorsp^_`abcdefghijklmnopqrTstuvwxyz{|U}~VWXYZ[\]            @^_`abcdefghijklmnopqrTstuvwxyz{|U }~VWXYZ[\]    None>FL;Decode an HDR (radiance) image, the resulting image can be:5Equivalent to decodeHDR but with aditional metadatas.FWrite an High dynamic range image into a radiance image file on disk.QWrite a RLE encoded High dynamic range image into a radiance image file on disk.^Encode an High dynamic range image into a radiance image file format. Alias for encodeRawHDRZEncode an High dynamic range image into a radiance image file format. without compression Encode an High dynamic range image into a radiance image file format using a light RLE compression. Some problems seem to arise with some image viewer.1   (  None 345>FILN!DThis typeclass determine if a pixel can be saved in the TGA format."TTransform a raw tga image to an image, without modifying the underlying pixel type..This function can output the following images:y~#1Equivalent to decodeTga but also provide metadata$9Helper function to directly write an image a tga on disk.%LTransform a compatible image to a raw bytestring representing a Targa file.Q!      Type witness Type witness"#$% !"#$%&'()!"#$%"#!%$*!      "#$% !"#$%&'()None!"*+,-./0123456736 *+,-./01234567None 34>LN&'Encode an image into a png if possible.'QTransform an image into a png encoded bytestring, ready to be written as a file.(TEncode a png using some metadatas. The following metadata keys will be stored in a tEXt field :  !"#$%&'(* using the key present in the constructor..the followings metadata will bes tored in the gAMA chunk.),The followings metadata will be stored in a pHYs chunk )<Helper function to directly write an image as a png on disk.*hWrite a dynamic image in a .png image file if possible. The same restriction as encodeDynamicPng apply.+lEncode a paletted image as a color indexed 8-bit PNG. the palette must have between 1 and 256 values in it.,Equivalent to + but allow writing of metadatas.-@Encode a dynamic image in PNG if possible, supported images are:yz|}~&'(8)9:;<=*+,->?@ABCDE&'()*+,-&'(8)9:;<=*+,->?@ABCDE None 34>FLN FTDirectly stolen from the definition in the standard (on W3C page), pixel predictor.GSimple structure used to hold information about Adam7 deinterlacing. A structure is used to avoid pollution of the module namespace.HThe real info about the matrix.IApply a filtering method on a reduced image. Apply the filter on each line, using the previous line (the one above it) to perform some prediction on the value.J!Unpack lines where bit depth is 8K!Unpack lines where bit depth is 1L Unpack lines when bit depth is 2M_Transform a scanline to a bunch of bytes. Bytes are then packed into pixels at a further step.N]Given data and image size, recreate an image with deinterlaced data for PNG's adam 7 method.ODdeinterlace picture in function of the method indicated in the iHDR.Transform a raw png image to an image, without modifying the underlying pixel type. If the image is greyscale and < 8 bits, a transformation to RGBA8 is performed. This should change in the future. The resulting image let you manage the pixel types..This function can output the following images:yz|}~/Same as .3 but also extract meta datas present in the files.'PQRFGSTUVWXYHZ[I Image sizeJKL\]M^_NO`abcdefghij./ &'()*+,-./ &'(./)-+,* PQRFGSTUVWXYHZ[IJKL\]M^_NO`abcdefghij./*None klmnopqrssklmnopqrs None 3457>FLN0CClass defining which pixel types can be serialized in a Tiff file.t.Final image and size, return offset and vectorudwhile mandatory some images don't put correct rowperstrip. So replacing 0 with actual image height.1wDecode a tiff encoded image while preserving the underlying pixel type (except for Y32 which is truncated to 16 bits)..This function can output the following images: yz|}~2Like 1: but also provides some metdata present in the Tiff file.The metadata extracted are the  & - information alongside the EXIF informations.3RTransform an image into a Tiff encoded bytestring, ready to be written as a file.4=Helper function to directly write an image as a tiff on disk.vHThe Word8 instance is just a passthrough, to avoid copying memory twiceP0wxyz{|}~t(Type witness, just for the type checker.(Temporary buffer handling decompression.Line size in pixelsWrite index, in bytessize, in bytesStride Final bufferu1234v0123412034.0wxyz{|}~tu1234v+None None 345>FLN 5:Helper type class describing all JPG-encodable pixel typesRecord gathering all information to encode a component from the source image. Previously was a huge tuple burried in the code1Assume the macro block is initialized with zeroesYDecompress a macroblock from a bitstream given the current configuration from the frame.nThis pseudo interpreter interpret the Jpg frame for the huffman, quant table and restart interval parameters.6Try to decompress and decode a jpeg file. The colorspace is still YCbCr if you want to perform computation on the luma part. You can convert it to RGB using K from the I typeclass..This function can output the following images:y|~7Equivalent to 6 but also extracts metadatas.4Extract the following metadatas from the JFIF block: ,Exif metadata are also extracted if present.8zEncode an image in jpeg at a reasonnable quality level. If you want better quality or reduced file size, you should use 99{Function to call to encode an image to jpeg. The quality factor should be between 0 and 100 (100 being the best quality).:Equivalent to 9I, but will store the following metadatas in the file using a JFIF block: ;Equivalent to 9I, but will store the following metadatas in the file using a JFIF block: NThis function also allow to create JPEG files with the following color space:Y (w) for greyscale.RGB (n)) with no color downsampling on any planeCMYK (d)) with no color downsampling on any planeG5component indexxyTree used for DC coefficientTree used for Ac coefficientCurrent quantization table#A zigzag table, to avoid allocationPrevious dc valuecomponent indexxyComponent indexxyComponent indexxyResult image to write into67 Source image*Mutable block where to put extracted blockPlaneX sampling factorY sampling factorSample per pixelBlock xBlock y89Quality factorImage to encode Encoded JPEG:Quality factorImage to encode Encoded JPEG;Quality factorImage to encode Encoded JPEG56789:;679:;85+56789:; None>L<zThis function will try to do anything to encode an image as RADIANCE, make all color conversion and such. Equivalent of  decodeImage for radiance encoding=vThis function will try to do anything to encode an image as JPEG, make all color conversion and such. Equivalent of  decodeImage for jpeg encoding Save Y or YCbCr Jpeg only, all other colorspaces are converted. To save a RGB or CMYK JPEG file, use the ; function>uThis function will try to do anything to encode an image as PNG, make all color conversion and such. Equivalent of  decodeImage for PNG encoding?xThis function will try to do anything to encode an image as a Tiff, make all color conversion and such. Equivalent of  decodeImage for Tiff encoding@xThis function will try to do anything to encode an image as bitmap, make all color conversion and such. Equivalent of  decodeImage for Bitmap encodingAThis function will try to do anything to encode an image as a gif, make all color conversion and quantization. Equivalent of  decodeImage for gif encodingBThis function will try to do anything to encode an image as a tga, make all color conversion and quantization. Equivalent of  decodeImage for tga encoding <=>?@AB<=>?@AB=>A@?<B <=>?@ABNone 3457>FL0Return the first Right thing, accumulating errorCUEncode a full color image to a gif by applying a color quantization algorithm on it.DTWrite a full color image to a gif by applying a color quantization algorithm on it.EmHelper function to create a gif animation. All the images of the animation are separated by the same delay.FIHelper function to write a gif animation on disk. See encodeGifAnimationGJLoad an image file without even thinking about it, it does everything as IHEquivalent to G but also providing metadatas.IIf you want to decode an image in a bytestring without even thinking in term of format or whatever, this is the function to use. It will try to decode in each known format and if one decoding succeeds, it will return the decoded image in it's own colorspace.JConvert by any mean possible a dynamic image to an image in RGBA. The process can lose precision while converting from 16bits pixels or Floating point pixels.KConvert by any mean possible a dynamic image to an image in RGB. The process can lose precision while converting from 16bits pixels or Floating point pixels. Any alpha layer will be droppedLEquivalent to IB, but also provide potential metadatas present in the given file.M>Helper function trying to load a png file from a file on disk.N>Helper function trying to load a gif file from a file on disk.O=Helper function trying to load tiff file from a file on disk.PGHelper function trying to load all the images of an animated gif file.QTry to load a jpeg file and decompress. The colorspace is still YCbCr if you want to perform computation on the luma part. You can convert it to RGB using colorSpaceConversion.R@Try to load a .bmp file. The colorspace would be RGB, RGBA or Y.SWTry to load a .pic file. The colorspace can only be RGB with floating point precision.T"Try to load a .tga file from disk.USave an image to a '.jpg'2 file, will do everything it can to save an image.VSave an image to a '.gif', file, will do everything it can to save it.WSave an image to a '.tiff'2 file, will do everything it can to save an image.XSave an image to a '.hdr'2 file, will do everything it can to save an image.YSave an image to a '.png'[ file, will do everything it can to save an image. For example, a simple transcoder to png transcodeToPng :: FilePath -> FilePath -> IO () transcodeToPng pathIn pathOut = do eitherImg <- readImage pathIn case eitherImg of Left _ -> return () Right img -> savePngImage pathOut imgZSave an image to a '.bmp'2 file, will do everything it can to save an image.% CDEF GHI  JKLMNOPQRSTUVWXYZ HOPQRSTUVWXYZ[\]^_`abcdefghijknopqrstvwxyz{|}~    !"$%&'()*+-.0134689CDEFGHIJKLMNOPQRSTUVWXYZGHILKJHZUVYWXRNPCD    EFQ689&'(M.)+-*T"!%$O0134Sxyz{|}~OPQRSTUVWXYZ[\]wvtrspqnojkhi`a^_fgdebc$ CDEF GHI  JKLMNOPQRSTUVWXYZ ,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~) !"#$%&'(      !"#$%& ' ( ) * +,-./0123 4 5 6 7 8 9 : ; < = > ? @ A B C D E F G HIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrsstuvwxyz{|}~      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLLMNOPQRR?STUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyyz{|}~s      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdeffghijklmnopqrstuvwxyz{|}~      !!"#$%&'()*+,-./0123456789:;<=>?@ABBCDEFFGHIJKLMMNOPQRRSTUVWXYZ[\9]^_`abcd:efghijklmnopqrstuvw)xyz{O|}~                                                     w                                           ! " # $ % & ' ( ) * + , - . / 0 1 2  3  4 5*6*7*8*9*:*;*<*=*> ? @ A B C D E E F G H I J J K K L L M N O P Q R R S T U V W X Y Z [ \ ] ^ _ ` a b c d e f g h i j k l m n o p q r s t u v w x y z { | } ~      +++                                             O                        Juicy_75qHy60wulk5MiLqOwbyH8Codec.Picture.Metadata.ExifCodec.Picture.TypesCodec.Picture.ColorQuant Codec.PictureCodec.Picture.MetadataCodec.Picture.BitmapCodec.Picture.GifCodec.Picture.HDRCodec.Picture.TgaCodec.Picture.PngCodec.Picture.TiffCodec.Picture.JpgCodec.Picture.SavingCodec.Picture.Tiff.TypesCodec.Picture.InternalHelper"Codec.Picture.VectorByteConversionCodec.Picture.BitWriterCodec.Picture.Gif.LZWCodec.Picture.Gif.LZWEncodingCodec.Picture.Jpg.DefaultTableCodec.Picture.Jpg.TypesCodec.Picture.Jpg.FastDctCodec.Picture.Jpg.FastIdctCodec.Picture.Png.TypeCodec.Picture.Jpg.CommonCodec.Picture.Jpg.ProgressiveDpiXDpiYCodec.Picture.Png.MetadataCodec.Picture.Png.ExportTitle DescriptionAuthor CopyrightSoftwareComment DisclaimerSourceWarningUnknownGammaCodec.Picture.Tiff.MetadataCodec.Picture.Jpg.MetadataExifDataExifNoneExifLong ExifShort ExifString ExifUndefined ExifShorts ExifLongs ExifRationalExifSignedRationalExifIFDExifTagTagPhotometricInterpretationTagCompression TagImageWidthTagImageLengthTagXResolutionTagYResolutionTagResolutionUnitTagRowPerStripTagStripByteCountsTagStripOffsetsTagBitsPerSample TagColorMap TagTileWidth TagTileLength TagTileOffsetTagTileByteCountTagSamplesPerPixel TagArtistTagDocumentName TagSoftwareTagPlanarConfigurationTagOrientationTagSampleFormat TagInkSetTagSubfileType TagFillOrder TagYCbCrCoeffTagYCbCrSubsamplingTagYCbCrPositioningTagReferenceBlackWhite TagXPosition TagYPositionTagExtraSampleTagImageDescription TagPredictor TagCopyrightTagMakeTagModel TagDateTime TagGPSInfo TagJpegProcTagJPEGInterchangeFormatTagJPEGInterchangeFormatLengthTagJPEGRestartIntervalTagJPEGLosslessPredictorsTagJPEGPointTransformsTagJPEGQTablesTagJPEGDCTablesTagJPEGACTables TagExifOffset TagUnknown tagOfWord16 word16OfTagPackeablePixelPackedRepresentation packPixel unpackPixelLumaPlaneExtractable computeLumaextractLumaPlane TraversalColorSpaceConvertible convertPixel convertImageColorConvertible promotePixel promoteImagePixelPixelBaseComponentmixWith mixWithAlpha pixelOpacitycomponentCountcolorMappixelBaseIndexmutablePixelBaseIndexpixelAt readPixel writePixel unsafePixelAtunsafeReadPixelunsafeWritePixel PixelRGBA16 PixelRGBA8 PixelCMYK16 PixelCMYK8 PixelYCbCr8 PixelRGBF PixelRGB16 PixelYCbCrK8 PixelRGB8 PixelYA16PixelYA8PixelFPixel32Pixel16Pixel8 DynamicImageImageY8ImageY16ImageYFImageYA8 ImageYA16 ImageRGB8 ImageRGB16 ImageRGBF ImageRGBA8 ImageRGBA16 ImageYCbCr8 ImageCMYK8 ImageCMYK16 MutableImagemutableImageWidthmutableImageHeightmutableImageDataTransparentPixeldropTransparencygetTransparency PlaneBlack PlaneYellow PlaneMagenta PlaneCyanPlaneCbPlaneCr PlaneLuma PlaneAlpha PlaneBlue PlaneGreenPlaneRed ColorPlanePaletteImage imageWidth imageHeight imageDataextractComponentunsafeExtractComponentdropAlphaLayer freezeImage thawImageunsafeThawImageunsafeFreezeImagecreateMutableImagenewMutableImage dynamicMapdynamicPixelMap generateImage withImagegenerateFoldImage pixelFold pixelFoldM pixelFoldMappixelMap imagePixels imageIPixels pixelMapXYzipPixelComponent3gammaCorrection toneMapping fillImageWithunsafeWritePixelBetweenAtreadPackedPixelAtwritePackedPixelAtPaletteOptionspaletteCreationMethodenableImageDitheringpaletteColorCountPaletteCreationMethod MedianMeanCutUniformdefaultPaletteOptions palettizeimageFromUnsafePtr MetadatasElem:=>ValueIntDoubleStringKeysFormatWidthHeightExif SourceFormat SourceJpeg SourceGif SourceBitmap SourceTiff SourcePng SourceHDR SourceTGAfoldl'foldMapdeletelookupinsert singletonemptydotsPerMeterToDotPerInchdotPerInchToDotsPerMeterdotsPerCentiMeterToDotPerInch mkDpiMetadatamkSizeMetadata basicMetadatasimpleMetadata BmpEncodable decodeBitmapdecodeBitmapWithMetadata writeBitmap encodeBitmapencodeBitmapWithMetadatawriteDynamicBitmapencodeDynamicBitmap"encodeBitmapWithPaletteAndMetadata GifLooping LoopingNeverLoopingForever LoopingRepeatGifDelay decodeGifdecodeGifWithMetadatadecodeGifImagesgetDelaysGifImages greyPaletteencodeGifImagesencodeGifImageencodeGifImageWithPalette writeGifImagewriteGifImageswriteGifImageWithPalette decodeHDRdecodeHDRWithMetadatawriteHDRwriteRLENewStyleHDR encodeHDR encodeRawHDRencodeRLENewStyleHDR TgaSaveable decodeTgadecodeTgaWithMetadatawriteTga encodeTga PngSavable encodePngencodePngWithMetadatawritePngwriteDynamicPngencodePalettedPngencodePalettedPngWithMetadataencodeDynamicPng decodePngdecodePngWithMetadata TiffSaveable decodeTiffdecodeTiffWithMetadata encodeTiff writeTiff JpgEncodable decodeJpegdecodeJpegWithMetadata encodeJpegencodeJpegAtQualityencodeJpegAtQualityWithMetadata%encodeDirectJpegAtQualityWithMetadataimageToRadiance imageToJpg imageToPng imageToTiff imageToBitmap imageToGif imageToTgaencodeColorReducedGifImagewriteColorReducedGifImageencodeGifAnimationwriteGifAnimation readImagereadImageWithMetadata decodeImage convertRGBA8 convertRGB8decodeImageWithMetadatareadPngreadGifreadTiff readGifImagesreadJpeg readBitmapreadHDRreadTGA saveJpgImage saveGifImage saveTiffImagesaveRadianceImage savePngImage saveBmpImage$fNFDataExifData$fNFDataExifTagExtraSampleUnspecifiedExtraSampleAssociatedAlphaExtraSampleUnassociatedAlphaTiffMonochromeWhite0TiffMonochromeTiffRGB TiffPaletedTiffTransparencyMaskTiffCMYK TiffYCbCr TiffCIELab BinaryParam orderIfdByTagsetupIfdOffsets ExtraSampleTiffColorspaceImageFileDirectory ifdIdentifierifdTypeifdCount ifdOffset ifdExtendedTiffSampleFormatTiffSampleUint TiffSampleIntTiffSampleDoubleTiffSampleUnknown Predictor PredictorNonePredictorHorizontalDifferencingIfdTypeTypeByte TypeAscii TypeShortTypeLong TypeRational TypeSByte TypeUndefinedTypeSignedShortTypeSignedLongTypeSignedRational TypeFloat TypeDoubleTiffCompressionCompressionNoneCompressionModifiedRLECompressionLZWCompressionJPEGCompressionPackBitTiffPlanarConfigurationPlanarConfigContigPlanarConfigSeparate TiffHeader hdrEndianness hdrOffsetgetPputP Endianness EndianLittle EndianBigplanarConfgOfConstantconstantToPlaneConfigurationpredictorOfConstantcleanImageFileDirectory fetchExtendedunpackSampleFormatpackPhotometricInterpretationunpackPhotometricInterpretationcodeOfExtraSampleunPackCompressionpackCompression$fBinaryParamEndianness[])$fBinaryParamEndiannessImageFileDirectory$fBinaryParamByteString(,)$fBinaryParam(,,)ExifData$fBinaryParamEndiannessExifTag$fBinaryParamEndiannessIfdType$fBinaryTiffHeader$fBinaryParamEndiannessWord32$fBinaryParamEndiannessInt32$fBinaryParamEndiannessWord16$fBinaryEndiannessdecoderunGet runGetStrictgetRemainingBytesgetRemainingLazyBytesGenSTtoComponentIndexbaseGHC.Errerror Data.FoldableFoldableGHC.Basemap$fColorConvertibleaa genActionlineFoldstridegenerateMutableImagewritePx freezeGenST scaleBitsoneHalffixrYTabgYTabbYTabrCbTabgCbTabbCbTabgCrTabbCrTabcrRTabcbBTabcrGTabcbGTabintegralRGBToCMYK$fPackeablePixelPixelYA8$fPackeablePixelPixelYA16$fPackeablePixelPixelCMYK16$fPackeablePixelPixelCMYK8$fPackeablePixelPixelRGBA16$fPackeablePixelPixelRGBA8$fPackeablePixelFloat$fPackeablePixelWord32$fPackeablePixelWord16$fPackeablePixelWord8!$fColorPlanePixelCMYK16PlaneBlack"$fColorPlanePixelCMYK16PlaneYellow#$fColorPlanePixelCMYK16PlaneMagenta $fColorPlanePixelCMYK16PlaneCyan,$fColorSpaceConvertiblePixelCMYK16PixelRGB16$fPixelPixelCMYK16 $fColorPlanePixelCMYK8PlaneBlack!$fColorPlanePixelCMYK8PlaneYellow"$fColorPlanePixelCMYK8PlaneMagenta$fColorPlanePixelCMYK8PlaneCyan*$fColorSpaceConvertiblePixelRGB8PixelCMYK8-$fColorSpaceConvertiblePixelYCbCrK8PixelCMYK8,$fColorSpaceConvertiblePixelYCbCrK8PixelRGB8$fPixelPixelYCbCrK8*$fColorSpaceConvertiblePixelCMYK8PixelRGB8$fPixelPixelCMYK8$fColorPlanePixelYCbCr8PlaneCr$fColorPlanePixelYCbCr8PlaneCb $fColorPlanePixelYCbCr8PlaneLuma+$fColorSpaceConvertiblePixelYCbCr8PixelRGB8+$fColorSpaceConvertiblePixelRGB8PixelYCbCr8$fColorSpaceConvertibleaa$fPixelPixelYCbCr8!$fColorPlanePixelRGBA16PlaneAlpha $fColorPlanePixelRGBA16PlaneBlue!$fColorPlanePixelRGBA16PlaneGreen$fColorPlanePixelRGBA16PlaneRed'$fTransparentPixelPixelRGBA16PixelRGB16$fPixelPixelRGBA16 $fColorPlanePixelRGBA8PlaneAlpha$fColorPlanePixelRGBA8PlaneBlue $fColorPlanePixelRGBA8PlaneGreen$fColorPlanePixelRGBA8PlaneRed'$fColorConvertiblePixelRGBA8PixelRGBA16$fPixelPixelRGBA8$fLumaPlaneExtractablePixelRGB8$fColorPlanePixelRGB8PlaneBlue$fColorPlanePixelRGB8PlaneGreen$fColorPlanePixelRGB8PlaneRed&$fColorConvertiblePixelRGB8PixelRGBA16%$fColorConvertiblePixelRGB8PixelRGB16$$fColorConvertiblePixelRGB8PixelRGBF%$fColorConvertiblePixelRGB8PixelRGBA8$fPixelPixelRGB8 $fLumaPlaneExtractablePixelRGB16'$fColorConvertiblePixelRGB16PixelRGBA16,$fColorSpaceConvertiblePixelRGB16PixelCMYK16$fColorPlanePixelRGB16PlaneBlue $fColorPlanePixelRGB16PlaneGreen$fColorPlanePixelRGB16PlaneRed$fPixelPixelRGB16$fColorPlanePixelRGBFPlaneBlue$fColorPlanePixelRGBFPlaneGreen$fColorPlanePixelRGBFPlaneRed$fPixelPixelRGBF!$fTransparentPixelPixelYA16Word16$fColorPlanePixelYA16PlaneAlpha$fColorPlanePixelYA16PlaneLuma&$fColorConvertiblePixelYA16PixelRGBA16$fPixelPixelYA16$fLumaPlaneExtractablePixelYA8$fTransparentPixelPixelYA8Word8$fColorPlanePixelYA8PlaneAlpha$fColorPlanePixelYA8PlaneLuma$$fColorConvertiblePixelYA8PixelRGBA8#$fColorConvertiblePixelYA8PixelRGB8$fPixelPixelYA8 $fColorConvertibleFloatPixelRGBF $fPixelFloat $fPixelWord32#$fColorConvertibleWord16PixelRGBA16"$fColorConvertibleWord16PixelRGB16!$fColorConvertibleWord16PixelYA16 $fPixelWord16!$fColorConvertibleWord8PixelRGBA8 $fColorConvertibleWord8PixelRGB8$fColorConvertibleWord8Word16$fColorConvertibleWord8Float$fColorConvertibleWord8PixelYA8 $fPixelWord8!$fLumaPlaneExtractablePixelYCbCr8 $fLumaPlaneExtractablePixelRGBA8$fLumaPlaneExtractablePixelRGBF$fLumaPlaneExtractableFloat$fLumaPlaneExtractableWord32$fLumaPlaneExtractableWord16$fLumaPlaneExtractableWord8$fNFDataDynamicImage$fNFDataMutableImage $fNFDataImage%$fTransparentPixelPixelRGBA8PixelRGB8Fold ApplicativemedianMeanCutQuantizationuniformQuantizationfold intLengthlengthGHC.NumNumAxisRAxisGAxisBAxisClustervalue meanColordimscolors PackedRGBPairisColorCountBelow vecToPalette listToPalettebitDiv3dither mkPaletteVecinffromRGB8toRGB8meanRGBminimalmaximalextrems volAndDims unpackFold mkClustermaxAxis subdivide rgbIntPack rgbIntUnpack initClustersplitclustersdist2PxnearestColorIdx $fOrdCluster $fEqCluster$fApplicativeFold $fFunctorFold blitVector toByteString BoolReader BoolStatesetDecodedString byteAlignJpgsetDecodedStringJpg writeBits' writeBitsGifBoolWriteStateRef bwsCurrBuffer bwsBufferListbwsWrittenWords bwsBitAcc bwsBitReadedemptyBoolState runBoolReaderrunBoolReaderWithexecBoolReader initBoolStateinitBoolStateJpg getNextBitJpg getNextIntJpgsetDecodedStringMSBgetNextBitsMSBFirstgetNextBitsLSBFirst getNextBitdefaultBufferSizenewWriteStateReffinalizeBoolWriterforceBufferFlushing'flushCurrentBuffer'byteStringFromVector setBitCount'resetBitCount' pushByte'lzw TiffVariant GifVariantOldTiffVariant.!!!...!!!...<-...<-.. duplicateData rangeSetter decodeLzw isOldTiffLZW decodeLzwTiffTrieNode trieIndextrieSubTrie emptyNode initialTrie lookupUpdate lzwEncode DctComponent MacroBlock HuffmanTreeBranchLeafEmptymakeMacroBlockbuildHuffmanTreedefaultDcLumaHuffmanTabledefaultDcChromaHuffmanTabledefaultAcLumaHuffmanTable HuffmanTable DcComponent AcComponentQuantificationTableHuffmanWriterCodeHuffmanPackedTreepackHuffmanTreemakeInverseTablescaleQuantisationMatrixhuffmanPackedDecodedefaultLumaQuantizationTabledefaultChromaQuantizationTabledefaultDcLumaHuffmanTreedefaultDcChromaHuffmanTreedefaultAcLumaHuffmanTreedefaultAcChromaHuffmanTreedefaultAcChromaHuffmanTablehuffmanTableClasshuffmanTableDest TableListquantPrecisionquantDestinationspectralSelectionsuccessiveApproxHighsuccessiveApproxLowdcEntropyCodingTableacEntropyCodingTablehorizontalSamplingFactorverticalSamplingFactorJFifUnitJFifUnitUnknownJFifPixelsPerInchJFifPixelsPerCentimeter AdobeUnknown AdobeYCbCr AdobeYCckMutableMacroBlock DcCoefficient parseJF__createEmptyMutableMacroBlockRestartInterval JpgImageKind BaseLineDCTProgressiveDCTJpgHuffmanTableSpec huffSizes huffCodesSizeCalculable calculateSizeJpgQuantTableSpec quantTable JpgScanHeader scanLengthscanComponentCountscansJpgScanSpecificationcomponentSelectorJpgImagejpgFrame JpgComponentcomponentIdentifierquantizationTableDestJpgFrameHeaderjpgFrameHeaderLengthjpgSamplePrecision jpgHeightjpgWidthjpgImageComponentCount jpgComponents JpgJFIFApp0 _jfifUnit _jfifDpiX _jfifDpiY_jfifThumbnail JpgAdobeApp14_adobeDctVersion _adobeFlag0 _adobeFlag1_adobeTransformAdobeTransform JpgColorSpaceJpgColorSpaceYCbCrJpgColorSpaceYCCJpgColorSpaceYJpgColorSpaceYAJpgColorSpaceYCCAJpgColorSpaceYCCKJpgColorSpaceCMYKJpgColorSpaceRGBJpgColorSpaceRGBAJpgFrame JpgAppFrame JpgAdobeAPP14JpgJFIFJpgExif JpgExtension JpgQuantTableJpgHuffmanTable JpgScanBlobJpgScansJpgIntervalRestart JpgFrameKindJpgBaselineDCTHuffmanJpgExtendedSequentialDCTHuffmanJpgProgressiveDCTHuffmanJpgLosslessHuffman#JpgDifferentialSequentialDCTHuffman$JpgDifferentialProgressiveDCTHuffmanJpgDifferentialLosslessHuffmanJpgExtendedSequentialArithmeticJpgProgressiveDCTArithmeticJpgLosslessArithmetic&JpgDifferentialSequentialDCTArithmetic'JpgDifferentialProgressiveDCTArithmetic!JpgDifferentialLosslessArithmeticJpgQuantizationTableJpgHuffmanTableMarkerJpgStartOfScan JpgEndOfImage JpgAppSegmentJpgExtensionSegmentJpgRestartIntervalJpgRestartIntervalEnd eatUntilCodetakeCurrentFrameputFramecommonMarkerFirstByte checkMarkerextractScanContent parseAdobe14 parseExif parseFramesbuildPackedHuffmanTreesecondStartOfFrameByteOfKindput4BitsOfEach get4BitOfEachprintMacroBlockprintPureMacroBlock dctBlockSize$fBinaryJpgScanHeader$fBinaryJpgScanSpecification$fBinaryJpgFrameHeader$fBinaryJpgComponent$fBinaryRestartInterval$fBinaryJpgFrameKind$fBinaryJpgImage$fBinaryJpgHuffmanTableSpec#$fSizeCalculableJpgHuffmanTableSpec$fBinaryJpgQuantTableSpec!$fSizeCalculableJpgQuantTableSpec$fBinaryTableList$fSizeCalculableJpgScanHeader$$fSizeCalculableJpgScanSpecification$fSizeCalculableJpgComponent$fSizeCalculableJpgFrameHeader$fBinaryJpgAdobeApp14$fBinaryAdobeTransform$fBinaryJpgJFIFApp0$fBinaryJFifUnit referenceDctfastDctLibJpeg cONST_BITS pASS1_BITSfIX_0_298631336fIX_0_390180644fIX_0_541196100fIX_0_765366865fIX_0_899976223fIX_1_175875602fIX_1_501321110fIX_1_847759065fIX_1_961570560fIX_2_053119869fIX_2_562915447fIX_3_072711026 cENTERJSAMPLEfastIdctmutableLevelShift IDctStagex0x1x2x3x4x5x6x7x8iclipw1w2w3w5w6w7idctRowidctColPngInterlaceMethodPngNoInterlacePngInterlaceAdam7 PngFilter FilterNone FilterSubFilterUp FilterAverage FilterPaeth PngLowLevelpngImage pngChunksPngChunk pngChunkDatapngChunkSignature PngRawChunk PngPalette PngRawImage PngImageTypeframeSequenceNum frameWidth frameHeight APngBlendOpAPngBlendSource APngBlendOverAPngFrameDisposalAPngDisposeNoneAPngDisposeBackgroundAPngDisposePreviousPngUnitUnknown PngUnitMeterPngIHdrwidthheightbitDepth colourTypecompressionMethod filterMethodinterlaceMethodChunkSignature parsePalette parseChunksparseRawPngImage pngSignature signature iHDRSignature pLTESignature iDATSignature iENDSignature pngCrcTable pngComputeCrc chunkLength chunkTypechunkCRC chunkDataheaderchunks PngGreyscale PngTrueColourPngIndexedColorPngGreyscaleWithAlphaPngTrueColourWithAlphaAPngFrameControl frameLeftframeTopframeDelayNumeratorframeDelayDenuminator frameDisposal frameBlendingAPngAnimationControlanimationFrameCountanimationPlayCountPngGamma getPngGammaPngPhysicalDimensionpngDpiXpngDpiYpngUnitPngUnit mkRawChunk tRNSSignaturegammaSignature pHYsSignature tEXtSignature zTXtSignatureanimationControlSignatureimageTypeOfCode chunksWithSig$fBinaryPngImageType$fBinaryPngInterlaceMethod$fBinaryPngIHdr$fBinaryPngRawChunk$fBinaryPngRawImage$fBinaryPngFilter$fBinaryPngGamma$fBinaryPngPhysicalDimension$fBinaryPngUnitDctCoefficients deQuantize unpackIntunpackMacroBlockdecodeMacroBlockJpgUnpackerParameter dcHuffmanTree acHuffmanTreecomponentIndexrestartIntervalcomponentWidthcomponentHeight subSamplingcoefficientRangesuccessiveApprox readerIndex indiceVector blockIndex blockMcuX blockMcuYdecodeRestartInterval decodeIntdecodeRrrrSsssdcCoefficientDecodeinverseDirectCosineTransform zigZagOrderzigZagReorderForwardvzigZagOrderForwardzigZagReorderForward zigZagReorder rasterMap pixelClamplineMap ComponentDatacomponentIndicescomponentBlocks componentIdcomponentBlockCountUnpackercreateMcuLineIndices decodeFirstDCdecodeRefineDc decodeFirstAcdecodeRefineAcprepareUnpackerprogressiveUnpackunionmempty getMetadatasEquivkeyEq$fMonoidMetadatas $fNFDataElem $fNFDataValue$fNFDataSourceFormat xResolution yResolutionfileSizeencodeBitmapWithPalette bitsPerPixel bmpEncodedefaultPalette BmpPalette BmpInfoHeadersizeplanes bitPerPixelbitmapCompression byteImageSize colorCountimportantColours BmpHeadermagicIdentifier reserved1 reserved2 dataOffsetbitmapMagicIdentifiersizeofBmpHeader sizeofBmpInfo putPalette stridePutdecodeImageRGBA8decodeImageRGB8 decodeImageY8pixelGetmetadataOfHeader getBitfield linePaddingextractDpiOfMetadata$fBmpEncodablePixelRGB8$fBmpEncodablePixelRGBA8$fBmpEncodableWord8$fBinaryBmpInfoHeader$fBinaryBmpHeadergceDisposalMethodImageDescriptorLogicalScreenDescriptor screenWidth screenHeightbackgroundIndex hasGlobalMapcolorResolutionisColorTableSortedcolorTableSizeGifFile gifHeader gifImagesgifLoopingBehaviour GifHeader gifVersiongifScreenDescriptor gifGlobalMapBlock BlockImageBlockGraphicControlGifImage imgDescriptorimgLocalPaletteimgLzwRootSizeimgDataGraphicControlExtensiongceUserInputFlaggceTransparentFlaggceDelaygceTransparentColorIndexDisposalMethod DisposalAny DisposalDoNotDisposalRestoreBackgroundDisposalRestorePreviousDisposalUnknowngDescPixelsFromLeftgDescPixelsFromTopgDescImageWidthgDescImageHeightgDescHasLocalMapgDescIsInterlacedgDescIsImgDescriptorSortedgDescLocalColorTableSize GifVersionGIF87aGIF89agif87aSignaturegif89aSignatureimageSeparatorextensionIntroducer gifTrailerplainTextLabelgraphicControlLabel commentLabelapplicationLabelparseDataBlocks putDataBlocksdisposalMethodOfCodecodeOfDisposalMethodskipSubDataBlocksparseGifBlocks getPalette putLoopingassociateDescrsubstituteColors substituteColorsWithTransparencydeinterlaceGifImagegifInterlacingIndices paletteOfgetFrameDelaystransparentColorOfhasTransparencydecodeAllGifImagesgifAnimationApplyerdecodeFirstGifImagecheckGifImageSizescheckPaletteValidityareIndexAbsentFromPalettecomputeMinimumLzwKeySize$fBinaryGifFile$fBinaryGifHeader$fBinaryImageDescriptor$fBinaryGifImage$fBinaryGraphicControlExtension$fBinaryLogicalScreenDescriptor$fBinaryGifVersionRadianceHeader radianceInfosradianceFormatradianceHeight radianceWidth radianceDataRadianceFormat FormatRGBE FormatXYZERGBE HDRReader.<<..>>.checkLineLengthisNewRunLengthMarkerradiance32bitRleRGBEFormatradiance32bitRleXYZEFromattoRGBE dropUntilgetUntilradianceFileSignature unpackColor storeColor parsePair decodeInfosgetChar8isSign isAxisLetter decodeNum copyPrevColor oldStyleRLE newStyleRLE decodeHeadertoFloatencodeScanlineColordecodeRadiancePicture$fBinaryRadianceHeader$fBinaryRadianceFormat $fBinaryRGBEtgaDataOfImagetgaPixelDepthOfImagetgaTypeOfImageTGAPixelUnpackedpackedByteSize tgaUnpackDepth32Depth24Depth15Depth8TgaFile_tgaFileHeader _tgaFileId _tgaPalette _tgaFileRest TgaHeader_tgaHdrIdLength_tgaHdrColorMapType_tgaHdrImageType_tgaHdrMapStart_tgaHdrMapLength_tgaHdrMapDepth_tgaHdrXOffset_tgaHdrYOffset _tgaHdrWidth _tgaHdrHeight_tgaHdrPixelDepth_tgaHdrImageDescriptionTgaImageDescription _tgaIdXOrigin _tgaIdYOrigin_tgaIdAttributeBits TgaImageTypeImageTypeNoDataImageTypeColorMappedImageTypeTrueColorImageTypeMonochromeTgaColorMapTypeColorMapWithoutTableColorMapWithTableColorMapUnknown isRleEncodedcodeOfImageType applyPaletteunparsewriteRuncopyDataunpackUncompressedTga isRleChunk runLength unpackRLETga flipImage validateTga$fTgaSaveablePixelRGBA8$fTgaSaveablePixelRGB8$fTgaSaveableWord8$fTGAPixelDepth32$fTGAPixelDepth24$fTGAPixelDepth15$fTGAPixelDepth8$fBinaryTgaFile$fBinaryTgaHeader$fBinaryTgaImageDescription$fBinaryTgaImageType$fBinaryTgaColorMapTypePngText pngKeywordpngData eitherFoldMapgetGammagetDpistextToMetadatagetTextsextractMetadatasencodePhysicalMetadataencodeSingleMetadataencodeMetadatas$fBinaryPngTextpreparePngHeaderendChunkprepareIDatChunkgenericEncode16BitsPngpreparePalettegenericEncodePng$fPngSavablePixelRGBA16$fPngSavablePixelRGB16$fPngSavableWord16$fPngSavablePixelYA16$fPngSavablePixelYA8$fPngSavableWord8$fPngSavablePixelRGB8$fPngSavablePixelRGBA8PngLineAdam7MatrixInfoadam7MatrixInfo pngFiltering byteUnpacker bitUnpackertwoBitsUnpackerscanlineUnpacker8 adam7Unpack deinterlacer BeginOffset StrideInfo LineUnpackeradam7StartingRowadam7StartingColadam7RowIncrementadam7ColIncrementadam7BlockHeightadam7BlockWidthunparsePngFilter getBoundshalfByteUnpacker shortUnpackerbyteSizeOfBitLengthscanLineInterleavinggenerateGreyscalePalettesampleCountOfImageType paletteRGB1 paletteRGB2 paletteRGB4boundsapplyPaletteWithTransparencytoImagepalette8TiffResolutionUnitResolutionUnitUnknownResolutionUnitInchResolutionUnitCentimeterextractTiffStringMetadatabyTag unitOfIfdextractTiffDpiMetadataextractTiffMetadataallocTempBufferunpack$fUnpackableWord8colorSpaceOfPixelextraSampleCodeOfPixelsubSamplingInfoYCbCrSubsampling ycbcrWidth ycbcrHeightycbcrImageWidthycbcrStripHeightPack12Pack2Pack4 Unpackable StorageTypeoutAlloc offsetStridemergeBackTempBufferTiffInfo tiffHeader tiffWidth tiffHeighttiffColorspacetiffSampleCounttiffRowPerStriptiffPlaneConfigurationtiffSampleFormattiffBitsPerSampletiffCompression tiffStripSize tiffOffsets tiffPalettetiffYCbCrSubsamplingtiffExtraSample tiffPredictor tiffMetadatasunLongfindIFD findPalette findIFDDatafindIFDDefaultData findIFDExtfindIFDExtDefaultDatacopyByteString unpackPackBit uncompressAt gatherStrips ifdSingleLongifdSingleShort ifdMultiLong ifdMultiShort$fTiffSaveablePixelYCbCr8$fTiffSaveablePixelRGBA16$fTiffSaveablePixelRGBA8$fTiffSaveablePixelRGB16$fTiffSaveablePixelRGB8$fTiffSaveablePixelCMYK16$fTiffSaveablePixelCMYK8$fTiffSaveablePixelYA16$fTiffSaveablePixelYA8$fTiffSaveableWord16$fTiffSaveableWord8$fBinaryParamByteStringTiffInfo$fUnpackableYCbCrSubsampling$fUnpackablePack12$fUnpackablePack2$fUnpackablePack4$fUnpackableWord32$fUnpackableWord16 scalerOfUnit EncoderStateacCoefficientsDecodedecompressMacroBlockjpgMachineStepadditionalBlockscomponentsOfColorSpace encodingStateimageHuffmanTablesscanSpecificationOfColorSpacequantTableSpecmaximumSubSamplingOf_encComponentIndex_encBlockWidth_encBlockHeight_encQuantTable _encDcHuffman _encAcHuffmanJpgDecoderStatedcDecoderTablesacDecoderTablesquantizationMatricescurrentRestartInterv currentFrame app14Markerapp0JFifMarkerapp1ExifMarkercomponentIndexMapping isProgressivemaximumHorizontalResolutionmaximumVerticalResolution seenBlobs JpgScripterquantizepowerOf encodeInt unpack444Yunpack444Ycbcrunpack421YcbcremptyDecoderStateunpackerDecisiongatherImageKindgatherScanInfodynamicOfColorSpacecolorSpaceOfAdobecolorSpaceOfStatecolorSpaceOfComponentStr extractBlockserializeMacroBlockencodeMacroBlock divUpwardprepareHuffmanTabledefaultHuffmanTableslumaQuantTableAtQualitychromaQuantTableAtQualityzigzaggedQuantificationSpec$fJpgEncodablePixelCMYK8$fJpgEncodablePixelRGB8$fJpgEncodablePixelYCbCr8$fJpgEncodableWord8componentToLDR toStandardDefgreyScaleToStandardDef from16to8 from16toFloat eitherLoad DecimabledecimateBitDepthwithImageDecoderdecimateWord16 decimateFloat$fDecimablePixelRGBFPixelRGB8$fDecimableFloatWord8 $fDecimablePixelCMYK16PixelCMYK8 $fDecimablePixelRGBA16PixelRGBA8$fDecimablePixelRGB16PixelRGB8$fDecimablePixelYA16PixelYA8$fDecimableWord16Word8