i've reimported the old code analysis module which seems to work properly but does not splits code blocks , so graphs are bigger to read, but are ..at least correct.
You can choose which algo you want to use with the graph.split eval variable (false is the default one which is the old algorithm). Esteve can you take a look on your code analysis loop (setting up graph.split=true in your .radarerc). I would like to maintain both algorithms, and make them work properly for 0.9.6. Thanks! btw, yesterday I commited a patch which adds support for a more human-friendly syntax for setting up the color palettes: here's a example: > eval scr.palette=.prompt=3.ret=a.address=4 Use eval scr.palette= to get a colorized help message: =>( Targets ): .prompt .address .default .changed .jumps .calls .push .trap .cmp .ret .nop .metadata .header .printable .lines0 .lines1 .lines2 .00 .7f .ff =>( Colors ): /*normal*/, black, = 0, gray, = 1, white, = 2, red, = 3, magenta, = 4, blue, = 5, green, = 6, yellow, = 7, turqoise, = 8, /*bold*/, bblack, = a, bgray, = b, bwhite, = c, bred, = d, bmagenta, = e, bblue, = f, bgreen, = g, byellow, = h, bturqoise, = i, /*special*/, reset, = r Example: eval scr.palette = .prompt=3.address=4 Enjoy! _______________________________________________ radare mailing list [email protected] http://lists.nopcode.org/listinfo.cgi/radare-nopcode.org
