Date: Sat, 6 Aug 2005 02:29:13 -0400 (EDT) From: "Mikhail T." <mi@aldan.algebra.com> To: java@FreeBSD.org, glewis@FreeBSD.org, phantom@FreeBSD.org Subject: more on the jint vs. intptr_t Message-ID: <200508060629.j766TDiU048417@blue.virtual-estates.net>
next in thread | raw e-mail | index | archive | help
More specifically, the LiveConnect part of Mozilla family of browsers is using Java's GetJavaWrapper() and UnwrapJavaWrapper() functions to store _pointers_ to the JavaScript objects -- see comments in work/mozilla/js/src/liveconnect/jsjava.h as extracted by firefox or mozilla ports. These pointers are presumed to be as wide as jint -- see calls to get_java_wrapper() in work/mozilla/js/src/liveconnect/jsj_JSObject.c Currently, the only Java (sort of) available for amd64 is jdk15, which does not build the browser plugin for some reason anyway :-( However, I hope, this will some day be fixed and the browsers better be ready. Defining jint as intptr_t just for this purpose is, probably, wasteful. Instead, the GetJavaWrapper() and UnwrapJavaWrapper() functions should, probably, be modified to accept either a jlong (a warning-triggering overkill on i386) or a true intptr_t. Any comments? -mi
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200508060629.j766TDiU048417>