gwlucastrig commented on a change in pull request #98:
URL: https://github.com/apache/commons-imaging/pull/98#discussion_r491751220



##########
File path: 
src/main/java/org/apache/commons/imaging/formats/tiff/datareaders/DataReaderStrips.java
##########
@@ -291,25 +318,50 @@ public BufferedImage readImageData(final Rectangle 
subImage)
         //        or working
         final ImageBuilder workingBuilder =
                 new ImageBuilder(width, workingHeight, false);
-
-        for (int strip = strip0; strip <= strip1; strip++) {
-            final long rowsPerStripLong = 0xFFFFffffL & rowsPerStrip;
-            final long rowsRemaining = height - (strip * rowsPerStripLong);
-            final long rowsInThisStrip = Math.min(rowsRemaining, 
rowsPerStripLong);
-            final long bytesPerRow = (bitsPerPixel * width + 7) / 8;
-            final long bytesPerStrip = rowsInThisStrip * bytesPerRow;
-            final long pixelsPerStrip = rowsInThisStrip * width;
-
-            final byte[] compressed = imageData.getImageData(strip).getData();
-
-            final byte[] decompressed = decompress(compressed, compression,
-                    (int) bytesPerStrip, width, (int) rowsInThisStrip);
-
-            interpretStrip(
-                    workingBuilder,
-                    decompressed,
-                    (int) pixelsPerStrip,
-                    yLimit);
+        if (planarConfiguration != 2) {
+            for (int strip = strip0; strip <= strip1; strip++) {
+                final long rowsPerStripLong = 0xFFFFffffL & rowsPerStrip;
+                final long rowsRemaining = height - (strip * rowsPerStripLong);
+                final long rowsInThisStrip = Math.min(rowsRemaining, 
rowsPerStripLong);
+                final long bytesPerRow = (bitsPerPixel * width + 7) / 8;
+                final long bytesPerStrip = rowsInThisStrip * bytesPerRow;
+                final long pixelsPerStrip = rowsInThisStrip * width;
+
+                final byte[] compressed = 
imageData.getImageData(strip).getData();
+
+                final byte[] decompressed = decompress(compressed, compression,
+                  (int) bytesPerStrip, width, (int) rowsInThisStrip);
+
+                interpretStrip(
+                  workingBuilder,
+                  decompressed,
+                  (int) pixelsPerStrip,
+                  yLimit);
+            }
+        } else {
+            int nStripsInPlane = imageData.getImageDataLength() / 3;
+            for (int strip = strip0; strip <= strip1; strip++) {
+                final long rowsPerStripLong = 0xFFFFffffL & rowsPerStrip;
+                final long rowsRemaining = height - (strip * rowsPerStripLong);
+                final long rowsInThisStrip = Math.min(rowsRemaining, 
rowsPerStripLong);
+                final long bytesPerRow = (bitsPerPixel * width + 7) / 8;
+                final long bytesPerStrip = rowsInThisStrip * bytesPerRow;
+                final long pixelsPerStrip = rowsInThisStrip * width;
+
+                byte[] b = new byte[(int) bytesPerStrip];
+                for (int iPlane = 0; iPlane < 3; iPlane++) {
+                    int planeStrip = iPlane * nStripsInPlane + strip;
+                    final byte[] compressed = 
imageData.getImageData(planeStrip).getData();
+                    final byte[] decompressed = decompress(compressed, 
compression,
+                      (int) bytesPerStrip, width, (int) rowsInThisStrip);
+                    int index = iPlane;
+                    for (int i = 0; i < decompressed.length; i++) {
+                        b[index] = decompressed[i];
+                        index += 3;
+                    }
+                }
+                interpretStrip(workingBuilder, b, (int) pixelsPerStrip, 
height);
+            }

Review comment:
       These are two methods have a lot of redundancy and it would be better to 
combine them into one.  Unfortunately, they are both public APIs, so I'm not 
sure we could change their external appearance.   Of course, we could create a 
single, consolidated method that does the processing. Then we could rewrite the 
existing methods to be "wrapper" methods that called the new consolidated 
method.
   
   I think it might be better to treat that change as a separate pull request.
   
   I contributed the "sub-image" logic a number of years ago. At the time, 
there was a certain amount of resistance to the idea of a sub-image API, but I 
thought it was necessary because may GeoTIFFs are huge (I'd been working with 
some that are 20000-by-20000 pixels large).  Anyway, I think I left the 
original method in place because the code was simpler than the sub-image code 
and I thought it would serve as a reference for anyone who was trying to figure 
out how it worked.  But I think you are right and the time has come to clean 
things up. 




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]


Reply via email to