libtiff(3) introduction to


#include <tiffio.h>

cc file.c -ltiff


libtiff is a library for reading and writing data files encoded with the Tag Image File format, Revision 6.0 (or revision 5.0 or revision 4.0). This file format is suitable for archiving multi-color and monochromatic image data.

The library supports several compression algorithms, as indicated by the Compression field, including: no compression (1), CCITT 1D Huffman compression (2), CCITT Group 3 Facsimile compression (3), CCITT Group 4 Facsimile compression (4), Lempel-Ziv & Welch compression (5), baseline JPEG compression (7), word-aligned 1D Huffman compression (32771), and PackBits compression (32773). In addition, several nonstandard compression algorithms are supported: the 4-bit compression algorithm used by the ThunderScan program (32809) (decompression only), NeXT's 2-bit compression algorithm (32766) (decompression only), an experimental LZ-style algorithm known as Deflate (32946), and an experimental CIE LogLuv compression scheme designed for images with high dynamic range (32845 for LogL and 32845 for LogLuv). Directory information may be in either little- or big-endian byte order-byte swapping is automatically done by the library. Data bit ordering may be either Most Significant Bit (MSB) to Least Significant Bit (LSB) or LSB to MSB. Finally, the library does not support files in which the BitsPerSample, Compression, MinSampleValue, or MaxSampleValue fields are defined differently on a per-sample basis (in Rev. 6.0 the Compression tag is not defined on a per-sample basis, so this is immaterial).


The library makes extensive use of C typedefs to promote portability. Two sets of typedefs are used, one for communication with clients of the library and one for internal data structures and parsing of the TIFF format. The following typedefs are exposed to users either through function definitions or through parameters passed through the varargs interfaces.

typedef unsigned short uint16;        16-bit unsigned integer
typedef unsigned <thing> uint32;      32-bit unsigned integer

typedef unsigned int ttag_t;          directory tag
typedef uint16 tdir_t;                directory index
typedef uint16 tsample_t;             sample number
typedef uint32 tstrip_t;              strip number
typedef uint32 ttile_t;               tile number
typedef int32 tsize_t;                i/o size in bytes
typedef void* tdata_t;                image data ref
typedef void* thandle_t;              client data handle
typedef int32 toff_t;                 file offset

Note that tstrip_t, ttile_t, and tsize_t are constrained to be no more than 32-bit quantities by 32-bit fields they are stored in in the TIFF image. Likewise tsample_t is limited by the 16-bit field used to store the SamplesPerPixel tag. tdir_t constrains the maximum number of IFDs that may appear in an image and may be an arbitrary size (w/o penalty). ttag_t must be either int, unsigned int, pointer, or double because the library uses a varargs interface and ANSI C restricts the type of the parameter before an ellipsis to be a promoted type. toff_t is defined as int32 because TIFF file offsets are (unsigned) 32-bit quantities. A signed value is used because some interfaces return -1 on error. Finally, note that user-specified data references are passed as opaque handles and only cast at the lowest layers where their type is presumed.


The following routines are part of the library. Consult specific manual pages for details on their operation; on most systems doing ``man function-name'' will work.

Name                     Description

TIFFCheckpointDirectory  writes the current state of the directory
TIFFCheckTile            very x,y,z,sample is within image
TIFFCIELabToRGBInit      initialize CIE L*a*b* 1976 to RGB conversion state
TIFFCIELabToXYZ          perform CIE L*a*b* 1976 to CIE XYZ conversion
TIFFClientOpen           open a file for reading or writing
TIFFClose                close an open file
TIFFComputeStrip         return strip containing y,sample
TIFFComputeTile          return tile containing x,y,z,sample
TIFFCurrentDirectory     return index of current directory
TIFFCurrentRow           return index of current scanline
TIFFCurrentStrip         return index of current strip
TIFFCurrentTile          return index of current tile
TIFFDataWidth            return the size of TIFF data types
TIFFError                library error handler
TIFFFdOpen               open a file for reading or writing
TIFFFileName             return name of open file
TIFFFileno               return open file descriptor
TIFFFindCODEC            find standard codec for the specific scheme
TIFFFlush                flush all pending writes
TIFFFlushData            flush pending data writes
TIFFGetBitRevTable       return bit reversal table
TIFFGetField             return tag value in current directory
TIFFGetFieldDefaulted    return tag value in current directory
TIFFGetMode              return open file mode
TIFFGetVersion           return library version string
TIFFIsCODECConfigured    check, whether we have working codec
TIFFIsMSB2LSB            return true if image data is being returned
                         with bit 0 as the most significant bit 
TIFFIsTiled              return true if image data is tiled
TIFFIsByteSwapped        return true if image data is byte-swapped
TIFFNumberOfStrips       return number of strips in an image
TIFFNumberOfTiles        return number of tiles in an image
TIFFOpen                 open a file for reading or writing
TIFFPrintDirectory       print description of the current directory
TIFFReadBufferSetup      specify i/o buffer for reading
TIFFReadDirectory        read the next directory
TIFFReadEncodedStrip     read and decode a strip of data
TIFFReadEncodedTile      read and decode a tile of data
TIFFReadRawStrip         read a raw strip of data
TIFFReadRawTile          read a raw tile of data
TIFFReadRGBAImage        read an image into a fixed format raster
TIFFReadScanline         read and decode a row of data
TIFFReadTile             read and decode a tile of data
TIFFRegisterCODEC        override standard codec for the specific scheme
TIFFReverseBits          reverse bits in an array of bytes
TIFFRGBAImageBegin       setup decoder state for TIFFRGBAImageGet
TIFFRGBAImageEnd         release TIFFRGBAImage decoder state
TIFFRGBAImageGet         read and decode an image
TIFFRGBAImageOK          is image readable by TIFFRGBAImageGet
TIFFScanlineSize         return size of a scanline
TIFFSetDirectory         set the current directory
TIFFSetSubDirectory      set the current directory
TIFFSetErrorHandler      set error handler function
TIFFSetField             set a tag's value in the current directory
TIFFSetWarningHandler    set warning handler function
TIFFStripSize            returns size of a strip
TIFFRawStripSize         returns the number of bytes in a raw strip
TIFFSwabShort            swap bytes of short
TIFFSwabLong             swap bytes of long
TIFFSwabArrayOfShort     swap bytes of an array of shorts
TIFFSwabArrayOfLong      swap bytes of an array of longs
TIFFTileRowSize          return size of a row in a tile
TIFFTileSize             return size of a tile
TIFFUnRegisterCODEC      unregisters the codec
TIFFVGetField            return tag value in current directory
TIFFVGetFieldDefaulted   return tag value in current directory
TIFFVSetField            set a tag's value in the current directory
TIFFVStripSize           returns the number of bytes in a strip
TIFFWarning              library warning handler
TIFFWriteDirectory       write the current directory
TIFFWriteEncodedStrip    compress and write a strip of data
TIFFWriteEncodedTile     compress and write a tile of data
TIFFWriteRawStrip        write a raw strip of data
TIFFWriteRawTile         write a raw tile of data
TIFFWriteScanline        write a scanline of data
TIFFWriteTile            compress and write a tile of data
TIFFXYZToRGB             perform CIE XYZ to RGB conversion
TIFFYCbCrToRGBInit       initialize YCbCr to RGB conversion state
TIFFYCbCrtoRGB           perform YCbCr to RGB conversion

Auxiliary functions:
_TIFFfree                free memory buffer
_TIFFmalloc              dynamically allocate memory buffer
_TIFFmemcmp              compare contents of the memory buffers
_TIFFmemcpy              copy contents of the one buffer to another
_TIFFmemset              fill memory buffer with a constant byte
_TIFFrealloc             dynamically reallocate memory buffer


The table below lists the TIFF tags that are recognized and handled by the library. If no use is indicated in the table, then the library reads and writes the tag, but does not use it internally. Note that some tags are meaningful only when a particular compression scheme is being used; e.g. Group3Options is only useful if Compression is set to CCITT Group 3 encoding. Tags of this sort are considered codec-specific tags and the library does not recognize them except when the Compression tag has been previously set to the relevant compression scheme.

Tag Name                Value  R/W  Library Use/Notes

Artist                  315    R/W
BadFaxLines             326    R/W
BitsPerSample           258    R/W  lots
CellLength              265         parsed but ignored
CellWidth               264         parsed but ignored
CleanFaxData            327    R/W
ColorMap                320    R/W
ColorResponseUnit       300         parsed but ignored
Compression             259    R/W  choosing codec
ConsecutiveBadFaxLines  328    R/W
Copyright       33432   R/W
DataType                32996  R    obsoleted by SampleFormat tag
DateTime                306    R/W
DocumentName            269    R/W
DotRange                336    R/W
ExtraSamples            338    R/W  lots
FaxRecvParams           34908  R/W
FaxSubAddress           34909  R/W
FaxRecvTime             34910  R/W
FillOrder               266    R/W  control bit order
FreeByteCounts          289         parsed but ignored
FreeOffsets             288         parsed but ignored
GrayResponseCurve       291         parsed but ignored
GrayResponseUnit        290         parsed but ignored
Group3Options           292    R/W  used by Group 3 codec
Group4Options           293    R/W
HostComputer            316    R/W
ImageDepth              32997  R/W  tile/strip calculations
ImageDescription        270    R/W
ImageLength             257    R/W  lots
ImageWidth              256    R/W  lots
InkNames                333    R/W
InkSet                  332    R/W
JPEGTables              347    R/W  used by JPEG codec
Make                    271    R/W
Matteing                32995  R    obsoleted by ExtraSamples tag
MaxSampleValue          281    R/W
MinSampleValue          280    R/W
Model                   272    R/W
NewSubFileType          254    R/W  called SubFileType in spec
NumberOfInks            334    R/W
Orientation             274    R/W
PageName                285    R/W
PageNumber              297    R/W
PhotometricInterpretation      262  R/Wused by Group 3 and JPEG codecs
PlanarConfiguration     284    R/W  data i/o
Predictor               317    R/W  used by LZW and Deflate codecs
PrimaryChromacities     319    R/W
ReferenceBlackWhite     532    R/W
ResolutionUnit          296    R/W  used by Group 3 codec
RowsPerStrip            278    R/W  data i/o
SampleFormat            339    R/W
SamplesPerPixel         277    R/W  lots
SMinSampleValue         340    R/W
SMaxSampleValue         341    R/W
Software                305    R/W
StoNits                 37439  R/W
StripByteCounts         279    R/W  data i/o
StripOffsets            273    R/W  data i/o
SubFileType             255    R/W  called OSubFileType in spec
TargetPrinter           337    R/W
Thresholding            263    R/W  
TileByteCounts          324    R/W  data i/o
TileDepth               32998  R/W  tile/strip calculations
TileLength              323    R/W  data i/o
TileOffsets             324    R/W  data i/o
TileWidth               322    R/W  data i/o
TransferFunction        301    R/W
WhitePoint              318    R/W
XPosition               286    R/W
XResolution             282    R/W
YCbCrCoefficients       529    R/W  used by TIFFRGBAImage support
YCbCrPositioning        531    R/W  tile/strip size calulcations
YCbCrSubsampling        530    R/W
YPosition               286    R/W
YResolution             283    R/W  used by Group 3 codec


In addition to the normal TIFF tags the library supports a collection of tags whose values lie in a range outside the valid range of TIFF tags. These tags are termed pseud-tags and are used to control various codec-specific functions within the library. The table below summarizes the defined pseudo-tags.

Tag Name                Codec  R/W  Library Use/Notes

TIFFTAG_FAXMODE         G3     R/W  general codec operation
TIFFTAG_FAXFILLFUNC     G3/G4  R/W  bitmap fill function
TIFFTAG_JPEGQUALITY     JPEG   R/W  compression quality control
TIFFTAG_JPEGCOLORMODE   JPEG   R/W  control colorspace conversions
TIFFTAG_JPEGTABLESMODE  JPEG   R/W  control contents of JPEGTables tag
TIFFTAG_ZIPQUALITY      Deflate     R/Wcompression quality level
TIFFTAG_PIXARLOGDATAFMT PixarLog    R/Wuser data format
TIFFTAG_PIXARLOGQUALITY PixarLog    R/Wcompression quality level
TIFFTAG_SGILOGDATAFMT   SGILog R/W  user data format
Control the operation of the Group 3 codec. Possible values (independent bits that can be combined by or'ing them together) are: FAXMODE_CLASSIC (enable old-style format in which the RTC is written at the end of the last strip), FAXMODE_NORTC (opposite of FAXMODE_CLASSIC; also called FAXMODE_CLASSF), FAXMODE_NOEOL (do not write EOL codes at the start of each row of data), FAXMODE_BYTEALIGN (align each encoded row to an 8-bit boundary), FAXMODE_WORDALIGN (align each encoded row to an 16-bit boundary), The default value is dependent on the compression scheme; this pseudo-tag is used by the various G3 and G4 codecs to share code.
Control the function used to convert arrays of black and white runs to packed bit arrays. This hook can be used to image decoded scanlines in multi-bit depth rasters (e.g. for display in colormap mode) or for other purposes. The default value is a pointer to a builtin function that images packed bilevel data.
Tag contaings image metadata per the IPTC newsphoto spec: Headline, captioning, credit, etc... Used by most wire services.
Tag contains Photoshop captioning information and metadata. Photoshop uses in parallel and redundantly alongside IPTCNEWSPHOTO information.
Control the compression quality level used in the baseline algorithm. Note that quality levels are in the range 0-100 with a default value of 75.
Control whether or not conversion is done between RGB and YCbCr colorspaces. Possible values are: JPEGCOLORMODE_RAW (do not convert), and JPEGCOLORMODE_RGB (convert to/from RGB) The default value is JPEGCOLORMODE_RAW.
Control the information written in the JPEGTables tag. Possible values (independent bits that can be combined by or'ing them together) are: JPEGTABLESMODE_QUANT (include quantization tables), and JPEGTABLESMODE_HUFF (include Huffman encoding tables). The default value is JPEGTABLESMODE_QUANT|JPEGTABLESMODE_HUFF.
Control the compression technique used by the Deflate codec. Quality levels are in the range 1-9 with larger numbers yielding better compression at the cost of more computation. The default quality level is 6 which yields a good time-space tradeoff.
Control the format of user data passed in to the PixarLog codec when encoding and passed out from when decoding. Possible values are: PIXARLOGDATAFMT_8BIT for 8-bit unsigned pixels, PIXARLOGDATAFMT_8BITABGR for 8-bit unsigned ABGR-ordered pixels, PIXARLOGDATAFMT_11BITLOG for 11-bit log-encoded raw data, PIXARLOGDATAFMT_12BITPICIO for 12-bit PICIO-compatible data, PIXARLOGDATAFMT_16BIT for 16-bit signed samples, and PIXARLOGDATAFMT_FLOAT for 32-bit IEEE floating point samples.
Control the compression technique used by the PixarLog codec. This value is treated identically to TIFFTAG_ZIPQUALITY; see the above description.
Control the format of client data passed in to the SGILog codec when encoding and passed out from when decoding. Possible values are: SGILOGDATAFMT_FLTXYZ for converting between LogLuv and 32-bit IEEE floating valued XYZ pixels, SGILOGDATAFMT_16BITLUV for 16-bit encoded Luv pixels, SGILOGDATAFMT_32BITRAW and SGILOGDATAFMT_24BITRAW for no conversion of data, SGILOGDATAFMT_8BITRGB for returning 8-bit RGB data (valid only when decoding LogLuv-encoded data), SGILOGDATAFMT_FLTY for converting between LogL and 32-bit IEEE floating valued Y pixels, SGILOGDATAFMT_16BITL for 16-bit encoded L pixels, and SGILOGDATAFMT_8BITGRY for returning 8-bit greyscale data (valid only when decoding LogL-encoded data).


All error messages are directed through the TIFFError routine. By default messages are directed to stderr in the form: module: message\n. Warning messages are likewise directed through the TIFFWarning routine.


The library does not support multi-sample images where some samples have different bits/sample.

The library does not support random access to compressed data that is organized with more than one row per tile or strip.