Hi, I logged a defect concerning the dynamic insertion of an image element using encoded content, and an error that sometimes occurs when the GVT rendering tree goes to decode it. Here's an example:
java.io.EOFException at java.io.DataInputStream.readInt(DataInputStream.java:397) at org.apache.batik.ext.awt.image.codec.PNGRed.readChunk(PNGRed.java:499) at org.apache.batik.ext.awt.image.codec.PNGRed.<init>(PNGRed.java:400) at org.apache.batik.ext.awt.image.spi.PNGRegistryEntry$1.run(PNGRegistryEntry.j ava:77) All input filters (PNG, JPEG, and TIFF) tend to fail with the same type of exception, which can be generalized as premature end-of-file errors. I believe this is a threading issue, but I don't know enough about the Batik rendering model (relating to the decoding of embedded images) to give you any more information than that. Please note that it does work some of the time; however, it seems to have more difficulty decoding larger files successfully. Once the defect is recreated, it is very reproducible for debugging purposes, but if the environment changes too much (a reboot for example) it may no longer be reproducible for that particular test image until another reboot. I am using Java's JDK v1.4.0 for the Windows OS, and a relatively recent snapshot of CVS (October 23rd). Can someone shed some light on this error? Michael --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
