Wireshark-commits: [Wireshark-commits] rev 50106: /trunk/epan/ftypes/ /trunk/epan/ftypes/: ftype-by

Date: Fri, 21 Jun 2013 05:25:00 GMT
http://anonsvn.wireshark.org/viewvc/viewvc.cgi?view=rev&revision=50106

User: guy
Date: 2013/06/20 10:25 PM

Log:
 get_ax25_name() is *NOT* the right routine to use to translate a
 human-readable representing an AX.25 subaddress (e.g. "KA9Q-01") into
 the binary form of an AX.25 address, because what it does is translate
 the binary form of an AX.25 address to the human-readable form!
 
 We currently have no routine that does the right thing and, even if we
 did, given that some bits in the AX.25 subaddress format are used for
 purposes other than representing the call sign and substation ID, so the
 matching routines for AX.25 addresses need to ignore certain bits.
 
 For now, we just remove the call to get_ax25_name() (which squelches the
 pointer-signedness warning that made me look at this code, and find the
 problem, in the first place) and replace it with a comment discussing
 the problem and a failure.
 
 The other pointer-signedness warning brought up a question of what to do
 with G_REGEX_RAW in the g_regex_match_full() call; it didn't bring up an
 immediately obvious *answer*, so we throw a cast at the warning and add
 another comment.  (We fix up alignment while we're at it.)

Directory: /trunk/epan/ftypes/
  Changes    Path             Action
  +61 -15    ftype-bytes.c    Modified