Re: [PATCH 21.4] Mixed signed/unsigned operations

Jerry> In the thread on the 21.4.9 release candidate, some Jerry> signed/unsigned operations were mentioned. Most turned out Jerry> to be easy to fix. Thanks! I'll put it into 21.4.10; fixing warnings after the release has been cut isn't a great idea. :) Jerry> There are still warnings from ccl_driver in mule-ccl.c. Jerry> There are two signed variables, i and j. They are compared Jerry> like this: I know. Yecccchhhh. Ben has completely redone 21.5 to eliminate signed/unsigned comparisons. You might look at mule-ccl.c from the trunk (if anyone wants to work on this, it's definitely on _my_ list though). Jerry> Something similar may be necessary for 21.5, but this patch Jerry> does not apply cleanly. Definitely not this patch. Eg, this: ＠＠ -675,7 +675,7 ＠＠ XtVaSetValues(w, XmNtraversalOn, True, 0) ; #endif - if( tab->tabs.row != tw->tabs.numRows-1 ) + if( (Cardinal)tab->tabs.row != tw->tabs.numRows-1 ) TabsShuffleRows(tw) ; is Evil and Wrong. The definition of the struct should be fixed for 21.5 (if Ben hasn't done so already). I would bet that Ben missed a lot of Motif stuff when he did this (I think that's why you're seeing most of these. I fixed all the Athena stuff I know how to fix). -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN My nostalgia for Icon makes me forget about any of the bad things. I don't have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py