summaryrefslogtreecommitdiff
path: root/java/lang/reflect/README
diff options
context:
space:
mode:
Diffstat (limited to 'java/lang/reflect/README')
-rw-r--r--java/lang/reflect/README23
1 files changed, 23 insertions, 0 deletions
diff --git a/java/lang/reflect/README b/java/lang/reflect/README
new file mode 100644
index 000000000..52efdc4bf
--- /dev/null
+++ b/java/lang/reflect/README
@@ -0,0 +1,23 @@
+README for java.lang.reflect:
+
+
+Jun 20 1998 - John Keiser - initial revision, 1.1.0
+
+OK, the Java side was easy.
+
+The difficult part is the native side, and while that has gone somewhere, there
+is so much integration between the VM and class libraries in Sun's version that
+a whole new API had to be created (the VMI being the prominent one). JCL
+Reflection will likely never work with Sun's JDK 1.1 unless we manage some real
+magic. Therefore I am targeting it to Japhar and Kaffe 1.1, where we at least
+have some ability to create VMI functionality.
+
+I am not even certain if Sun's JDK 1.2 will be able to support JCL Reflection.
+I am not getting my hopes up. Unless their JVMDI spec improves, it will be
+impossible to do so efficiently, and probably impossible in any case, without
+knowing their internal structures so that we can .
+
+This whole thing has been compiled but has never run. I am checking this in
+right now so that Geoffrey Berry, who is working on Serialization, will have
+some starting points, especially in regards to his native code.
+