Subversion Repositories Kolibri OS

libpng-manual.txt - A description on how to use and modify libpng

 

 libpng version 1.5.1 - February 3, 2011

 Updated and distributed by Glenn Randers-Pehrson

 <glennrp at users.sourceforge.net>

 Copyright (c) 1998-2011 Glenn Randers-Pehrson

 

 This document is released under the libpng license.

 For conditions of distribution and use, see the disclaimer

 and license in png.h

 

 Based on:

 

 libpng versions 0.97, January 1998, through 1.5.1 - February 3, 2011

 Updated and distributed by Glenn Randers-Pehrson

 Copyright (c) 1998-2011 Glenn Randers-Pehrson

 

 libpng 1.0 beta 6  version 0.96 May 28, 1997

 Updated and distributed by Andreas Dilger

 Copyright (c) 1996, 1997 Andreas Dilger

 

 libpng 1.0 beta 2 - version 0.88  January 26, 1996

 For conditions of distribution and use, see copyright

 notice in png.h. Copyright (c) 1995, 1996 Guy Eric

 Schalnat, Group 42, Inc.

 

 Updated/rewritten per request in the libpng FAQ

 Copyright (c) 1995, 1996 Frank J. T. Wojcik

 December 18, 1995 & January 20, 1996

 

I. Introduction

 

This file describes how to use and modify the PNG reference library

(known as libpng) for your own use.  There are five sections to this

file: introduction, structures, reading, writing, and modification and

configuration notes for various special platforms.  In addition to this

file, example.c is a good starting point for using the library, as

it is heavily commented and should include everything most people

will need.  We assume that libpng is already installed; see the

INSTALL file for instructions on how to install libpng.

 

For examples of libpng usage, see the files "example.c", "pngtest.c",

and the files in the "contrib" directory, all of which are included in

the libpng distribution.

 

Libpng was written as a companion to the PNG specification, as a way

of reducing the amount of time and effort it takes to support the PNG

file format in application programs.

 

The PNG specification (second edition), November 2003, is available as

a W3C Recommendation and as an ISO Standard (ISO/IEC 15948:2003 (E)) at

<http://www.w3.org/TR/2003/REC-PNG-20031110/

The W3C and ISO documents have identical technical content.

 

The PNG-1.2 specification is available at

<http://www.libpng.org/pub/png/documents/>.  It is technically equivalent

to the PNG specification (second edition) but has some additional material.

 

The PNG-1.0 specification is available

as RFC 2083 <http://www.libpng.org/pub/png/documents/> and as a

W3C Recommendation <http://www.w3.org/TR/REC.png.html>.

 

Some additional chunks are described in the special-purpose public chunks

documents at <http://www.libpng.org/pub/png/documents/>.

 

Other information

about PNG, and the latest version of libpng, can be found at the PNG home

page, <http://www.libpng.org/pub/png/>.

 

Most users will not have to modify the library significantly; advanced

users may want to modify it more.  All attempts were made to make it as

complete as possible, while keeping the code easy to understand.

Currently, this library only supports C.  Support for other languages

is being considered.

 

Libpng has been designed to handle multiple sessions at one time,

to be easily modifiable, to be portable to the vast majority of

machines (ANSI, K&R, 16-, 32-, and 64-bit) available, and to be easy

to use.  The ultimate goal of libpng is to promote the acceptance of

the PNG file format in whatever way possible.  While there is still

work to be done (see the TODO file), libpng should cover the

majority of the needs of its users.

 

Libpng uses zlib for its compression and decompression of PNG files.

Further information about zlib, and the latest version of zlib, can

be found at the zlib home page, <http://www.info-zip.org/pub/infozip/zlib/>.

The zlib compression utility is a general purpose utility that is

useful for more than PNG files, and can be used without libpng.

See the documentation delivered with zlib for more details.

You can usually find the source files for the zlib utility wherever you

find the libpng source files.

 

Libpng is thread safe, provided the threads are using different

instances of the structures.  Each thread should have its own

png_struct and png_info instances, and thus its own image.

Libpng does not protect itself against two threads using the

same instance of a structure.

 

II. Structures

 

There are two main structures that are important to libpng, png_struct

and png_info.  Both are internal structures that are no longer exposed

in the libpng interface (as of libpng 1.5.0).

 

The png_info structure is designed to provide information about the

PNG file.  At one time, the fields of png_info were intended to be

directly accessible to the user.  However, this tended to cause problems

with applications using dynamically loaded libraries, and as a result

a set of interface functions for png_info (the png_get_*() and png_set_*()

functions) was developed.

 

The png_struct structure is the object used by the library to decode a

single image.  As of 1.5.0 this structure is also not exposed.

 

Almost all libpng APIs require a pointer to a png_struct as the first argument.

Many (in particular the png_set and png_get APIs) also require a pointer

to png_info as the second argument.  Some application visible macros

defined in png.h designed for basic data access (reading and writing

integers in the PNG format) break this rule, but it's almost always safe

to assume that a (png_struct*) has to be passed to call an API function.

 

The png.h header file is an invaluable reference for programming with libpng.

And while I'm on the topic, make sure you include the libpng header file:

 

#include <png.h>

 

Types

 

The png.h header file defines a number of integral types used by the

APIs.  Most of these are fairly obvious; for example types corresponding

to integers of particular sizes and types for passing color values.

 

One exception is how non-integral numbers are handled.  For application

convenience most APIs that take such numbers have C (double) arguments,

however internally PNG, and libpng, use 32 bit signed integers and encode

the value by multiplying by 100,000.  As of libpng 1.5.0 a convenience

macro PNG_FP_1 is defined in png.h along with a type (png_fixed_point)

which is simply (png_int_32).

 

All APIs that take (double) arguments also have an matching API that

takes the corresponding fixed point integer arguments.  The fixed point

API has the same name as the floating point one with _fixed appended.

The actual range of values permitted in the APIs is frequently less than

the full range of (png_fixed_point) (-21474 to +21474).  When APIs require

a non-negative argument the type is recorded as png_uint_32 above.  Consult

the header file and the text below for more information.

 

Special care must be take with sCAL chunk handling because the chunk itself

uses non-integral values encoded as strings containing decimal floating point

numbers.  See the comments in the header file.

 

Configuration

 

The main header file function declarations are frequently protected by C

preprocessing directives of the form:

 

    #ifdef PNG_feature_SUPPORTED

    declare-function

    #endif

 

The library can be built without support for these APIs, although a

standard build will have all implemented APIs.  Application programs

should check the feature macros before using an API for maximum

portability.  From libpng 1.5.0 the feature macros set during the build

of libpng are recorded in the header file "pnglibconf.h" and this file

is always included by png.h.

 

If you don't need to change the library configuration from the default skip to

the next section ("Reading").

 

Notice that some of the makefiles in the 'scripts' directory and (in 1.5.0) all

of the build project files in the 'projects' directory simply copy

scripts/pnglibconf.h.prebuilt to pnglibconf.h.  This means that these build

systems do not permit easy auto-configuration of the library - they only

support the default configuration.

 

The easiest way to make minor changes to the libpng configuration when

auto-configuration is supported is to add definitions to the command line

using (typically) CPPFLAGS.  For example:

 

CPPFLAGS=-DPNG_NO_FLOATING_ARITHMETIC

 

will change the internal libpng math implementation for gamma correction and

other arithmetic calculations to fixed point, avoiding the need for fast

floating point support.  The result can be seen in the generated pnglibconf.h -

make sure it contains the changed feature macro setting.

 

If you need to make more extensive configuration changes - more than one or two

feature macro settings - you can either add -DPNG_USER_CONFIG to the build

command line and put a list of feature macro settings in pngusr.h or you can set

DFA_XTRA (a makefile variable) to a file containing the same information in the

form of 'option' settings.

 

A. Changing pnglibconf.h

 

A variety of methods exist to build libpng.  Not all of these support

reconfiguration of pnglibconf.h.  To reconfigure pnglibconf.h it must either be

rebuilt from scripts/pnglibconf.dfa using awk or it must be edited by hand.

 

Hand editing is achieved by copying scripts/pnglibconf.h.prebuilt and changing

the lines defining the supported features, paying very close attention to the

'option' information in scripts/pnglibconf.dfa that describes those features and

their requirements.  This is easy to get wrong.

 

B. Configuration using DFA_XTRA

 

Rebuilding from pnglibconf.dfa is easy if a functioning 'awk', or a later

variant such as 'nawk' or 'gawk', is available.  The configure build will

automatically find an appropriate awk and build pnglibconf.h.

scripts/pnglibconf.mak contains a set of make rules for doing the same thing if

configure is not used, and many of the makefiles in the scripts directory use

this approach.

 

When rebuilding simply write new file containing changed options and set

DFA_XTRA to the name of this file.  This causes the build to append the new file

to the end of scripts/pnglibconf.dfa.  pngusr.dfa should contain lines of the

following forms:

 

everything = off

 

This turns all optional features off.  Include it at the start of pngusr.dfa to

make it easier to build a minimal configuration.  You will need to turn at least

some features on afterward to enable either reading or writing code, or both.

 

option feature on

option feature off

 

Enable or disable a single feature.  This will automatically enable other

features required by a feature that is turned on or disable other features that

require a feature which is turned off.  Conflicting settings will cause an error

message to be emitted by awk.

 

setting feature default value

 

Changes the default value of setting 'feature' to 'value'.  There are a small

number of settings listed at the top of pnglibconf.h, they are documented in the

source code.  Most of these values have performance implications for the library

but most of them have no visible effect on the API.  Some can also be overridden

from the API.

 

C. Configuration using PNG_USR_CONFIG

 

If -DPNG_USR_CONFIG is added to the CFLAGS when pnglibconf.h is built the file

pngusr.h will automatically be included before the options in

scripts/pnglibconf.dfa are processed.  pngusr.h should contain only macro

definitions turning features on or off or setting settings.

 

Apart from the global setting "everything = off" all the options listed above

can be set using macros in pngusr.h:

 

#define PNG_feature_SUPPORTED

 

is equivalent to:

 

option feature on

 

#define PNG_NO_feature

 

is equivalent to:

 

option feature off

 

#define PNG_feature value

 

is equivalent to:

 

setting feature default value

 

Notice that in both cases, pngusr.dfa and pngusr.h, the contents of the

pngusr file you supply override the contents of scripts/pnglibconf.dfa

 

If confusing or incomprehensible behavior results it is possible to

examine the intermediate file pnglibconf.dfn to find the full set of

dependency information for each setting and option.  Simply locate the

feature in the file and read the C comments that precede it.

 

III. Reading

 

We'll now walk you through the possible functions to call when reading

in a PNG file sequentially, briefly explaining the syntax and purpose

of each one.  See example.c and png.h for more detail.  While

progressive reading is covered in the next section, you will still

need some of the functions discussed in this section to read a PNG

file.

 

Setup

 

You will want to do the I/O initialization(*) before you get into libpng,

so if it doesn't work, you don't have much to undo.  Of course, you

will also want to insure that you are, in fact, dealing with a PNG

file.  Libpng provides a simple check to see if a file is a PNG file.

To use it, pass in the first 1 to 8 bytes of the file to the function

png_sig_cmp(), and it will return 0 (false) if the bytes match the

corresponding bytes of the PNG signature, or nonzero (true) otherwise.

Of course, the more bytes you pass in, the greater the accuracy of the

prediction.

 

If you are intending to keep the file pointer open for use in libpng,

you must ensure you don't read more than 8 bytes from the beginning

of the file, and you also have to make a call to png_set_sig_bytes_read()

with the number of bytes you read from the beginning.  Libpng will

then only check the bytes (if any) that your program didn't read.

 

(*): If you are not using the standard I/O functions, you will need

to replace them with custom functions.  See the discussion under

Customizing libpng.

 

 

    FILE *fp = fopen(file_name, "rb");

    if (!fp)

    {

       return (ERROR);

    }

 

    fread(header, 1, number, fp);

    is_png = !png_sig_cmp(header, 0, number);

 

    if (!is_png)

    {

       return (NOT_PNG);

    }

 

 

Next, png_struct and png_info need to be allocated and initialized.  In

order to ensure that the size of these structures is correct even with a

dynamically linked libpng, there are functions to initialize and

allocate the structures.  We also pass the library version, optional

pointers to error handling functions, and a pointer to a data struct for

use by the error functions, if necessary (the pointer and functions can

be NULL if the default error handlers are to be used).  See the section

on Changes to Libpng below regarding the old initialization functions.

The structure allocation functions quietly return NULL if they fail to

create the structure, so your application should check for that.

 

    png_structp png_ptr = png_create_read_struct

        (PNG_LIBPNG_VER_STRING, (png_voidp)user_error_ptr,

        user_error_fn, user_warning_fn);

 

    if (!png_ptr)

       return (ERROR);

 

    png_infop info_ptr = png_create_info_struct(png_ptr);

 

    if (!info_ptr)

    {

       png_destroy_read_struct(&png_ptr,

           (png_infopp)NULL, (png_infopp)NULL);

       return (ERROR);

    }

 

    png_infop end_info = png_create_info_struct(png_ptr);

 

    if (!end_info)

    {

       png_destroy_read_struct(&png_ptr, &info_ptr,

          (png_infopp)NULL);

       return (ERROR);

    }

 

If you want to use your own memory allocation routines,

use a libpng that was built with PNG_USER_MEM_SUPPORTED defined, and use

png_create_read_struct_2() instead of png_create_read_struct():

 

    png_structp png_ptr = png_create_read_struct_2

       (PNG_LIBPNG_VER_STRING, (png_voidp)user_error_ptr,

        user_error_fn, user_warning_fn, (png_voidp)

        user_mem_ptr, user_malloc_fn, user_free_fn);

 

The error handling routines passed to png_create_read_struct()

and the memory alloc/free routines passed to png_create_struct_2()

are only necessary if you are not using the libpng supplied error

handling and memory alloc/free functions.

 

When libpng encounters an error, it expects to longjmp back

to your routine.  Therefore, you will need to call setjmp and pass

your png_jmpbuf(png_ptr).  If you read the file from different

routines, you will need to update the jmpbuf field every time you enter

a new routine that will call a png_*() function.

 

See your documentation of setjmp/longjmp for your compiler for more

information on setjmp/longjmp.  See the discussion on libpng error

handling in the Customizing Libpng section below for more information

on the libpng error handling.  If an error occurs, and libpng longjmp's

back to your setjmp, you will want to call png_destroy_read_struct() to

free any memory.

 

    if (setjmp(png_jmpbuf(png_ptr)))

    {

       png_destroy_read_struct(&png_ptr, &info_ptr,

           &end_info);

       fclose(fp);

       return (ERROR);

    }

 

If you would rather avoid the complexity of setjmp/longjmp issues,

you can compile libpng with PNG_NO_SETJMP, in which case

errors will result in a call to PNG_ABORT() which defaults to abort().

 

You can #define PNG_ABORT() to a function that does something

more useful than abort(), as long as your function does not

return.

 

Now you need to set up the input code.  The default for libpng is to

use the C function fread().  If you use this, you will need to pass a

valid FILE * in the function png_init_io().  Be sure that the file is

opened in binary mode.  If you wish to handle reading data in another

way, you need not call the png_init_io() function, but you must then

implement the libpng I/O methods discussed in the Customizing Libpng

section below.

 

    png_init_io(png_ptr, fp);

 

If you had previously opened the file and read any of the signature from

the beginning in order to see if this was a PNG file, you need to let

libpng know that there are some bytes missing from the start of the file.

 

    png_set_sig_bytes(png_ptr, number);

 

You can change the zlib compression buffer size to be used while

reading compressed data with

 

    png_set_compression_buffer_size(png_ptr, buffer_size);

 

where the default size is 8192 bytes.  Note that the buffer size

is changed immediately and the buffer is reallocated immediately,

instead of setting a flag to be acted upon later.

 

If you want CRC errors to be handled in a different manner than

the default, use

 

    png_set_crc_action(png_ptr, crit_action, ancil_action);

 

The values for png_set_crc_action() say how libpng is to handle CRC errors in

ancillary and critical chunks, and whether to use the data contained

therein.  Note that it is impossible to "discard" data in a critical

chunk.

 

Choices for (int) crit_action are

   PNG_CRC_DEFAULT      0  error/quit

   PNG_CRC_ERROR_QUIT   1  error/quit

   PNG_CRC_WARN_USE     3  warn/use data

   PNG_CRC_QUIET_USE    4  quiet/use data

   PNG_CRC_NO_CHANGE    5  use the current value

 

Choices for (int) ancil_action are

   PNG_CRC_DEFAULT      0  error/quit

   PNG_CRC_ERROR_QUIT   1  error/quit

   PNG_CRC_WARN_DISCARD 2  warn/discard data

   PNG_CRC_WARN_USE     3  warn/use data

   PNG_CRC_QUIET_USE    4  quiet/use data

   PNG_CRC_NO_CHANGE    5  use the current value

 

Setting up callback code

 

You can set up a callback function to handle any unknown chunks in the

input stream. You must supply the function

 

    read_chunk_callback(png_structp png_ptr,

         png_unknown_chunkp chunk);

    {

       /* The unknown chunk structure contains your

          chunk data, along with similar data for any other

          unknown chunks: */

 

           png_byte name[5];

           png_byte *data;

           png_size_t size;

 

       /* Note that libpng has already taken care of

          the CRC handling */

 

       /* put your code here.  Search for your chunk in the

          unknown chunk structure, process it, and return one

          of the following: */

 

       return (-n); /* chunk had an error */

       return (0); /* did not recognize */

       return (n); /* success */

    }

 

(You can give your function another name that you like instead of

"read_chunk_callback")

 

To inform libpng about your function, use

 

    png_set_read_user_chunk_fn(png_ptr, user_chunk_ptr,

        read_chunk_callback);

 

This names not only the callback function, but also a user pointer that

you can retrieve with

 

    png_get_user_chunk_ptr(png_ptr);

 

If you call the png_set_read_user_chunk_fn() function, then all unknown

chunks will be saved when read, in case your callback function will need

one or more of them.  This behavior can be changed with the

png_set_keep_unknown_chunks() function, described below.

 

At this point, you can set up a callback function that will be

called after each row has been read, which you can use to control

a progress meter or the like.  It's demonstrated in pngtest.c.

You must supply a function

 

    void read_row_callback(png_structp png_ptr,

       png_uint_32 row, int pass);

    {

      /* put your code here */

    }

 

(You can give it another name that you like instead of "read_row_callback")

 

To inform libpng about your function, use

 

    png_set_read_status_fn(png_ptr, read_row_callback);

 

Unknown-chunk handling

 

Now you get to set the way the library processes unknown chunks in the

input PNG stream. Both known and unknown chunks will be read.  Normal

behavior is that known chunks will be parsed into information in

various info_ptr members while unknown chunks will be discarded. This

behavior can be wasteful if your application will never use some known

chunk types. To change this, you can call:

 

    png_set_keep_unknown_chunks(png_ptr, keep,

        chunk_list, num_chunks);

    keep       - 0: default unknown chunk handling

                 1: ignore; do not keep

                 2: keep only if safe-to-copy

                 3: keep even if unsafe-to-copy

 

               You can use these definitions:

                 PNG_HANDLE_CHUNK_AS_DEFAULT   0

                 PNG_HANDLE_CHUNK_NEVER        1

                 PNG_HANDLE_CHUNK_IF_SAFE      2

                 PNG_HANDLE_CHUNK_ALWAYS       3

 

    chunk_list - list of chunks affected (a byte string,

                 five bytes per chunk, NULL or '\0' if

                 num_chunks is 0)

 

    num_chunks - number of chunks affected; if 0, all

                 unknown chunks are affected.  If nonzero,

                 only the chunks in the list are affected

 

Unknown chunks declared in this way will be saved as raw data onto a

list of png_unknown_chunk structures.  If a chunk that is normally

known to libpng is named in the list, it will be handled as unknown,

according to the "keep" directive.  If a chunk is named in successive

instances of png_set_keep_unknown_chunks(), the final instance will

take precedence.  The IHDR and IEND chunks should not be named in

chunk_list; if they are, libpng will process them normally anyway.

 

Here is an example of the usage of png_set_keep_unknown_chunks(),

where the private "vpAg" chunk will later be processed by a user chunk

callback function:

 

    png_byte vpAg[5]={118, 112,  65, 103, (png_byte) '\0'};

 

    #if defined(PNG_UNKNOWN_CHUNKS_SUPPORTED)

      png_byte unused_chunks[]=

      {

        104,  73,  83,  84, (png_byte) '\0',   /* hIST */

        105,  84,  88, 116, (png_byte) '\0',   /* iTXt */

        112,  67,  65,  76, (png_byte) '\0',   /* pCAL */

        115,  67,  65,  76, (png_byte) '\0',   /* sCAL */

        115,  80,  76,  84, (png_byte) '\0',   /* sPLT */

        116,  73,  77,  69, (png_byte) '\0',   /* tIME */

      };

    #endif

 

    ...

 

    #if defined(PNG_UNKNOWN_CHUNKS_SUPPORTED)

      /* ignore all unknown chunks: */

      png_set_keep_unknown_chunks(read_ptr, 1, NULL, 0);

 

      /* except for vpAg: */

      png_set_keep_unknown_chunks(read_ptr, 2, vpAg, 1);

 

      /* also ignore unused known chunks: */

      png_set_keep_unknown_chunks(read_ptr, 1, unused_chunks,

         (int)sizeof(unused_chunks)/5);

    #endif

 

User limits

 

The PNG specification allows the width and height of an image to be as

large as 2^31-1 (0x7fffffff), or about 2.147 billion rows and columns.

Since very few applications really need to process such large images,

we have imposed an arbitrary 1-million limit on rows and columns.

Larger images will be rejected immediately with a png_error() call. If

you wish to override this limit, you can use

 

   png_set_user_limits(png_ptr, width_max, height_max);

 

to set your own limits, or use width_max = height_max = 0x7fffffffL

to allow all valid dimensions (libpng may reject some very large images

anyway because of potential buffer overflow conditions).

 

You should put this statement after you create the PNG structure and

before calling png_read_info(), png_read_png(), or png_process_data().

If you need to retrieve the limits that are being applied, use

 

   width_max = png_get_user_width_max(png_ptr);

   height_max = png_get_user_height_max(png_ptr);

 

The PNG specification sets no limit on the number of ancillary chunks

allowed in a PNG datastream.  You can impose a limit on the total number

of sPLT, tEXt, iTXt, zTXt, and unknown chunks that will be stored, with

 

   png_set_chunk_cache_max(png_ptr, user_chunk_cache_max);

 

where 0x7fffffffL means unlimited.  You can retrieve this limit with

 

   chunk_cache_max = png_get_chunk_cache_max(png_ptr);

 

This limit also applies to the number of buffers that can be allocated

by png_decompress_chunk() while decompressing iTXt, zTXt, and iCCP chunks.

 

You can also set a limit on the amount of memory that a compressed chunk

other than IDAT can occupy, with

 

   png_set_chunk_malloc_max(png_ptr, user_chunk_malloc_max);

 

and you can retrieve the limit with

 

   chunk_malloc_max = png_get_chunk_malloc_max(png_ptr);

 

Any chunks that would cause either of these limits to be exceeded will

be ignored.

 

The high-level read interface

 

At this point there are two ways to proceed; through the high-level

read interface, or through a sequence of low-level read operations.

You can use the high-level interface if (a) you are willing to read

the entire image into memory, and (b) the input transformations

you want to do are limited to the following set:

 

    PNG_TRANSFORM_IDENTITY      No transformation

    PNG_TRANSFORM_STRIP_16      Strip 16-bit samples to

                                8 bits

    PNG_TRANSFORM_STRIP_ALPHA   Discard the alpha channel

    PNG_TRANSFORM_PACKING       Expand 1, 2 and 4-bit

                                samples to bytes

    PNG_TRANSFORM_PACKSWAP      Change order of packed

                                pixels to LSB first

    PNG_TRANSFORM_EXPAND        Perform set_expand()

    PNG_TRANSFORM_INVERT_MONO   Invert monochrome images

    PNG_TRANSFORM_SHIFT         Normalize pixels to the

                                sBIT depth

    PNG_TRANSFORM_BGR           Flip RGB to BGR, RGBA

                                to BGRA

    PNG_TRANSFORM_SWAP_ALPHA    Flip RGBA to ARGB or GA

                                to AG

    PNG_TRANSFORM_INVERT_ALPHA  Change alpha from opacity

                                to transparency

    PNG_TRANSFORM_SWAP_ENDIAN   Byte-swap 16-bit samples

    PNG_TRANSFORM_GRAY_TO_RGB   Expand grayscale samples

                                to RGB (or GA to RGBA)

 

(This excludes setting a background color, doing gamma transformation,

quantizing, and setting filler.)  If this is the case, simply do this:

 

    png_read_png(png_ptr, info_ptr, png_transforms, NULL)

 

where png_transforms is an integer containing the bitwise OR of some

set of transformation flags.  This call is equivalent to png_read_info(),

followed the set of transformations indicated by the transform mask,

then png_read_image(), and finally png_read_end().

 

(The final parameter of this call is not yet used.  Someday it might point

to transformation parameters required by some future input transform.)

 

You must use png_transforms and not call any png_set_transform() functions

when you use png_read_png().

 

After you have called png_read_png(), you can retrieve the image data

with

 

   row_pointers = png_get_rows(png_ptr, info_ptr);

 

where row_pointers is an array of pointers to the pixel data for each row:

 

   png_bytep row_pointers[height];

 

If you know your image size and pixel size ahead of time, you can allocate

row_pointers prior to calling png_read_png() with

 

   if (height > PNG_UINT_32_MAX/png_sizeof(png_byte))

      png_error (png_ptr,

          "Image is too tall to process in memory");

 

   if (width > PNG_UINT_32_MAX/pixel_size)

      png_error (png_ptr,

          "Image is too wide to process in memory");

 

   row_pointers = png_malloc(png_ptr,

       height*png_sizeof(png_bytep));

 

   for (int i=0; i<height, i++)

      row_pointers[i]=NULL;  /* security precaution */

 

   for (int i=0; i<height, i++)

      row_pointers[i]=png_malloc(png_ptr,

          width*pixel_size);

 

   png_set_rows(png_ptr, info_ptr, &row_pointers);

 

Alternatively you could allocate your image in one big block and define

row_pointers[i] to point into the proper places in your block.

 

If you use png_set_rows(), the application is responsible for freeing

row_pointers (and row_pointers[i], if they were separately allocated).

 

If you don't allocate row_pointers ahead of time, png_read_png() will

do it, and it'll be free'ed by libpng when you call png_destroy_*().

 

The low-level read interface

 

If you are going the low-level route, you are now ready to read all

the file information up to the actual image data.  You do this with a

call to png_read_info().

 

    png_read_info(png_ptr, info_ptr);

 

This will process all chunks up to but not including the image data.

 

Querying the info structure

 

Functions are used to get the information from the info_ptr once it

has been read.  Note that these fields may not be completely filled

in until png_read_end() has read the chunk data following the image.

 

    png_get_IHDR(png_ptr, info_ptr, &width, &height,

       &bit_depth, &color_type, &interlace_type,

       &compression_type, &filter_method);

 

    width          - holds the width of the image

                     in pixels (up to 2^31).

 

    height         - holds the height of the image

                     in pixels (up to 2^31).

 

    bit_depth      - holds the bit depth of one of the

                     image channels.  (valid values are

                     1, 2, 4, 8, 16 and depend also on

                     the color_type.  See also

                     significant bits (sBIT) below).

 

    color_type     - describes which color/alpha channels

                         are present.

                     PNG_COLOR_TYPE_GRAY

                        (bit depths 1, 2, 4, 8, 16)

                     PNG_COLOR_TYPE_GRAY_ALPHA

                        (bit depths 8, 16)

                     PNG_COLOR_TYPE_PALETTE

                        (bit depths 1, 2, 4, 8)

                     PNG_COLOR_TYPE_RGB

                        (bit_depths 8, 16)

                     PNG_COLOR_TYPE_RGB_ALPHA

                        (bit_depths 8, 16)

 

                     PNG_COLOR_MASK_PALETTE

                     PNG_COLOR_MASK_COLOR

                     PNG_COLOR_MASK_ALPHA

 

    interlace_type - (PNG_INTERLACE_NONE or

                     PNG_INTERLACE_ADAM7)

 

    compression_type - (must be PNG_COMPRESSION_TYPE_BASE

                     for PNG 1.0)

 

    filter_method  - (must be PNG_FILTER_TYPE_BASE

                     for PNG 1.0, and can also be

                     PNG_INTRAPIXEL_DIFFERENCING if

                     the PNG datastream is embedded in

                     a MNG-1.0 datastream)

 

    Any or all of interlace_type, compression_type, or

    filter_method can be NULL if you are

    not interested in their values.

 

    Note that png_get_IHDR() returns 32-bit data into

    the application's width and height variables.

    This is an unsafe situation if these are 16-bit

    variables.  In such situations, the

    png_get_image_width() and png_get_image_height()

    functions described below are safer.

 

    width            = png_get_image_width(png_ptr,

                         info_ptr);

 

    height           = png_get_image_height(png_ptr,

                         info_ptr);

 

    bit_depth        = png_get_bit_depth(png_ptr,

                         info_ptr);

 

    color_type       = png_get_color_type(png_ptr,

                         info_ptr);

 

    interlace_type   = png_get_interlace_type(png_ptr,

                         info_ptr);

 

    compression_type = png_get_compression_type(png_ptr,

                         info_ptr);

 

    filter_method    = png_get_filter_type(png_ptr,

                         info_ptr);

 

    channels = png_get_channels(png_ptr, info_ptr);

 

    channels       - number of channels of info for the

                     color type (valid values are 1 (GRAY,

                     PALETTE), 2 (GRAY_ALPHA), 3 (RGB),

                     4 (RGB_ALPHA or RGB + filler byte))

 

    rowbytes = png_get_rowbytes(png_ptr, info_ptr);

 

    rowbytes       - number of bytes needed to hold a row

 

    signature = png_get_signature(png_ptr, info_ptr);

 

    signature      - holds the signature read from the

                     file (if any).  The data is kept in

                     the same offset it would be if the

                     whole signature were read (i.e. if an

                     application had already read in 4

                     bytes of signature before starting

                     libpng, the remaining 4 bytes would

                     be in signature[4] through signature[7]

                     (see png_set_sig_bytes())).

 

These are also important, but their validity depends on whether the chunk

has been read.  The png_get_valid(png_ptr, info_ptr, PNG_INFO_<chunk>) and

png_get_<chunk>(png_ptr, info_ptr, ...) functions return non-zero if the

data has been read, or zero if it is missing.  The parameters to the

png_get_<chunk> are set directly if they are simple data types, or a

pointer into the info_ptr is returned for any complex types.

 

    png_get_PLTE(png_ptr, info_ptr, &palette,

                     &num_palette);

 

    palette        - the palette for the file

                     (array of png_color)

 

    num_palette    - number of entries in the palette

 

    png_get_gAMA(png_ptr, info_ptr, &file_gamma);

    png_get_gAMA_fixed(png_ptr, info_ptr, &int_file_gamma);

 

    file_gamma     - the gamma at which the file is

                     written (PNG_INFO_gAMA)

 

    int_file_gamma - 100,000 times the gamma at which the

                     file is written

 

    png_get_sRGB(png_ptr, info_ptr, &srgb_intent);

 

    file_srgb_intent - the rendering intent (PNG_INFO_sRGB)

                     The presence of the sRGB chunk

                     means that the pixel data is in the

                     sRGB color space.  This chunk also

                     implies specific values of gAMA and

                     cHRM.

 

    png_get_iCCP(png_ptr, info_ptr, &name,

       &compression_type, &profile, &proflen);

 

    name             - The profile name.

 

    compression_type - The compression type; always

                       PNG_COMPRESSION_TYPE_BASE for PNG 1.0.

                       You may give NULL to this argument to

                       ignore it.

 

    profile          - International Color Consortium color

                       profile data. May contain NULs.

 

    proflen          - length of profile data in bytes.

 

    png_get_sBIT(png_ptr, info_ptr, &sig_bit);

 

    sig_bit        - the number of significant bits for

                     (PNG_INFO_sBIT) each of the gray,

                     red, green, and blue channels,

                     whichever are appropriate for the

                     given color type (png_color_16)

 

    png_get_tRNS(png_ptr, info_ptr, &trans_alpha,

                     &num_trans, &trans_color);

 

    trans_alpha    - array of alpha (transparency)

                     entries for palette (PNG_INFO_tRNS)

 

    num_trans      - number of transparent entries

                     (PNG_INFO_tRNS)

 

    trans_color    - graylevel or color sample values of

                     the single transparent color for

                     non-paletted images (PNG_INFO_tRNS)

 

    png_get_hIST(png_ptr, info_ptr, &hist);

                     (PNG_INFO_hIST)

 

    hist           - histogram of palette (array of

                     png_uint_16)

 

    png_get_tIME(png_ptr, info_ptr, &mod_time);

 

    mod_time       - time image was last modified

                    (PNG_VALID_tIME)

 

    png_get_bKGD(png_ptr, info_ptr, &background);

 

    background     - background color (PNG_VALID_bKGD)

                     valid 16-bit red, green and blue

                     values, regardless of color_type

 

    num_comments   = png_get_text(png_ptr, info_ptr,

                     &text_ptr, &num_text);

 

    num_comments   - number of comments

 

    text_ptr       - array of png_text holding image

                     comments

 

    text_ptr[i].compression - type of compression used

                 on "text" PNG_TEXT_COMPRESSION_NONE

                           PNG_TEXT_COMPRESSION_zTXt

                           PNG_ITXT_COMPRESSION_NONE

                           PNG_ITXT_COMPRESSION_zTXt

 

    text_ptr[i].key   - keyword for comment.  Must contain

                         1-79 characters.

 

    text_ptr[i].text  - text comments for current

                         keyword.  Can be empty.

 

    text_ptr[i].text_length - length of text string,

                 after decompression, 0 for iTXt

 

    text_ptr[i].itxt_length - length of itxt string,

                 after decompression, 0 for tEXt/zTXt

 

    text_ptr[i].lang  - language of comment (empty

                         string for unknown).

 

    text_ptr[i].lang_key  - keyword in UTF-8

                         (empty string for unknown).

 

    Note that the itxt_length, lang, and lang_key

    members of the text_ptr structure only exist

    when the library is built with iTXt chunk support.

 

    num_text       - number of comments (same as

                     num_comments; you can put NULL here

                     to avoid the duplication)

 

    Note while png_set_text() will accept text, language,

    and translated keywords that can be NULL pointers, the

    structure returned by png_get_text will always contain

    regular zero-terminated C strings.  They might be

    empty strings but they will never be NULL pointers.

 

    num_spalettes = png_get_sPLT(png_ptr, info_ptr,

       &palette_ptr);

 

    num_spalettes  - number of sPLT chunks read.

 

    palette_ptr    - array of palette structures holding

                     contents of one or more sPLT chunks

                     read.

 

    png_get_oFFs(png_ptr, info_ptr, &offset_x, &offset_y,

       &unit_type);

 

    offset_x       - positive offset from the left edge

                     of the screen

 

    offset_y       - positive offset from the top edge

                     of the screen

 

    unit_type      - PNG_OFFSET_PIXEL, PNG_OFFSET_MICROMETER

 

    png_get_pHYs(png_ptr, info_ptr, &res_x, &res_y,

       &unit_type);

 

    res_x          - pixels/unit physical resolution in

                     x direction

 

    res_y          - pixels/unit physical resolution in

                     x direction

 

    unit_type      - PNG_RESOLUTION_UNKNOWN,

                     PNG_RESOLUTION_METER

 

    png_get_sCAL(png_ptr, info_ptr, &unit, &width,

       &height)

 

    unit        - physical scale units (an integer)

 

    width       - width of a pixel in physical scale units

 

    height      - height of a pixel in physical scale units

                 (width and height are doubles)

 

    png_get_sCAL_s(png_ptr, info_ptr, &unit, &width,

       &height)

 

    unit        - physical scale units (an integer)

 

    width       - width of a pixel in physical scale units

 

    height      - height of a pixel in physical scale units

                 (width and height are strings like "2.54")

 

    num_unknown_chunks = png_get_unknown_chunks(png_ptr,

       info_ptr, &unknowns)

 

    unknowns          - array of png_unknown_chunk

                        structures holding unknown chunks

 

    unknowns[i].name  - name of unknown chunk

 

    unknowns[i].data  - data of unknown chunk

 

    unknowns[i].size  - size of unknown chunk's data

 

    unknowns[i].location - position of chunk in file

 

    The value of "i" corresponds to the order in which the

    chunks were read from the PNG file or inserted with the

    png_set_unknown_chunks() function.

 

The data from the pHYs chunk can be retrieved in several convenient

forms:

 

    res_x = png_get_x_pixels_per_meter(png_ptr,

       info_ptr)

 

    res_y = png_get_y_pixels_per_meter(png_ptr,

       info_ptr)

 

    res_x_and_y = png_get_pixels_per_meter(png_ptr,

       info_ptr)

 

    res_x = png_get_x_pixels_per_inch(png_ptr,

       info_ptr)

 

    res_y = png_get_y_pixels_per_inch(png_ptr,

       info_ptr)

 

    res_x_and_y = png_get_pixels_per_inch(png_ptr,

       info_ptr)

 

    aspect_ratio = png_get_pixel_aspect_ratio(png_ptr,

       info_ptr)

 

    Each of these returns 0 [signifying "unknown"] if

       the data is not present or if res_x is 0;

       res_x_and_y is 0 if res_x != res_y

 

    Note that because of the way the resolutions are

       stored internally, the inch conversions won't

       come out to exactly even number.  For example,

       72 dpi is stored as 0.28346 pixels/meter, and

       when this is retrieved it is 71.9988 dpi, so

       be sure to round the returned value appropriately

       if you want to display a reasonable-looking result.

 

The data from the oFFs chunk can be retrieved in several convenient

forms:

 

    x_offset = png_get_x_offset_microns(png_ptr, info_ptr);

 

    y_offset = png_get_y_offset_microns(png_ptr, info_ptr);

 

    x_offset = png_get_x_offset_inches(png_ptr, info_ptr);

 

    y_offset = png_get_y_offset_inches(png_ptr, info_ptr);

 

    Each of these returns 0 [signifying "unknown" if both

       x and y are 0] if the data is not present or if the

       chunk is present but the unit is the pixel.  The

       remark about inexact inch conversions applies here

       as well, because a value in inches can't always be

       converted to microns and back without some loss

       of precision.

 

For more information, see the png_info definition in png.h and the

PNG specification for chunk contents.  Be careful with trusting

rowbytes, as some of the transformations could increase the space

needed to hold a row (expand, filler, gray_to_rgb, etc.).

See png_read_update_info(), below.

 

A quick word about text_ptr and num_text.  PNG stores comments in

keyword/text pairs, one pair per chunk, with no limit on the number

of text chunks, and a 2^31 byte limit on their size.  While there are

suggested keywords, there is no requirement to restrict the use to these

strings.  It is strongly suggested that keywords and text be sensible

to humans (that's the point), so don't use abbreviations.  Non-printing

symbols are not allowed.  See the PNG specification for more details.

There is also no requirement to have text after the keyword.

 

Keywords should be limited to 79 Latin-1 characters without leading or

trailing spaces, but non-consecutive spaces are allowed within the

keyword.  It is possible to have the same keyword any number of times.

The text_ptr is an array of png_text structures, each holding a

pointer to a language string, a pointer to a keyword and a pointer to

a text string.  The text string, language code, and translated

keyword may be empty or NULL pointers.  The keyword/text

pairs are put into the array in the order that they are received.

However, some or all of the text chunks may be after the image, so, to

make sure you have read all the text chunks, don't mess with these

until after you read the stuff after the image.  This will be

mentioned again below in the discussion that goes with png_read_end().

 

Input transformations

 

After you've read the header information, you can set up the library

to handle any special transformations of the image data.  The various

ways to transform the data will be described in the order that they

should occur.  This is important, as some of these change the color

type and/or bit depth of the data, and some others only work on

certain color types and bit depths.  Even though each transformation

checks to see if it has data that it can do something with, you should

make sure to only enable a transformation if it will be valid for the

data.  For example, don't swap red and blue on grayscale data.

 

The colors used for the background and transparency values should be

supplied in the same format/depth as the current image data.  They

are stored in the same format/depth as the image data in a bKGD or tRNS

chunk, so this is what libpng expects for this data.  The colors are

transformed to keep in sync with the image data when an application

calls the png_read_update_info() routine (see below).

 

Data will be decoded into the supplied row buffers packed into bytes

unless the library has been told to transform it into another format.

For example, 4 bit/pixel paletted or grayscale data will be returned

2 pixels/byte with the leftmost pixel in the high-order bits of the

byte, unless png_set_packing() is called.  8-bit RGB data will be stored

in RGB RGB RGB format unless png_set_filler() or png_set_add_alpha()

is called to insert filler bytes, either before or after each RGB triplet.

16-bit RGB data will be returned RRGGBB RRGGBB, with the most significant

byte of the color value first, unless png_set_strip_16() is called to

transform it to regular RGB RGB triplets, or png_set_filler() or

png_set_add alpha() is called to insert filler bytes, either before or

after each RRGGBB triplet.  Similarly, 8-bit or 16-bit grayscale data can

be modified with

png_set_filler(), png_set_add_alpha(), or png_set_strip_16().

 

The following code transforms grayscale images of less than 8 to 8 bits,

changes paletted images to RGB, and adds a full alpha channel if there is

transparency information in a tRNS chunk.  This is most useful on

grayscale images with bit depths of 2 or 4 or if there is a multiple-image

viewing application that wishes to treat all images in the same way.

 

    if (color_type == PNG_COLOR_TYPE_PALETTE)

        png_set_palette_to_rgb(png_ptr);

 

    if (color_type == PNG_COLOR_TYPE_GRAY &&

        bit_depth < 8) png_set_expand_gray_1_2_4_to_8(png_ptr);

 

    if (png_get_valid(png_ptr, info_ptr,

        PNG_INFO_tRNS)) png_set_tRNS_to_alpha(png_ptr);

 

These three functions are actually aliases for png_set_expand(), added

in libpng version 1.0.4, with the function names expanded to improve code

readability.  In some future version they may actually do different

things.

 

As of libpng version 1.2.9, png_set_expand_gray_1_2_4_to_8() was

added.  It expands the sample depth without changing tRNS to alpha.

 

As of libpng version 1.5.1, not all possible expansions are supported.

 

In the following table, the 01 means grayscale with depth<8, 31 means

indexed with depth<8, other numerals represent the color type, "T" means

the tRNS chunk is present, A means an alpha channel is present, and O

means tRNS or alpha is present but all pixels in the image are opaque.

 

  FROM  01  31   0  0T  0O   2  2T  2O   3  3T  3O  4A  4O  6A  6O

   TO

   01    -                  

   31        -

    0    1       -          

   0T                -

   0O                    -

    2           GX           -

   2T                            -

   2O                                -

    3        1                           -

   3T                                        -

   3O                                            -

   4A                T                               -

   4O                                                    -

   6A               GX         TX           TX               -

   6O                   GX                      TX               -

 

Within the matrix,

     "-" means the transformation is not supported.

     "X" means the transformation is obtained by png_set_expand().

     "1" means the transformation is obtained by

         png_set_expand_gray_1_2_4_to_8

     "G" means the transformation is obtained by

         png_set_gray_to_rgb().

     "P" means the transformation is obtained by

         png_set_expand_palette_to_rgb().

     "T" means the transformation is obtained by

         png_set_tRNS_to_alpha().

 

PNG can have files with 16 bits per channel.  If you only can handle

8 bits per channel, this will strip the pixels down to 8 bit.

 

    if (bit_depth == 16)

       png_set_strip_16(png_ptr);

 

If, for some reason, you don't need the alpha channel on an image,

and you want to remove it rather than combining it with the background

(but the image author certainly had in mind that you *would* combine

it with the background, so that's what you should probably do):

 

    if (color_type & PNG_COLOR_MASK_ALPHA)

       png_set_strip_alpha(png_ptr);

 

In PNG files, the alpha channel in an image

is the level of opacity.  If you need the alpha channel in an image to

be the level of transparency instead of opacity, you can invert the

alpha channel (or the tRNS chunk data) after it's read, so that 0 is

fully opaque and 255 (in 8-bit or paletted images) or 65535 (in 16-bit

images) is fully transparent, with

 

    png_set_invert_alpha(png_ptr);

 

PNG files pack pixels of bit depths 1, 2, and 4 into bytes as small as

they can, resulting in, for example, 8 pixels per byte for 1 bit

files.  This code expands to 1 pixel per byte without changing the

values of the pixels:

 

    if (bit_depth < 8)

       png_set_packing(png_ptr);

 

PNG files have possible bit depths of 1, 2, 4, 8, and 16.  All pixels

stored in a PNG image have been "scaled" or "shifted" up to the next

higher possible bit depth (e.g. from 5 bits/sample in the range [0,31]

to 8 bits/sample in the range [0, 255]).  However, it is also possible

to convert the PNG pixel data back to the original bit depth of the

image.  This call reduces the pixels back down to the original bit depth:

 

    png_color_8p sig_bit;

 

    if (png_get_sBIT(png_ptr, info_ptr, &sig_bit))

       png_set_shift(png_ptr, sig_bit);

 

PNG files store 3-color pixels in red, green, blue order.  This code

changes the storage of the pixels to blue, green, red:

 

    if (color_type == PNG_COLOR_TYPE_RGB ||

        color_type == PNG_COLOR_TYPE_RGB_ALPHA)

       png_set_bgr(png_ptr);

 

PNG files store RGB pixels packed into 3 or 6 bytes. This code expands them

into 4 or 8 bytes for windowing systems that need them in this format:

 

    if (color_type == PNG_COLOR_TYPE_RGB)

       png_set_filler(png_ptr, filler, PNG_FILLER_BEFORE);

 

where "filler" is the 8 or 16-bit number to fill with, and the location is

either PNG_FILLER_BEFORE or PNG_FILLER_AFTER, depending upon whether

you want the filler before the RGB or after.  This transformation

does not affect images that already have full alpha channels.  To add an

opaque alpha channel, use filler=0xff or 0xffff and PNG_FILLER_AFTER which

will generate RGBA pixels.

 

Note that png_set_filler() does not change the color type.  If you want

to do that, you can add a true alpha channel with

 

    if (color_type == PNG_COLOR_TYPE_RGB ||

       color_type == PNG_COLOR_TYPE_GRAY)

       png_set_add_alpha(png_ptr, filler, PNG_FILLER_AFTER);

 

where "filler" contains the alpha value to assign to each pixel.

This function was added in libpng-1.2.7.

 

If you are reading an image with an alpha channel, and you need the

data as ARGB instead of the normal PNG format RGBA:

 

    if (color_type == PNG_COLOR_TYPE_RGB_ALPHA)

       png_set_swap_alpha(png_ptr);

 

For some uses, you may want a grayscale image to be represented as

RGB.  This code will do that conversion:

 

    if (color_type == PNG_COLOR_TYPE_GRAY ||

        color_type == PNG_COLOR_TYPE_GRAY_ALPHA)

       png_set_gray_to_rgb(png_ptr);

 

Conversely, you can convert an RGB or RGBA image to grayscale or grayscale

with alpha.

 

    if (color_type == PNG_COLOR_TYPE_RGB ||

        color_type == PNG_COLOR_TYPE_RGB_ALPHA)

       png_set_rgb_to_gray_fixed(png_ptr, error_action,

           int red_weight, int green_weight);

 

    error_action = 1: silently do the conversion

 

    error_action = 2: issue a warning if the original

                      image has any pixel where

                      red != green or red != blue

 

    error_action = 3: issue an error and abort the

                      conversion if the original

                      image has any pixel where

                      red != green or red != blue

 

    red_weight:       weight of red component times 100000

 

    green_weight:     weight of green component times 100000

                      If either weight is negative, default

                      weights (21268, 71514) are used.

 

If you have set error_action = 1 or 2, you can

later check whether the image really was gray, after processing

the image rows, with the png_get_rgb_to_gray_status(png_ptr) function.

It will return a png_byte that is zero if the image was gray or

1 if there were any non-gray pixels.  bKGD and sBIT data

will be silently converted to grayscale, using the green channel

data, regardless of the error_action setting.

 

With red_weight+green_weight<=100000,

the normalized graylevel is computed:

 

    int rw = red_weight * 65536;

    int gw = green_weight * 65536;

    int bw = 65536 - (rw + gw);

    gray = (rw*red + gw*green + bw*blue)/65536;

 

The default values approximate those recommended in the Charles

Poynton's Color FAQ, <http://www.inforamp.net/~poynton/>

Copyright (c) 1998-01-04 Charles Poynton <poynton at inforamp.net>

 

    Y = 0.212671 * R + 0.715160 * G + 0.072169 * B

 

Libpng approximates this with integers scaled by 32768:

 

    Y = (6968 * R + 23434 * G + 2366 * B)/32768

 

The calculation is done in a linear colorspace, if the image gamma

can be determined.

 

If you have a grayscale and you are using png_set_expand_depth(),

png_set_expand(), or png_set_gray_to_rgb to change to truecolor or to

a higher bit-depth, you must either supply the background color as a gray

value at the original file bit-depth (need_expand = 1) or else supply the

background color as an RGB triplet at the final, expanded bit depth

(need_expand = 0).  Similarly, if you are reading a paletted image, you

must either supply the background color as a palette index (need_expand = 1)

or as an RGB triplet that may or may not be in the palette (need_expand = 0).

 

    png_color_16 my_background;

    png_color_16p image_background;

 

    if (png_get_bKGD(png_ptr, info_ptr, &image_background))

       png_set_background(png_ptr, image_background,

           PNG_BACKGROUND_GAMMA_FILE, 1, 1.0);

    else

       png_set_background(png_ptr, &my_background,

           PNG_BACKGROUND_GAMMA_SCREEN, 0, 1.0);

 

The png_set_background() function tells libpng to composite images

with alpha or simple transparency against the supplied background

color.  If the PNG file contains a bKGD chunk (PNG_INFO_bKGD valid),

you may use this color, or supply another color more suitable for

the current display (e.g., the background color from a web page).  You

need to tell libpng whether the color is in the gamma space of the

display (PNG_BACKGROUND_GAMMA_SCREEN for colors you supply), the file

(PNG_BACKGROUND_GAMMA_FILE for colors from the bKGD chunk), or one

that is neither of these gammas (PNG_BACKGROUND_GAMMA_UNIQUE - I don't

know why anyone would use this, but it's here).

 

To properly display PNG images on any kind of system, the application needs

to know what the display gamma is.  Ideally, the user will know this, and

the application will allow them to set it.  One method of allowing the user

to set the display gamma separately for each system is to check for a

SCREEN_GAMMA or DISPLAY_GAMMA environment variable, which will hopefully be

correctly set.

 

Note that display_gamma is the overall gamma correction required to produce

pleasing results, which depends on the lighting conditions in the surrounding

environment.  In a dim or brightly lit room, no compensation other than

the physical gamma exponent of the monitor is needed, while in a dark room

a slightly smaller exponent is better.

 

   double gamma, screen_gamma;

 

   if (/* We have a user-defined screen

       gamma value */)

   {

      screen_gamma = user_defined_screen_gamma;

   }

 

   /* One way that applications can share the same

      screen gamma value */

   else if ((gamma_str = getenv("SCREEN_GAMMA"))

      != NULL)

   {

      screen_gamma = (double)atof(gamma_str);

   }

 

   /* If we don't have another value */

   else

   {

      screen_gamma = 2.2; /* A good guess for a

           PC monitor in a bright office or a dim room */

 

      screen_gamma = 2.0; /* A good guess for a

           PC monitor in a dark room */

 

      screen_gamma = 1.7 or 1.0;  /* A good

           guess for Mac systems */

   }

 

The functions png_set_gamma() and its fixed point equivalent

png_set_gamma_fixed() handle gamma transformations of the data.

Pass both the file gamma and the current screen_gamma.  If the file does

not have a gamma value, you can pass one anyway if you have an idea what

it is (usually 0.45455 is a good guess for GIF images on PCs).  Note

that file gammas are inverted from screen gammas.  See the discussions

on gamma in the PNG specification for an excellent description of what

gamma is, and why all applications should support it.  It is strongly

recommended that PNG viewers support gamma correction.

 

   if (png_get_gAMA(png_ptr, info_ptr, &file_gamma))

      png_set_gamma(png_ptr, screen_gamma, file_gamma);

 

   else

      png_set_gamma(png_ptr, screen_gamma, 0.45455);

 

If you need to reduce an RGB file to a paletted file, or if a paletted

file has more entries then will fit on your screen, png_set_quantize()

will do that.  Note that this is a simple match quantization that merely

finds the closest color available.  This should work fairly well with

optimized palettes, but fairly badly with linear color cubes.  If you

pass a palette that is larger then maximum_colors, the file will

reduce the number of colors in the palette so it will fit into

maximum_colors.  If there is a histogram, it will use it to make

more intelligent choices when reducing the palette.  If there is no

histogram, it may not do as good a job.

 

   if (color_type & PNG_COLOR_MASK_COLOR)

   {

      if (png_get_valid(png_ptr, info_ptr,

          PNG_INFO_PLTE))

      {

         png_uint_16p histogram = NULL;

 

         png_get_hIST(png_ptr, info_ptr,

             &histogram);

         png_set_quantize(png_ptr, palette, num_palette,

            max_screen_colors, histogram, 1);

      }

 

      else

      {

         png_color std_color_cube[MAX_SCREEN_COLORS] =

            { ... colors ... };

 

         png_set_quantize(png_ptr, std_color_cube,

            MAX_SCREEN_COLORS, MAX_SCREEN_COLORS,

            NULL,0);

      }

   }

 

PNG files describe monochrome as black being zero and white being one.

The following code will reverse this (make black be one and white be

zero):

 

   if (bit_depth == 1 && color_type == PNG_COLOR_TYPE_GRAY)

      png_set_invert_mono(png_ptr);

 

This function can also be used to invert grayscale and gray-alpha images:

 

   if (color_type == PNG_COLOR_TYPE_GRAY ||

       color_type == PNG_COLOR_TYPE_GRAY_ALPHA)

      png_set_invert_mono(png_ptr);

 

PNG files store 16 bit pixels in network byte order (big-endian,

ie. most significant bits first).  This code changes the storage to the

other way (little-endian, i.e. least significant bits first, the

way PCs store them):

 

    if (bit_depth == 16)

       png_set_swap(png_ptr);

 

If you are using packed-pixel images (1, 2, or 4 bits/pixel), and you

need to change the order the pixels are packed into bytes, you can use:

 

    if (bit_depth < 8)

       png_set_packswap(png_ptr);

 

Finally, you can write your own transformation function if none of

the existing ones meets your needs.  This is done by setting a callback

with

 

    png_set_read_user_transform_fn(png_ptr,

        read_transform_fn);

 

You must supply the function

 

    void read_transform_fn(png_structp png_ptr, row_info_ptr

        row_info, png_bytep data)

 

See pngtest.c for a working example.  Your function will be called

after all of the other transformations have been processed.  Take care with

interlaced images if you do the interlace yourself - the width of the row is the

width in 'row_info', not the overall image width.

 

If supported libpng provides two information routines that you can use to find

where you are in processing the image:

 

   png_get_current_pass_number(png_structp png_ptr);

   png_get_current_row_number(png_structp png_ptr);

 

Don't try using these outside a transform callback - firstly they are only

supported if user transforms are supported, secondly they may well return

unexpected results unless the row is actually being processed at the moment they

are called.

 

You can also set up a pointer to a user structure for use by your

callback function, and you can inform libpng that your transform

function will change the number of channels or bit depth with the

function

 

    png_set_user_transform_info(png_ptr, user_ptr,

        user_depth, user_channels);

 

The user's application, not libpng, is responsible for allocating and

freeing any memory required for the user structure.

 

You can retrieve the pointer via the function

png_get_user_transform_ptr().  For example:

 

    voidp read_user_transform_ptr =

        png_get_user_transform_ptr(png_ptr);

 

The last thing to handle is interlacing; this is covered in detail below,

but you must call the function here if you want libpng to handle expansion

of the interlaced image.

 

    number_of_passes = png_set_interlace_handling(png_ptr);

 

After setting the transformations, libpng can update your png_info

structure to reflect any transformations you've requested with this

call.  This is most useful to update the info structure's rowbytes

field so you can use it to allocate your image memory.  This function

will also update your palette with the correct screen_gamma and

background if these have been given with the calls above.

 

    png_read_update_info(png_ptr, info_ptr);

 

After you call png_read_update_info(), you can allocate any

memory you need to hold the image.  The row data is simply

raw byte data for all forms of images.  As the actual allocation

varies among applications, no example will be given.  If you

are allocating one large chunk, you will need to build an

array of pointers to each row, as it will be needed for some

of the functions below.

 

Remember: Before you call png_read_update_info(), the png_get_

functions return the values corresponding to the original PNG image.

After you call png_read_update_info the values refer to the image

that libpng will output.  Consequently you must call all the png_set_

functions before you call png_read_update_info().  This is particularly

important for png_set_interlace_handling() - if you are going to call

png_read_update_info() you must call png_set_interlace_handling() before

it unless you want to receive interlaced output.

 

Reading image data

 

After you've allocated memory, you can read the image data.

The simplest way to do this is in one function call.  If you are

allocating enough memory to hold the whole image, you can just

call png_read_image() and libpng will read in all the image data

and put it in the memory area supplied.  You will need to pass in

an array of pointers to each row.

 

This function automatically handles interlacing, so you don't

need to call png_set_interlace_handling() (unless you call

png_read_update_info()) or call this function multiple times, or any

of that other stuff necessary with png_read_rows().

 

   png_read_image(png_ptr, row_pointers);

 

where row_pointers is:

 

   png_bytep row_pointers[height];

 

You can point to void or char or whatever you use for pixels.

 

If you don't want to read in the whole image at once, you can

use png_read_rows() instead.  If there is no interlacing (check

interlace_type == PNG_INTERLACE_NONE), this is simple:

 

    png_read_rows(png_ptr, row_pointers, NULL,

        number_of_rows);

 

where row_pointers is the same as in the png_read_image() call.

 

If you are doing this just one row at a time, you can do this with

a single row_pointer instead of an array of row_pointers:

 

    png_bytep row_pointer = row;

    png_read_row(png_ptr, row_pointer, NULL);

 

If the file is interlaced (interlace_type != 0 in the IHDR chunk), things

get somewhat harder.  The only current (PNG Specification version 1.2)

interlacing type for PNG is (interlace_type == PNG_INTERLACE_ADAM7);

a somewhat complicated 2D interlace scheme, known as Adam7, that

breaks down an image into seven smaller images of varying size, based

on an 8x8 grid.  This number is defined (from libpng 1.5) as

PNG_INTERLACE_ADAM7_PASSES in png.h

 

libpng can fill out those images or it can give them to you "as is".

It is almost always better to have libpng handle the interlacing for you.

If you want the images filled out, there are two ways to do that.  The one

mentioned in the PNG specification is to expand each pixel to cover

those pixels that have not been read yet (the "rectangle" method).

This results in a blocky image for the first pass, which gradually

smooths out as more pixels are read.  The other method is the "sparkle"

method, where pixels are drawn only in their final locations, with the

rest of the image remaining whatever colors they were initialized to

before the start of the read.  The first method usually looks better,

but tends to be slower, as there are more pixels to put in the rows.

 

If, as is likely, you want libpng to expand the images, call this before

calling png_start_read_image() or png_read_update_info():

 

    if (interlace_type == PNG_INTERLACE_ADAM7)

       number_of_passes

           = png_set_interlace_handling(png_ptr);

 

This will return the number of passes needed.  Currently, this is seven,

but may change if another interlace type is added.  This function can be

called even if the file is not interlaced, where it will return one pass.

You then need to read the whole image 'number_of_passes' times.  Each time

will distribute the pixels from the current pass to the correct place in

the output image, so you need to supply the same rows to png_read_rows in

each pass.

 

If you are not going to display the image after each pass, but are

going to wait until the entire image is read in, use the sparkle

effect.  This effect is faster and the end result of either method

is exactly the same.  If you are planning on displaying the image

after each pass, the "rectangle" effect is generally considered the

better looking one.

 

If you only want the "sparkle" effect, just call png_read_rows() as

normal, with the third parameter NULL.  Make sure you make pass over

the image number_of_passes times, and you don't change the data in the

rows between calls.  You can change the locations of the data, just

not the data.  Each pass only writes the pixels appropriate for that

pass, and assumes the data from previous passes is still valid.

 

    png_read_rows(png_ptr, row_pointers, NULL,

        number_of_rows);

 

If you only want the first effect (the rectangles), do the same as

before except pass the row buffer in the third parameter, and leave

the second parameter NULL.

 

    png_read_rows(png_ptr, NULL, row_pointers,

        number_of_rows);

 

If you don't want libpng to handle the interlacing details, just call

png_read_rows() PNG_INTERLACE_ADAM7_PASSES times to read in all the images.

Each of the images is a valid image by itself, however you will almost

certainly need to distribute the pixels from each sub-image to the

correct place.  This is where everything gets very tricky.

 

If you want to retrieve the separate images you must pass the correct

number of rows to each successive call of png_read_rows().  The calculation

gets pretty complicated for small images, where some sub-images may

not even exist because either their width or height ends up zero.

libpng provides two macros to help you in 1.5 and later versions:

 

   png_uint_32 width = PNG_PASS_COLS(image_width, pass_number);

   png_uint_32 height = PNG_PASS_ROWS(image_height, pass_number);

 

Respectively these tell you the width and height of the sub-image

corresponding to the numbered pass.  'pass' is in in the range 0 to 6 -

this can be confusing because the specification refers to the same passes

as 1 to 7!  Be careful, you must check both the width and height before

calling png_read_rows() and not call it for that pass if either is zero.

 

You can, of course, read each sub-image row by row.  If you want to

produce optimal code to make a pixel-by-pixel transformation of an

interlaced image this is the best approach; read each row of each pass,

transform it, and write it out to a new interlaced image.

 

If you want to de-interlace the image yourself libpng provides further

macros to help that tell you where to place the pixels in the output image.

Because the interlacing scheme is rectangular - sub-image pixels are always

arranged on a rectangular grid - all you need to know for each pass is the

starting column and row in the output image of the first pixel plus the

spacing between each pixel.  As of libpng 1.5 there are four macros to

retrieve this information:

 

   png_uint_32 x = PNG_PASS_START_COL(pass);

   png_uint_32 y = PNG_PASS_START_ROW(pass);

   png_uint_32 xStep = 1U << PNG_PASS_COL_SHIFT(pass);

   png_uint_32 yStep = 1U << PNG_PASS_ROW_SHIFT(pass);

 

These allow you to write the obvious loop:

 

   png_uint_32 input_y = 0;

   png_uint_32 output_y = PNG_PASS_START_ROW(pass);

 

   while (output_y < output_image_height)

   {

      png_uint_32 input_x = 0;

      png_uint_32 output_x = PNG_PASS_START_COL(pass);

 

      while (output_x < output_image_width)

      {

         image[output_y][output_x] =

             subimage[pass][input_y][input_x++];

 

         output_x += xStep;

      }

 

      ++input_y;

      output_y += yStep;

   }

 

Notice that the steps between successive output rows and columns are

returned as shifts.  This is possible because the pixels in the subimages

are always a power of 2 apart - 1, 2, 4 or 8 pixels - in the original

image.  In practice you may need to directly calculate the output coordinate

given an input coordinate.  libpng provides two further macros for this

purpose:

 

   png_uint_32 output_x = PNG_COL_FROM_PASS_COL(input_x, pass);

   png_uint_32 output_y = PNG_ROW_FROM_PASS_ROW(input_y, pass);

 

Finally a pair of macros are provided to tell you if a particular image

row or column appears in a given pass:

 

   int col_in_pass = PNG_COL_IN_INTERLACE_PASS(output_x, pass);

   int row_in_pass = PNG_ROW_IN_INTERLACE_PASS(output_y, pass);

 

Bear in mind that you will probably also need to check the width and height

of the pass in addition to the above to be sure the pass even exists!

 

With any luck you are convinced by now that you don't want to do your own

interlace handling.  In reality normally the only good reason for doing this

is if you are processing PNG files on a pixel-by-pixel basis and don't want

to load the whole file into memory when it is interlaced.

 

libpng includes a test program, pngvalid, that illustrates reading and

writing of interlaced images.  If you can't get interlacing to work in your

code and don't want to leave it to libpng (the recommended approach) see

how pngvalid.c does it.

 

Finishing a sequential read

 

After you are finished reading the image through the

low-level interface, you can finish reading the file.  If you are

interested in comments or time, which may be stored either before or

after the image data, you should pass the separate png_info struct if

you want to keep the comments from before and after the image

separate.  If you are not interested, you can pass NULL.

 

   png_read_end(png_ptr, end_info);

 

When you are done, you can free all memory allocated by libpng like this:

 

   png_destroy_read_struct(&png_ptr, &info_ptr,

       &end_info);

 

It is also possible to individually free the info_ptr members that

point to libpng-allocated storage with the following function:

 

    png_free_data(png_ptr, info_ptr, mask, seq)

 

    mask - identifies data to be freed, a mask

           containing the bitwise OR of one or

           more of

             PNG_FREE_PLTE, PNG_FREE_TRNS,

             PNG_FREE_HIST, PNG_FREE_ICCP,

             PNG_FREE_PCAL, PNG_FREE_ROWS,

             PNG_FREE_SCAL, PNG_FREE_SPLT,

             PNG_FREE_TEXT, PNG_FREE_UNKN,

           or simply PNG_FREE_ALL

 

    seq  - sequence number of item to be freed

           (-1 for all items)

 

This function may be safely called when the relevant storage has

already been freed, or has not yet been allocated, or was allocated

by the user and not by libpng,  and will in those cases do nothing.

The "seq" parameter is ignored if only one item of the selected data

type, such as PLTE, is allowed.  If "seq" is not -1, and multiple items

are allowed for the data type identified in the mask, such as text or

sPLT, only the n'th item in the structure is freed, where n is "seq".

 

The default behavior is only to free data that was allocated internally

by libpng.  This can be changed, so that libpng will not free the data,

or so that it will free data that was allocated by the user with png_malloc()

or png_zalloc() and passed in via a png_set_*() function, with

 

    png_data_freer(png_ptr, info_ptr, freer, mask)

 

    freer  - one of

               PNG_DESTROY_WILL_FREE_DATA

               PNG_SET_WILL_FREE_DATA

               PNG_USER_WILL_FREE_DATA

 

    mask   - which data elements are affected

             same choices as in png_free_data()

 

This function only affects data that has already been allocated.

You can call this function after reading the PNG data but before calling

any png_set_*() functions, to control whether the user or the png_set_*()

function is responsible for freeing any existing data that might be present,

and again after the png_set_*() functions to control whether the user

or png_destroy_*() is supposed to free the data.  When the user assumes

responsibility for libpng-allocated data, the application must use

png_free() to free it, and when the user transfers responsibility to libpng

for data that the user has allocated, the user must have used png_malloc()

or png_zalloc() to allocate it.

 

If you allocated your row_pointers in a single block, as suggested above in

the description of the high level read interface, you must not transfer

responsibility for freeing it to the png_set_rows or png_read_destroy function,

because they would also try to free the individual row_pointers[i].

 

If you allocated text_ptr.text, text_ptr.lang, and text_ptr.translated_keyword

separately, do not transfer responsibility for freeing text_ptr to libpng,

because when libpng fills a png_text structure it combines these members with

the key member, and png_free_data() will free only text_ptr.key.  Similarly,

if you transfer responsibility for free'ing text_ptr from libpng to your

application, your application must not separately free those members.

 

The png_free_data() function will turn off the "valid" flag for anything

it frees.  If you need to turn the flag off for a chunk that was freed by

your application instead of by libpng, you can use

 

    png_set_invalid(png_ptr, info_ptr, mask);

 

    mask - identifies the chunks to be made invalid,

           containing the bitwise OR of one or

           more of

             PNG_INFO_gAMA, PNG_INFO_sBIT,

             PNG_INFO_cHRM, PNG_INFO_PLTE,

             PNG_INFO_tRNS, PNG_INFO_bKGD,

             PNG_INFO_hIST, PNG_INFO_pHYs,

             PNG_INFO_oFFs, PNG_INFO_tIME,

             PNG_INFO_pCAL, PNG_INFO_sRGB,

             PNG_INFO_iCCP, PNG_INFO_sPLT,

             PNG_INFO_sCAL, PNG_INFO_IDAT

 

For a more compact example of reading a PNG image, see the file example.c.

 

Reading PNG files progressively

 

The progressive reader is slightly different then the non-progressive

reader.  Instead of calling png_read_info(), png_read_rows(), and

png_read_end(), you make one call to png_process_data(), which calls

callbacks when it has the info, a row, or the end of the image.  You

set up these callbacks with png_set_progressive_read_fn().  You don't

have to worry about the input/output functions of libpng, as you are

giving the library the data directly in png_process_data().  I will

assume that you have read the section on reading PNG files above,

so I will only highlight the differences (although I will show

all of the code).

 

png_structp png_ptr;

png_infop info_ptr;

 

 /*  An example code fragment of how you would

     initialize the progressive reader in your

     application. */

 int

 initialize_png_reader()

 {

    png_ptr = png_create_read_struct

        (PNG_LIBPNG_VER_STRING, (png_voidp)user_error_ptr,

         user_error_fn, user_warning_fn);

 

    if (!png_ptr)

        return (ERROR);

 

    info_ptr = png_create_info_struct(png_ptr);

 

    if (!info_ptr)

    {

       png_destroy_read_struct(&png_ptr,

          (png_infopp)NULL, (png_infopp)NULL);

       return (ERROR);

    }

 

    if (setjmp(png_jmpbuf(png_ptr)))

    {

       png_destroy_read_struct(&png_ptr, &info_ptr,

          (png_infopp)NULL);

       return (ERROR);

    }

 

    /* This one's new.  You can provide functions

       to be called when the header info is valid,

       when each row is completed, and when the image

       is finished.  If you aren't using all functions,

       you can specify NULL parameters.  Even when all

       three functions are NULL, you need to call

       png_set_progressive_read_fn().  You can use

       any struct as the user_ptr (cast to a void pointer

       for the function call), and retrieve the pointer

       from inside the callbacks using the function

 

          png_get_progressive_ptr(png_ptr);

 

       which will return a void pointer, which you have

       to cast appropriately.

     */

    png_set_progressive_read_fn(png_ptr, (void *)user_ptr,

        info_callback, row_callback, end_callback);

 

    return 0;

 }

 

 /* A code fragment that you call as you receive blocks

   of data */

 int

 process_data(png_bytep buffer, png_uint_32 length)

 {

    if (setjmp(png_jmpbuf(png_ptr)))

    {

       png_destroy_read_struct(&png_ptr, &info_ptr,

           (png_infopp)NULL);

       return (ERROR);

    }

 

    /* This one's new also.  Simply give it a chunk

       of data from the file stream (in order, of

       course).  On machines with segmented memory

       models machines, don't give it any more than

       64K.  The library seems to run fine with sizes

       of 4K. Although you can give it much less if

       necessary (I assume you can give it chunks of

       1 byte, I haven't tried less then 256 bytes

       yet).  When this function returns, you may

       want to display any rows that were generated

       in the row callback if you don't already do

       so there.

     */

    png_process_data(png_ptr, info_ptr, buffer, length);

 

    /* At this point you can call png_process_data_skip if

       you want to handle data the library will skip yourself;

       it simply returns the number of bytes to skip (and stops

       libpng skipping that number of bytes on the next

       png_process_data call).

    return 0;

 }

 

 /* This function is called (as set by

    png_set_progressive_read_fn() above) when enough data

    has been supplied so all of the header has been

    read.

 */

 void

 info_callback(png_structp png_ptr, png_infop info)

 {

    /* Do any setup here, including setting any of

       the transformations mentioned in the Reading

       PNG files section.  For now, you _must_ call

       either png_start_read_image() or

       png_read_update_info() after all the

       transformations are set (even if you don't set

       any).  You may start getting rows before

       png_process_data() returns, so this is your

       last chance to prepare for that.

 

       This is where you turn on interlace handling,

       assuming you don't want to do it yourself.

 

       If you need to you can stop the processing of

       your original input data at this point by calling

       png_process_data_pause.  This returns the number

       of unprocessed bytes from the last png_process_data

       call - it is up to you to ensure that the next call

       sees these bytes again.  If you don't want to bother

       with this you can get libpng to cache the unread

       bytes by setting the 'save' parameter (see png.h) but

       then libpng will have to copy the data internally.

     */

 }

 

 /* This function is called when each row of image

    data is complete */

 void

 row_callback(png_structp png_ptr, png_bytep new_row,

    png_uint_32 row_num, int pass)

 {

    /* If the image is interlaced, and you turned

       on the interlace handler, this function will

       be called for every row in every pass.  Some

       of these rows will not be changed from the

       previous pass.  When the row is not changed,

       the new_row variable will be NULL.  The rows

       and passes are called in order, so you don't

       really need the row_num and pass, but I'm

       supplying them because it may make your life

       easier.

 

       If you did not turn on interlace handling then

       the callback is called for each row of each

       sub-image when the image is interlaced.  In this

       case 'row_num' is the row in the sub-image, not

       the row in the output image as it is in all other

       cases.

 

       For the non-NULL rows of interlaced images when

       you have switched on libpng interlace handling,

       you must call png_progressive_combine_row()

       passing in the row and the old row.  You can

       call this function for NULL rows (it will just

       return) and for non-interlaced images (it just

       does the memcpy for you) if it will make the

       code easier.  Thus, you can just do this for

       all cases if you switch on interlace handling;

     */

 

        png_progressive_combine_row(png_ptr, old_row,

          new_row);

 

    /* where old_row is what was displayed for

       previously for the row.  Note that the first

       pass (pass == 0, really) will completely cover

       the old row, so the rows do not have to be

       initialized.  After the first pass (and only

       for interlaced images), you will have to pass

       the current row, and the function will combine

       the old row and the new row.

 

       You can also call png_process_data_pause in this

       callback - see above.

    */

 }

 

 void

 end_callback(png_structp png_ptr, png_infop info)

 {

    /* This function is called after the whole image

       has been read, including any chunks after the

       image (up to and including the IEND).  You

       will usually have the same info chunk as you

       had in the header, although some data may have

       been added to the comments and time fields.

 

       Most people won't do much here, perhaps setting

       a flag that marks the image as finished.

     */

 }

 

 

 

IV. Writing

 

Much of this is very similar to reading.  However, everything of

importance is repeated here, so you won't have to constantly look

back up in the reading section to understand writing.

 

Setup

 

You will want to do the I/O initialization before you get into libpng,

so if it doesn't work, you don't have anything to undo. If you are not

using the standard I/O functions, you will need to replace them with

custom writing functions.  See the discussion under Customizing libpng.

 

    FILE *fp = fopen(file_name, "wb");

 

    if (!fp)

       return (ERROR);

 

Next, png_struct and png_info need to be allocated and initialized.

As these can be both relatively large, you may not want to store these

on the stack, unless you have stack space to spare.  Of course, you

will want to check if they return NULL.  If you are also reading,

you won't want to name your read structure and your write structure

both "png_ptr"; you can call them anything you like, such as

"read_ptr" and "write_ptr".  Look at pngtest.c, for example.

 

    png_structp png_ptr = png_create_write_struct

       (PNG_LIBPNG_VER_STRING, (png_voidp)user_error_ptr,

        user_error_fn, user_warning_fn);

 

    if (!png_ptr)

       return (ERROR);

 

    png_infop info_ptr = png_create_info_struct(png_ptr);

    if (!info_ptr)

    {

       png_destroy_write_struct(&png_ptr,

           (png_infopp)NULL);

       return (ERROR);

    }

 

If you want to use your own memory allocation routines,

define PNG_USER_MEM_SUPPORTED and use

png_create_write_struct_2() instead of png_create_write_struct():

 

    png_structp png_ptr = png_create_write_struct_2

       (PNG_LIBPNG_VER_STRING, (png_voidp)user_error_ptr,

        user_error_fn, user_warning_fn, (png_voidp)

        user_mem_ptr, user_malloc_fn, user_free_fn);

 

After you have these structures, you will need to set up the

error handling.  When libpng encounters an error, it expects to

longjmp() back to your routine.  Therefore, you will need to call

setjmp() and pass the png_jmpbuf(png_ptr).  If you

write the file from different routines, you will need to update

the png_jmpbuf(png_ptr) every time you enter a new routine that will

call a png_*() function.  See your documentation of setjmp/longjmp

for your compiler for more information on setjmp/longjmp.  See

the discussion on libpng error handling in the Customizing Libpng

section below for more information on the libpng error handling.

 

    if (setjmp(png_jmpbuf(png_ptr)))

    {

    png_destroy_write_struct(&png_ptr, &info_ptr);

       fclose(fp);

       return (ERROR);

    }

    ...

    return;

 

If you would rather avoid the complexity of setjmp/longjmp issues,

you can compile libpng with PNG_NO_SETJMP, in which case

errors will result in a call to PNG_ABORT() which defaults to abort().

 

You can #define PNG_ABORT() to a function that does something

more useful than abort(), as long as your function does not

return.

 

Now you need to set up the output code.  The default for libpng is to

use the C function fwrite().  If you use this, you will need to pass a

valid FILE * in the function png_init_io().  Be sure that the file is

opened in binary mode.  Again, if you wish to handle writing data in

another way, see the discussion on libpng I/O handling in the Customizing

Libpng section below.

 

    png_init_io(png_ptr, fp);

 

If you are embedding your PNG into a datastream such as MNG, and don't

want libpng to write the 8-byte signature, or if you have already

written the signature in your application, use

 

    png_set_sig_bytes(png_ptr, 8);

 

to inform libpng that it should not write a signature.

 

Write callbacks

 

At this point, you can set up a callback function that will be

called after each row has been written, which you can use to control

a progress meter or the like.  It's demonstrated in pngtest.c.

You must supply a function

 

    void write_row_callback(png_structp png_ptr, png_uint_32 row,

       int pass);

    {

      /* put your code here */

    }

 

(You can give it another name that you like instead of "write_row_callback")

 

To inform libpng about your function, use

 

    png_set_write_status_fn(png_ptr, write_row_callback);

 

You now have the option of modifying how the compression library will

run.  The following functions are mainly for testing, but may be useful

in some cases, like if you need to write PNG files extremely fast and

are willing to give up some compression, or if you want to get the

maximum possible compression at the expense of slower writing.  If you

have no special needs in this area, let the library do what it wants by

not calling this function at all, as it has been tuned to deliver a good

speed/compression ratio. The second parameter to png_set_filter() is

the filter method, for which the only valid values are 0 (as of the

July 1999 PNG specification, version 1.2) or 64 (if you are writing

a PNG datastream that is to be embedded in a MNG datastream).  The third

parameter is a flag that indicates which filter type(s) are to be tested

for each scanline.  See the PNG specification for details on the specific

filter types.

 

 

    /* turn on or off filtering, and/or choose

       specific filters.  You can use either a single

       PNG_FILTER_VALUE_NAME or the bitwise OR of one

       or more PNG_FILTER_NAME masks.

     */

    png_set_filter(png_ptr, 0,

       PNG_FILTER_NONE  | PNG_FILTER_VALUE_NONE |

       PNG_FILTER_SUB   | PNG_FILTER_VALUE_SUB  |

       PNG_FILTER_UP    | PNG_FILTER_VALUE_UP   |

       PNG_FILTER_AVG   | PNG_FILTER_VALUE_AVG  |

       PNG_FILTER_PAETH | PNG_FILTER_VALUE_PAETH|

       PNG_ALL_FILTERS);

 

If an application wants to start and stop using particular filters during

compression, it should start out with all of the filters (to ensure that

the previous row of pixels will be stored in case it's needed later),

and then add and remove them after the start of compression.

 

If you are writing a PNG datastream that is to be embedded in a MNG

datastream, the second parameter can be either 0 or 64.

 

The png_set_compression_*() functions interface to the zlib compression

library, and should mostly be ignored unless you really know what you are

doing.  The only generally useful call is png_set_compression_level()

which changes how much time zlib spends on trying to compress the image

data.  See the Compression Library (zlib.h and algorithm.txt, distributed

with zlib) for details on the compression levels.

 

    /* set the zlib compression level */

    png_set_compression_level(png_ptr,

        Z_BEST_COMPRESSION);

 

    /* set other zlib parameters */

    png_set_compression_mem_level(png_ptr, 8);

    png_set_compression_strategy(png_ptr,

        Z_DEFAULT_STRATEGY);

    png_set_compression_window_bits(png_ptr, 15);

    png_set_compression_method(png_ptr, 8);

    png_set_compression_buffer_size(png_ptr, 8192)

 

extern PNG_EXPORT(void,png_set_zbuf_size)

 

Setting the contents of info for output

 

You now need to fill in the png_info structure with all the data you

wish to write before the actual image.  Note that the only thing you

are allowed to write after the image is the text chunks and the time

chunk (as of PNG Specification 1.2, anyway).  See png_write_end() and

the latest PNG specification for more information on that.  If you

wish to write them before the image, fill them in now, and flag that

data as being valid.  If you want to wait until after the data, don't

fill them until png_write_end().  For all the fields in png_info and

their data types, see png.h.  For explanations of what the fields

contain, see the PNG specification.

 

Some of the more important parts of the png_info are:

 

    png_set_IHDR(png_ptr, info_ptr, width, height,

       bit_depth, color_type, interlace_type,

       compression_type, filter_method)

 

    width          - holds the width of the image

                     in pixels (up to 2^31).

 

    height         - holds the height of the image

                     in pixels (up to 2^31).

 

    bit_depth      - holds the bit depth of one of the

                     image channels.

                     (valid values are 1, 2, 4, 8, 16

                     and depend also on the

                     color_type.  See also significant

                     bits (sBIT) below).

 

    color_type     - describes which color/alpha

                     channels are present.

                     PNG_COLOR_TYPE_GRAY

                        (bit depths 1, 2, 4, 8, 16)

                     PNG_COLOR_TYPE_GRAY_ALPHA

                        (bit depths 8, 16)

                     PNG_COLOR_TYPE_PALETTE

                        (bit depths 1, 2, 4, 8)

                     PNG_COLOR_TYPE_RGB

                        (bit_depths 8, 16)

                     PNG_COLOR_TYPE_RGB_ALPHA

                        (bit_depths 8, 16)

 

                     PNG_COLOR_MASK_PALETTE

                     PNG_COLOR_MASK_COLOR

                     PNG_COLOR_MASK_ALPHA

 

    interlace_type - PNG_INTERLACE_NONE or

                     PNG_INTERLACE_ADAM7

 

    compression_type - (must be

                     PNG_COMPRESSION_TYPE_DEFAULT)

 

    filter_method  - (must be PNG_FILTER_TYPE_DEFAULT

                     or, if you are writing a PNG to

                     be embedded in a MNG datastream,

                     can also be

                     PNG_INTRAPIXEL_DIFFERENCING)

 

If you call png_set_IHDR(), the call must appear before any of the

other png_set_*() functions, because they might require access to some of

the IHDR settings.  The remaining png_set_*() functions can be called

in any order.

 

If you wish, you can reset the compression_type, interlace_type, or

filter_method later by calling png_set_IHDR() again; if you do this, the

width, height, bit_depth, and color_type must be the same in each call.

 

    png_set_PLTE(png_ptr, info_ptr, palette,

       num_palette);

 

    palette        - the palette for the file

                     (array of png_color)

    num_palette    - number of entries in the palette

 

    png_set_gAMA(png_ptr, info_ptr, file_gamma);

    png_set_gAMA_fixed(png_ptr, info_ptr, int_file_gamma);

 

    file_gamma     - the gamma at which the image was

                     created (PNG_INFO_gAMA)

 

    int_file_gamma - 100,000 times the gamma at which

                     the image was created

 

    png_set_sRGB(png_ptr, info_ptr, srgb_intent);

 

    srgb_intent    - the rendering intent

                     (PNG_INFO_sRGB) The presence of

                     the sRGB chunk means that the pixel

                     data is in the sRGB color space.

                     This chunk also implies specific

                     values of gAMA and cHRM.  Rendering

                     intent is the CSS-1 property that

                     has been defined by the International

                     Color Consortium

                     (http://www.color.org).

                     It can be one of

                     PNG_sRGB_INTENT_SATURATION,

                     PNG_sRGB_INTENT_PERCEPTUAL,

                     PNG_sRGB_INTENT_ABSOLUTE, or

                     PNG_sRGB_INTENT_RELATIVE.

 

 

    png_set_sRGB_gAMA_and_cHRM(png_ptr, info_ptr,

       srgb_intent);

 

    srgb_intent    - the rendering intent

                     (PNG_INFO_sRGB) The presence of the

                     sRGB chunk means that the pixel

                     data is in the sRGB color space.

                     This function also causes gAMA and

                     cHRM chunks with the specific values

                     that are consistent with sRGB to be

                     written.

 

    png_set_iCCP(png_ptr, info_ptr, name, compression_type,

                       profile, proflen);

 

    name             - The profile name.

 

    compression_type - The compression type; always

                       PNG_COMPRESSION_TYPE_BASE for PNG 1.0.

                       You may give NULL to this argument to

                       ignore it.

 

    profile          - International Color Consortium color

                       profile data. May contain NULs.

 

    proflen          - length of profile data in bytes.

 

    png_set_sBIT(png_ptr, info_ptr, sig_bit);

 

    sig_bit        - the number of significant bits for

                     (PNG_INFO_sBIT) each of the gray, red,

                     green, and blue channels, whichever are

                     appropriate for the given color type

                     (png_color_16)

 

    png_set_tRNS(png_ptr, info_ptr, trans_alpha,

       num_trans, trans_color);

 

    trans_alpha    - array of alpha (transparency)

                     entries for palette (PNG_INFO_tRNS)

 

    trans_color    - graylevel or color sample values

                     (in order red, green, blue) of the

                     single transparent color for

                     non-paletted images (PNG_INFO_tRNS)

 

    num_trans      - number of transparent entries

                     (PNG_INFO_tRNS)

 

    png_set_hIST(png_ptr, info_ptr, hist);

 

    hist           - histogram of palette (array of

                     png_uint_16) (PNG_INFO_hIST)

 

    png_set_tIME(png_ptr, info_ptr, mod_time);

 

    mod_time       - time image was last modified

                     (PNG_VALID_tIME)

 

    png_set_bKGD(png_ptr, info_ptr, background);

 

    background     - background color (PNG_VALID_bKGD)

 

    png_set_text(png_ptr, info_ptr, text_ptr, num_text);

 

    text_ptr       - array of png_text holding image

                     comments

 

    text_ptr[i].compression - type of compression used

                 on "text" PNG_TEXT_COMPRESSION_NONE

                           PNG_TEXT_COMPRESSION_zTXt

                           PNG_ITXT_COMPRESSION_NONE

                           PNG_ITXT_COMPRESSION_zTXt

    text_ptr[i].key   - keyword for comment.  Must contain

                 1-79 characters.

    text_ptr[i].text  - text comments for current

                         keyword.  Can be NULL or empty.

    text_ptr[i].text_length - length of text string,

                 after decompression, 0 for iTXt

    text_ptr[i].itxt_length - length of itxt string,

                 after decompression, 0 for tEXt/zTXt

    text_ptr[i].lang  - language of comment (NULL or

                         empty for unknown).

    text_ptr[i].translated_keyword  - keyword in UTF-8 (NULL

                         or empty for unknown).

    Note that the itxt_length, lang, and lang_key

    members of the text_ptr structure only exist

    when the library is built with iTXt chunk support.

 

    num_text       - number of comments

 

    png_set_sPLT(png_ptr, info_ptr, &palette_ptr,

       num_spalettes);

 

    palette_ptr    - array of png_sPLT_struct structures

                     to be added to the list of palettes

                     in the info structure.

    num_spalettes  - number of palette structures to be

                     added.

 

    png_set_oFFs(png_ptr, info_ptr, offset_x, offset_y,

        unit_type);

 

    offset_x  - positive offset from the left

                     edge of the screen

 

    offset_y  - positive offset from the top

                     edge of the screen

 

    unit_type - PNG_OFFSET_PIXEL, PNG_OFFSET_MICROMETER

 

    png_set_pHYs(png_ptr, info_ptr, res_x, res_y,

        unit_type);

 

    res_x       - pixels/unit physical resolution

                  in x direction

 

    res_y       - pixels/unit physical resolution

                  in y direction

 

    unit_type   - PNG_RESOLUTION_UNKNOWN,

                  PNG_RESOLUTION_METER

 

    png_set_sCAL(png_ptr, info_ptr, unit, width, height)

 

    unit        - physical scale units (an integer)

 

    width       - width of a pixel in physical scale units

 

    height      - height of a pixel in physical scale units

                  (width and height are doubles)

 

    png_set_sCAL_s(png_ptr, info_ptr, unit, width, height)

 

    unit        - physical scale units (an integer)

 

    width       - width of a pixel in physical scale units

 

    height      - height of a pixel in physical scale units

                 (width and height are strings like "2.54")

 

    png_set_unknown_chunks(png_ptr, info_ptr, &unknowns,

       num_unknowns)

 

    unknowns          - array of png_unknown_chunk

                        structures holding unknown chunks

    unknowns[i].name  - name of unknown chunk

    unknowns[i].data  - data of unknown chunk

    unknowns[i].size  - size of unknown chunk's data

    unknowns[i].location - position to write chunk in file

                           0: do not write chunk

                           PNG_HAVE_IHDR: before PLTE

                           PNG_HAVE_PLTE: before IDAT

                           PNG_AFTER_IDAT: after IDAT

 

The "location" member is set automatically according to

what part of the output file has already been written.

You can change its value after calling png_set_unknown_chunks()

as demonstrated in pngtest.c.  Within each of the "locations",

the chunks are sequenced according to their position in the

structure (that is, the value of "i", which is the order in which

the chunk was either read from the input file or defined with

png_set_unknown_chunks).

 

A quick word about text and num_text.  text is an array of png_text

structures.  num_text is the number of valid structures in the array.

Each png_text structure holds a language code, a keyword, a text value,

and a compression type.

 

The compression types have the same valid numbers as the compression

types of the image data.  Currently, the only valid number is zero.

However, you can store text either compressed or uncompressed, unlike

images, which always have to be compressed.  So if you don't want the

text compressed, set the compression type to PNG_TEXT_COMPRESSION_NONE.

Because tEXt and zTXt chunks don't have a language field, if you

specify PNG_TEXT_COMPRESSION_NONE or PNG_TEXT_COMPRESSION_zTXt

any language code or translated keyword will not be written out.

 

Until text gets around 1000 bytes, it is not worth compressing it.

After the text has been written out to the file, the compression type

is set to PNG_TEXT_COMPRESSION_NONE_WR or PNG_TEXT_COMPRESSION_zTXt_WR,

so that it isn't written out again at the end (in case you are calling

png_write_end() with the same struct).

 

The keywords that are given in the PNG Specification are:

 

    Title            Short (one line) title or

                     caption for image

 

    Author           Name of image's creator

 

    Description      Description of image (possibly long)

 

    Copyright        Copyright notice

 

    Creation Time    Time of original image creation

                     (usually RFC 1123 format, see below)

 

    Software         Software used to create the image

 

    Disclaimer       Legal disclaimer

 

    Warning          Warning of nature of content

 

    Source           Device used to create the image

 

    Comment          Miscellaneous comment; conversion

                     from other image format

 

The keyword-text pairs work like this.  Keywords should be short

simple descriptions of what the comment is about.  Some typical

keywords are found in the PNG specification, as is some recommendations

on keywords.  You can repeat keywords in a file.  You can even write

some text before the image and some after.  For example, you may want

to put a description of the image before the image, but leave the

disclaimer until after, so viewers working over modem connections

don't have to wait for the disclaimer to go over the modem before

they start seeing the image.  Finally, keywords should be full

words, not abbreviations.  Keywords and text are in the ISO 8859-1

(Latin-1) character set (a superset of regular ASCII) and can not

contain NUL characters, and should not contain control or other

unprintable characters.  To make the comments widely readable, stick

with basic ASCII, and avoid machine specific character set extensions

like the IBM-PC character set.  The keyword must be present, but

you can leave off the text string on non-compressed pairs.

Compressed pairs must have a text string, as only the text string

is compressed anyway, so the compression would be meaningless.

 

PNG supports modification time via the png_time structure.  Two

conversion routines are provided, png_convert_from_time_t() for

time_t and png_convert_from_struct_tm() for struct tm.  The

time_t routine uses gmtime().  You don't have to use either of

these, but if you wish to fill in the png_time structure directly,

you should provide the time in universal time (GMT) if possible

instead of your local time.  Note that the year number is the full

year (e.g. 1998, rather than 98 - PNG is year 2000 compliant!), and

that months start with 1.

 

If you want to store the time of the original image creation, you should

use a plain tEXt chunk with the "Creation Time" keyword.  This is

necessary because the "creation time" of a PNG image is somewhat vague,

depending on whether you mean the PNG file, the time the image was

created in a non-PNG format, a still photo from which the image was

scanned, or possibly the subject matter itself.  In order to facilitate

machine-readable dates, it is recommended that the "Creation Time"

tEXt chunk use RFC 1123 format dates (e.g. "22 May 1997 18:07:10 GMT"),

although this isn't a requirement.  Unlike the tIME chunk, the

"Creation Time" tEXt chunk is not expected to be automatically changed

by the software.  To facilitate the use of RFC 1123 dates, a function

png_convert_to_rfc1123(png_timep) is provided to convert from PNG

time to an RFC 1123 format string.

 

Writing unknown chunks

 

You can use the png_set_unknown_chunks function to queue up chunks

for writing.  You give it a chunk name, raw data, and a size; that's

all there is to it.  The chunks will be written by the next following

png_write_info_before_PLTE, png_write_info, or png_write_end function.

Any chunks previously read into the info structure's unknown-chunk

list will also be written out in a sequence that satisfies the PNG

specification's ordering rules.

 

The high-level write interface

 

At this point there are two ways to proceed; through the high-level

write interface, or through a sequence of low-level write operations.

You can use the high-level interface if your image data is present

in the info structure.  All defined output

transformations are permitted, enabled by the following masks.

 

    PNG_TRANSFORM_IDENTITY      No transformation

    PNG_TRANSFORM_PACKING       Pack 1, 2 and 4-bit samples

    PNG_TRANSFORM_PACKSWAP      Change order of packed

                                pixels to LSB first

    PNG_TRANSFORM_INVERT_MONO   Invert monochrome images

    PNG_TRANSFORM_SHIFT         Normalize pixels to the

                                sBIT depth

    PNG_TRANSFORM_BGR           Flip RGB to BGR, RGBA

                                to BGRA

    PNG_TRANSFORM_SWAP_ALPHA    Flip RGBA to ARGB or GA

                                to AG

    PNG_TRANSFORM_INVERT_ALPHA  Change alpha from opacity

                                to transparency

    PNG_TRANSFORM_SWAP_ENDIAN   Byte-swap 16-bit samples

    PNG_TRANSFORM_STRIP_FILLER        Strip out filler

                                      bytes (deprecated).

    PNG_TRANSFORM_STRIP_FILLER_BEFORE Strip out leading

                                      filler bytes

    PNG_TRANSFORM_STRIP_FILLER_AFTER  Strip out trailing

                                      filler bytes

 

If you have valid image data in the info structure (you can use

png_set_rows() to put image data in the info structure), simply do this:

 

    png_write_png(png_ptr, info_ptr, png_transforms, NULL)

 

where png_transforms is an integer containing the bitwise OR of some set of

transformation flags.  This call is equivalent to png_write_info(),

followed the set of transformations indicated by the transform mask,

then png_write_image(), and finally png_write_end().

 

(The final parameter of this call is not yet used.  Someday it might point

to transformation parameters required by some future output transform.)

 

You must use png_transforms and not call any png_set_transform() functions

when you use png_write_png().

 

The low-level write interface

 

If you are going the low-level route instead, you are now ready to

write all the file information up to the actual image data.  You do

this with a call to png_write_info().

 

    png_write_info(png_ptr, info_ptr);

 

Note that there is one transformation you may need to do before

png_write_info().  In PNG files, the alpha channel in an image is the

level of opacity.  If your data is supplied as a level of transparency,

you can invert the alpha channel before you write it, so that 0 is

fully transparent and 255 (in 8-bit or paletted images) or 65535

(in 16-bit images) is fully opaque, with

 

    png_set_invert_alpha(png_ptr);

 

This must appear before png_write_info() instead of later with the

other transformations because in the case of paletted images the tRNS

chunk data has to be inverted before the tRNS chunk is written.  If

your image is not a paletted image, the tRNS data (which in such cases

represents a single color to be rendered as transparent) won't need to

be changed, and you can safely do this transformation after your

png_write_info() call.

 

If you need to write a private chunk that you want to appear before

the PLTE chunk when PLTE is present, you can write the PNG info in

two steps, and insert code to write your own chunk between them:

 

    png_write_info_before_PLTE(png_ptr, info_ptr);

    png_set_unknown_chunks(png_ptr, info_ptr, ...);

    png_write_info(png_ptr, info_ptr);

 

After you've written the file information, you can set up the library

to handle any special transformations of the image data.  The various

ways to transform the data will be described in the order that they

should occur.  This is important, as some of these change the color

type and/or bit depth of the data, and some others only work on

certain color types and bit depths.  Even though each transformation

checks to see if it has data that it can do something with, you should

make sure to only enable a transformation if it will be valid for the

data.  For example, don't swap red and blue on grayscale data.

 

PNG files store RGB pixels packed into 3 or 6 bytes.  This code tells

the library to strip input data that has 4 or 8 bytes per pixel down

to 3 or 6 bytes (or strip 2 or 4-byte grayscale+filler data to 1 or 2

bytes per pixel).

 

    png_set_filler(png_ptr, 0, PNG_FILLER_BEFORE);

 

where the 0 is unused, and the location is either PNG_FILLER_BEFORE or

PNG_FILLER_AFTER, depending upon whether the filler byte in the pixel

is stored XRGB or RGBX.

 

PNG files pack pixels of bit depths 1, 2, and 4 into bytes as small as

they can, resulting in, for example, 8 pixels per byte for 1 bit files.

If the data is supplied at 1 pixel per byte, use this code, which will

correctly pack the pixels into a single byte:

 

    png_set_packing(png_ptr);

 

PNG files reduce possible bit depths to 1, 2, 4, 8, and 16.  If your

data is of another bit depth, you can write an sBIT chunk into the

file so that decoders can recover the original data if desired.

 

    /* Set the true bit depth of the image data */

    if (color_type & PNG_COLOR_MASK_COLOR)

    {

       sig_bit.red = true_bit_depth;

       sig_bit.green = true_bit_depth;

       sig_bit.blue = true_bit_depth;

    }

 

    else

    {

       sig_bit.gray = true_bit_depth;

    }

 

    if (color_type & PNG_COLOR_MASK_ALPHA)

    {

       sig_bit.alpha = true_bit_depth;

    }

 

    png_set_sBIT(png_ptr, info_ptr, &sig_bit);

 

If the data is stored in the row buffer in a bit depth other than

one supported by PNG (e.g. 3 bit data in the range 0-7 for a 4-bit PNG),

this will scale the values to appear to be the correct bit depth as

is required by PNG.

 

    png_set_shift(png_ptr, &sig_bit);

 

PNG files store 16 bit pixels in network byte order (big-endian,

ie. most significant bits first).  This code would be used if they are

supplied the other way (little-endian, i.e. least significant bits

first, the way PCs store them):

 

    if (bit_depth > 8)

       png_set_swap(png_ptr);

 

If you are using packed-pixel images (1, 2, or 4 bits/pixel), and you

need to change the order the pixels are packed into bytes, you can use:

 

    if (bit_depth < 8)

       png_set_packswap(png_ptr);

 

PNG files store 3 color pixels in red, green, blue order.  This code

would be used if they are supplied as blue, green, red:

 

    png_set_bgr(png_ptr);

 

PNG files describe monochrome as black being zero and white being

one. This code would be used if the pixels are supplied with this reversed

(black being one and white being zero):

 

    png_set_invert_mono(png_ptr);

 

Finally, you can write your own transformation function if none of

the existing ones meets your needs.  This is done by setting a callback

with

 

    png_set_write_user_transform_fn(png_ptr,

       write_transform_fn);

 

You must supply the function

 

    void write_transform_fn(png_structp png_ptr ptr,

       row_info_ptr row_info, png_bytep data)

 

See pngtest.c for a working example.  Your function will be called

before any of the other transformations are processed.  If supported

libpng also supplies an information routine that may be called from

your callback:

 

   png_get_current_row_number(png_ptr);

 

This returns the current row passed to the transform.  Even with interlaced

images the value returned is the row in the final output image.

 

You can also set up a pointer to a user structure for use by your

callback function.

 

    png_set_user_transform_info(png_ptr, user_ptr, 0, 0);

 

The user_channels and user_depth parameters of this function are ignored

when writing; you can set them to zero as shown.

 

You can retrieve the pointer via the function png_get_user_transform_ptr().

For example:

 

    voidp write_user_transform_ptr =

       png_get_user_transform_ptr(png_ptr);

 

It is possible to have libpng flush any pending output, either manually,

or automatically after a certain number of lines have been written.  To

flush the output stream a single time call:

 

    png_write_flush(png_ptr);

 

and to have libpng flush the output stream periodically after a certain

number of scanlines have been written, call:

 

    png_set_flush(png_ptr, nrows);

 

Note that the distance between rows is from the last time png_write_flush()

was called, or the first row of the image if it has never been called.

So if you write 50 lines, and then png_set_flush 25, it will flush the

output on the next scanline, and every 25 lines thereafter, unless

png_write_flush() is called before 25 more lines have been written.

If nrows is too small (less than about 10 lines for a 640 pixel wide

RGB image) the image compression may decrease noticeably (although this

may be acceptable for real-time applications).  Infrequent flushing will

only degrade the compression performance by a few percent over images

that do not use flushing.

 

Writing the image data

 

That's it for the transformations.  Now you can write the image data.

The simplest way to do this is in one function call.  If you have the

whole image in memory, you can just call png_write_image() and libpng

will write the image.  You will need to pass in an array of pointers to

each row.  This function automatically handles interlacing, so you don't

need to call png_set_interlace_handling() or call this function multiple

times, or any of that other stuff necessary with png_write_rows().

 

    png_write_image(png_ptr, row_pointers);

 

where row_pointers is:

 

    png_byte *row_pointers[height];

 

You can point to void or char or whatever you use for pixels.

 

If you don't want to write the whole image at once, you can

use png_write_rows() instead.  If the file is not interlaced,

this is simple:

 

    png_write_rows(png_ptr, row_pointers,

       number_of_rows);

 

row_pointers is the same as in the png_write_image() call.

 

If you are just writing one row at a time, you can do this with

a single row_pointer instead of an array of row_pointers:

 

    png_bytep row_pointer = row;

 

    png_write_row(png_ptr, row_pointer);

 

When the file is interlaced, things can get a good deal more complicated.

The only currently (as of the PNG Specification version 1.2, dated July

1999) defined interlacing scheme for PNG files is the "Adam7" interlace

scheme, that breaks down an image into seven smaller images of varying

size.  libpng will build these images for you, or you can do them

yourself.  If you want to build them yourself, see the PNG specification

for details of which pixels to write when.

 

If you don't want libpng to handle the interlacing details, just

use png_set_interlace_handling() and call png_write_rows() the

correct number of times to write all the sub-images

(png_set_interlace_handling() returns the number of sub-images.)

 

If you want libpng to build the sub-images, call this before you start

writing any rows:

 

    number_of_passes = png_set_interlace_handling(png_ptr);

 

This will return the number of passes needed.  Currently, this is seven,

but may change if another interlace type is added.

 

Then write the complete image number_of_passes times.

 

    png_write_rows(png_ptr, row_pointers, number_of_rows);

 

Think carefully before you write an interlaced image.  Typically code that

reads such images reads all the image data into memory, uncompressed, before

doing any processing.  Only code that can display an image on the fly can

take advantage of the interlacing and even then the image has to be exactly

the correct size for the output device, because scaling an image requires

adjacent pixels and these are not available until all the passes have been

read.

 

If you do write an interlaced image you will hardly ever need to handle

the interlacing yourself.  Call png_set_interlace_handling() and use the

approach described above.

 

The only time it is conceivable that you will really need to write an

interlaced image pass-by-pass is when you have read one pass by pass and

made some pixel-by-pixel transformation to it, as described in the read

code above.  In this case use the PNG_PASS_ROWS and PNG_PASS_COLS macros

to determine the size of each sub-image in turn and simply write the rows

you obtained from the read code.

 

Finishing a sequential write

 

After you are finished writing the image, you should finish writing

the file.  If you are interested in writing comments or time, you should

pass an appropriately filled png_info pointer.  If you are not interested,

you can pass NULL.

 

    png_write_end(png_ptr, info_ptr);

 

When you are done, you can free all memory used by libpng like this:

 

    png_destroy_write_struct(&png_ptr, &info_ptr);

 

It is also possible to individually free the info_ptr members that

point to libpng-allocated storage with the following function:

 

    png_free_data(png_ptr, info_ptr, mask, seq)

 

    mask  - identifies data to be freed, a mask

            containing the bitwise OR of one or

            more of

              PNG_FREE_PLTE, PNG_FREE_TRNS,

              PNG_FREE_HIST, PNG_FREE_ICCP,

              PNG_FREE_PCAL, PNG_FREE_ROWS,

              PNG_FREE_SCAL, PNG_FREE_SPLT,

              PNG_FREE_TEXT, PNG_FREE_UNKN,

            or simply PNG_FREE_ALL

 

    seq   - sequence number of item to be freed

            (-1 for all items)

 

This function may be safely called when the relevant storage has

already been freed, or has not yet been allocated, or was allocated

by the user  and not by libpng,  and will in those cases do nothing.

The "seq" parameter is ignored if only one item of the selected data

type, such as PLTE, is allowed.  If "seq" is not -1, and multiple items

are allowed for the data type identified in the mask, such as text or

sPLT, only the n'th item in the structure is freed, where n is "seq".

 

If you allocated data such as a palette that you passed in to libpng

with png_set_*, you must not free it until just before the call to

png_destroy_write_struct().

 

The default behavior is only to free data that was allocated internally

by libpng.  This can be changed, so that libpng will not free the data,

or so that it will free data that was allocated by the user with png_malloc()

or png_zalloc() and passed in via a png_set_*() function, with

 

    png_data_freer(png_ptr, info_ptr, freer, mask)

 

    freer  - one of

               PNG_DESTROY_WILL_FREE_DATA

               PNG_SET_WILL_FREE_DATA

               PNG_USER_WILL_FREE_DATA

 

    mask   - which data elements are affected

             same choices as in png_free_data()

 

For example, to transfer responsibility for some data from a read structure

to a write structure, you could use

 

    png_data_freer(read_ptr, read_info_ptr,

       PNG_USER_WILL_FREE_DATA,

       PNG_FREE_PLTE|PNG_FREE_tRNS|PNG_FREE_hIST)

 

    png_data_freer(write_ptr, write_info_ptr,

       PNG_DESTROY_WILL_FREE_DATA,

       PNG_FREE_PLTE|PNG_FREE_tRNS|PNG_FREE_hIST)

 

thereby briefly reassigning responsibility for freeing to the user but

immediately afterwards reassigning it once more to the write_destroy

function.  Having done this, it would then be safe to destroy the read

structure and continue to use the PLTE, tRNS, and hIST data in the write

structure.

 

This function only affects data that has already been allocated.

You can call this function before calling after the png_set_*() functions

to control whether the user or png_destroy_*() is supposed to free the data.

When the user assumes responsibility for libpng-allocated data, the

application must use

png_free() to free it, and when the user transfers responsibility to libpng

for data that the user has allocated, the user must have used png_malloc()

or png_zalloc() to allocate it.

 

If you allocated text_ptr.text, text_ptr.lang, and text_ptr.translated_keyword

separately, do not transfer responsibility for freeing text_ptr to libpng,

because when libpng fills a png_text structure it combines these members with

the key member, and png_free_data() will free only text_ptr.key.  Similarly,

if you transfer responsibility for free'ing text_ptr from libpng to your

application, your application must not separately free those members.

For a more compact example of writing a PNG image, see the file example.c.

 

V. Modifying/Customizing libpng:

 

There are two issues here.  The first is changing how libpng does

standard things like memory allocation, input/output, and error handling.

The second deals with more complicated things like adding new chunks,

adding new transformations, and generally changing how libpng works.

Both of those are compile-time issues; that is, they are generally

determined at the time the code is written, and there is rarely a need

to provide the user with a means of changing them.

 

Memory allocation, input/output, and error handling

 

All of the memory allocation, input/output, and error handling in libpng

goes through callbacks that are user-settable.  The default routines are

in pngmem.c, pngrio.c, pngwio.c, and pngerror.c, respectively.  To change

these functions, call the appropriate png_set_*_fn() function.

 

Memory allocation is done through the functions png_malloc(), png_calloc(),

and png_free().  These currently just call the standard C functions.

png_calloc() calls png_malloc() and then clears the newly

allocated memory to zero.  There is limited support for certain systems

with segmented memory architectures and the types of pointers declared by

png.h match this; you will have to use appropriate pointers in your

application.  Since it is

unlikely that the method of handling memory allocation on a platform

will change between applications, these functions must be modified in

the library at compile time.  If you prefer to use a different method

of allocating and freeing data, you can use png_create_read_struct_2() or

png_create_write_struct_2() to register your own functions as described

above.  These functions also provide a void pointer that can be retrieved

via

 

    mem_ptr=png_get_mem_ptr(png_ptr);

 

Your replacement memory functions must have prototypes as follows:

 

    png_voidp malloc_fn(png_structp png_ptr,

       png_alloc_size_t size);

 

    void free_fn(png_structp png_ptr, png_voidp ptr);

 

Your malloc_fn() must return NULL in case of failure.  The png_malloc()

function will normally call png_error() if it receives a NULL from the

system memory allocator or from your replacement malloc_fn().

 

Your free_fn() will never be called with a NULL ptr, since libpng's

png_free() checks for NULL before calling free_fn().

 

Input/Output in libpng is done through png_read() and png_write(),

which currently just call fread() and fwrite().  The FILE * is stored in

png_struct and is initialized via png_init_io().  If you wish to change

the method of I/O, the library supplies callbacks that you can set

through the function png_set_read_fn() and png_set_write_fn() at run

time, instead of calling the png_init_io() function.  These functions

also provide a void pointer that can be retrieved via the function

png_get_io_ptr().  For example:

 

    png_set_read_fn(png_structp read_ptr,

        voidp read_io_ptr, png_rw_ptr read_data_fn)

 

    png_set_write_fn(png_structp write_ptr,

        voidp write_io_ptr, png_rw_ptr write_data_fn,

        png_flush_ptr output_flush_fn);

 

    voidp read_io_ptr = png_get_io_ptr(read_ptr);

    voidp write_io_ptr = png_get_io_ptr(write_ptr);

 

The replacement I/O functions must have prototypes as follows:

 

    void user_read_data(png_structp png_ptr,

        png_bytep data, png_size_t length);

 

    void user_write_data(png_structp png_ptr,

        png_bytep data, png_size_t length);

 

    void user_flush_data(png_structp png_ptr);

 

The user_read_data() function is responsible for detecting and

handling end-of-data errors.

 

Supplying NULL for the read, write, or flush functions sets them back

to using the default C stream functions, which expect the io_ptr to

point to a standard *FILE structure.  It is probably a mistake

to use NULL for one of write_data_fn and output_flush_fn but not both

of them, unless you have built libpng with PNG_NO_WRITE_FLUSH defined.

It is an error to read from a write stream, and vice versa.

 

Error handling in libpng is done through png_error() and png_warning().

Errors handled through png_error() are fatal, meaning that png_error()

should never return to its caller.  Currently, this is handled via

setjmp() and longjmp() (unless you have compiled libpng with

PNG_NO_SETJMP, in which case it is handled via PNG_ABORT()),

but you could change this to do things like exit() if you should wish,

as long as your function does not return.

 

On non-fatal errors, png_warning() is called

to print a warning message, and then control returns to the calling code.

By default png_error() and png_warning() print a message on stderr via

fprintf() unless the library is compiled with PNG_NO_CONSOLE_IO defined

(because you don't want the messages) or PNG_NO_STDIO defined (because

fprintf() isn't available).  If you wish to change the behavior of the error

functions, you will need to set up your own message callbacks.  These

functions are normally supplied at the time that the png_struct is created.

It is also possible to redirect errors and warnings to your own replacement

functions after png_create_*_struct() has been called by calling:

 

    png_set_error_fn(png_structp png_ptr,

        png_voidp error_ptr, png_error_ptr error_fn,

        png_error_ptr warning_fn);

 

    png_voidp error_ptr = png_get_error_ptr(png_ptr);

 

If NULL is supplied for either error_fn or warning_fn, then the libpng

default function will be used, calling fprintf() and/or longjmp() if a

problem is encountered.  The replacement error functions should have

parameters as follows:

 

    void user_error_fn(png_structp png_ptr,

        png_const_charp error_msg);

 

    void user_warning_fn(png_structp png_ptr,

        png_const_charp warning_msg);

 

The motivation behind using setjmp() and longjmp() is the C++ throw and

catch exception handling methods.  This makes the code much easier to write,

as there is no need to check every return code of every function call.

However, there are some uncertainties about the status of local variables

after a longjmp, so the user may want to be careful about doing anything

after setjmp returns non-zero besides returning itself.  Consult your

compiler documentation for more details.  For an alternative approach, you

may wish to use the "cexcept" facility (see http://cexcept.sourceforge.net).

 

Custom chunks

 

If you need to read or write custom chunks, you may need to get deeper

into the libpng code.  The library now has mechanisms for storing

and writing chunks of unknown type; you can even declare callbacks

for custom chunks.  However, this may not be good enough if the

library code itself needs to know about interactions between your

chunk and existing `intrinsic' chunks.

 

If you need to write a new intrinsic chunk, first read the PNG

specification. Acquire a first level of understanding of how it works.

Pay particular attention to the sections that describe chunk names,

and look at how other chunks were designed, so you can do things

similarly.  Second, check out the sections of libpng that read and

write chunks.  Try to find a chunk that is similar to yours and use

it as a template.  More details can be found in the comments inside

the code.  It is best to handle unknown chunks in a generic method,

via callback functions, instead of by modifying libpng functions.

 

If you wish to write your own transformation for the data, look through

the part of the code that does the transformations, and check out some of

the simpler ones to get an idea of how they work.  Try to find a similar

transformation to the one you want to add and copy off of it.  More details

can be found in the comments inside the code itself.

 

Configuring for 16 bit platforms

 

You will want to look into zconf.h to tell zlib (and thus libpng) that

it cannot allocate more then 64K at a time.  Even if you can, the memory

won't be accessible.  So limit zlib and libpng to 64K by defining MAXSEG_64K.

 

Configuring for DOS

 

For DOS users who only have access to the lower 640K, you will

have to limit zlib's memory usage via a png_set_compression_mem_level()

call.  See zlib.h or zconf.h in the zlib library for more information.

 

Configuring for Medium Model

 

Libpng's support for medium model has been tested on most of the popular

compilers.  Make sure MAXSEG_64K gets defined, USE_FAR_KEYWORD gets

defined, and FAR gets defined to far in pngconf.h, and you should be

all set.  Everything in the library (except for zlib's structure) is

expecting far data.  You must use the typedefs with the p or pp on

the end for pointers (or at least look at them and be careful).  Make

note that the rows of data are defined as png_bytepp, which is an

unsigned char far * far *.

 

Configuring for gui/windowing platforms:

 

You will need to write new error and warning functions that use the GUI

interface, as described previously, and set them to be the error and

warning functions at the time that png_create_*_struct() is called,

in order to have them available during the structure initialization.

They can be changed later via png_set_error_fn().  On some compilers,

you may also have to change the memory allocators (png_malloc, etc.).

 

Configuring for compiler xxx:

 

All includes for libpng are in pngconf.h.  If you need to add, change

or delete an include, this is the place to do it.

The includes that are not needed outside libpng are placed in pngpriv.h,

which is only used by the routines inside libpng itself.

The files in libpng proper only include pngpriv.h and png.h, which

in turn includes pngconf.h.

 

Configuring zlib:

 

There are special functions to configure the compression.  Perhaps the

most useful one changes the compression level, which currently uses

input compression values in the range 0 - 9.  The library normally

uses the default compression level (Z_DEFAULT_COMPRESSION = 6).  Tests

have shown that for a large majority of images, compression values in

the range 3-6 compress nearly as well as higher levels, and do so much

faster.  For online applications it may be desirable to have maximum speed

(Z_BEST_SPEED = 1).  With versions of zlib after v0.99, you can also

specify no compression (Z_NO_COMPRESSION = 0), but this would create

files larger than just storing the raw bitmap.  You can specify the

compression level by calling:

 

    png_set_compression_level(png_ptr, level);

 

Another useful one is to reduce the memory level used by the library.

The memory level defaults to 8, but it can be lowered if you are

short on memory (running DOS, for example, where you only have 640K).

Note that the memory level does have an effect on compression; among

other things, lower levels will result in sections of incompressible

data being emitted in smaller stored blocks, with a correspondingly

larger relative overhead of up to 15% in the worst case.

 

    png_set_compression_mem_level(png_ptr, level);

 

The other functions are for configuring zlib.  They are not recommended

for normal use and may result in writing an invalid PNG file.  See

zlib.h for more information on what these mean.

 

    png_set_compression_strategy(png_ptr,

        strategy);

 

    png_set_compression_window_bits(png_ptr,

        window_bits);

 

    png_set_compression_method(png_ptr, method);

    png_set_compression_buffer_size(png_ptr, size);

 

Controlling row filtering

 

If you want to control whether libpng uses filtering or not, which

filters are used, and how it goes about picking row filters, you

can call one of these functions.  The selection and configuration

of row filters can have a significant impact on the size and

encoding speed and a somewhat lesser impact on the decoding speed

of an image.  Filtering is enabled by default for RGB and grayscale

images (with and without alpha), but not for paletted images nor

for any images with bit depths less than 8 bits/pixel.

 

The 'method' parameter sets the main filtering method, which is

currently only '0' in the PNG 1.2 specification.  The 'filters'

parameter sets which filter(s), if any, should be used for each

scanline.  Possible values are PNG_ALL_FILTERS and PNG_NO_FILTERS

to turn filtering on and off, respectively.

 

Individual filter types are PNG_FILTER_NONE, PNG_FILTER_SUB,

PNG_FILTER_UP, PNG_FILTER_AVG, PNG_FILTER_PAETH, which can be bitwise

ORed together with '|' to specify one or more filters to use.

These filters are described in more detail in the PNG specification.

If you intend to change the filter type during the course of writing

the image, you should start with flags set for all of the filters

you intend to use so that libpng can initialize its internal

structures appropriately for all of the filter types.  (Note that this

means the first row must always be adaptively filtered, because libpng

currently does not allocate the filter buffers until png_write_row()

is called for the first time.)

 

    filters = PNG_FILTER_NONE | PNG_FILTER_SUB

              PNG_FILTER_UP | PNG_FILTER_AVG |

              PNG_FILTER_PAETH | PNG_ALL_FILTERS;

 

    png_set_filter(png_ptr, PNG_FILTER_TYPE_BASE,

       filters);

              The second parameter can also be

              PNG_INTRAPIXEL_DIFFERENCING if you are

              writing a PNG to be embedded in a MNG

              datastream.  This parameter must be the

              same as the value of filter_method used

              in png_set_IHDR().

 

It is also possible to influence how libpng chooses from among the

available filters.  This is done in one or both of two ways - by

telling it how important it is to keep the same filter for successive

rows, and by telling it the relative computational costs of the filters.

 

    double weights[3] = {1.5, 1.3, 1.1},

       costs[PNG_FILTER_VALUE_LAST] =

       {1.0, 1.3, 1.3, 1.5, 1.7};

 

    png_set_filter_heuristics(png_ptr,

       PNG_FILTER_HEURISTIC_WEIGHTED, 3,

       weights, costs);

 

The weights are multiplying factors that indicate to libpng that the

row filter should be the same for successive rows unless another row filter

is that many times better than the previous filter.  In the above example,

if the previous 3 filters were SUB, SUB, NONE, the SUB filter could have a

"sum of absolute differences" 1.5 x 1.3 times higher than other filters

and still be chosen, while the NONE filter could have a sum 1.1 times

higher than other filters and still be chosen.  Unspecified weights are

taken to be 1.0, and the specified weights should probably be declining

like those above in order to emphasize recent filters over older filters.

 

The filter costs specify for each filter type a relative decoding cost

to be considered when selecting row filters.  This means that filters

with higher costs are less likely to be chosen over filters with lower

costs, unless their "sum of absolute differences" is that much smaller.

The costs do not necessarily reflect the exact computational speeds of

the various filters, since this would unduly influence the final image

size.

 

Note that the numbers above were invented purely for this example and

are given only to help explain the function usage.  Little testing has

been done to find optimum values for either the costs or the weights.

 

Removing unwanted object code

 

There are a bunch of #define's in pngconf.h that control what parts of

libpng are compiled.  All the defines end in _SUPPORTED.  If you are

never going to use a capability, you can change the #define to #undef

before recompiling libpng and save yourself code and data space, or

you can turn off individual capabilities with defines that begin with

PNG_NO_.

 

In libpng-1.5.0 and later, the #define's are in pnglibconf.h instead.

 

You can also turn all of the transforms and ancillary chunk capabilities

off en masse with compiler directives that define

PNG_NO_READ[or WRITE]_TRANSFORMS, or PNG_NO_READ[or WRITE]_ANCILLARY_CHUNKS,

or all four,

along with directives to turn on any of the capabilities that you do

want.  The PNG_NO_READ[or WRITE]_TRANSFORMS directives disable the extra

transformations but still leave the library fully capable of reading

and writing PNG files with all known public chunks. Use of the

PNG_NO_READ[or WRITE]_ANCILLARY_CHUNKS directive produces a library

that is incapable of reading or writing ancillary chunks.  If you are

not using the progressive reading capability, you can turn that off

with PNG_NO_PROGRESSIVE_READ (don't confuse this with the INTERLACING

capability, which you'll still have).

 

All the reading and writing specific code are in separate files, so the

linker should only grab the files it needs.  However, if you want to

make sure, or if you are building a stand alone library, all the

reading files start with pngr and all the writing files start with

pngw.  The files that don't match either (like png.c, pngtrans.c, etc.)

are used for both reading and writing, and always need to be included.

The progressive reader is in pngpread.c

 

If you are creating or distributing a dynamically linked library (a .so

or DLL file), you should not remove or disable any parts of the library,

as this will cause applications linked with different versions of the

library to fail if they call functions not available in your library.

The size of the library itself should not be an issue, because only

those sections that are actually used will be loaded into memory.

 

Requesting debug printout

 

The macro definition PNG_DEBUG can be used to request debugging

printout.  Set it to an integer value in the range 0 to 3.  Higher

numbers result in increasing amounts of debugging information.  The

information is printed to the "stderr" file, unless another file

name is specified in the PNG_DEBUG_FILE macro definition.

 

When PNG_DEBUG > 0, the following functions (macros) become available:

 

   png_debug(level, message)

   png_debug1(level, message, p1)

   png_debug2(level, message, p1, p2)

 

in which "level" is compared to PNG_DEBUG to decide whether to print

the message, "message" is the formatted string to be printed,

and p1 and p2 are parameters that are to be embedded in the string

according to printf-style formatting directives.  For example,

 

   png_debug1(2, "foo=%d\n", foo);

 

is expanded to

 

   if (PNG_DEBUG > 2)

      fprintf(PNG_DEBUG_FILE, "foo=%d\n", foo);

 

When PNG_DEBUG is defined but is zero, the macros aren't defined, but you

can still use PNG_DEBUG to control your own debugging:

 

   #ifdef PNG_DEBUG

       fprintf(stderr, ...

   #endif

 

When PNG_DEBUG = 1, the macros are defined, but only png_debug statements

having level = 0 will be printed.  There aren't any such statements in

this version of libpng, but if you insert some they will be printed.

 

VI.  MNG support

 

The MNG specification (available at http://www.libpng.org/pub/mng) allows

certain extensions to PNG for PNG images that are embedded in MNG datastreams.

Libpng can support some of these extensions.  To enable them, use the

png_permit_mng_features() function:

 

   feature_set = png_permit_mng_features(png_ptr, mask)

 

   mask is a png_uint_32 containing the bitwise OR of the

        features you want to enable.  These include

        PNG_FLAG_MNG_EMPTY_PLTE

        PNG_FLAG_MNG_FILTER_64

        PNG_ALL_MNG_FEATURES

 

   feature_set is a png_uint_32 that is the bitwise AND of

      your mask with the set of MNG features that is

      supported by the version of libpng that you are using.

 

It is an error to use this function when reading or writing a standalone

PNG file with the PNG 8-byte signature.  The PNG datastream must be wrapped

in a MNG datastream.  As a minimum, it must have the MNG 8-byte signature

and the MHDR and MEND chunks.  Libpng does not provide support for these

or any other MNG chunks; your application must provide its own support for

them.  You may wish to consider using libmng (available at

http://www.libmng.com) instead.

 

VII.  Changes to Libpng from version 0.88

 

It should be noted that versions of libpng later than 0.96 are not

distributed by the original libpng author, Guy Schalnat, nor by

Andreas Dilger, who had taken over from Guy during 1996 and 1997, and

distributed versions 0.89 through 0.96, but rather by another member

of the original PNG Group, Glenn Randers-Pehrson.  Guy and Andreas are

still alive and well, but they have moved on to other things.

 

The old libpng functions png_read_init(), png_write_init(),

png_info_init(), png_read_destroy(), and png_write_destroy() have been

moved to PNG_INTERNAL in version 0.95 to discourage their use.  These

functions will be removed from libpng version 1.4.0.

 

The preferred method of creating and initializing the libpng structures is

via the png_create_read_struct(), png_create_write_struct(), and

png_create_info_struct() because they isolate the size of the structures

from the application, allow version error checking, and also allow the

use of custom error handling routines during the initialization, which

the old functions do not.  The functions png_read_destroy() and

png_write_destroy() do not actually free the memory that libpng

allocated for these structs, but just reset the data structures, so they

can be used instead of png_destroy_read_struct() and

png_destroy_write_struct() if you feel there is too much system overhead

allocating and freeing the png_struct for each image read.

 

Setting the error callbacks via png_set_message_fn() before

png_read_init() as was suggested in libpng-0.88 is no longer supported

because this caused applications that do not use custom error functions

to fail if the png_ptr was not initialized to zero.  It is still possible

to set the error callbacks AFTER png_read_init(), or to change them with

png_set_error_fn(), which is essentially the same function, but with a new

name to force compilation errors with applications that try to use the old

method.

 

Starting with version 1.0.7, you can find out which version of the library

you are using at run-time:

 

   png_uint_32 libpng_vn = png_access_version_number();

 

The number libpng_vn is constructed from the major version, minor

version with leading zero, and release number with leading zero,

(e.g., libpng_vn for version 1.0.7 is 10007).

 

You can also check which version of png.h you used when compiling your

application:

 

   png_uint_32 application_vn = PNG_LIBPNG_VER;

 

VIII.  Changes to Libpng from version 1.0.x to 1.2.x

 

Support for user memory management was enabled by default.  To

accomplish this, the functions png_create_read_struct_2(),

png_create_write_struct_2(), png_set_mem_fn(), png_get_mem_ptr(),

png_malloc_default(), and png_free_default() were added.

 

Support for the iTXt chunk has been enabled by default as of

version 1.2.41.

 

Support for certain MNG features was enabled.

 

Support for numbered error messages was added.  However, we never got

around to actually numbering the error messages.  The function

png_set_strip_error_numbers() was added (Note: the prototype for this

function was inadvertently removed from png.h in PNG_NO_ASSEMBLER_CODE

builds of libpng-1.2.15.  It was restored in libpng-1.2.36).

 

The png_malloc_warn() function was added at libpng-1.2.3.  This issues

a png_warning and returns NULL instead of aborting when it fails to

acquire the requested memory allocation.

 

Support for setting user limits on image width and height was enabled

by default.  The functions png_set_user_limits(), png_get_user_width_max(),

and png_get_user_height_max() were added at libpng-1.2.6.

 

The png_set_add_alpha() function was added at libpng-1.2.7.

 

The function png_set_expand_gray_1_2_4_to_8() was added at libpng-1.2.9.

Unlike png_set_gray_1_2_4_to_8(), the new function does not expand the

tRNS chunk to alpha. The png_set_gray_1_2_4_to_8() function is

deprecated.

 

A number of macro definitions in support of runtime selection of

assembler code features (especially Intel MMX code support) were

added at libpng-1.2.0:

 

    PNG_ASM_FLAG_MMX_SUPPORT_COMPILED

    PNG_ASM_FLAG_MMX_SUPPORT_IN_CPU

    PNG_ASM_FLAG_MMX_READ_COMBINE_ROW

    PNG_ASM_FLAG_MMX_READ_INTERLACE

    PNG_ASM_FLAG_MMX_READ_FILTER_SUB

    PNG_ASM_FLAG_MMX_READ_FILTER_UP

    PNG_ASM_FLAG_MMX_READ_FILTER_AVG

    PNG_ASM_FLAG_MMX_READ_FILTER_PAETH

    PNG_ASM_FLAGS_INITIALIZED

    PNG_MMX_READ_FLAGS

    PNG_MMX_FLAGS

    PNG_MMX_WRITE_FLAGS

    PNG_MMX_FLAGS

 

We added the following functions in support of runtime

selection of assembler code features:

 

    png_get_mmx_flagmask()

    png_set_mmx_thresholds()

    png_get_asm_flags()

    png_get_mmx_bitdepth_threshold()

    png_get_mmx_rowbytes_threshold()

    png_set_asm_flags()

 

We replaced all of these functions with simple stubs in libpng-1.2.20,

when the Intel assembler code was removed due to a licensing issue.

 

These macros are deprecated:

 

    PNG_READ_TRANSFORMS_NOT_SUPPORTED

    PNG_PROGRESSIVE_READ_NOT_SUPPORTED

    PNG_NO_SEQUENTIAL_READ_SUPPORTED

    PNG_WRITE_TRANSFORMS_NOT_SUPPORTED

    PNG_READ_ANCILLARY_CHUNKS_NOT_SUPPORTED

    PNG_WRITE_ANCILLARY_CHUNKS_NOT_SUPPORTED

 

They have been replaced, respectively, by:

 

    PNG_NO_READ_TRANSFORMS

    PNG_NO_PROGRESSIVE_READ

    PNG_NO_SEQUENTIAL_READ

    PNG_NO_WRITE_TRANSFORMS

    PNG_NO_READ_ANCILLARY_CHUNKS

    PNG_NO_WRITE_ANCILLARY_CHUNKS

 

PNG_MAX_UINT was replaced with PNG_UINT_31_MAX.  It has been

deprecated since libpng-1.0.16 and libpng-1.2.6.

 

The function

    png_check_sig(sig, num)

was replaced with

    !png_sig_cmp(sig, 0, num)

It has been deprecated since libpng-0.90.

 

The function

    png_set_gray_1_2_4_to_8()

which also expands tRNS to alpha was replaced with

    png_set_expand_gray_1_2_4_to_8()

which does not. It has been deprecated since libpng-1.0.18 and 1.2.9.

 

IX.  Changes to Libpng from version 1.0.x/1.2.x to 1.4.x

 

Private libpng prototypes and macro definitions were moved from

png.h and pngconf.h into a new pngpriv.h header file.

 

Functions png_set_benign_errors(), png_benign_error(), and

png_chunk_benign_error() were added.

 

Support for setting the maximum amount of memory that the application

will allocate for reading chunks was added, as a security measure.

The functions png_set_chunk_cache_max() and png_get_chunk_cache_max()

were added to the library.

 

We implemented support for I/O states by adding png_ptr member io_state

and functions png_get_io_chunk_name() and png_get_io_state() in pngget.c

 

We added PNG_TRANSFORM_GRAY_TO_RGB to the available high-level

input transforms.

 

Checking for and reporting of errors in the IHDR chunk is more thorough.

 

Support for global arrays was removed, to improve thread safety.

 

Some obsolete/deprecated macros and functions have been removed.

 

Typecasted NULL definitions such as

   #define png_voidp_NULL            (png_voidp)NULL

were eliminated.  If you used these in your application, just use

NULL instead.

 

The png_struct and info_struct members "trans" and "trans_values" were

changed to "trans_alpha" and "trans_color", respectively.

 

The obsolete, unused pnggccrd.c and pngvcrd.c files and related makefiles

were removed.

 

The PNG_1_0_X and PNG_1_2_X macros were eliminated.

 

The PNG_LEGACY_SUPPORTED macro was eliminated.

 

Many WIN32_WCE #ifdefs were removed.

 

The functions png_read_init(info_ptr), png_write_init(info_ptr),

png_info_init(info_ptr), png_read_destroy(), and png_write_destroy()

have been removed.  They have been deprecated since libpng-0.95.

 

The png_permit_empty_plte() was removed. It has been deprecated

since libpng-1.0.9.  Use png_permit_mng_features() instead.

 

We removed the obsolete stub functions png_get_mmx_flagmask(),

png_set_mmx_thresholds(), png_get_asm_flags(),

png_get_mmx_bitdepth_threshold(), png_get_mmx_rowbytes_threshold(),

png_set_asm_flags(), and png_mmx_supported()

 

We removed the obsolete png_check_sig(), png_memcpy_check(), and

png_memset_check() functions.  Instead use !png_sig_cmp(), memcpy(),

and memset(), respectively.

 

The function png_set_gray_1_2_4_to_8() was removed. It has been

deprecated since libpng-1.0.18 and 1.2.9, when it was replaced with

png_set_expand_gray_1_2_4_to_8() because the former function also

expanded palette images.

 

Macros for png_get_uint_16, png_get_uint_32, and png_get_int_32

were added and are used by default instead of the corresponding

functions. Unfortunately,

from libpng-1.4.0 until 1.4.4, the png_get_uint_16 macro (but not the  

function) incorrectly returned a value of type png_uint_32.

 

We changed the prototype for png_malloc() from

    png_malloc(png_structp png_ptr, png_uint_32 size)

to

    png_malloc(png_structp png_ptr, png_alloc_size_t size)

 

This also applies to the prototype for the user replacement malloc_fn().

 

The png_calloc() function was added and is used in place of

of "png_malloc(); memset();" except in the case in png_read_png()

where the array consists of pointers; in this case a "for" loop is used

after the png_malloc() to set the pointers to NULL, to give robust.

behavior in case the application runs out of memory part-way through

the process.

 

We changed the prototypes of png_get_compression_buffer_size() and

png_set_compression_buffer_size() to work with png_size_t instead of

png_uint_32.

 

Support for numbered error messages was removed by default, since we

never got around to actually numbering the error messages. The function

png_set_strip_error_numbers() was removed from the library by default.

 

The png_zalloc() and png_zfree() functions are no longer exported.

The png_zalloc() function no longer zeroes out the memory that it

allocates.

 

Support for dithering was disabled by default in libpng-1.4.0, because

been well tested and doesn't actually "dither".  The code was not

removed, however, and could be enabled by building libpng with

PNG_READ_DITHER_SUPPORTED defined.  In libpng-1.4.2, this support

was reenabled, but the function was renamed png_set_quantize() to

reflect more accurately what it actually does.  At the same time,

the PNG_DITHER_[RED,GREEN_BLUE]_BITS macros were also renamed to

PNG_QUANTIZE_[RED,GREEN,BLUE]_BITS, and PNG_READ_DITHER_SUPPORTED

was renamed to PNG_READ_QUANTIZE_SUPPORTED.

 

We removed the trailing '.' from the warning and error messages.

 

X.  Changes to Libpng from version 1.4.x to 1.5.x

 

From libpng-1.4.0 until 1.4.4, the png_get_uint_16 macro (but not the  

function) incorrectly returned a value of type png_uint_32.

 

A. Changes that affect users of libpng

 

There are no substantial API changes between the non-deprecated parts of

the 1.4.5 API and the 1.5.0 API, however the ability to directly access

the main libpng control structures, png_struct and png_info, deprecated

in earlier versions of libpng, has been completely removed from

libpng 1.5.

 

We no longer include zlib.h in png.h.  Applications that need access

to information in zlib.h will need to add the '#include "zlib.h"'

directive.  It does not matter whether it is placed prior to or after

the '"#include png.h"' directive.

 

We moved the png_strcpy(), png_strncpy(), png_strlen(), png_memcpy(),

png_memcmp(), png_sprintf, and png_memcpy() macros into a private

header file (pngpriv.h) that is not accessible to applications.

 

In png_get_iCCP, the type of "profile" was changed from png_charpp

to png_bytepp, and in png_set_iCCP, from png_charp to png_const_bytep.

 

There are changes of form in png.h, including new and changed macros to

declare

parts of the API.  Some API functions with arguments that are pointers to

data not modified within the function have been corrected to declare

these arguments with PNG_CONST.

 

Much of the internal use of C macros to control the library build has also

changed and some of this is visible in the exported header files, in

particular the use of macros to control data and API elements visible

during application compilation may require significant revision to

application code.  (It is extremely rare for an application to do this.)

 

Any program that compiled against libpng 1.4 and did not use deprecated

features or access internal library structures should compile and work

against libpng 1.5.

 

libpng 1.5.0 adds PNG_ PASS macros to help in the reading and writing of

interlaced images.  The macros return the number of rows and columns in

each pass and information that can be used to de-interlace and (if

absolutely necessary) interlace an image.

 

libpng 1.5.0 adds an API png_longjmp(png_ptr, value).  This API calls

the application provided png_longjmp_ptr on the internal, but application

initialized, jmpbuf.  It is provided as a convenience to avoid the need

to use the png_jmpbuf macro, which had the unnecessary side effect of

resetting the internal png_longjmp_ptr value.

 

libpng 1.5.0 includes a complete fixed point API.  By default this is

present along with the corresponding floating point API.  In general the

fixed point API is faster and smaller than the floating point one because

the PNG file format used fixed point, not floating point.  This applies

even if the library uses floating point in internal calculations.  A new

macro, PNG_FLOATING_ARITHMETIC_SUPPORTED, reveals whether the library

uses floating point arithmetic (the default) or fixed point arithmetic

internally for performance critical calculations such as gamma correction.

In some cases, the gamma calculations may produce slightly different

results.  This has changed the results in png_rgb_to_gray and in alpha

composition (png_set_background for example). This applies even if the

original image was already linear (gamma == 1.0) and, therefore, it is

not necessary to linearize the image.  This is because libpng has *not*

been changed to optimize that case correctly, yet.

 

Fixed point support for the sCAL chunk comes with an important caveat;

the sCAL specification uses a decimal encoding of floating point values

and the accuracy of PNG fixed point values is insufficient for

representation of these values. Consequently a "string" API

(png_get_sCAL_s and png_set_sCAL_s) is the only reliable way of reading

arbitrary sCAL chunks in the absence of either the floating point API or

internal floating point calculations.

 

Applications no longer need to include the optional distribution header

file pngusr.h or define the corresponding macros during application

build in order to see the correct variant of the libpng API.  From 1.5.0

application code can check for the corresponding _SUPPORTED macro:

 

#ifdef PNG_INCH_CONVERSIONS_SUPPORTED

   /* code that uses the inch conversion APIs. */

#endif

 

This macro will only be defined if the inch conversion functions have been

compiled into libpng.  The full set of macros, and whether or not support

has been compiled in, are available in the header file pnglibconf.h.

This header file is specific to the libpng build.  Notice that prior to

1.5.0 the _SUPPORTED macros would always have the default definition unless

reset by pngusr.h or by explicit settings on the compiler command line.

These settings may produce compiler warnings or errors in 1.5.0 because

of macro redefinition.

 

From libpng-1.4.0 until 1.4.4, the png_get_uint_16 macro (but not the  

function) incorrectly returned a value of type png_uint_32.  libpng 1.5.0

is consistent with the implementation in 1.4.5 and 1.2.x (where the macro

did not exist.)

 

Applications can now choose whether to use these macros or to call the

corresponding function by defining PNG_USE_READ_MACROS or

PNG_NO_USE_READ_MACROS before including png.h.  Notice that this is

only supported from 1.5.0 -defining PNG_NO_USE_READ_MACROS prior to 1.5.0

 will lead to a link failure.

 

B. Changes to the build and configuration of libpng

 

Details of internal changes to the library code can be found in the CHANGES

file.  These will be of no concern to the vast majority of library users or

builders, however the few who configure libpng to a non-default feature

set may need to change how this is done.

 

There should be no need for library builders to alter build scripts if

these use the distributed build support - configure or the makefiles -

however users of the makefiles may care to update their build scripts

to build pnglibconf.h where the corresponding makefile does not do so.

 

Building libpng with a non-default configuration has changed completely.

The old method using pngusr.h should still work correctly even though the

way pngusr.h is used in the build has been changed, however library

builders will probably want to examine the changes to take advantage of

new capabilities and to simplify their build system.

 

B.1 Specific changes to library configuration capabilities

 

The library now supports a complete fixed point implementation and can

thus be used on systems which have no floating point support or very

limited or slow support.  Previously gamma correction, an essential part

of complete PNG support, required reasonably fast floating point.

 

As part of this the choice of internal implementation has been made

independent of the choice of fixed versus floating point APIs and all the

missing fixed point APIs have been implemented.

 

The exact mechanism used to control attributes of API functions has

changed.  A single set of operating system independent macro definitions

is used and operating system specific directives are defined in

pnglibconf.h

 

As part of this the mechanism used to choose procedure call standards on

those systems that allow a choice has been changed.  At present this only

affects certain Microsoft (DOS, Windows) and IBM (OS/2) operating systems

running on Intel processors.  As before PNGAPI is defined where required

to control the exported API functions; however, two new macros, PNGCBAPI

and PNGCAPI, are used instead for callback functions (PNGCBAPI) and

(PNGCAPI) for functions that must match a C library prototype (currently

only png_longjmp_ptr, which must match the C longjmp function.)  The new

approach is documented in pngconf.h

 

Despite these changes libpng 1.5.0 only supports the native C function

calling standard on those platforms tested so far (__cdecl on Microsoft

Windows).  This is because the support requirements for alternative

calling conventions seem to no longer exist.  Developers who find it

necessary to set PNG_API_RULE to 1 should advise the mailing list

(png-mng-implement) of this and library builders who use Openwatcom and

therefore set PNG_API_RULE to 2 should also contact the mailing list.

 

A new test program, pngvalid, is provided in addition to pngtest.

pngvalid validates the arithmetic accuracy of the gamma correction

calculations and includes a number of validations of the file format.

A subset of the full range of tests is run when "make check" is done

(in the 'configure' build.)  pngvalid also allows total allocated memory

usage to be evaluated and performs additional memory overwrite validation.

 

Many changes to individual feature macros have been made. The following

are the changes most likely to be noticed by library builders who

configure libpng:

 

1) All feature macros now have consistent naming:

 

#define PNG_NO_feature turns the feature off

#define PNG_feature_SUPPORTED turns the feature on

 

pnglibconf.h contains one line for each feature macro which is either:

 

#define PNG_feature_SUPPORTED

 

if the feature is supported or:

 

/*#undef PNG_feature_SUPPORTED*/

 

if it is not.  Library code consistently checks for the 'SUPPORTED' macro.

It does not, and should not, check for the 'NO' macro which will not

normally be defined even if the feature is not supported.

 

Compatibility with the old names is provided as follows:

 

PNG_INCH_CONVERSIONS turns on PNG_INCH_CONVERSIONS_SUPPORTED

 

And the following definitions disable the corresponding feature:

 

PNG_SETJMP_NOT_SUPPORTED disables SETJMP

PNG_READ_TRANSFORMS_NOT_SUPPORTED disables READ_TRANSFORMS

PNG_NO_READ_COMPOSITED_NODIV disables READ_COMPOSITE_NODIV

PNG_WRITE_TRANSFORMS_NOT_SUPPORTED disables WRITE_TRANSFORMS

PNG_READ_ANCILLARY_CHUNKS_NOT_SUPPORTED disables READ_ANCILLARY_CHUNKS

PNG_WRITE_ANCILLARY_CHUNKS_NOT_SUPPORTED disables WRITE_ANCILLARY_CHUNKS

 

Library builders should remove use of the above, inconsistent, names.

 

2) Warning and error message formatting was previously conditional on

the STDIO feature. The library has been changed to use the

CONSOLE_IO feature instead. This means that if CONSOLE_IO is disabled

the library no longer uses the printf(3) functions, even though the

default read/write implementations use (FILE) style stdio.h functions.

 

3) Three feature macros now control the fixed/floating point decisions:

 

PNG_FLOATING_POINT_SUPPORTED enables the floating point APIs

 

PNG_FIXED_POINT_SUPPORTED enables the fixed point APIs; however, in

practice these are normally required internally anyway (because the PNG

file format is fixed point), therefore in most cases PNG_NO_FIXED_POINT

merely stops the function from being exported.

 

PNG_FLOATING_ARITHMETIC_SUPPORTED chooses between the internal floating

point implementation or the fixed point one.  Typically the fixed point

implementation is larger and slower than the floating point implementation

on a system that supports floating point, however it may be faster on a

system which lacks floating point hardware and therefore uses a software

emulation.

 

4) Added PNG_{READ,WRITE}_INT_FUNCTIONS_SUPPORTED.  This allows the

functions to read and write ints to be disabled independently of

PNG_USE_READ_MACROS, which allows libpng to be built with the functions

even though the default is to use the macros - this allows applications

to choose at app buildtime whether or not to use macros (previously

impossible because the functions weren't in the default build.)

 

B.2 Changes to the configuration mechanism

 

Prior to libpng-1.5.0 library builders who needed to configure libpng

had either to modify the exported pngconf.h header file to add system

specific configuration or had to write feature selection macros into

pngusr.h and cause this to be included into pngconf.h by defining

PNG_USER_CONFIG. The latter mechanism had the disadvantage that an

application built without PNG_USER_CONFIG defined would see the

unmodified, default, libpng API and thus would probably fail to link.

 

These mechanisms still work in the configure build and in any makefile

build that builds pnglibconf.h although the feature selection macros

have changed somewhat as described above.  In 1.5.0, however, pngusr.h is

processed only once, when the exported header file pnglibconf.h is built.

pngconf.h no longer includes pngusr.h, therefore it is ignored after the

build of pnglibconf.h and it is never included in an application build.

 

The rarely used alternative of adding a list of feature macros to the

CFLAGS setting in the build also still works, however the macros will be

copied to pnglibconf.h and this may produce macro redefinition warnings

when the individual C files are compiled.

 

All configuration now only works if pnglibconf.h is built from

scripts/pnglibconf.dfa.  This requires the program awk.  Brian Kernighan

(the original author of awk) maintains C source code of that awk and this

and all known later implementations (often called by subtly different

names - nawk and gawk for example) are adequate to build pnglibconf.h.

The Sun Microsystems (now Oracle) program 'awk' is an earlier version

and does not work, this may also apply to other systems that have a

functioning awk called 'nawk'.

 

Configuration options are now documented in scripts/pnglibconf.dfa.  This

file also includes dependency information that ensures a configuration is

consistent; that is, if a feature is switched off dependent features are

also removed.  As a recommended alternative to using feature macros in

pngusr.h a system builder may also define equivalent options in pngusr.dfa

(or, indeed, any file) and add that to the configuration by setting

DFA_XTRA to the file name.  The makefiles in contrib/pngminim illustrate

how to do this, and a case where pngusr.h is still required.

 

XI. Detecting libpng

 

The png_get_io_ptr() function has been present since libpng-0.88, has never

changed, and is unaffected by conditional compilation macros.  It is the

best choice for use in configure scripts for detecting the presence of any

libpng version since 0.88.  In an autoconf "configure.in" you could use

 

    AC_CHECK_LIB(png, png_get_io_ptr, ...

 

XII. Source code repository

 

Since about February 2009, version 1.2.34, libpng has been under "git" source

control.  The git repository was built from old libpng-x.y.z.tar.gz files

going back to version 0.70.  You can access the git repository (read only)

at

 

    git://libpng.git.sourceforge.net/gitroot/libpng

 

or you can browse it via "gitweb" at

 

    http://libpng.git.sourceforge.net/git/gitweb.cgi?p=libpng