summaryrefslogtreecommitdiffstats
path: root/graphics/exact-image
diff options
context:
space:
mode:
Diffstat (limited to 'graphics/exact-image')
-rw-r--r--graphics/exact-image/Provide-gif-quantization-for-giflib-5.2.patch384
-rw-r--r--graphics/exact-image/exact-image.SlackBuild24
2 files changed, 403 insertions, 5 deletions
diff --git a/graphics/exact-image/Provide-gif-quantization-for-giflib-5.2.patch b/graphics/exact-image/Provide-gif-quantization-for-giflib-5.2.patch
new file mode 100644
index 0000000000..9f9210abf5
--- /dev/null
+++ b/graphics/exact-image/Provide-gif-quantization-for-giflib-5.2.patch
@@ -0,0 +1,384 @@
+From: Sven Eckelmann <sven@narfation.org>
+Date: Sat, 31 Aug 2019 14:13:11 +0200
+Subject: Provide gif quantization for giflib >= 5.2
+
+Bug-Debian: https://bugs.debian.org/939031
+
+diff -Naur exact-image-1.0.2.orig/codecs/Makefile exact-image-1.0.2/codecs/Makefile
+--- exact-image-1.0.2.orig/codecs/Makefile 2016-06-18 21:49:25.000000000 +0200
++++ exact-image-1.0.2/codecs/Makefile 2020-12-11 15:30:44.374504000 +0100
+@@ -21,7 +21,7 @@
+ ifeq "$(WITHLIBGIF)" "1"
+ LDFLAGS += -lgif
+ else
+-NOT_SRCS += gif.cc
++NOT_SRCS += gif.cc gif_quantization.c
+ endif
+
+ ifeq "$(WITHJASPER)" "1"
+diff -Naur exact-image-1.0.2.orig/codecs/gif.cc exact-image-1.0.2/codecs/gif.cc
+--- exact-image-1.0.2.orig/codecs/gif.cc 2017-07-21 16:19:01.000000000 +0200
++++ exact-image-1.0.2/codecs/gif.cc 2020-12-11 15:34:55.340504000 +0100
+@@ -160,6 +160,17 @@
+ return true;
+ }
+
++extern "C" int
++eiGifQuantizeBuffer(unsigned int Width,
++ unsigned int Height,
++ int *ColorMapSize,
++ GifByteType * RedInput,
++ GifByteType * GreenInput,
++ GifByteType * BlueInput,
++ GifByteType * OutputBuffer,
++ GifColorType * OutputColorMap);
++
++
+ bool GIFCodec::writeImage (std::ostream* stream, Image& image, int quality,
+ const std::string& compress)
+ {
+@@ -203,8 +214,7 @@
+ *bptr++ = b;
+ }
+
+-
+- if (GifQuantizeBuffer(image.w, image.h, &ColorMapSize,
++ if (eiGifQuantizeBuffer(image.w, image.h, &ColorMapSize,
+ RedBuffer, GreenBuffer, BlueBuffer,
+ OutputBuffer, OutputColorMap->Colors) == GIF_ERROR) {
+ return false;
+diff -Naur exact-image-1.0.2.orig/codecs/gif_quantization.c exact-image-1.0.2/codecs/gif_quantization.c
+--- exact-image-1.0.2.orig/codecs/gif_quantization.c 1970-01-01 01:00:00.000000000 +0100
++++ exact-image-1.0.2/codecs/gif_quantization.c 2020-12-11 15:30:19.181504000 +0100
+@@ -0,0 +1,331 @@
++/*****************************************************************************
++
++ quantize.c - quantize a high resolution image into lower one
++
++ Based on: "Color Image Quantization for frame buffer Display", by
++ Paul Heckbert SIGGRAPH 1982 page 297-307.
++
++ This doesn't really belong in the core library, was undocumented,
++ and was removed in 4.2. Then it turned out some client apps were
++ actually using it, so it was restored in 5.0.
++
++SPDX-License-Identifier: MIT
++
++******************************************************************************/
++
++#include <stdlib.h>
++#include <stdio.h>
++#include "gif_lib.h"
++
++#define ABS(x) ((x) > 0 ? (x) : (-(x)))
++
++#define COLOR_ARRAY_SIZE 32768
++#define BITS_PER_PRIM_COLOR 5
++#define MAX_PRIM_COLOR 0x1f
++
++static int SortRGBAxis;
++
++typedef struct QuantizedColorType {
++ GifByteType RGB[3];
++ GifByteType NewColorIndex;
++ long Count;
++ struct QuantizedColorType *Pnext;
++} QuantizedColorType;
++
++typedef struct NewColorMapType {
++ GifByteType RGBMin[3], RGBWidth[3];
++ unsigned int NumEntries; /* # of QuantizedColorType in linked list below */
++ unsigned long Count; /* Total number of pixels in all the entries */
++ QuantizedColorType *QuantizedColors;
++} NewColorMapType;
++
++static int SubdivColorMap(NewColorMapType * NewColorSubdiv,
++ unsigned int ColorMapSize,
++ unsigned int *NewColorMapSize);
++static int SortCmpRtn(const void *Entry1, const void *Entry2);
++
++/******************************************************************************
++ Quantize high resolution image into lower one. Input image consists of a
++ 2D array for each of the RGB colors with size Width by Height. There is no
++ Color map for the input. Output is a quantized image with 2D array of
++ indexes into the output color map.
++ Note input image can be 24 bits at the most (8 for red/green/blue) and
++ the output has 256 colors at the most (256 entries in the color map.).
++ ColorMapSize specifies size of color map up to 256 and will be updated to
++ real size before returning.
++ Also non of the parameter are allocated by this routine.
++ This function returns GIF_OK if successful, GIF_ERROR otherwise.
++******************************************************************************/
++int
++eiGifQuantizeBuffer(unsigned int Width,
++ unsigned int Height,
++ int *ColorMapSize,
++ GifByteType * RedInput,
++ GifByteType * GreenInput,
++ GifByteType * BlueInput,
++ GifByteType * OutputBuffer,
++ GifColorType * OutputColorMap) {
++
++ unsigned int Index, NumOfEntries;
++ int i, j, MaxRGBError[3];
++ unsigned int NewColorMapSize;
++ long Red, Green, Blue;
++ NewColorMapType NewColorSubdiv[256];
++ QuantizedColorType *ColorArrayEntries, *QuantizedColor;
++
++ ColorArrayEntries = (QuantizedColorType *)malloc(
++ sizeof(QuantizedColorType) * COLOR_ARRAY_SIZE);
++ if (ColorArrayEntries == NULL) {
++ return GIF_ERROR;
++ }
++
++ for (i = 0; i < COLOR_ARRAY_SIZE; i++) {
++ ColorArrayEntries[i].RGB[0] = i >> (2 * BITS_PER_PRIM_COLOR);
++ ColorArrayEntries[i].RGB[1] = (i >> BITS_PER_PRIM_COLOR) &
++ MAX_PRIM_COLOR;
++ ColorArrayEntries[i].RGB[2] = i & MAX_PRIM_COLOR;
++ ColorArrayEntries[i].Count = 0;
++ }
++
++ /* Sample the colors and their distribution: */
++ for (i = 0; i < (int)(Width * Height); i++) {
++ Index = ((RedInput[i] >> (8 - BITS_PER_PRIM_COLOR)) <<
++ (2 * BITS_PER_PRIM_COLOR)) +
++ ((GreenInput[i] >> (8 - BITS_PER_PRIM_COLOR)) <<
++ BITS_PER_PRIM_COLOR) +
++ (BlueInput[i] >> (8 - BITS_PER_PRIM_COLOR));
++ ColorArrayEntries[Index].Count++;
++ }
++
++ /* Put all the colors in the first entry of the color map, and call the
++ * recursive subdivision process. */
++ for (i = 0; i < 256; i++) {
++ NewColorSubdiv[i].QuantizedColors = NULL;
++ NewColorSubdiv[i].Count = NewColorSubdiv[i].NumEntries = 0;
++ for (j = 0; j < 3; j++) {
++ NewColorSubdiv[i].RGBMin[j] = 0;
++ NewColorSubdiv[i].RGBWidth[j] = 255;
++ }
++ }
++
++ /* Find the non empty entries in the color table and chain them: */
++ for (i = 0; i < COLOR_ARRAY_SIZE; i++)
++ if (ColorArrayEntries[i].Count > 0)
++ break;
++ QuantizedColor = NewColorSubdiv[0].QuantizedColors = &ColorArrayEntries[i];
++ NumOfEntries = 1;
++ while (++i < COLOR_ARRAY_SIZE)
++ if (ColorArrayEntries[i].Count > 0) {
++ QuantizedColor->Pnext = &ColorArrayEntries[i];
++ QuantizedColor = &ColorArrayEntries[i];
++ NumOfEntries++;
++ }
++ QuantizedColor->Pnext = NULL;
++
++ NewColorSubdiv[0].NumEntries = NumOfEntries; /* Different sampled colors */
++ NewColorSubdiv[0].Count = ((long)Width) * Height; /* Pixels */
++ NewColorMapSize = 1;
++ if (SubdivColorMap(NewColorSubdiv, *ColorMapSize, &NewColorMapSize) !=
++ GIF_OK) {
++ free((char *)ColorArrayEntries);
++ return GIF_ERROR;
++ }
++ if (NewColorMapSize < *ColorMapSize) {
++ /* And clear rest of color map: */
++ for (i = NewColorMapSize; i < *ColorMapSize; i++)
++ OutputColorMap[i].Red = OutputColorMap[i].Green =
++ OutputColorMap[i].Blue = 0;
++ }
++
++ /* Average the colors in each entry to be the color to be used in the
++ * output color map, and plug it into the output color map itself. */
++ for (i = 0; i < NewColorMapSize; i++) {
++ if ((j = NewColorSubdiv[i].NumEntries) > 0) {
++ QuantizedColor = NewColorSubdiv[i].QuantizedColors;
++ Red = Green = Blue = 0;
++ while (QuantizedColor) {
++ QuantizedColor->NewColorIndex = i;
++ Red += QuantizedColor->RGB[0];
++ Green += QuantizedColor->RGB[1];
++ Blue += QuantizedColor->RGB[2];
++ QuantizedColor = QuantizedColor->Pnext;
++ }
++ OutputColorMap[i].Red = (Red << (8 - BITS_PER_PRIM_COLOR)) / j;
++ OutputColorMap[i].Green = (Green << (8 - BITS_PER_PRIM_COLOR)) / j;
++ OutputColorMap[i].Blue = (Blue << (8 - BITS_PER_PRIM_COLOR)) / j;
++ }
++ }
++
++ /* Finally scan the input buffer again and put the mapped index in the
++ * output buffer. */
++ MaxRGBError[0] = MaxRGBError[1] = MaxRGBError[2] = 0;
++ for (i = 0; i < (int)(Width * Height); i++) {
++ Index = ((RedInput[i] >> (8 - BITS_PER_PRIM_COLOR)) <<
++ (2 * BITS_PER_PRIM_COLOR)) +
++ ((GreenInput[i] >> (8 - BITS_PER_PRIM_COLOR)) <<
++ BITS_PER_PRIM_COLOR) +
++ (BlueInput[i] >> (8 - BITS_PER_PRIM_COLOR));
++ Index = ColorArrayEntries[Index].NewColorIndex;
++ OutputBuffer[i] = Index;
++ if (MaxRGBError[0] < ABS(OutputColorMap[Index].Red - RedInput[i]))
++ MaxRGBError[0] = ABS(OutputColorMap[Index].Red - RedInput[i]);
++ if (MaxRGBError[1] < ABS(OutputColorMap[Index].Green - GreenInput[i]))
++ MaxRGBError[1] = ABS(OutputColorMap[Index].Green - GreenInput[i]);
++ if (MaxRGBError[2] < ABS(OutputColorMap[Index].Blue - BlueInput[i]))
++ MaxRGBError[2] = ABS(OutputColorMap[Index].Blue - BlueInput[i]);
++ }
++
++#ifdef DEBUG
++ fprintf(stderr,
++ "Quantization L(0) errors: Red = %d, Green = %d, Blue = %d.\n",
++ MaxRGBError[0], MaxRGBError[1], MaxRGBError[2]);
++#endif /* DEBUG */
++
++ free((char *)ColorArrayEntries);
++
++ *ColorMapSize = NewColorMapSize;
++
++ return GIF_OK;
++}
++
++/******************************************************************************
++ Routine to subdivide the RGB space recursively using median cut in each
++ axes alternatingly until ColorMapSize different cubes exists.
++ The biggest cube in one dimension is subdivide unless it has only one entry.
++ Returns GIF_ERROR if failed, otherwise GIF_OK.
++*******************************************************************************/
++static int
++SubdivColorMap(NewColorMapType * NewColorSubdiv,
++ unsigned int ColorMapSize,
++ unsigned int *NewColorMapSize) {
++
++ unsigned int i, j, Index = 0;
++ QuantizedColorType *QuantizedColor, **SortArray;
++
++ while (ColorMapSize > *NewColorMapSize) {
++ /* Find candidate for subdivision: */
++ long Sum, Count;
++ int MaxSize = -1;
++ unsigned int NumEntries, MinColor, MaxColor;
++ for (i = 0; i < *NewColorMapSize; i++) {
++ for (j = 0; j < 3; j++) {
++ if ((((int)NewColorSubdiv[i].RGBWidth[j]) > MaxSize) &&
++ (NewColorSubdiv[i].NumEntries > 1)) {
++ MaxSize = NewColorSubdiv[i].RGBWidth[j];
++ Index = i;
++ SortRGBAxis = j;
++ }
++ }
++ }
++
++ if (MaxSize == -1)
++ return GIF_OK;
++
++ /* Split the entry Index into two along the axis SortRGBAxis: */
++
++ /* Sort all elements in that entry along the given axis and split at
++ * the median. */
++ SortArray = (QuantizedColorType **)malloc(
++ sizeof(QuantizedColorType *) *
++ NewColorSubdiv[Index].NumEntries);
++ if (SortArray == NULL)
++ return GIF_ERROR;
++ for (j = 0, QuantizedColor = NewColorSubdiv[Index].QuantizedColors;
++ j < NewColorSubdiv[Index].NumEntries && QuantizedColor != NULL;
++ j++, QuantizedColor = QuantizedColor->Pnext)
++ SortArray[j] = QuantizedColor;
++
++ /*
++ * Because qsort isn't stable, this can produce differing
++ * results for the order of tuples depending on platform
++ * details of how qsort() is implemented.
++ *
++ * We mitigate this problem by sorting on all three axes rather
++ * than only the one specied by SortRGBAxis; that way the instability
++ * can only become an issue if there are multiple color indices
++ * referring to identical RGB tuples. Older versions of this
++ * sorted on only the one axis.
++ */
++ qsort(SortArray, NewColorSubdiv[Index].NumEntries,
++ sizeof(QuantizedColorType *), SortCmpRtn);
++
++ /* Relink the sorted list into one: */
++ for (j = 0; j < NewColorSubdiv[Index].NumEntries - 1; j++)
++ SortArray[j]->Pnext = SortArray[j + 1];
++ SortArray[NewColorSubdiv[Index].NumEntries - 1]->Pnext = NULL;
++ NewColorSubdiv[Index].QuantizedColors = QuantizedColor = SortArray[0];
++ free((char *)SortArray);
++
++ /* Now simply add the Counts until we have half of the Count: */
++ Sum = NewColorSubdiv[Index].Count / 2 - QuantizedColor->Count;
++ NumEntries = 1;
++ Count = QuantizedColor->Count;
++ while (QuantizedColor->Pnext != NULL &&
++ (Sum -= QuantizedColor->Pnext->Count) >= 0 &&
++ QuantizedColor->Pnext->Pnext != NULL) {
++ QuantizedColor = QuantizedColor->Pnext;
++ NumEntries++;
++ Count += QuantizedColor->Count;
++ }
++ /* Save the values of the last color of the first half, and first
++ * of the second half so we can update the Bounding Boxes later.
++ * Also as the colors are quantized and the BBoxes are full 0..255,
++ * they need to be rescaled.
++ */
++ MaxColor = QuantizedColor->RGB[SortRGBAxis]; /* Max. of first half */
++ /* coverity[var_deref_op] */
++ MinColor = QuantizedColor->Pnext->RGB[SortRGBAxis]; /* of second */
++ MaxColor <<= (8 - BITS_PER_PRIM_COLOR);
++ MinColor <<= (8 - BITS_PER_PRIM_COLOR);
++
++ /* Partition right here: */
++ NewColorSubdiv[*NewColorMapSize].QuantizedColors =
++ QuantizedColor->Pnext;
++ QuantizedColor->Pnext = NULL;
++ NewColorSubdiv[*NewColorMapSize].Count = Count;
++ NewColorSubdiv[Index].Count -= Count;
++ NewColorSubdiv[*NewColorMapSize].NumEntries =
++ NewColorSubdiv[Index].NumEntries - NumEntries;
++ NewColorSubdiv[Index].NumEntries = NumEntries;
++ for (j = 0; j < 3; j++) {
++ NewColorSubdiv[*NewColorMapSize].RGBMin[j] =
++ NewColorSubdiv[Index].RGBMin[j];
++ NewColorSubdiv[*NewColorMapSize].RGBWidth[j] =
++ NewColorSubdiv[Index].RGBWidth[j];
++ }
++ NewColorSubdiv[*NewColorMapSize].RGBWidth[SortRGBAxis] =
++ NewColorSubdiv[*NewColorMapSize].RGBMin[SortRGBAxis] +
++ NewColorSubdiv[*NewColorMapSize].RGBWidth[SortRGBAxis] - MinColor;
++ NewColorSubdiv[*NewColorMapSize].RGBMin[SortRGBAxis] = MinColor;
++
++ NewColorSubdiv[Index].RGBWidth[SortRGBAxis] =
++ MaxColor - NewColorSubdiv[Index].RGBMin[SortRGBAxis];
++
++ (*NewColorMapSize)++;
++ }
++
++ return GIF_OK;
++}
++
++/****************************************************************************
++ Routine called by qsort to compare two entries.
++*****************************************************************************/
++
++static int
++SortCmpRtn(const void *Entry1,
++ const void *Entry2) {
++ QuantizedColorType *entry1 = (*((QuantizedColorType **) Entry1));
++ QuantizedColorType *entry2 = (*((QuantizedColorType **) Entry2));
++
++ /* sort on all axes of the color space! */
++ int hash1 = entry1->RGB[SortRGBAxis] * 256 * 256
++ + entry1->RGB[(SortRGBAxis+1) % 3] * 256
++ + entry1->RGB[(SortRGBAxis+2) % 3];
++ int hash2 = entry2->RGB[SortRGBAxis] * 256 * 256
++ + entry2->RGB[(SortRGBAxis+1) % 3] * 256
++ + entry2->RGB[(SortRGBAxis+2) % 3];
++
++ return hash1 - hash2;
++}
++
++/* end */
diff --git a/graphics/exact-image/exact-image.SlackBuild b/graphics/exact-image/exact-image.SlackBuild
index bc64a31f5b..5ef236189f 100644
--- a/graphics/exact-image/exact-image.SlackBuild
+++ b/graphics/exact-image/exact-image.SlackBuild
@@ -1,8 +1,8 @@
-#!/bin/sh
+#!/bin/bash
# Slackware build script for exact-image
-# Copyright 2010-2019 Benjamin Trigona-Harany <slackbuilds@jaxartes.net>
+# Copyright 2010-2021 Benjamin Trigona-Harany <slackbuilds@jaxartes.net>
# All rights reserved.
#
# Redistribution and use of this script, with or without modification, is
@@ -22,10 +22,13 @@
# OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
# ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+cd $(dirname $0) ; CWD=$(pwd)
+
PRGNAM=exact-image
VERSION=${VERSION:-1.0.2}
BUILD=${BUILD:-1}
TAG=${TAG:-_SBo}
+PKGTYPE=${PKGTYPE:-tgz}
if [ -z "$ARCH" ]; then
case "$( uname -m )" in
@@ -35,7 +38,14 @@ if [ -z "$ARCH" ]; then
esac
fi
-CWD=$(pwd)
+# If the variable PRINT_PACKAGE_NAME is set, then this script will report what
+# the name of the created package would be, and then exit. This information
+# could be useful to other scripts.
+if [ ! -z "${PRINT_PACKAGE_NAME}" ]; then
+ echo "$PRGNAM-$VERSION-$ARCH-$BUILD$TAG.$PKGTYPE"
+ exit 0
+fi
+
TMP=${TMP:-/tmp/SBo}
PKG=$TMP/package-$PRGNAM
OUTPUT=${OUTPUT:-/tmp}
@@ -74,11 +84,15 @@ eval $(perl -V:vendorlib)
sed "s,@vendorlib@,$vendorlib,g" $CWD/fixup-perl_paths.diff | patch -p1 --verbose
patch -p1 < $CWD/c98.patch
+patch -p1 < $CWD/Provide-gif-quantization-for-giflib-5.2.patch
./configure \
--bindir=/usr/bin \
--libdir=/usr/lib${LIBDIRSUFFIX} \
- --without-php
+ --without-libgif \
+ --without-openexr \
+ --without-php \
+ --without-python
make CFLAGS="$SLKCFLAGS" CXXFLAGS="$SLKCFLAGS -Wno-narrowing -fpermissive"
make install DESTDIR=$PKG
@@ -96,4 +110,4 @@ mkdir -p $PKG/install
cat $CWD/slack-desc > $PKG/install/slack-desc
cd $PKG
-/sbin/makepkg -l y -c n $OUTPUT/$PRGNAM-$VERSION-$ARCH-$BUILD$TAG.${PKGTYPE:-tgz}
+/sbin/makepkg -l y -c n $OUTPUT/$PRGNAM-$VERSION-$ARCH-$BUILD$TAG.$PKGTYPE