Fix whitespace errors

jpeg_crop_scanlines: Handle gray images w/ samp!=1
Since the sampling factor has no meaning for single-component images, the decompressor ignores it, and jpeg_crop_scanlines() should as well. Fixes #195
2017-12-13 21:48:54 -06:00 · 2017-12-06 11:59:04 -06:00 · 2017-12-05 16:08:01 -06:00 · 2017-11-29 14:23:31 -06:00 · 2017-11-17 18:45:08 -06:00 · 2017-11-17 18:39:53 -06:00
38 changed files with 515 additions and 379 deletions
--- a/.travis.yml
+++ b/.travis.yml
@@ -43,6 +43,7 @@ matrix:

 before_install:
  - if [ "$TRAVIS_OS_NAME" = "osx" ]; then
+      brew update &&
      brew install nasm homebrew/versions/gcc5 md5sha1sum Caskroom/versions/java6 &&
      ln -fs /usr/local/bin/gpg1 /usr/local/bin/gpg &&
      git clone --depth=1 https://github.com/libjpeg-turbo/gas-preprocessor.git ~/src/gas-preprocessor &&
@@ -84,7 +85,8 @@ script:
      mkdir build &&
      pushd build &&
      ../configure ${CONFIGURE_FLAGS} &&
-      make -j &&
+      export NUMCPUS=`grep -c '^processor' /proc/cpuinfo` &&
+      make -j$NUMCPUS --load-average=$NUMCPUS &&
      if [[ "${CONFIGURE_FLAGS}" =~ "with-12bit" ||
            "${CONFIGURE_FLAGS}" =~ "without-simd" ]]; then
        make test FLOATTEST=32bit;
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -9,7 +9,7 @@ if(POLICY CMP0022)
 endif()

 project(libjpeg-turbo C)
-set(VERSION 1.5.2)
+set(VERSION 1.5.3)
 string(REPLACE "." ";" VERSION_TRIPLET ${VERSION})
 list(GET VERSION_TRIPLET 0 VERSION_MAJOR)
 list(GET VERSION_TRIPLET 1 VERSION_MINOR)
--- a/ChangeLog.md
+++ b/ChangeLog.md
@@ -1,3 +1,47 @@
+1.5.3
+=====
+
+### Significant changes relative to 1.5.2:
+
+1. Fixed a NullPointerException in the TurboJPEG Java wrapper that occurred
+when using the YUVImage constructor that creates an instance backed by separate
+image planes and allocates memory for the image planes.
+
+2. Fixed an issue whereby the Java version of TJUnitTest would fail when
+testing BufferedImage encoding/decoding on big endian systems.
+
+3. Fixed a segfault in djpeg that would occur if an output format other than
+PPM/PGM was selected along with the `-crop` option.  The `-crop` option now
+works with the GIF and Targa formats as well (unfortunately, it cannot be made
+to work with the BMP and RLE formats due to the fact that those output engines
+write scanlines in bottom-up order.)  djpeg will now exit gracefully if an
+output format other than PPM/PGM, GIF, or Targa is selected along with the
+`-crop` option.
+
+4. Fixed an issue whereby `jpeg_skip_scanlines()` would segfault if color
+quantization was enabled.
+
+5. TJBench (both C and Java versions) will now display usage information if any
+command-line argument is unrecognized.  This prevents the program from silently
+ignoring typos.
+
+6. Fixed an access violation in tjbench.exe (Windows) that occurred when the
+program was used to decompress an existing JPEG image.
+
+7. Fixed an ArrayIndexOutOfBoundsException in the TJExample Java program that
+occurred when attempting to decompress a JPEG image that had been compressed
+with 4:1:1 chrominance subsampling.
+
+8. Fixed an issue whereby, when using `jpeg_skip_scanlines()` to skip to the
+end of a single-scan (non-progressive) image, subsequent calls to
+`jpeg_consume_input()` would return `JPEG_SUSPENDED` rather than
+`JPEG_REACHED_EOI`.
+
+9. `jpeg_crop_scanlines()` now works correctly when decompressing grayscale
+JPEG images that were compressed with a sampling factor other than 1 (for
+instance, with `cjpeg -grayscale -sample 2x2`).
+
+
 1.5.2
 =====

--- a/Makefile.am
+++ b/Makefile.am
@@ -173,7 +173,7 @@ dist_example_DATA = example.c
 EXTRA_DIST = win release $(DOCS) testimages CMakeLists.txt \
 	sharedlib/CMakeLists.txt cmakescripts libjpeg.map.in doc doxygen.config \
 	doxygen-extra.css jccolext.c jdcolext.c jdcol565.c jdmrgext.c jdmrg565.c \
-	jstdhuff.c jdcoefct.h jdmainct.h jdmaster.h jdsample.h wrppm.h \
+	jstdhuff.c jdcoefct.h jdmainct.h jdmaster.h jdsample.h \
 	md5/CMakeLists.txt

 dist-hook:
@@ -697,6 +697,8 @@ testclean:
 	rm -f *_411_*.ppm
 	rm -f *_411_*.jpg
 	rm -f *_411.yuv
+	rm -f tjbenchtest*.log
+	rm -f tjexampletest*.log


 tjtest:
--- a/acinclude.m4
+++ b/acinclude.m4
@@ -252,3 +252,36 @@ AC_DEFUN([AC_CHECK_COMPATIBLE_ARM64_ASSEMBLER_IFELSE],[
    $2
  fi
 ])
+
+# AC_CHECK_ALTIVEC
+# ----------------
+# Test whether AltiVec intrinsics are supported
+AC_DEFUN([AC_CHECK_ALTIVEC],[
+  ac_save_CFLAGS="$CFLAGS"
+  CFLAGS="$CFLAGS -maltivec"
+  AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+    #include <altivec.h>
+    int main(void) {
+      __vector int vi = { 0, 0, 0, 0 };
+      int i[4];
+      vec_st(vi, 0, i);
+      return i[0];
+    }]])], ac_has_altivec=yes)
+  CFLAGS="$ac_save_CFLAGS"
+  if test "x$ac_has_altivec" = "xyes" ; then
+    $1
+  else
+    $2
+  fi
+])
+
+AC_DEFUN([AC_NO_SIMD],[
+  AC_MSG_RESULT([no ("$1")])
+  with_simd=no;
+  if test "x${require_simd}" = "xyes"; then
+    AC_MSG_ERROR([SIMD support not available for this CPU.])
+  else
+    AC_MSG_WARN([SIMD support not available for this CPU.  Performance will\
+ suffer.])
+  fi
+])
--- a/cdjpeg.h
+++ b/cdjpeg.h
@@ -3,8 +3,8 @@
 *
 * This file was part of the Independent JPEG Group's software:
 * Copyright (C) 1994-1997, Thomas G. Lane.
- * It was modified by The libjpeg-turbo Project to include only code relevant
- * to libjpeg-turbo.
+ * libjpeg-turbo Modifications:
+ * Copyright (C) 2017, D. R. Commander.
 * For conditions of distribution and use, see the accompanying README.ijg
 * file.
 *
@@ -54,6 +54,14 @@ struct djpeg_dest_struct {
                          JDIMENSION rows_supplied);
  /* Finish up at the end of the image. */
  void (*finish_output) (j_decompress_ptr cinfo, djpeg_dest_ptr dinfo);
+  /* Re-calculate buffer dimensions based on output dimensions (for use with
+     partial image decompression.)  If this is NULL, then the output format
+     does not support partial image decompression (BMP and RLE, in particular,
+     cannot support partial decompression because they use an inversion buffer
+     to write the image in bottom-up order.) */
+  void (*calc_buffer_dimensions) (j_decompress_ptr cinfo,
+                                  djpeg_dest_ptr dinfo);
+

  /* Target file spec; filled in by djpeg.c after object is created. */
  FILE *output_file;
--- a/configure.ac
+++ b/configure.ac
@@ -2,7 +2,7 @@
 # Process this file with autoconf to produce a configure script.

 AC_PREREQ([2.56])
-AC_INIT([libjpeg-turbo], [1.5.2])
+AC_INIT([libjpeg-turbo], [1.5.3])

 AM_INIT_AUTOMAKE([-Wall foreign dist-bzip2])
 AC_PREFIX_DEFAULT(/opt/libjpeg-turbo)
@@ -361,6 +361,7 @@ if test "x$JAVAC" = "x"; then
 fi
 AC_SUBST(JAVAC)
 AC_ARG_VAR(JAVACFLAGS, [Java compiler flags])
+JAVACFLAGS="$JAVACFLAGS -J-Dfile.encoding=UTF8"
 AC_SUBST(JAVACFLAGS)
 AC_ARG_VAR(JAR, [Java archive command (default: jar)])
 if test "x$JAR" = "x"; then
@@ -517,17 +518,13 @@ if test "x${with_simd}" != "xno"; then
      fi
      ;;
    powerpc*)
-      AC_MSG_RESULT([yes (powerpc)])
-      simd_arch=powerpc
+      AC_CHECK_ALTIVEC(
+        [AC_MSG_RESULT([yes (powerpc)])
+         simd_arch=powerpc],
+        [AC_NO_SIMD(PowerPC SPE)])
      ;;
    *)
-      AC_MSG_RESULT([no ("$host_cpu")])
-      with_simd=no;
-      if test "x${require_simd}" = "xyes"; then
-        AC_MSG_ERROR([SIMD support not available for this CPU.])
-      else
-        AC_MSG_WARN([SIMD support not available for this CPU.  Performance will suffer.])
-      fi
+      AC_NO_SIMD($host_cpu)
      ;;
  esac

@@ -565,6 +562,14 @@ case "$host_cpu" in
    RPMARCH=i386
    DEBARCH=i386
    ;;
+  powerpc64le)
+    RPMARCH=`uname -m`
+    DEBARCH=ppc64el
+    ;;
+  powerpc)
+    RPMARCH=ppc
+    DEBARCH=ppc
+    ;;
  *)
    RPMARCH=`uname -m`
    DEBARCH=$RPMARCH
--- a/djpeg.1
+++ b/djpeg.1
@@ -1,4 +1,4 @@
-.TH DJPEG 1 "18 March 2017"
+.TH DJPEG 1 "13 November 2017"
 .SH NAME
 djpeg \- decompress a JPEG file to an image file
 .SH SYNOPSIS
@@ -204,7 +204,8 @@ Decompress only a rectangular subregion of the image, starting at point X,Y
 with width W and height H.  If necessary, X will be shifted left to the nearest
 iMCU boundary, and the width will be increased accordingly.  Note that if
 decompression scaling is being used, then X, Y, W, and H are relative to the
-scaled image dimensions.
+scaled image dimensions.  Currently this option only works with the
+PBMPLUS (PPM/PGM), GIF, and Targa output formats.
 .TP
 .B \-verbose
 Enable debug printout.  More
--- a/djpeg.c
+++ b/djpeg.c
@@ -5,7 +5,7 @@
 * Copyright (C) 1991-1997, Thomas G. Lane.
 * Modified 2013 by Guido Vollbeding.
 * libjpeg-turbo Modifications:
- * Copyright (C) 2010-2011, 2013-2016, D. R. Commander.
+ * Copyright (C) 2010-2011, 2013-2017, D. R. Commander.
 * Copyright (C) 2015, Google, Inc.
 * For conditions of distribution and use, see the accompanying README.ijg
 * file.
@@ -31,7 +31,6 @@
 #include "cdjpeg.h"             /* Common decls for cjpeg/djpeg applications */
 #include "jversion.h"           /* for version message */
 #include "jconfigint.h"
-#include "wrppm.h"

 #include <ctype.h>              /* to declare isprint() */

@@ -173,6 +172,7 @@ usage (void)

  fprintf(stderr, "  -skip Y0,Y1    Decompress all rows except those between Y0 and Y1 (inclusive)\n");
  fprintf(stderr, "  -crop WxH+X+Y  Decompress only a rectangular subregion of the image\n");
+  fprintf(stderr, "                 [requires PBMPLUS (PPM/PGM), GIF, or Targa output format]\n");
  fprintf(stderr, "  -verbose  or  -debug   Emit debug output\n");
  fprintf(stderr, "  -version       Print version information and exit\n");
  exit(EXIT_FAILURE);
@@ -713,9 +713,10 @@ main (int argc, char **argv)
    }

    jpeg_crop_scanline(&cinfo, &crop_x, &crop_width);
-    ((ppm_dest_ptr) dest_mgr)->buffer_width = cinfo.output_width *
-                                              cinfo.out_color_components *
-                                              sizeof(JSAMPLE);
+    if (dest_mgr->calc_buffer_dimensions)
+      (*dest_mgr->calc_buffer_dimensions) (&cinfo, dest_mgr);
+    else
+      ERREXIT(&cinfo, JERR_UNSUPPORTED_FORMAT);

    /* Write output file header.  This is a hack to ensure that the destination
     * manager creates an output image of the proper size.
--- a/doc/html/group___turbo_j_p_e_g.html
+++ b/doc/html/group___turbo_j_p_e_g.html
@@ -1014,7 +1014,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
    <tr><td class="paramname">jpegSize</td><td>pointer to an unsigned long variable that holds the size of the JPEG image buffer. If <code>*jpegBuf</code> points to a pre-allocated buffer, then <code>*jpegSize</code> should be set to the size of the buffer. Upon return, <code>*jpegSize</code> will contain the size of the JPEG image (in bytes.) If <code>*jpegBuf</code> points to a JPEG image buffer that is being reused from a previous call to one of the JPEG compression functions, then <code>*jpegSize</code> is ignored.</td></tr>
    <tr><td class="paramname">jpegSubsamp</td><td>the level of chrominance subsampling to be used when generating the JPEG image (see <a class="el" href="group___turbo_j_p_e_g.html#ga1d047060ea80bb9820d540bb928e9074">Chrominance subsampling options</a>.)</td></tr>
    <tr><td class="paramname">jpegQual</td><td>the image quality of the generated JPEG image (1 = worst, 100 = best)</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1111,7 +1111,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
 If you choose option 1, <code>*jpegSize</code> should be set to the size of your pre-allocated buffer. In any case, unless you have set <a class="el" href="group___turbo_j_p_e_g.html#ga8808d403c68b62aaa58a4c1e58e98963" title="Disable buffer (re)allocation.">TJFLAG_NOREALLOC</a>, you should always check <code>*jpegBuf</code> upon return from this function, as it may have changed.</td></tr>
    <tr><td class="paramname">jpegSize</td><td>pointer to an unsigned long variable that holds the size of the JPEG image buffer. If <code>*jpegBuf</code> points to a pre-allocated buffer, then <code>*jpegSize</code> should be set to the size of the buffer. Upon return, <code>*jpegSize</code> will contain the size of the JPEG image (in bytes.) If <code>*jpegBuf</code> points to a JPEG image buffer that is being reused from a previous call to one of the JPEG compression functions, then <code>*jpegSize</code> is ignored.</td></tr>
    <tr><td class="paramname">jpegQual</td><td>the image quality of the generated JPEG image (1 = worst, 100 = best)</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1208,7 +1208,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
 If you choose option 1, <code>*jpegSize</code> should be set to the size of your pre-allocated buffer. In any case, unless you have set <a class="el" href="group___turbo_j_p_e_g.html#ga8808d403c68b62aaa58a4c1e58e98963" title="Disable buffer (re)allocation.">TJFLAG_NOREALLOC</a>, you should always check <code>*jpegBuf</code> upon return from this function, as it may have changed.</td></tr>
    <tr><td class="paramname">jpegSize</td><td>pointer to an unsigned long variable that holds the size of the JPEG image buffer. If <code>*jpegBuf</code> points to a pre-allocated buffer, then <code>*jpegSize</code> should be set to the size of the buffer. Upon return, <code>*jpegSize</code> will contain the size of the JPEG image (in bytes.) If <code>*jpegBuf</code> points to a JPEG image buffer that is being reused from a previous call to one of the JPEG compression functions, then <code>*jpegSize</code> is ignored.</td></tr>
    <tr><td class="paramname">jpegQual</td><td>the image quality of the generated JPEG image (1 = worst, 100 = best)</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1301,7 +1301,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
    <tr><td class="paramname">pitch</td><td>bytes per line in the destination image. Normally, this should be <code>width * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat]</code> if the destination image is unpadded, or <code><a class="el" href="group___turbo_j_p_e_g.html#ga0aba955473315e405295d978f0c16511" title="Pad the given width to the nearest 32-bit boundary.">TJPAD</a>(width * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat])</code> if each line of the destination image should be padded to the nearest 32-bit boundary, as is the case for Windows bitmaps. You can also be clever and use the pitch parameter to skip lines, etc. Setting this parameter to 0 is the equivalent of setting it to <code>width * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat]</code>.</td></tr>
    <tr><td class="paramname">height</td><td>height (in pixels) of the source and destination images</td></tr>
    <tr><td class="paramname">pixelFormat</td><td>pixel format of the destination image (see <a class="el" href="group___turbo_j_p_e_g.html#gac916144e26c3817ac514e64ae5d12e2a">Pixel formats</a>.)</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1394,7 +1394,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
    <tr><td class="paramname">pitch</td><td>bytes per line in the destination image. Normally, this should be <code>width * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat]</code> if the destination image is unpadded, or <code><a class="el" href="group___turbo_j_p_e_g.html#ga0aba955473315e405295d978f0c16511" title="Pad the given width to the nearest 32-bit boundary.">TJPAD</a>(width * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat])</code> if each line of the destination image should be padded to the nearest 32-bit boundary, as is the case for Windows bitmaps. You can also be clever and use the pitch parameter to skip lines, etc. Setting this parameter to 0 is the equivalent of setting it to <code>width * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat]</code>.</td></tr>
    <tr><td class="paramname">height</td><td>height (in pixels) of the source and destination images</td></tr>
    <tr><td class="paramname">pixelFormat</td><td>pixel format of the destination image (see <a class="el" href="group___turbo_j_p_e_g.html#gac916144e26c3817ac514e64ae5d12e2a">Pixel formats</a>.)</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1479,7 +1479,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
    <tr><td class="paramname">pitch</td><td>bytes per line in the destination image. Normally, this is <code>scaledWidth * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat]</code> if the decompressed image is unpadded, else <code><a class="el" href="group___turbo_j_p_e_g.html#ga0aba955473315e405295d978f0c16511" title="Pad the given width to the nearest 32-bit boundary.">TJPAD</a>(scaledWidth * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat])</code> if each line of the decompressed image is padded to the nearest 32-bit boundary, as is the case for Windows bitmaps. (NOTE: <code>scaledWidth</code> can be determined by calling <a class="el" href="group___turbo_j_p_e_g.html#ga84878bb65404204743aa18cac02781df" title="Compute the scaled value of dimension using the given scaling factor.">TJSCALED()</a> with the JPEG image width and one of the scaling factors returned by <a class="el" href="group___turbo_j_p_e_g.html#ga6449044b9af402999ccf52f401333be8" title="Returns a list of fractional scaling factors that the JPEG decompressor in this implementation of Tur...">tjGetScalingFactors()</a>.) You can also be clever and use the pitch parameter to skip lines, etc. Setting this parameter to 0 is the equivalent of setting it to <code>scaledWidth * <a class="el" href="group___turbo_j_p_e_g.html#gad77cf8fe5b2bfd3cb3f53098146abb4c" title="Pixel size (in bytes) for a given pixel format.">tjPixelSize</a>[pixelFormat]</code>.</td></tr>
    <tr><td class="paramname">height</td><td>desired height (in pixels) of the destination image. If this is different than the height of the JPEG image being decompressed, then TurboJPEG will use scaling in the JPEG decompressor to generate the largest possible image that will fit within the desired height. If <code>height</code> is set to 0, then only the width will be considered when determining the scaled image size.</td></tr>
    <tr><td class="paramname">pixelFormat</td><td>pixel format of the destination image (see <a class="el" href="group___turbo_j_p_e_g.html#gac916144e26c3817ac514e64ae5d12e2a">Pixel formats</a>.)</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1629,7 +1629,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
    <tr><td class="paramname">width</td><td>desired width (in pixels) of the YUV image. If this is different than the width of the JPEG image being decompressed, then TurboJPEG will use scaling in the JPEG decompressor to generate the largest possible image that will fit within the desired width. If <code>width</code> is set to 0, then only the height will be considered when determining the scaled image size. If the scaled width is not an even multiple of the MCU block width (see <a class="el" href="group___turbo_j_p_e_g.html#ga9e61e7cd47a15a173283ba94e781308c" title="MCU block width (in pixels) for a given level of chrominance subsampling.">tjMCUWidth</a>), then an intermediate buffer copy will be performed within TurboJPEG.</td></tr>
    <tr><td class="paramname">pad</td><td>the width of each line in each plane of the YUV image will be padded to the nearest multiple of this number of bytes (must be a power of 2.) To generate images suitable for X Video, <code>pad</code> should be set to 4.</td></tr>
    <tr><td class="paramname">height</td><td>desired height (in pixels) of the YUV image. If this is different than the height of the JPEG image being decompressed, then TurboJPEG will use scaling in the JPEG decompressor to generate the largest possible image that will fit within the desired height. If <code>height</code> is set to 0, then only the width will be considered when determining the scaled image size. If the scaled height is not an even multiple of the MCU block height (see <a class="el" href="group___turbo_j_p_e_g.html#gabd247bb9fecb393eca57366feb8327bf" title="MCU block height (in pixels) for a given level of chrominance subsampling.">tjMCUHeight</a>), then an intermediate buffer copy will be performed within TurboJPEG.</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1708,7 +1708,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
    <tr><td class="paramname">width</td><td>desired width (in pixels) of the YUV image. If this is different than the width of the JPEG image being decompressed, then TurboJPEG will use scaling in the JPEG decompressor to generate the largest possible image that will fit within the desired width. If <code>width</code> is set to 0, then only the height will be considered when determining the scaled image size. If the scaled width is not an even multiple of the MCU block width (see <a class="el" href="group___turbo_j_p_e_g.html#ga9e61e7cd47a15a173283ba94e781308c" title="MCU block width (in pixels) for a given level of chrominance subsampling.">tjMCUWidth</a>), then an intermediate buffer copy will be performed within TurboJPEG.</td></tr>
    <tr><td class="paramname">strides</td><td>an array of integers, each specifying the number of bytes per line in the corresponding plane of the output image. Setting the stride for any plane to 0 is the same as setting it to the scaled plane width (see <a class="el" href="group___turbo_j_p_e_g.html#YUVnotes">YUV Image Format Notes</a>.) If <code>strides</code> is NULL, then the strides for all planes will be set to their respective scaled plane widths. You can adjust the strides in order to add an arbitrary amount of line padding to each plane or to decompress the JPEG image into a subregion of a larger YUV planar image.</td></tr>
    <tr><td class="paramname">height</td><td>desired height (in pixels) of the YUV image. If this is different than the height of the JPEG image being decompressed, then TurboJPEG will use scaling in the JPEG decompressor to generate the largest possible image that will fit within the desired height. If <code>height</code> is set to 0, then only the width will be considered when determining the scaled image size. If the scaled height is not an even multiple of the MCU block height (see <a class="el" href="group___turbo_j_p_e_g.html#gabd247bb9fecb393eca57366feb8327bf" title="MCU block height (in pixels) for a given level of chrominance subsampling.">tjMCUHeight</a>), then an intermediate buffer copy will be performed within TurboJPEG.</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1826,7 +1826,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
    <tr><td class="paramname">dstBuf</td><td>pointer to an image buffer that will receive the YUV image. Use <a class="el" href="group___turbo_j_p_e_g.html#gaf451664a62c1f6c7cc5a6401f32908c9" title="The size of the buffer (in bytes) required to hold a YUV planar image with the given parameters...">tjBufSizeYUV2()</a> to determine the appropriate size for this buffer based on the image width, height, padding, and level of chrominance subsampling. The Y, U (Cb), and V (Cr) image planes will be stored sequentially in the buffer (refer to <a class="el" href="group___turbo_j_p_e_g.html#YUVnotes">YUV Image Format Notes</a>.)</td></tr>
    <tr><td class="paramname">pad</td><td>the width of each line in each plane of the YUV image will be padded to the nearest multiple of this number of bytes (must be a power of 2.) To generate images suitable for X Video, <code>pad</code> should be set to 4.</td></tr>
    <tr><td class="paramname">subsamp</td><td>the level of chrominance subsampling to be used when generating the YUV image (see <a class="el" href="group___turbo_j_p_e_g.html#ga1d047060ea80bb9820d540bb928e9074">Chrominance subsampling options</a>.) To generate images suitable for X Video, <code>subsamp</code> should be set to <a class="el" href="group___turbo_j_p_e_g.html#gga1d047060ea80bb9820d540bb928e9074a63085dbf683cfe39e513cdb6343e3737">TJSAMP_420</a>. This produces an image compatible with the I420 (AKA "YUV420P") format.</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -1919,7 +1919,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
    <tr><td class="paramname">dstPlanes</td><td>an array of pointers to Y, U (Cb), and V (Cr) image planes (or just a Y plane, if generating a grayscale image) that will receive the encoded image. These planes can be contiguous or non-contiguous in memory. Use <a class="el" href="group___turbo_j_p_e_g.html#ga6f98d977bfa9d167c97172e876ba61e2" title="The size of the buffer (in bytes) required to hold a YUV image plane with the given parameters...">tjPlaneSizeYUV()</a> to determine the appropriate size for each plane based on the image width, height, strides, and level of chrominance subsampling. Refer to <a class="el" href="group___turbo_j_p_e_g.html#YUVnotes">YUV Image Format Notes</a> for more details.</td></tr>
    <tr><td class="paramname">strides</td><td>an array of integers, each specifying the number of bytes per line in the corresponding plane of the output image. Setting the stride for any plane to 0 is the same as setting it to the plane width (see <a class="el" href="group___turbo_j_p_e_g.html#YUVnotes">YUV Image Format Notes</a>.) If <code>strides</code> is NULL, then the strides for all planes will be set to their respective plane widths. You can adjust the strides in order to add an arbitrary amount of line padding to each plane or to encode an RGB or grayscale image into a subregion of a larger YUV planar image.</td></tr>
    <tr><td class="paramname">subsamp</td><td>the level of chrominance subsampling to be used when generating the YUV image (see <a class="el" href="group___turbo_j_p_e_g.html#ga1d047060ea80bb9820d540bb928e9074">Chrominance subsampling options</a>.) To generate images suitable for X Video, <code>subsamp</code> should be set to <a class="el" href="group___turbo_j_p_e_g.html#gga1d047060ea80bb9820d540bb928e9074a63085dbf683cfe39e513cdb6343e3737">TJSAMP_420</a>. This produces an image compatible with the I420 (AKA "YUV420P") format.</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
@@ -2275,7 +2275,7 @@ If you choose option 1, <code>*jpegSize</code> should be set to the size of your
 If you choose option 1, <code>dstSizes[i]</code> should be set to the size of your pre-allocated buffer. In any case, unless you have set <a class="el" href="group___turbo_j_p_e_g.html#ga8808d403c68b62aaa58a4c1e58e98963" title="Disable buffer (re)allocation.">TJFLAG_NOREALLOC</a>, you should always check <code>dstBufs[i]</code> upon return from this function, as it may have changed.</td></tr>
    <tr><td class="paramname">dstSizes</td><td>pointer to an array of n unsigned long variables that will receive the actual sizes (in bytes) of each transformed JPEG image. If <code>dstBufs[i]</code> points to a pre-allocated buffer, then <code>dstSizes[i]</code> should be set to the size of the buffer. Upon return, <code>dstSizes[i]</code> will contain the size of the JPEG image (in bytes.)</td></tr>
    <tr><td class="paramname">transforms</td><td>pointer to an array of n <a class="el" href="structtjtransform.html" title="Lossless transform.">tjtransform</a> structures, each of which specifies the transform parameters and/or cropping region for the corresponding transformed output image.</td></tr>
-    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#ga72ecf4ebe6eb702d3c6f5ca27455e1ec">flags</a></td></tr>
+    <tr><td class="paramname">flags</td><td>the bitwise OR of one or more of the <a class="el" href="group___turbo_j_p_e_g.html#gacb233cfd722d66d1ccbf48a7de81f0e0">flags</a></td></tr>
  </table>
  </dd>
 </dl>
--- a/doxygen-extra.css
+++ b/doxygen-extra.css
@@ -1,3 +1,3 @@
 code {
-	color: #4665A2; 
+	color: #4665A2;
 }
--- a/java/TJBench.java
+++ b/java/TJBench.java
@@ -721,37 +721,37 @@ class TJBench {
          if (argv[i].equalsIgnoreCase("-tile")) {
            doTile = true;  xformOpt |= TJTransform.OPT_CROP;
          }
-          if (argv[i].equalsIgnoreCase("-fastupsample")) {
+          else if (argv[i].equalsIgnoreCase("-fastupsample")) {
            System.out.println("Using fast upsampling code\n");
            flags |= TJ.FLAG_FASTUPSAMPLE;
          }
-          if (argv[i].equalsIgnoreCase("-fastdct")) {
+          else if (argv[i].equalsIgnoreCase("-fastdct")) {
            System.out.println("Using fastest DCT/IDCT algorithm\n");
            flags |= TJ.FLAG_FASTDCT;
          }
-          if (argv[i].equalsIgnoreCase("-accuratedct")) {
+          else if (argv[i].equalsIgnoreCase("-accuratedct")) {
            System.out.println("Using most accurate DCT/IDCT algorithm\n");
            flags |= TJ.FLAG_ACCURATEDCT;
          }
-          if (argv[i].equalsIgnoreCase("-rgb"))
+          else if (argv[i].equalsIgnoreCase("-rgb"))
            pf = TJ.PF_RGB;
-          if (argv[i].equalsIgnoreCase("-rgbx"))
+          else if (argv[i].equalsIgnoreCase("-rgbx"))
            pf = TJ.PF_RGBX;
-          if (argv[i].equalsIgnoreCase("-bgr"))
+          else if (argv[i].equalsIgnoreCase("-bgr"))
            pf = TJ.PF_BGR;
-          if (argv[i].equalsIgnoreCase("-bgrx"))
+          else if (argv[i].equalsIgnoreCase("-bgrx"))
            pf = TJ.PF_BGRX;
-          if (argv[i].equalsIgnoreCase("-xbgr"))
+          else if (argv[i].equalsIgnoreCase("-xbgr"))
            pf = TJ.PF_XBGR;
-          if (argv[i].equalsIgnoreCase("-xrgb"))
+          else if (argv[i].equalsIgnoreCase("-xrgb"))
            pf = TJ.PF_XRGB;
-          if (argv[i].equalsIgnoreCase("-bottomup"))
+          else if (argv[i].equalsIgnoreCase("-bottomup"))
            flags |= TJ.FLAG_BOTTOMUP;
-          if (argv[i].equalsIgnoreCase("-quiet"))
+          else if (argv[i].equalsIgnoreCase("-quiet"))
            quiet = 1;
-          if (argv[i].equalsIgnoreCase("-qq"))
+          else if (argv[i].equalsIgnoreCase("-qq"))
            quiet = 2;
-          if (argv[i].equalsIgnoreCase("-scale") && i < argv.length - 1) {
+          else if (argv[i].equalsIgnoreCase("-scale") && i < argv.length - 1) {
            int temp1 = 0, temp2 = 0;
            boolean match = false, scanned = true;
            Scanner scanner = new Scanner(argv[++i]).useDelimiter("/");
@@ -774,25 +774,25 @@ class TJBench {
            } else
              usage();
          }
-          if (argv[i].equalsIgnoreCase("-hflip"))
+          else if (argv[i].equalsIgnoreCase("-hflip"))
            xformOp = TJTransform.OP_HFLIP;
-          if (argv[i].equalsIgnoreCase("-vflip"))
+          else if (argv[i].equalsIgnoreCase("-vflip"))
            xformOp = TJTransform.OP_VFLIP;
-          if (argv[i].equalsIgnoreCase("-transpose"))
+          else if (argv[i].equalsIgnoreCase("-transpose"))
            xformOp = TJTransform.OP_TRANSPOSE;
-          if (argv[i].equalsIgnoreCase("-transverse"))
+          else if (argv[i].equalsIgnoreCase("-transverse"))
            xformOp = TJTransform.OP_TRANSVERSE;
-          if (argv[i].equalsIgnoreCase("-rot90"))
+          else if (argv[i].equalsIgnoreCase("-rot90"))
            xformOp = TJTransform.OP_ROT90;
-          if (argv[i].equalsIgnoreCase("-rot180"))
+          else if (argv[i].equalsIgnoreCase("-rot180"))
            xformOp = TJTransform.OP_ROT180;
-          if (argv[i].equalsIgnoreCase("-rot270"))
+          else if (argv[i].equalsIgnoreCase("-rot270"))
            xformOp = TJTransform.OP_ROT270;
-          if (argv[i].equalsIgnoreCase("-grayscale"))
+          else if (argv[i].equalsIgnoreCase("-grayscale"))
            xformOpt |= TJTransform.OPT_GRAY;
-          if (argv[i].equalsIgnoreCase("-nooutput"))
+          else if (argv[i].equalsIgnoreCase("-nooutput"))
            xformOpt |= TJTransform.OPT_NOOUTPUT;
-          if (argv[i].equalsIgnoreCase("-benchtime") && i < argv.length - 1) {
+          else if (argv[i].equalsIgnoreCase("-benchtime") && i < argv.length - 1) {
            double temp = -1;
            try {
              temp = Double.parseDouble(argv[++i]);
@@ -802,11 +802,11 @@ class TJBench {
            else
              usage();
          }
-          if (argv[i].equalsIgnoreCase("-yuv")) {
+          else if (argv[i].equalsIgnoreCase("-yuv")) {
            System.out.println("Testing YUV planar encoding/decoding\n");
            doYUV = true;
          }
-          if (argv[i].equalsIgnoreCase("-yuvpad") && i < argv.length - 1) {
+          else if (argv[i].equalsIgnoreCase("-yuvpad") && i < argv.length - 1) {
            int temp = 0;
            try {
             temp = Integer.parseInt(argv[++i]);
@@ -814,7 +814,7 @@ class TJBench {
            if (temp >= 1)
              yuvpad = temp;
          }
-          if (argv[i].equalsIgnoreCase("-subsamp") && i < argv.length - 1) {
+          else if (argv[i].equalsIgnoreCase("-subsamp") && i < argv.length - 1) {
            i++;
            if (argv[i].toUpperCase().startsWith("G"))
              subsamp = TJ.SAMP_GRAY;
@@ -829,11 +829,11 @@ class TJBench {
            else if (argv[i].equals("411"))
              subsamp = TJ.SAMP_411;
          }
-          if (argv[i].equalsIgnoreCase("-componly"))
+          else if (argv[i].equalsIgnoreCase("-componly"))
            compOnly = true;
-          if (argv[i].equalsIgnoreCase("-nowrite"))
+          else if (argv[i].equalsIgnoreCase("-nowrite"))
            write = false;
-          if (argv[i].equalsIgnoreCase("-warmup") && i < argv.length - 1) {
+          else if (argv[i].equalsIgnoreCase("-warmup") && i < argv.length - 1) {
            double temp = -1;
            try {
             temp = Double.parseDouble(argv[++i]);
@@ -844,8 +844,7 @@ class TJBench {
            } else
              usage();
          }
-          if (argv[i].equalsIgnoreCase("-?"))
-            usage();
+          else usage();
        }
      }

--- a/java/TJExample.java
+++ b/java/TJExample.java
@@ -1,5 +1,6 @@
 /*
- * Copyright (C)2011-2012, 2014-2015 D. R. Commander.  All Rights Reserved.
+ * Copyright (C)2011-2012, 2014-2015, 2017 D. R. Commander.
+ *                                         All Rights Reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
@@ -94,7 +95,7 @@ public class TJExample implements TJCustomFilter {
  }

  private static final String[] sampName = {
-    "4:4:4", "4:2:2", "4:2:0", "Grayscale", "4:4:0"
+    "4:4:4", "4:2:2", "4:2:0", "Grayscale", "4:4:0", "4:1:1"
  };

  public static void main(String[] argv) {
@@ -117,114 +118,114 @@ public class TJExample implements TJCustomFilter {
      int outSubsamp = -1, outQual = 95;
      boolean display = false;

-      if (argv.length > 1) {
-        for (int i = 1; i < argv.length; i++) {
-          if (argv[i].length() < 2)
-            continue;
-          if (argv[i].length() > 2 &&
-              argv[i].substring(0, 3).equalsIgnoreCase("-sc")) {
-            int match = 0;
-            if (i < argv.length - 1) {
-              String[] scaleArg = argv[++i].split("/");
-              if (scaleArg.length == 2) {
-                TJScalingFactor tempsf =
-                  new TJScalingFactor(Integer.parseInt(scaleArg[0]),
-                                      Integer.parseInt(scaleArg[1]));
-                for (int j = 0; j < sf.length; j++) {
-                  if (tempsf.equals(sf[j])) {
-                    scaleFactor = sf[j];
-                    match = 1;
-                    break;
-                  }
+      if (argv[1].substring(0, 2).equalsIgnoreCase("-d"))
+        display = true;
+
+      for (int i = 2; i < argv.length; i++) {
+        if (argv[i].length() < 2)
+          continue;
+        else if (argv[i].length() > 2 &&
+            argv[i].substring(0, 3).equalsIgnoreCase("-sc")) {
+          int match = 0;
+          if (i < argv.length - 1) {
+            String[] scaleArg = argv[++i].split("/");
+            if (scaleArg.length == 2) {
+              TJScalingFactor tempsf =
+                new TJScalingFactor(Integer.parseInt(scaleArg[0]),
+                                    Integer.parseInt(scaleArg[1]));
+              for (int j = 0; j < sf.length; j++) {
+                if (tempsf.equals(sf[j])) {
+                  scaleFactor = sf[j];
+                  match = 1;
+                  break;
                }
              }
            }
-            if (match != 1) usage();
-          }
-          if (argv[i].equalsIgnoreCase("-h") || argv[i].equalsIgnoreCase("-?"))
-            usage();
-          if (argv[i].length() > 2 &&
-              argv[i].substring(0, 3).equalsIgnoreCase("-sa")) {
-            if (i < argv.length - 1) {
-              i++;
-              if (argv[i].substring(0, 1).equalsIgnoreCase("g"))
-                outSubsamp = TJ.SAMP_GRAY;
-              else if (argv[i].equals("444"))
-                outSubsamp = TJ.SAMP_444;
-              else if (argv[i].equals("422"))
-                outSubsamp = TJ.SAMP_422;
-              else if (argv[i].equals("420"))
-                outSubsamp = TJ.SAMP_420;
-              else
-                usage();
-            } else
-              usage();
-          }
-          if (argv[i].substring(0, 2).equalsIgnoreCase("-q")) {
-            if (i < argv.length - 1) {
-              int qual = Integer.parseInt(argv[++i]);
-              if (qual >= 1 && qual <= 100)
-                outQual = qual;
-              else
-                usage();
-            } else
-              usage();
-          }
-          if (argv[i].substring(0, 2).equalsIgnoreCase("-g"))
-            xform.options |= TJTransform.OPT_GRAY;
-          if (argv[i].equalsIgnoreCase("-hflip"))
-            xform.op = TJTransform.OP_HFLIP;
-          if (argv[i].equalsIgnoreCase("-vflip"))
-            xform.op = TJTransform.OP_VFLIP;
-          if (argv[i].equalsIgnoreCase("-transpose"))
-            xform.op = TJTransform.OP_TRANSPOSE;
-          if (argv[i].equalsIgnoreCase("-transverse"))
-            xform.op = TJTransform.OP_TRANSVERSE;
-          if (argv[i].equalsIgnoreCase("-rot90"))
-            xform.op = TJTransform.OP_ROT90;
-          if (argv[i].equalsIgnoreCase("-rot180"))
-            xform.op = TJTransform.OP_ROT180;
-          if (argv[i].equalsIgnoreCase("-rot270"))
-            xform.op = TJTransform.OP_ROT270;
-          if (argv[i].equalsIgnoreCase("-custom"))
-            xform.cf = new TJExample();
-          else if (argv[i].length() > 2 &&
-                   argv[i].substring(0, 2).equalsIgnoreCase("-c")) {
-            if (i >= argv.length - 1)
-              usage();
-            String[] cropArg = argv[++i].split(",");
-            if (cropArg.length != 3)
-              usage();
-            String[] dimArg = cropArg[2].split("[xX]");
-            if (dimArg.length != 2)
-              usage();
-            int tempx = Integer.parseInt(cropArg[0]);
-            int tempy = Integer.parseInt(cropArg[1]);
-            int tempw = Integer.parseInt(dimArg[0]);
-            int temph = Integer.parseInt(dimArg[1]);
-            if (tempx < 0 || tempy < 0 || tempw < 0 || temph < 0)
-              usage();
-            xform.x = tempx;
-            xform.y = tempy;
-            xform.width = tempw;
-            xform.height = temph;
-            xform.options |= TJTransform.OPT_CROP;
-          }
-          if (argv[i].substring(0, 2).equalsIgnoreCase("-d"))
-            display = true;
-          if (argv[i].equalsIgnoreCase("-fastupsample")) {
-            System.out.println("Using fast upsampling code");
-            flags |= TJ.FLAG_FASTUPSAMPLE;
-          }
-          if (argv[i].equalsIgnoreCase("-fastdct")) {
-            System.out.println("Using fastest DCT/IDCT algorithm");
-            flags |= TJ.FLAG_FASTDCT;
-          }
-          if (argv[i].equalsIgnoreCase("-accuratedct")) {
-            System.out.println("Using most accurate DCT/IDCT algorithm");
-            flags |= TJ.FLAG_ACCURATEDCT;
          }
+          if (match != 1) usage();
        }
+        else if (argv[i].length() > 2 &&
+            argv[i].substring(0, 3).equalsIgnoreCase("-sa")) {
+          if (i < argv.length - 1) {
+            i++;
+            if (argv[i].substring(0, 1).equalsIgnoreCase("g"))
+              outSubsamp = TJ.SAMP_GRAY;
+            else if (argv[i].equals("444"))
+              outSubsamp = TJ.SAMP_444;
+            else if (argv[i].equals("422"))
+              outSubsamp = TJ.SAMP_422;
+            else if (argv[i].equals("420"))
+              outSubsamp = TJ.SAMP_420;
+            else
+              usage();
+          } else
+            usage();
+        }
+        else if (argv[i].substring(0, 2).equalsIgnoreCase("-q")) {
+          if (i < argv.length - 1) {
+            int qual = Integer.parseInt(argv[++i]);
+            if (qual >= 1 && qual <= 100)
+              outQual = qual;
+            else
+              usage();
+          } else
+            usage();
+        }
+        else if (argv[i].substring(0, 2).equalsIgnoreCase("-g"))
+          xform.options |= TJTransform.OPT_GRAY;
+        else if (argv[i].equalsIgnoreCase("-hflip"))
+          xform.op = TJTransform.OP_HFLIP;
+        else if (argv[i].equalsIgnoreCase("-vflip"))
+          xform.op = TJTransform.OP_VFLIP;
+        else if (argv[i].equalsIgnoreCase("-transpose"))
+          xform.op = TJTransform.OP_TRANSPOSE;
+        else if (argv[i].equalsIgnoreCase("-transverse"))
+          xform.op = TJTransform.OP_TRANSVERSE;
+        else if (argv[i].equalsIgnoreCase("-rot90"))
+          xform.op = TJTransform.OP_ROT90;
+        else if (argv[i].equalsIgnoreCase("-rot180"))
+          xform.op = TJTransform.OP_ROT180;
+        else if (argv[i].equalsIgnoreCase("-rot270"))
+          xform.op = TJTransform.OP_ROT270;
+        else if (argv[i].equalsIgnoreCase("-custom"))
+          xform.cf = new TJExample();
+        else if (argv[i].length() > 2 &&
+                 argv[i].substring(0, 2).equalsIgnoreCase("-c")) {
+          if (i >= argv.length - 1)
+            usage();
+          String[] cropArg = argv[++i].split(",");
+          if (cropArg.length != 3)
+            usage();
+          String[] dimArg = cropArg[2].split("[xX]");
+          if (dimArg.length != 2)
+            usage();
+          int tempx = Integer.parseInt(cropArg[0]);
+          int tempy = Integer.parseInt(cropArg[1]);
+          int tempw = Integer.parseInt(dimArg[0]);
+          int temph = Integer.parseInt(dimArg[1]);
+          if (tempx < 0 || tempy < 0 || tempw < 0 || temph < 0)
+            usage();
+          xform.x = tempx;
+          xform.y = tempy;
+          xform.width = tempw;
+          xform.height = temph;
+          xform.options |= TJTransform.OPT_CROP;
+        }
+        else if (argv[i].substring(0, 2).equalsIgnoreCase("-d"))
+          display = true;
+        else if (argv[i].equalsIgnoreCase("-fastupsample")) {
+          System.out.println("Using fast upsampling code");
+          flags |= TJ.FLAG_FASTUPSAMPLE;
+        }
+        else if (argv[i].equalsIgnoreCase("-fastdct")) {
+          System.out.println("Using fastest DCT/IDCT algorithm");
+          flags |= TJ.FLAG_FASTDCT;
+        }
+        else if (argv[i].equalsIgnoreCase("-accuratedct")) {
+          System.out.println("Using most accurate DCT/IDCT algorithm");
+          flags |= TJ.FLAG_ACCURATEDCT;
+        }
+        else usage();
      }
      String[] inFileTokens = argv[0].split("\\.");
      if (inFileTokens.length > 1)
--- a/java/TJUnitTest.java
+++ b/java/TJUnitTest.java
@@ -1,5 +1,5 @@
 /*
- * Copyright (C)2011-2016 D. R. Commander.  All Rights Reserved.
+ * Copyright (C)2011-2017 D. R. Commander.  All Rights Reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
@@ -44,10 +44,10 @@ public class TJUnitTest {

  private static void usage() {
    System.out.println("\nUSAGE: java " + classname + " [options]\n");
-    System.out.println("Options:\n");
-    System.out.println("-yuv = test YUV encoding/decoding support\n");
-    System.out.println("-noyuvpad = do not pad each line of each Y, U, and V plane to the nearest\n");
-    System.out.println("            4-byte boundary\n");
+    System.out.println("Options:");
+    System.out.println("-yuv = test YUV encoding/decoding support");
+    System.out.println("-noyuvpad = do not pad each line of each Y, U, and V plane to the nearest");
+    System.out.println("            4-byte boundary");
    System.out.println("-bi = test BufferedImage support\n");
    System.exit(1);
  }
@@ -109,21 +109,12 @@ public class TJUnitTest {
      case BufferedImage.TYPE_BYTE_GRAY:
        return TJ.PF_GRAY;
      case BufferedImage.TYPE_INT_BGR:
-        if (byteOrder == ByteOrder.BIG_ENDIAN)
-          return TJ.PF_XBGR;
-        else
-          return TJ.PF_RGBX;
+        return TJ.PF_RGBX;
      case BufferedImage.TYPE_INT_RGB:
-        if (byteOrder == ByteOrder.BIG_ENDIAN)
-          return TJ.PF_XRGB;
-        else
-          return TJ.PF_BGRX;
+        return TJ.PF_BGRX;
      case BufferedImage.TYPE_INT_ARGB:
      case BufferedImage.TYPE_INT_ARGB_PRE:
-        if (byteOrder == ByteOrder.BIG_ENDIAN)
-          return TJ.PF_ARGB;
-        else
-          return TJ.PF_BGRA;
+        return TJ.PF_BGRA;
    }
    return 0;
  }
@@ -911,15 +902,13 @@ public class TJUnitTest {
      for (int i = 0; i < argv.length; i++) {
        if (argv[i].equalsIgnoreCase("-yuv"))
          doYUV = true;
-        if (argv[i].equalsIgnoreCase("-noyuvpad"))
+        else if (argv[i].equalsIgnoreCase("-noyuvpad"))
          pad = 1;
-        if (argv[i].substring(0, 1).equalsIgnoreCase("-h") ||
-            argv[i].equalsIgnoreCase("-?"))
-          usage();
-        if (argv[i].equalsIgnoreCase("-bi")) {
+        else if (argv[i].equalsIgnoreCase("-bi")) {
          bi = true;
          testName = "javabitest";
-        }
+        } else
+          usage();
      }
      if (doYUV)
        _4byteFormats[4] = -1;
--- a/java/org/libjpegturbo/turbojpeg/YUVImage.java
+++ b/java/org/libjpegturbo/turbojpeg/YUVImage.java
@@ -1,5 +1,5 @@
 /*
- * Copyright (C)2014 D. R. Commander.  All Rights Reserved.
+ * Copyright (C)2014, 2017 D. R. Commander.  All Rights Reserved.
 * Copyright (C)2015 Viktor Szathmáry.  All Rights Reserved.
 *
 * Redistribution and use in source and binary forms, with or without
@@ -220,10 +220,13 @@ public class YUVImage {
      throw new IllegalArgumentException("Invalid argument in YUVImage::setBuf()");

    int nc = (subsamp == TJ.SAMP_GRAY ? 1 : 3);
-    if (planes.length != nc || (offsets != null && offsets.length != nc) ||
+    if ((planes != null && planes.length != nc) ||
+        (offsets != null && offsets.length != nc) ||
        (strides != null && strides.length != nc))
      throw new IllegalArgumentException("YUVImage::setBuf(): planes, offsets, or strides array is the wrong size");

+    if (planes == null)
+      planes = new byte[nc][];
    if (offsets == null)
      offsets = new int[nc];
    if (strides == null)
--- a/jcdctmgr.c
+++ b/jcdctmgr.c
@@ -216,7 +216,7 @@ compute_reciprocal (UINT16 divisor, DCTELEM *dtbl)
 #endif
  dtbl[DCTSIZE2 * 3] = (DCTELEM) r - sizeof(DCTELEM)*8; /* shift */

-  if(r <= 16) return 0;
+  if (r <= 16) return 0;
  else return 1;
 }

--- a/jcstest.c
+++ b/jcstest.c
@@ -77,7 +77,7 @@ int main(void)
  jerr.pub.error_exit = my_error_exit;
  jerr.pub.output_message = my_output_message;

-  if(setjmp(jerr.jb)) {
+  if (setjmp(jerr.jb)) {
    /* this will execute if libjpeg has an error */
    jcs_valid = 0;
    goto done;
@@ -104,7 +104,7 @@ int main(void)
  printf("  Not present at compile time\n");
  #endif

-  if(setjmp(jerr.jb)) {
+  if (setjmp(jerr.jb)) {
    /* this will execute if libjpeg has an error */
    jcs_alpha_valid = 0;
    goto done2;
--- a/jdapistd.c
+++ b/jdapistd.c
@@ -4,7 +4,7 @@
 * This file was part of the Independent JPEG Group's software:
 * Copyright (C) 1994-1996, Thomas G. Lane.
 * libjpeg-turbo Modifications:
- * Copyright (C) 2010, 2015-2016, D. R. Commander.
+ * Copyright (C) 2010, 2015-2017, D. R. Commander.
 * Copyright (C) 2015, Google, Inc.
 * For conditions of distribution and use, see the accompanying README.ijg
 * file.
@@ -190,7 +190,10 @@ jpeg_crop_scanline (j_decompress_ptr cinfo, JDIMENSION *xoffset,
   * single-pass decompression case, allowing us to use the same MCU column
   * width for all of the components.
   */
-  align = cinfo->_min_DCT_scaled_size * cinfo->max_h_samp_factor;
+  if (cinfo->comps_in_scan == 1 && cinfo->num_components == 1)
+    align = cinfo->_min_DCT_scaled_size;
+  else
+    align = cinfo->_min_DCT_scaled_size * cinfo->max_h_samp_factor;

  /* Adjust xoffset to the nearest iMCU boundary <= the requested value */
  input_xoffset = *xoffset;
@@ -215,6 +218,9 @@ jpeg_crop_scanline (j_decompress_ptr cinfo, JDIMENSION *xoffset,

  for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components;
       ci++, compptr++) {
+    int hsf = (cinfo->comps_in_scan == 1 && cinfo->num_components == 1) ?
+              1 : compptr->h_samp_factor;
+
    /* Set downsampled_width to the new output width. */
    orig_downsampled_width = compptr->downsampled_width;
    compptr->downsampled_width =
@@ -228,11 +234,10 @@ jpeg_crop_scanline (j_decompress_ptr cinfo, JDIMENSION *xoffset,
     * values will be used in multi-scan decompressions.
     */
    cinfo->master->first_MCU_col[ci] =
-      (JDIMENSION) (long) (*xoffset * compptr->h_samp_factor) /
-                   (long) align;
+      (JDIMENSION) (long) (*xoffset * hsf) / (long) align;
    cinfo->master->last_MCU_col[ci] =
      (JDIMENSION) jdiv_round_up((long) ((*xoffset + cinfo->output_width) *
-                                         compptr->h_samp_factor),
+                                         hsf),
                                 (long) align) - 1;
  }

@@ -293,6 +298,14 @@ noop_convert (j_decompress_ptr cinfo, JSAMPIMAGE input_buf,
 }


+/* Dummy quantize function used by jpeg_skip_scanlines() */
+LOCAL(void)
+noop_quantize (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
+               JSAMPARRAY output_buf, int num_rows)
+{
+}
+
+
 /*
 * In some cases, it is best to call jpeg_read_scanlines() and discard the
 * output, rather than skipping the scanlines, because this allows us to
@@ -308,14 +321,22 @@ read_and_discard_scanlines (j_decompress_ptr cinfo, JDIMENSION num_lines)
  void (*color_convert) (j_decompress_ptr cinfo, JSAMPIMAGE input_buf,
                         JDIMENSION input_row, JSAMPARRAY output_buf,
                         int num_rows);
+  void (*color_quantize) (j_decompress_ptr cinfo, JSAMPARRAY input_buf,
+                          JSAMPARRAY output_buf, int num_rows) = NULL;

  color_convert = cinfo->cconvert->color_convert;
  cinfo->cconvert->color_convert = noop_convert;
+  if (cinfo->cquantize && cinfo->cquantize->color_quantize) {
+    color_quantize = cinfo->cquantize->color_quantize;
+    cinfo->cquantize->color_quantize = noop_quantize;
+  }

  for (n = 0; n < num_lines; n++)
    jpeg_read_scanlines(cinfo, NULL, 1);

  cinfo->cconvert->color_convert = color_convert;
+  if (color_quantize)
+    cinfo->cquantize->color_quantize = color_quantize;
 }


@@ -370,6 +391,8 @@ jpeg_skip_scanlines (j_decompress_ptr cinfo, JDIMENSION num_lines)
  /* Do not skip past the bottom of the image. */
  if (cinfo->output_scanline + num_lines >= cinfo->output_height) {
    cinfo->output_scanline = cinfo->output_height;
+    (*cinfo->inputctl->finish_input_pass) (cinfo);
+    cinfo->inputctl->eoi_reached = TRUE;
    return cinfo->output_height - cinfo->output_scanline;
  }

--- a/jdatadst-tj.c
+++ b/jdatadst-tj.c
@@ -130,7 +130,7 @@ term_mem_destination (j_compress_ptr cinfo)
 {
  my_mem_dest_ptr dest = (my_mem_dest_ptr) cinfo->dest;

-  if(dest->alloc) *dest->outbuffer = dest->buffer;
+  if (dest->alloc) *dest->outbuffer = dest->buffer;
  *dest->outsize = (unsigned long)(dest->bufsize - dest->pub.free_in_buffer);
 }

--- a/jdcolor.c
+++ b/jdcolor.c
@@ -616,7 +616,7 @@ static const JLONG dither_matrix[4] = {
 static INLINE boolean is_big_endian(void)
 {
  int test_value = 1;
-  if(*(char *)&test_value != 1)
+  if (*(char *)&test_value != 1)
    return TRUE;
  return FALSE;
 }
--- a/jdmerge.c
+++ b/jdmerge.c
@@ -503,7 +503,7 @@ static const JLONG dither_matrix[4] = {
 static INLINE boolean is_big_endian(void)
 {
  int test_value = 1;
-  if(*(char *)&test_value != 1)
+  if (*(char *)&test_value != 1)
    return TRUE;
  return FALSE;
 }
--- a/release/libjpeg-turbo.spec.in
+++ b/release/libjpeg-turbo.spec.in
@@ -13,12 +13,22 @@
 # Path under which headers should be installed
 %define _includedir %{__includedir}

-# _libdir is set to %{_prefix}/%{_lib} by default
-%ifarch x86_64
-%define _lib lib64
+%if "%{?__isa_bits:1}" == "1"
+%define _bits %{__isa_bits}
+%else
+# RPM < 4.6
+%if "%{_lib}" == "lib64"
+%define _bits 64
+%else
+%define _bits 32
+%endif
+%endif
+
+%if "%{_bits}" == "64"
+%define _libdir %{_exec_prefix}/lib64
 %else
 %if "%{_prefix}" == "/opt/libjpeg-turbo"
-%define _lib lib32
+%define _libdir %{_exec_prefix}/lib32
 %endif
 %endif

@@ -36,7 +46,7 @@ Release: @BUILD@
 License: BSD-style
 BuildRoot: %{_blddir}/%{name}-buildroot-%{version}-%{release}
 Prereq: /sbin/ldconfig
-%ifarch x86_64
+%if "%{_bits}" == "64"
 Provides: %{name} = %{version}-%{release}, @PACKAGE_NAME@ = %{version}-%{release}, libturbojpeg.so()(64bit)
 %else
 Provides: %{name} = %{version}-%{release}, @PACKAGE_NAME@ = %{version}-%{release}, libturbojpeg.so
@@ -73,7 +83,8 @@ broader range of users and developers.
 #-->	mandir=%{_mandir} JPEG_LIB_VERSION=@JPEG_LIB_VERSION@ \
 #-->	SO_MAJOR_VERSION=@SO_MAJOR_VERSION@ SO_MINOR_VERSION=@SO_MINOR_VERSION@ \
 #-->	--with-pic @RPM_CONFIG_ARGS@
-#-->make DESTDIR=$RPM_BUILD_ROOT
+#-->export NUMCPUS=`grep -c '^processor' /proc/cpuinfo`
+#-->make -j$NUMCPUS --load-average=$NUMCPUS DESTDIR=$RPM_BUILD_ROOT

 %install

@@ -86,7 +97,7 @@ rm -f $RPM_BUILD_ROOT%{_libdir}/*.la

 LJT_LIBDIR=%{__libdir}
 if [ ! "$LJT_LIBDIR" = "%{_libdir}" ]; then
-	echo ERROR: libjpeg-turbo must be configured with libdir=%{_prefix}/%{_lib} when generating an in-tree RPM for this architecture.
+	echo ERROR: libjpeg-turbo must be configured with libdir=%{_libdir} when generating an in-tree RPM for this architecture.
 	exit 1
 fi

--- a/simd/jsimd_mips_dspr2.S
+++ b/simd/jsimd_mips_dspr2.S
@@ -4484,4 +4484,3 @@ LEAF_MIPS_DSPR2(jsimd_convsamp_float_mips_dspr2)
 END(jsimd_convsamp_float_mips_dspr2)

 /*****************************************************************************/
-
--- a/simd/jsimd_mips_dspr2_asm.h
+++ b/simd/jsimd_mips_dspr2_asm.h
@@ -281,5 +281,3 @@ LEAF_MIPS32R2(symbol)                                   \
    addiu           sp, sp, \stack_offset
    .endif
 .endm
-
-
--- a/tjbench.c
+++ b/tjbench.c
@@ -502,7 +502,7 @@ int decompTest(char *filename)
 	char *temp=NULL, tempstr[80], tempstr2[80];
 	int row, col, i, iter, tilew, tileh, ntilesw=1, ntilesh=1, retval=0;
 	double start, elapsed;
-	int ps=tjPixelSize[pf], tile;
+	int ps=tjPixelSize[pf], tile, decompsrc=0;

 	if((file=fopen(filename, "rb"))==NULL)
 		_throwunix("opening file");
@@ -682,18 +682,17 @@ int decompTest(char *filename)
 		else
 		{
 			if(quiet==1) printf("N/A     N/A     ");
-			jpegsize[0]=srcsize;
-			free(jpegbuf[0]);
-			jpegbuf[0]=srcbuf;
-			srcbuf=NULL;
+			tjFree(jpegbuf[0]);
+			jpegbuf[0]=NULL;
+			decompsrc=1;
 		}

 		if(w==tilew) _tilew=_w;
 		if(h==tileh) _tileh=_h;
 		if(!(xformopt&TJXOPT_NOOUTPUT))
 		{
-			if(decomp(NULL, jpegbuf, jpegsize, NULL, _w, _h, _subsamp, 0,
-				filename, _tilew, _tileh)==-1)
+			if(decomp(NULL, decompsrc? &srcbuf:jpegbuf, decompsrc? &srcsize:jpegsize,
+					NULL, _w, _h, _subsamp, 0, filename, _tilew, _tileh)==-1)
 				goto bailout;
 		}
 		else if(quiet==1) printf("N/A\n");
@@ -835,32 +834,32 @@ int main(int argc, char *argv[])
 			{
 				dotile=1;  xformopt|=TJXOPT_CROP;
 			}
-			if(!strcasecmp(argv[i], "-fastupsample"))
+			else if(!strcasecmp(argv[i], "-fastupsample"))
 			{
 				printf("Using fast upsampling code\n\n");
 				flags|=TJFLAG_FASTUPSAMPLE;
 			}
-			if(!strcasecmp(argv[i], "-fastdct"))
+			else if(!strcasecmp(argv[i], "-fastdct"))
 			{
 				printf("Using fastest DCT/IDCT algorithm\n\n");
 				flags|=TJFLAG_FASTDCT;
 			}
-			if(!strcasecmp(argv[i], "-accuratedct"))
+			else if(!strcasecmp(argv[i], "-accuratedct"))
 			{
 				printf("Using most accurate DCT/IDCT algorithm\n\n");
 				flags|=TJFLAG_ACCURATEDCT;
 			}
-			if(!strcasecmp(argv[i], "-rgb")) pf=TJPF_RGB;
-			if(!strcasecmp(argv[i], "-rgbx")) pf=TJPF_RGBX;
-			if(!strcasecmp(argv[i], "-bgr")) pf=TJPF_BGR;
-			if(!strcasecmp(argv[i], "-bgrx")) pf=TJPF_BGRX;
-			if(!strcasecmp(argv[i], "-xbgr")) pf=TJPF_XBGR;
-			if(!strcasecmp(argv[i], "-xrgb")) pf=TJPF_XRGB;
-			if(!strcasecmp(argv[i], "-cmyk")) pf=TJPF_CMYK;
-			if(!strcasecmp(argv[i], "-bottomup")) flags|=TJFLAG_BOTTOMUP;
-			if(!strcasecmp(argv[i], "-quiet")) quiet=1;
-			if(!strcasecmp(argv[i], "-qq")) quiet=2;
-			if(!strcasecmp(argv[i], "-scale") && i<argc-1)
+			else if(!strcasecmp(argv[i], "-rgb")) pf=TJPF_RGB;
+			else if(!strcasecmp(argv[i], "-rgbx")) pf=TJPF_RGBX;
+			else if(!strcasecmp(argv[i], "-bgr")) pf=TJPF_BGR;
+			else if(!strcasecmp(argv[i], "-bgrx")) pf=TJPF_BGRX;
+			else if(!strcasecmp(argv[i], "-xbgr")) pf=TJPF_XBGR;
+			else if(!strcasecmp(argv[i], "-xrgb")) pf=TJPF_XRGB;
+			else if(!strcasecmp(argv[i], "-cmyk")) pf=TJPF_CMYK;
+			else if(!strcasecmp(argv[i], "-bottomup")) flags|=TJFLAG_BOTTOMUP;
+			else if(!strcasecmp(argv[i], "-quiet")) quiet=1;
+			else if(!strcasecmp(argv[i], "-qq")) quiet=2;
+			else if(!strcasecmp(argv[i], "-scale") && i<argc-1)
 			{
 				int temp1=0, temp2=0, match=0;
 				if(sscanf(argv[++i], "%d/%d", &temp1, &temp2)==2)
@@ -878,43 +877,42 @@ int main(int argc, char *argv[])
 				}
 				else usage(argv[0]);
 			}
-			if(!strcasecmp(argv[i], "-hflip")) xformop=TJXOP_HFLIP;
-			if(!strcasecmp(argv[i], "-vflip")) xformop=TJXOP_VFLIP;
-			if(!strcasecmp(argv[i], "-transpose")) xformop=TJXOP_TRANSPOSE;
-			if(!strcasecmp(argv[i], "-transverse")) xformop=TJXOP_TRANSVERSE;
-			if(!strcasecmp(argv[i], "-rot90")) xformop=TJXOP_ROT90;
-			if(!strcasecmp(argv[i], "-rot180")) xformop=TJXOP_ROT180;
-			if(!strcasecmp(argv[i], "-rot270")) xformop=TJXOP_ROT270;
-			if(!strcasecmp(argv[i], "-grayscale")) xformopt|=TJXOPT_GRAY;
-			if(!strcasecmp(argv[i], "-custom")) customFilter=dummyDCTFilter;
-			if(!strcasecmp(argv[i], "-nooutput")) xformopt|=TJXOPT_NOOUTPUT;
-			if(!strcasecmp(argv[i], "-benchtime") && i<argc-1)
+			else if(!strcasecmp(argv[i], "-hflip")) xformop=TJXOP_HFLIP;
+			else if(!strcasecmp(argv[i], "-vflip")) xformop=TJXOP_VFLIP;
+			else if(!strcasecmp(argv[i], "-transpose")) xformop=TJXOP_TRANSPOSE;
+			else if(!strcasecmp(argv[i], "-transverse")) xformop=TJXOP_TRANSVERSE;
+			else if(!strcasecmp(argv[i], "-rot90")) xformop=TJXOP_ROT90;
+			else if(!strcasecmp(argv[i], "-rot180")) xformop=TJXOP_ROT180;
+			else if(!strcasecmp(argv[i], "-rot270")) xformop=TJXOP_ROT270;
+			else if(!strcasecmp(argv[i], "-grayscale")) xformopt|=TJXOPT_GRAY;
+			else if(!strcasecmp(argv[i], "-custom")) customFilter=dummyDCTFilter;
+			else if(!strcasecmp(argv[i], "-nooutput")) xformopt|=TJXOPT_NOOUTPUT;
+			else if(!strcasecmp(argv[i], "-benchtime") && i<argc-1)
 			{
 				double temp=atof(argv[++i]);
 				if(temp>0.0) benchtime=temp;
 				else usage(argv[0]);
 			}
-			if(!strcasecmp(argv[i], "-warmup") && i<argc-1)
+			else if(!strcasecmp(argv[i], "-warmup") && i<argc-1)
 			{
 				double temp=atof(argv[++i]);
 				if(temp>=0.0) warmup=temp;
 				else usage(argv[0]);
 				printf("Warmup time = %.1f seconds\n\n", warmup);
 			}
-			if(!strcmp(argv[i], "-?")) usage(argv[0]);
-			if(!strcasecmp(argv[i], "-alloc")) flags&=(~TJFLAG_NOREALLOC);
-			if(!strcasecmp(argv[i], "-bmp")) ext="bmp";
-			if(!strcasecmp(argv[i], "-yuv"))
+			else if(!strcasecmp(argv[i], "-alloc")) flags&=(~TJFLAG_NOREALLOC);
+			else if(!strcasecmp(argv[i], "-bmp")) ext="bmp";
+			else if(!strcasecmp(argv[i], "-yuv"))
 			{
 				printf("Testing YUV planar encoding/decoding\n\n");
 				doyuv=1;
 			}
-			if(!strcasecmp(argv[i], "-yuvpad") && i<argc-1)
+			else if(!strcasecmp(argv[i], "-yuvpad") && i<argc-1)
 			{
 				int temp=atoi(argv[++i]);
 				if(temp>=1) yuvpad=temp;
 			}
-			if(!strcasecmp(argv[i], "-subsamp") && i<argc-1)
+			else if(!strcasecmp(argv[i], "-subsamp") && i<argc-1)
 			{
 				i++;
 				if(toupper(argv[i][0])=='G') subsamp=TJSAMP_GRAY;
@@ -931,8 +929,9 @@ int main(int argc, char *argv[])
 					}
 				}
 			}
-			if(!strcasecmp(argv[i], "-componly")) componly=1;
-			if(!strcasecmp(argv[i], "-nowrite")) dowrite=0;
+			else if(!strcasecmp(argv[i], "-componly")) componly=1;
+			else if(!strcasecmp(argv[i], "-nowrite")) dowrite=0;
+			else usage(argv[0]);
 		}
 	}

--- a/tjbenchtest.in
+++ b/tjbenchtest.in
@@ -36,10 +36,9 @@ if [ -d $OUTDIR ]; then
 fi
 mkdir -p $OUTDIR

-exec >$EXEDIR/tjbenchtest.log
-
-if [ $# -gt 0 ]; then
-	if [ "$1" = "-yuv" ]; then
+while [ $# -gt 0 ]; do
+	case "$1" in
+	-yuv)
 		NSARG=-nosmooth
 		YUVARG=-yuv

@@ -60,12 +59,16 @@ if [ $# -gt 0 ]; then
 # phenomenon is not yet fully understood but is also believed to be some sort
 # of round-off error.)
 		IMAGES="vgl_6548_0026a.${EXT}"
-	fi
-	if [ "$1" = "-alloc" ]; then
+		;;
+	-alloc)
 		ALLOCARG=-alloc
 		ALLOC=1
-	fi
-fi
+		;;
+	esac
+	shift
+done
+
+exec >$EXEDIR/tjbenchtest$YUVARG$ALLOCARG.log

 # Standard tests
 for image in $IMAGES; do
--- a/tjbenchtest.java.in
+++ b/tjbenchtest.java.in
@@ -33,8 +33,6 @@ if [ -d $OUTDIR ]; then
 fi
 mkdir -p $OUTDIR

-exec >$EXEDIR/tjbenchtest-java.log
-
 if [ $# -gt 0 ]; then
 	if [ "$1" = "-yuv" ]; then
 		NSARG=-nosmooth
@@ -60,6 +58,8 @@ if [ $# -gt 0 ]; then
 	fi
 fi

+exec >$EXEDIR/tjbenchtest-java$YUVARG.log
+
 # Standard tests
 for image in $IMAGES; do

--- a/tjexampletest.in
+++ b/tjexampletest.in
@@ -21,7 +21,7 @@ runme()

 IMAGES="vgl_5674_0098.bmp vgl_6434_0018a.bmp vgl_6548_0026a.bmp nightshot_iso_100.bmp"
 IMGDIR=@srcdir@/testimages
-OUTDIR=__tjexampletest_output
+OUTDIR=`mktemp -d /tmp/__tjexampletest_output.XXXXXX`
 EXEDIR=.
 JAVA="@JAVA@ -cp java/turbojpeg.jar -Djava.library.path=.libs"

@@ -36,23 +36,23 @@ for image in $IMAGES; do

 	cp $IMGDIR/$image $OUTDIR
 	basename=`basename $image .bmp`
-	$EXEDIR/cjpeg -quality 95 -dct fast -grayscale $IMGDIR/${basename}.bmp >$OUTDIR/${basename}_GRAY_fast_cjpeg.jpg
-	$EXEDIR/cjpeg -quality 95 -dct fast -sample 2x2 $IMGDIR/${basename}.bmp >$OUTDIR/${basename}_420_fast_cjpeg.jpg
-	$EXEDIR/cjpeg -quality 95 -dct fast -sample 2x1 $IMGDIR/${basename}.bmp >$OUTDIR/${basename}_422_fast_cjpeg.jpg
-	$EXEDIR/cjpeg -quality 95 -dct fast -sample 1x1 $IMGDIR/${basename}.bmp >$OUTDIR/${basename}_444_fast_cjpeg.jpg
-	$EXEDIR/cjpeg -quality 95 -dct int -grayscale $IMGDIR/${basename}.bmp >$OUTDIR/${basename}_GRAY_accurate_cjpeg.jpg
-	$EXEDIR/cjpeg -quality 95 -dct int -sample 2x2 $IMGDIR/${basename}.bmp >$OUTDIR/${basename}_420_accurate_cjpeg.jpg
-	$EXEDIR/cjpeg -quality 95 -dct int -sample 2x1 $IMGDIR/${basename}.bmp >$OUTDIR/${basename}_422_accurate_cjpeg.jpg
-	$EXEDIR/cjpeg -quality 95 -dct int -sample 1x1 $IMGDIR/${basename}.bmp >$OUTDIR/${basename}_444_accurate_cjpeg.jpg
+	runme $EXEDIR/cjpeg -quality 95 -dct fast -grayscale -outfile $OUTDIR/${basename}_GRAY_fast_cjpeg.jpg $IMGDIR/${basename}.bmp
+	runme $EXEDIR/cjpeg -quality 95 -dct fast -sample 2x2 -outfile $OUTDIR/${basename}_420_fast_cjpeg.jpg $IMGDIR/${basename}.bmp
+	runme $EXEDIR/cjpeg -quality 95 -dct fast -sample 2x1 -outfile $OUTDIR/${basename}_422_fast_cjpeg.jpg $IMGDIR/${basename}.bmp
+	runme $EXEDIR/cjpeg -quality 95 -dct fast -sample 1x1 -outfile $OUTDIR/${basename}_444_fast_cjpeg.jpg $IMGDIR/${basename}.bmp
+	runme $EXEDIR/cjpeg -quality 95 -dct int -grayscale -outfile $OUTDIR/${basename}_GRAY_accurate_cjpeg.jpg $IMGDIR/${basename}.bmp
+	runme $EXEDIR/cjpeg -quality 95 -dct int -sample 2x2 -outfile $OUTDIR/${basename}_420_accurate_cjpeg.jpg $IMGDIR/${basename}.bmp
+	runme $EXEDIR/cjpeg -quality 95 -dct int -sample 2x1 -outfile $OUTDIR/${basename}_422_accurate_cjpeg.jpg $IMGDIR/${basename}.bmp
+	runme $EXEDIR/cjpeg -quality 95 -dct int -sample 1x1 -outfile $OUTDIR/${basename}_444_accurate_cjpeg.jpg $IMGDIR/${basename}.bmp
 	for samp in GRAY 420 422 444; do
-		$EXEDIR/djpeg -rgb -bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg >$OUTDIR/${basename}_${samp}_default_djpeg.bmp
-		$EXEDIR/djpeg -dct fast -rgb -bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg >$OUTDIR/${basename}_${samp}_fast_djpeg.bmp
-		$EXEDIR/djpeg -dct int -rgb -bmp $OUTDIR/${basename}_${samp}_accurate_cjpeg.jpg >$OUTDIR/${basename}_${samp}_accurate_djpeg.bmp
+		runme $EXEDIR/djpeg -rgb -bmp -outfile $OUTDIR/${basename}_${samp}_default_djpeg.bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg
+		runme $EXEDIR/djpeg -dct fast -rgb -bmp -outfile $OUTDIR/${basename}_${samp}_fast_djpeg.bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg
+		runme $EXEDIR/djpeg -dct int -rgb -bmp -outfile $OUTDIR/${basename}_${samp}_accurate_djpeg.bmp $OUTDIR/${basename}_${samp}_accurate_cjpeg.jpg
 	done
 	for samp in 420 422; do
-		$EXEDIR/djpeg -nosmooth -bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg >$OUTDIR/${basename}_${samp}_default_nosmooth_djpeg.bmp
-		$EXEDIR/djpeg -dct fast -nosmooth -bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg >$OUTDIR/${basename}_${samp}_fast_nosmooth_djpeg.bmp
-		$EXEDIR/djpeg -dct int -nosmooth -bmp $OUTDIR/${basename}_${samp}_accurate_cjpeg.jpg >$OUTDIR/${basename}_${samp}_accurate_nosmooth_djpeg.bmp
+		runme $EXEDIR/djpeg -nosmooth -bmp -outfile $OUTDIR/${basename}_${samp}_default_nosmooth_djpeg.bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg
+		runme $EXEDIR/djpeg -dct fast -nosmooth -bmp -outfile $OUTDIR/${basename}_${samp}_fast_nosmooth_djpeg.bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg
+		runme $EXEDIR/djpeg -dct int -nosmooth -bmp -outfile $OUTDIR/${basename}_${samp}_accurate_nosmooth_djpeg.bmp $OUTDIR/${basename}_${samp}_accurate_cjpeg.jpg
 	done

 	# Compression
@@ -87,7 +87,7 @@ for image in $IMAGES; do
 	for scale in 2_1 15_8 7_4 13_8 3_2 11_8 5_4 9_8 7_8 3_4 5_8 1_2 3_8 1_4 1_8; do
 		scalearg=`echo $scale | sed s@_@/@g`
 		for samp in GRAY 420 422 444; do
-			$EXEDIR/djpeg -rgb -bmp -scale ${scalearg} $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg >$OUTDIR/${basename}_${samp}_${scale}_djpeg.bmp
+			runme $EXEDIR/djpeg -rgb -bmp -scale ${scalearg} -outfile $OUTDIR/${basename}_${samp}_${scale}_djpeg.bmp $OUTDIR/${basename}_${samp}_fast_cjpeg.jpg
 			runme $JAVA TJExample $OUTDIR/${basename}_${samp}_fast.jpg $OUTDIR/${basename}_${samp}_${scale}.bmp -scale ${scalearg}
 			runme cmp -i 54:54 $OUTDIR/${basename}_${samp}_${scale}.bmp $OUTDIR/${basename}_${samp}_${scale}_djpeg.bmp
 			rm $OUTDIR/${basename}_${samp}_${scale}.bmp
@@ -96,25 +96,25 @@ for image in $IMAGES; do

 	# Transforms
 	for samp in GRAY 420 422 444; do
-		$EXEDIR/jpegtran -crop 70x60+16+16 -flip horizontal -trim $OUTDIR/${basename}_${samp}_fast.jpg >$OUTDIR/${basename}_${samp}_hflip_jpegtran.jpg
-		$EXEDIR/jpegtran -crop 70x60+16+16 -flip vertical -trim $OUTDIR/${basename}_${samp}_fast.jpg >$OUTDIR/${basename}_${samp}_vflip_jpegtran.jpg
-		$EXEDIR/jpegtran -crop 70x60+16+16 -transpose -trim $OUTDIR/${basename}_${samp}_fast.jpg >$OUTDIR/${basename}_${samp}_transpose_jpegtran.jpg
-		$EXEDIR/jpegtran -crop 70x60+16+16 -transverse -trim $OUTDIR/${basename}_${samp}_fast.jpg >$OUTDIR/${basename}_${samp}_transverse_jpegtran.jpg
-		$EXEDIR/jpegtran -crop 70x60+16+16 -rotate 90 -trim $OUTDIR/${basename}_${samp}_fast.jpg >$OUTDIR/${basename}_${samp}_rot90_jpegtran.jpg
-		$EXEDIR/jpegtran -crop 70x60+16+16 -rotate 180 -trim $OUTDIR/${basename}_${samp}_fast.jpg >$OUTDIR/${basename}_${samp}_rot180_jpegtran.jpg
-		$EXEDIR/jpegtran -crop 70x60+16+16 -rotate 270 -trim $OUTDIR/${basename}_${samp}_fast.jpg >$OUTDIR/${basename}_${samp}_rot270_jpegtran.jpg
+		runme $EXEDIR/jpegtran -crop 70x60+16+16 -flip horizontal -trim -outfile $OUTDIR/${basename}_${samp}_hflip_jpegtran.jpg $OUTDIR/${basename}_${samp}_fast.jpg
+		runme $EXEDIR/jpegtran -crop 70x60+16+16 -flip vertical -trim -outfile $OUTDIR/${basename}_${samp}_vflip_jpegtran.jpg $OUTDIR/${basename}_${samp}_fast.jpg
+		runme $EXEDIR/jpegtran -crop 70x60+16+16 -transpose -trim -outfile $OUTDIR/${basename}_${samp}_transpose_jpegtran.jpg $OUTDIR/${basename}_${samp}_fast.jpg
+		runme $EXEDIR/jpegtran -crop 70x60+16+16 -transverse -trim -outfile $OUTDIR/${basename}_${samp}_transverse_jpegtran.jpg $OUTDIR/${basename}_${samp}_fast.jpg
+		runme $EXEDIR/jpegtran -crop 70x60+16+16 -rotate 90 -trim -outfile $OUTDIR/${basename}_${samp}_rot90_jpegtran.jpg $OUTDIR/${basename}_${samp}_fast.jpg
+		runme $EXEDIR/jpegtran -crop 70x60+16+16 -rotate 180 -trim -outfile $OUTDIR/${basename}_${samp}_rot180_jpegtran.jpg $OUTDIR/${basename}_${samp}_fast.jpg
+		runme $EXEDIR/jpegtran -crop 70x60+16+16 -rotate 270 -trim -outfile $OUTDIR/${basename}_${samp}_rot270_jpegtran.jpg $OUTDIR/${basename}_${samp}_fast.jpg
 	done
 	for xform in hflip vflip transpose transverse rot90 rot180 rot270; do
 		for samp in GRAY 420 422 444; do
 			runme $JAVA TJExample $OUTDIR/${basename}_${samp}_fast.jpg $OUTDIR/${basename}_${samp}_${xform}.jpg -$xform -crop 16,16,70x60
 			runme cmp $OUTDIR/${basename}_${samp}_${xform}.jpg $OUTDIR/${basename}_${samp}_${xform}_jpegtran.jpg
-			$EXEDIR/djpeg -rgb -bmp $OUTDIR/${basename}_${samp}_${xform}_jpegtran.jpg >$OUTDIR/${basename}_${samp}_${xform}_jpegtran.bmp
+			runme $EXEDIR/djpeg -rgb -bmp -outfile $OUTDIR/${basename}_${samp}_${xform}_jpegtran.bmp $OUTDIR/${basename}_${samp}_${xform}_jpegtran.jpg
 			runme $JAVA TJExample $OUTDIR/${basename}_${samp}_fast.jpg $OUTDIR/${basename}_${samp}_${xform}.bmp -$xform -crop 16,16,70x60
 			runme cmp -i 54:54 $OUTDIR/${basename}_${samp}_${xform}.bmp $OUTDIR/${basename}_${samp}_${xform}_jpegtran.bmp
 			rm $OUTDIR/${basename}_${samp}_${xform}.bmp
 		done
 		for samp in 420 422; do
-			$EXEDIR/djpeg -nosmooth -rgb -bmp $OUTDIR/${basename}_${samp}_${xform}_jpegtran.jpg >$OUTDIR/${basename}_${samp}_${xform}_jpegtran.bmp
+			runme $EXEDIR/djpeg -nosmooth -rgb -bmp -outfile $OUTDIR/${basename}_${samp}_${xform}_jpegtran.bmp $OUTDIR/${basename}_${samp}_${xform}_jpegtran.jpg
 			runme $JAVA TJExample $OUTDIR/${basename}_${samp}_fast.jpg $OUTDIR/${basename}_${samp}_${xform}.bmp -$xform -crop 16,16,70x60 -fastupsample
 			runme cmp -i 54:54 $OUTDIR/${basename}_${samp}_${xform}.bmp $OUTDIR/${basename}_${samp}_${xform}_jpegtran.bmp
 			rm $OUTDIR/${basename}_${samp}_${xform}.bmp
@@ -137,7 +137,7 @@ for image in $IMAGES; do
 		for samp in GRAY 444 422 420; do
 			for scale in 2_1 15_8 7_4 13_8 3_2 11_8 5_4 9_8 7_8 3_4 5_8 1_2 3_8 1_4 1_8; do
 				scalearg=`echo $scale | sed s@_@/@g`
-				$EXEDIR/djpeg -rgb -bmp -scale ${scalearg} $OUTDIR/${basename}_${samp}_${xform}_jpegtran.jpg >$OUTDIR/${basename}_${samp}_${xform}_${scale}_jpegtran.bmp
+				runme $EXEDIR/djpeg -rgb -bmp -scale ${scalearg} -outfile $OUTDIR/${basename}_${samp}_${xform}_${scale}_jpegtran.bmp $OUTDIR/${basename}_${samp}_${xform}_jpegtran.jpg
 				runme $JAVA TJExample $OUTDIR/${basename}_${samp}_fast.jpg $OUTDIR/${basename}_${samp}_${xform}_${scale}.bmp -$xform -scale ${scalearg} -crop 16,16,70x60
 				runme cmp -i 54:54 $OUTDIR/${basename}_${samp}_${xform}_${scale}.bmp $OUTDIR/${basename}_${samp}_${xform}_${scale}_jpegtran.bmp
 				rm $OUTDIR/${basename}_${samp}_${xform}_${scale}.bmp
--- a/tjunittest.c
+++ b/tjunittest.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C)2009-2014 D. R. Commander.  All Rights Reserved.
+ * Copyright (C)2009-2014, 2017 D. R. Commander.  All Rights Reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
@@ -44,12 +44,12 @@

 void usage(char *progName)
 {
-	printf("\nUSAGE: %s [options]\n", progName);
+	printf("\nUSAGE: %s [options]\n\n", progName);
 	printf("Options:\n");
 	printf("-yuv = test YUV encoding/decoding support\n");
 	printf("-noyuvpad = do not pad each line of each Y, U, and V plane to the nearest\n");
 	printf("            4-byte boundary\n");
-	printf("-alloc = test automatic buffer allocation\n");
+	printf("-alloc = test automatic buffer allocation\n\n");
 	exit(1);
 }

@@ -697,10 +697,9 @@ int main(int argc, char *argv[])
 		for(i=1; i<argc; i++)
 		{
 			if(!strcasecmp(argv[i], "-yuv")) doyuv=1;
-			if(!strcasecmp(argv[i], "-noyuvpad")) pad=1;
-			if(!strcasecmp(argv[i], "-alloc")) alloc=1;
-			if(!strncasecmp(argv[i], "-h", 2) || !strcasecmp(argv[i], "-?"))
-				usage(argv[0]);
+			else if(!strcasecmp(argv[i], "-noyuvpad")) pad=1;
+			else if(!strcasecmp(argv[i], "-alloc")) alloc=1;
+			else usage(argv[0]);
 		}
 	}
 	if(alloc) printf("Testing automatic buffer allocation\n");
--- a/turbojpeg.c
+++ b/turbojpeg.c
@@ -222,7 +222,7 @@ static int setCompDefaults(struct jpeg_compress_struct *cinfo,
 #ifndef NO_GETENV
 	if((env=getenv("TJ_OPTIMIZE"))!=NULL && strlen(env)>0 && !strcmp(env, "1"))
 		cinfo->optimize_coding=TRUE;
-	if((env=getenv("TJ_ARITHMETIC"))!=NULL && strlen(env)>0	&& !strcmp(env, "1"))
+	if((env=getenv("TJ_ARITHMETIC"))!=NULL && strlen(env)>0 && !strcmp(env, "1"))
 		cinfo->arith_code=TRUE;
 	if((env=getenv("TJ_RESTART"))!=NULL && strlen(env)>0)
 	{
@@ -790,8 +790,7 @@ DLLEXPORT int DLLCALL tjCompress2(tjhandle handle, const unsigned char *srcBuf,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	cinfo->image_width=width;
@@ -907,8 +906,7 @@ DLLEXPORT int DLLCALL tjEncodeYUVPlanes(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	cinfo->image_width=width;
@@ -990,8 +988,7 @@ DLLEXPORT int DLLCALL tjEncodeYUVPlanes(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	for(row=0; row<ph0; row+=cinfo->max_v_samp_factor)
@@ -1108,8 +1105,7 @@ DLLEXPORT int DLLCALL tjCompressFromYUVPlanes(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	cinfo->image_width=width;
@@ -1171,8 +1167,7 @@ DLLEXPORT int DLLCALL tjCompressFromYUVPlanes(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	for(row=0; row<(int)cinfo->image_height;
@@ -1404,8 +1399,7 @@ DLLEXPORT int DLLCALL tjDecompress2(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	jpeg_mem_src_tj(dinfo, jpegBuf, jpegSize);
@@ -1456,8 +1450,7 @@ DLLEXPORT int DLLCALL tjDecompress2(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}
 	for(i=0; i<(int)dinfo->output_height; i++)
 	{
@@ -1589,8 +1582,7 @@ DLLEXPORT int DLLCALL tjDecodeYUVPlanes(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	if(pixelFormat==TJPF_CMYK)
@@ -1684,8 +1676,7 @@ DLLEXPORT int DLLCALL tjDecodeYUVPlanes(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	for(row=0; row<ph0; row+=dinfo->max_v_samp_factor)
@@ -1789,8 +1780,7 @@ DLLEXPORT int DLLCALL tjDecompressToYUVPlanes(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	if(!this->headerRead)
@@ -1871,8 +1861,7 @@ DLLEXPORT int DLLCALL tjDecompressToYUVPlanes(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	if(flags&TJFLAG_FASTUPSAMPLE) dinfo->do_fancy_upsampling=FALSE;
@@ -2060,8 +2049,7 @@ DLLEXPORT int DLLCALL tjTransform(tjhandle handle,
 	if(setjmp(this->jerr.setjmp_buffer))
 	{
 		/* If we get here, the JPEG code has signaled an error. */
-		retval=-1;
-		goto bailout;
+		retval=-1;  goto bailout;
 	}

 	jpeg_mem_src_tj(dinfo, jpegBuf, jpegSize);
--- a/turbojpeg.h
+++ b/turbojpeg.h
@@ -275,7 +275,6 @@ static const int tjGreenOffset[TJ_NUMPF] = {1, 1, 1, 1, 2, 2, 0, 1, 1, 2, 2, -1}
 * then the blue component will be <tt>pixel[tjBlueOffset[TJ_BGRX]]</tt>.
 */
 static const int tjBlueOffset[TJ_NUMPF] = {2, 0, 2, 0, 1, 3, 0, 2, 0, 1, 3, -1};
-
 /**
 * Pixel size (in bytes) for a given pixel format.
 */
@@ -667,7 +666,7 @@ DLLEXPORT tjhandle DLLCALL tjInitCompress(void);
 * @param jpegQual the image quality of the generated JPEG image (1 = worst,
 * 100 = best)
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -731,7 +730,7 @@ DLLEXPORT int DLLCALL tjCompress2(tjhandle handle, const unsigned char *srcBuf,
 * @param jpegQual the image quality of the generated JPEG image (1 = worst,
 * 100 = best)
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -801,7 +800,7 @@ DLLEXPORT int DLLCALL tjCompressFromYUV(tjhandle handle,
 * @param jpegQual the image quality of the generated JPEG image (1 = worst,
 * 100 = best)
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -961,7 +960,7 @@ DLLEXPORT int tjPlaneHeight(int componentID, int height, int subsamp);
 * Video, <tt>subsamp</tt> should be set to @ref TJSAMP_420.  This produces an
 * image compatible with the I420 (AKA "YUV420P") format.
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -1019,7 +1018,7 @@ DLLEXPORT int DLLCALL tjEncodeYUV3(tjhandle handle,
 * Video, <tt>subsamp</tt> should be set to @ref TJSAMP_420.  This produces an
 * image compatible with the I420 (AKA "YUV420P") format.
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -1126,7 +1125,7 @@ DLLEXPORT tjscalingfactor* DLLCALL tjGetScalingFactors(int *numscalingfactors);
 * @param pixelFormat pixel format of the destination image (see @ref
 * TJPF "Pixel formats".)
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -1176,7 +1175,7 @@ DLLEXPORT int DLLCALL tjDecompress2(tjhandle handle,
 * block height (see #tjMCUHeight), then an intermediate buffer copy will be
 * performed within TurboJPEG.
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -1232,7 +1231,7 @@ DLLEXPORT int DLLCALL tjDecompressToYUV2(tjhandle handle,
 * block height (see #tjMCUHeight), then an intermediate buffer copy will be
 * performed within TurboJPEG.
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -1284,7 +1283,7 @@ DLLEXPORT int DLLCALL tjDecompressToYUVPlanes(tjhandle handle,
 * @param pixelFormat pixel format of the destination image (see @ref TJPF
 * "Pixel formats".)
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -1341,7 +1340,7 @@ DLLEXPORT int DLLCALL tjDecodeYUV(tjhandle handle, const unsigned char *srcBuf,
 * @param pixelFormat pixel format of the destination image (see @ref TJPF
 * "Pixel formats".)
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
@@ -1415,7 +1414,7 @@ DLLEXPORT tjhandle DLLCALL tjInitTransform(void);
 * which specifies the transform parameters and/or cropping region for the
 * corresponding transformed output image.
 *
- * @param flags the bitwise OR of one or more of the @ref TJFLAG_BOTTOMUP
+ * @param flags the bitwise OR of one or more of the @ref TJFLAG_ACCURATEDCT
 * "flags"
 *
 * @return 0 if successful, or -1 if an error occurred (see #tjGetErrorStr().)
--- a/wrbmp.c
+++ b/wrbmp.c
@@ -5,7 +5,7 @@
 * Copyright (C) 1994-1996, Thomas G. Lane.
 * libjpeg-turbo Modifications:
 * Copyright (C) 2013, Linaro Limited.
- * Copyright (C) 2014-2015, D. R. Commander.
+ * Copyright (C) 2014-2015, 2017, D. R. Commander.
 * For conditions of distribution and use, see the accompanying README.ijg
 * file.
 *
@@ -70,7 +70,7 @@ LOCAL(void) write_colormap
 static INLINE boolean is_big_endian(void)
 {
  int test_value = 1;
-  if(*(char *)&test_value != 1)
+  if (*(char *)&test_value != 1)
    return TRUE;
  return FALSE;
 }
@@ -104,7 +104,7 @@ put_pixel_rows (j_decompress_ptr cinfo, djpeg_dest_ptr dinfo,
  inptr = dest->pub.buffer[0];
  outptr = image_ptr[0];

-  if(cinfo->out_color_space == JCS_RGB565) {
+  if (cinfo->out_color_space == JCS_RGB565) {
    boolean big_endian = is_big_endian();
    unsigned short *inptr2 = (unsigned short *)inptr;
    for (col = cinfo->output_width; col > 0; col--) {
@@ -437,6 +437,7 @@ jinit_write_bmp (j_decompress_ptr cinfo, boolean is_os2)
                                  sizeof(bmp_dest_struct));
  dest->pub.start_output = start_output_bmp;
  dest->pub.finish_output = finish_output_bmp;
+  dest->pub.calc_buffer_dimensions = NULL;
  dest->is_os2 = is_os2;

  if (cinfo->out_color_space == JCS_GRAYSCALE) {
@@ -446,7 +447,7 @@ jinit_write_bmp (j_decompress_ptr cinfo, boolean is_os2)
      dest->pub.put_pixel_rows = put_gray_rows;
    else
      dest->pub.put_pixel_rows = put_pixel_rows;
-  } else if(cinfo->out_color_space == JCS_RGB565 ) {
+  } else if (cinfo->out_color_space == JCS_RGB565) {
      dest->pub.put_pixel_rows = put_pixel_rows;
  } else {
    ERREXIT(cinfo, JERR_BMP_COLORSPACE);
--- a/wrgif.c
+++ b/wrgif.c
@@ -4,7 +4,7 @@
 * This file was part of the Independent JPEG Group's software:
 * Copyright (C) 1991-1997, Thomas G. Lane.
 * libjpeg-turbo Modifications:
- * Copyright (C) 2015, D. R. Commander.
+ * Copyright (C) 2015, 2017, D. R. Commander.
 * For conditions of distribution and use, see the accompanying README.ijg
 * file.
 *
@@ -355,6 +355,16 @@ finish_output_gif (j_decompress_ptr cinfo, djpeg_dest_ptr dinfo)
 }


+/*
+ * Re-calculate buffer dimensions based on output dimensions.
+ */
+
+METHODDEF(void)
+calc_buffer_dimensions_gif (j_decompress_ptr cinfo, djpeg_dest_ptr dinfo)
+{
+}
+
+
 /*
 * The module selection routine for GIF format output.
 */
@@ -372,6 +382,7 @@ jinit_write_gif (j_decompress_ptr cinfo)
  dest->pub.start_output = start_output_gif;
  dest->pub.put_pixel_rows = put_pixel_rows;
  dest->pub.finish_output = finish_output_gif;
+  dest->pub.calc_buffer_dimensions = calc_buffer_dimensions_gif;

  if (cinfo->out_color_space != JCS_GRAYSCALE &&
      cinfo->out_color_space != JCS_RGB)
--- a/wrjpgcom.c
+++ b/wrjpgcom.c
@@ -247,7 +247,7 @@ copy_variable (void)
  if (length < 2)
    ERREXIT("Erroneous JPEG marker length");
  length -= 2;
-  /* Skip over the remaining bytes */
+  /* Copy the remaining bytes */
  while (length > 0) {
    write_1_byte(read_1_byte());
    length--;
--- a/wrppm.c
+++ b/wrppm.c
@@ -4,8 +4,8 @@
 * This file was part of the Independent JPEG Group's software:
 * Copyright (C) 1991-1996, Thomas G. Lane.
 * Modified 2009 by Guido Vollbeding.
- * It was modified by The libjpeg-turbo Project to include only code and
- * information relevant to libjpeg-turbo.
+ * libjpeg-turbo Modifications:
+ * Copyright (C) 2017, D. R. Commander.
 * For conditions of distribution and use, see the accompanying README.ijg
 * file.
 *
@@ -20,7 +20,6 @@
 */

 #include "cdjpeg.h"             /* Common decls for cjpeg/djpeg applications */
-#include "wrppm.h"

 #ifdef PPM_SUPPORTED

@@ -63,6 +62,21 @@
 */


+/* Private version of data destination object */
+
+typedef struct {
+  struct djpeg_dest_struct pub; /* public fields */
+
+  /* Usually these two pointers point to the same place: */
+  char *iobuffer;               /* fwrite's I/O buffer */
+  JSAMPROW pixrow;              /* decompressor output buffer */
+  size_t buffer_width;          /* width of I/O buffer */
+  JDIMENSION samples_per_row;   /* JSAMPLEs per output row */
+} ppm_dest_struct;
+
+typedef ppm_dest_struct *ppm_dest_ptr;
+
+
 /*
 * Write some pixel data.
 * In this module rows_supplied will always be 1.
@@ -196,6 +210,20 @@ finish_output_ppm (j_decompress_ptr cinfo, djpeg_dest_ptr dinfo)
 }


+/*
+ * Re-calculate buffer dimensions based on output dimensions.
+ */
+
+METHODDEF(void)
+calc_buffer_dimensions_ppm (j_decompress_ptr cinfo, djpeg_dest_ptr dinfo)
+{
+  ppm_dest_ptr dest = (ppm_dest_ptr) dinfo;
+
+  dest->samples_per_row = cinfo->output_width * cinfo->out_color_components;
+  dest->buffer_width = dest->samples_per_row * (BYTESPERSAMPLE * sizeof(char));
+}
+
+
 /*
 * The module selection routine for PPM format output.
 */
@@ -211,13 +239,13 @@ jinit_write_ppm (j_decompress_ptr cinfo)
                                  sizeof(ppm_dest_struct));
  dest->pub.start_output = start_output_ppm;
  dest->pub.finish_output = finish_output_ppm;
+  dest->pub.calc_buffer_dimensions = calc_buffer_dimensions_ppm;

  /* Calculate output image dimensions so we can allocate space */
  jpeg_calc_output_dimensions(cinfo);

  /* Create physical I/O buffer */
-  dest->samples_per_row = cinfo->output_width * cinfo->out_color_components;
-  dest->buffer_width = dest->samples_per_row * (BYTESPERSAMPLE * sizeof(char));
+  dest->pub.calc_buffer_dimensions (cinfo, (djpeg_dest_ptr) dest);
  dest->iobuffer = (char *) (*cinfo->mem->alloc_small)
    ((j_common_ptr) cinfo, JPOOL_IMAGE, dest->buffer_width);

--- a/wrppm.h
+++ b/wrppm.h
@@ -1,26 +0,0 @@
-/*
- * wrppm.h
- *
- * This file was part of the Independent JPEG Group's software:
- * Copyright (C) 1994, Thomas G. Lane.
- * For conditions of distribution and use, see the accompanying README.ijg
- * file.
- */
-
-#ifdef PPM_SUPPORTED
-
-/* Private version of data destination object */
-
-typedef struct {
-  struct djpeg_dest_struct pub; /* public fields */
-
-  /* Usually these two pointers point to the same place: */
-  char *iobuffer;               /* fwrite's I/O buffer */
-  JSAMPROW pixrow;              /* decompressor output buffer */
-  size_t buffer_width;          /* width of I/O buffer */
-  JDIMENSION samples_per_row;   /* JSAMPLEs per output row */
-} ppm_dest_struct;
-
-typedef ppm_dest_struct *ppm_dest_ptr;
-
-#endif
--- a/wrrle.c
+++ b/wrrle.c
@@ -3,8 +3,8 @@
 *
 * This file was part of the Independent JPEG Group's software:
 * Copyright (C) 1991-1996, Thomas G. Lane.
- * It was modified by The libjpeg-turbo Project to include only code and
- * information relevant to libjpeg-turbo.
+ * libjpeg-turbo Modifications:
+ * Copyright (C) 2017, D. R. Commander.
 * For conditions of distribution and use, see the accompanying README.ijg
 * file.
 *
@@ -286,6 +286,7 @@ jinit_write_rle (j_decompress_ptr cinfo)
                                  sizeof(rle_dest_struct));
  dest->pub.start_output = start_output_rle;
  dest->pub.finish_output = finish_output_rle;
+  dest->pub.calc_buffer_dimensions = NULL;

  /* Calculate output image dimensions so we can allocate space */
  jpeg_calc_output_dimensions(cinfo);
--- a/wrtarga.c
+++ b/wrtarga.c
@@ -3,8 +3,8 @@
 *
 * This file was part of the Independent JPEG Group's software:
 * Copyright (C) 1991-1996, Thomas G. Lane.
- * It was modified by The libjpeg-turbo Project to include only code and
- * information relevant to libjpeg-turbo.
+ * libjpeg-turbo Modifications:
+ * Copyright (C) 2017, D. R. Commander.
 * For conditions of distribution and use, see the accompanying README.ijg
 * file.
 *
@@ -211,6 +211,19 @@ finish_output_tga (j_decompress_ptr cinfo, djpeg_dest_ptr dinfo)
 }


+/*
+ * Re-calculate buffer dimensions based on output dimensions.
+ */
+
+METHODDEF(void)
+calc_buffer_dimensions_tga (j_decompress_ptr cinfo, djpeg_dest_ptr dinfo)
+{
+  tga_dest_ptr dest = (tga_dest_ptr) dinfo;
+
+  dest->buffer_width = cinfo->output_width * cinfo->output_components;
+}
+
+
 /*
 * The module selection routine for Targa format output.
 */
@@ -226,12 +239,13 @@ jinit_write_targa (j_decompress_ptr cinfo)
                                  sizeof(tga_dest_struct));
  dest->pub.start_output = start_output_tga;
  dest->pub.finish_output = finish_output_tga;
+  dest->pub.calc_buffer_dimensions = calc_buffer_dimensions_tga;

  /* Calculate output image dimensions so we can allocate space */
  jpeg_calc_output_dimensions(cinfo);

  /* Create I/O buffer. */
-  dest->buffer_width = cinfo->output_width * cinfo->output_components;
+  dest->pub.calc_buffer_dimensions (cinfo, (djpeg_dest_ptr) dest);
  dest->iobuffer = (char *)
    (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE,
                                (size_t) (dest->buffer_width * sizeof(char)));
Author	SHA1	Message	Date
DRC	bf6c774305	Fix whitespace errors	2017-12-13 21:48:54 -06:00
DRC	c308d43416	jpeg_crop_scanlines: Handle gray images w/ samp!=1 Since the sampling factor has no meaning for single-component images, the decompressor ignores it, and jpeg_crop_scanlines() should as well. Fixes #195	2017-12-06 11:59:04 -06:00
DRC	773040f9d9	Fix lib state when skipping to end of 1-scan image If jpeg_skip_scanlines() is used to skip to the end of a single-scan image, then we need to change the library state such that subsequent calls to jpeg_consume_input() will return JPEG_REACHED_EOI rather than JPEG_SUSPENDED. (NOTE: not necessary for multi-scan images, since the scans are processed prior to any call to jpeg_skip_scanlines().) Unless I miss my guess, using jpeg_skip_scanlines() in this manner will prevent any markers at the end of the JPEG image from being read, but I don't think there is any way around that without actually reading the data, which would defeat the purpose of jpeg_skip_scanlines(). Fixes #194	2017-12-05 16:08:01 -06:00
DRC	a831b5a986	Travis: Work around xcode7.3 image bug Refer to travis-ci/travis-ci#8552. This was supposed to be fixed on November 15, then on November 28. Travis blew through both deadlines, so I have no confidence that the issue will be fixed as promised in a timely manner. Adding 'brew update' to .travis.yml slows the OS X build, but there is no choice at the moment.	2017-11-29 14:23:31 -06:00
DRC	19b393b624	TJExample: Fix array index OOB w/ 4:1:1 JPEG input	2017-11-17 18:45:08 -06:00
DRC	9d9d8fe658	Code formatting tweaks	2017-11-17 18:39:53 -06:00
DRC	78e97e38cb	Uniquify tjbenchtest log file names based on args + clean up log files when 'make testclean' is invoked + fix 'tjbenchtest -yuv -alloc' + fix tjexampletest so that it creates images under /tmp + clean up tjexampletest	2017-11-17 18:39:53 -06:00
DRC	468f2fed27	TJExample.java: Don't ignore mistyped args	2017-11-17 18:39:48 -06:00
DRC	5abf2536f6	Doc tweak: TJFLAG_ACCURATEDCT is the first flag	2017-11-14 16:12:13 -06:00
DRC	5fce2e9421	tjbench.exe: Fix decompression access violation The program crashed when a JPEG image was passed on the command line, because we were mixing our metaphors vis-a-vis malloc()/free() and tjAlloc()/tjFree() (malloc()/free() uses the tjbench.exe heap, whereas tjAlloc()/tjFree() uses the turbojpeg.dll heap.)	2017-11-14 15:41:30 -06:00
DRC	907dd683a4	ChangeLog.md: buglet	2017-11-13 21:46:07 -06:00
DRC	f008876cc1	Build: Fix `make dist` Broken by previous commit	2017-11-13 21:24:05 -06:00
DRC	5bc43c7821	Further partial image decompression fixes - Referring to `073b0e88a1` and #185, the reason why BMP and RLE didn't (and won't) work with partial image decompression is that the output engines for both formats maintain a whole-image buffer, which is used to reverse the order of scanlines. However, it was straightforward to add -crop support for GIF and Targa, which is useful for testing partial image decompression along with color quantization. - Such testing reproduced a bug reported by Mozilla (refer to PR #182) whereby jpeg_skip_scanlines() would segfault if color quantization was enabled. To fix this issue, read_and_discard_scanlines() now sets up a dummy quantize function in the same manner that it sets up a dummy color conversion function. Closes #182	2017-11-13 21:01:53 -06:00
DRC	f3ad13e3d8	TJBench/TJUnitTest: Don't ignore mistyped args	2017-11-13 16:00:35 -06:00
DRC	94e152b186	TurboJPEG C: Code formatting tweaks	2017-11-13 15:14:12 -06:00
DRC	073b0e88a1	djpeg -crop: Exit gracefully with non-PPM formats ... and document that only PPM/PGM output images are supported with the -crop option for the moment. I investigated the possibility of supporting -crop with -bmp, but even after resetting the buffer dimensions, I still kept getting virtual array access errors. It seems that doing this the "right way" would require creating a re-initialization function for each image format's destination manager. I'm disinclined to do that right now, given that this feature was Google's baby (developed as a prerequisite for including libjpeg-turbo in Android), and the -crop option in djpeg is intended only as an example of how to use the partial image decompression API. Real-world applications would need to handle this in their own destination managers. It would probably be possible to make this work with Targa by employing a similar hack to the one we used with PPM, but Targa isn't popular enough to bother. Fixes #185	2017-11-08 21:30:19 -06:00
DRC	1b385f37f3	Prevent "unmappable character" error in Java build This was causing the build to fail when rebuilding libjpeg-turbo from a source RPM.	2017-09-20 17:05:15 +00:00
DRC	fd778bba46	Fix PowerPC 32-bit RPM build	2017-09-20 16:43:27 +00:00
DRC	8d403aeb6a	Fix 32-bit RPM build w/ newer RHEL/Fedora releases The version of RPM on RHEL 5 and older platforms defines _libdir as %{_exec_prefix}/%{_lib}, so defining _lib in the spec file redefined _libdir. However, newer versions of RPM (probably >= 4.6, since that was the version that introduced the ISA macros) define _libdir as either %{_prefix}/lib or %{_prefix}/lib64. Thus, we need to explicitly override _libdir in our spec file.	2017-09-19 23:15:46 +00:00
DRC	01b74c101e	Packaging: Use parallel make when rebuilding SRPM	2017-09-11 10:06:22 -05:00
DRC	db0dec3c5a	Travis: Limit parallel make jobs to # of CPU cores This tends to be faster than 'make -j', since it is making more wise use of the available resources.	2017-09-11 09:48:33 -05:00
nyg	ba6e9d8a11	wrjpgcom: Fix comment typo Comment was copied/pasted from skip_variable() without making the necessary modifications.	2017-09-05 11:23:56 -05:00
DRC	5426a4cb16	TJUnitTest: Usage formatting tweaks	2017-09-02 04:08:06 +00:00
DRC	d0bac69a8a	Java: Fix TJUnitTest on big endian platforms It is necessary for the C code to be aware of the machine's endianness, which is why the TurboJPEG Java wrapper sets a different pixel format for integer BufferedImages depending on ByteOrder.nativeOrder(). However, it isn't necessary to handle endianness in pure Java code such as TJUnitTest (d'oh!) This was a product of porting the C version of TJUnitTest too literally, and of insufficient testing (historically, the big endian systems I had available for testing didn't have Java.)	2017-09-02 03:47:56 +00:00
DRC	59322e0973	Build: Change ppc64le DEB arch to ppc64el This is the convention among Debian-based distros.	2017-09-01 21:57:48 +00:00
DRC	4ded4dfac6	Build: Fix AltiVec detection on OS X Leopard The ability to directly access elements of an AltiVec vector is apparently a more recent thing.	2017-09-01 09:01:00 -05:00
DRC	02fa8f244e	Fix build on PowerPC SPE systems SPE systems don't support AltiVec instructions. Fixes #172	2017-08-31 20:32:48 -05:00
DRC	32120054c2	Java: Fix NullPointerException in YUVImage planes == null is a valid argument to setBuf() if alloc == true, so we need to make sure that planes is non-null before validating its length. We also need to allocate one dimension of the planes array if it's null. Fixes #168	2017-08-14 11:46:26 -05:00
DRC	1478361204	Bump version to 1.5.3 to prepare for new commits	2017-08-14 11:20:19 -05:00