give the error: SCRIPT ENGINE: C:\Program
Files\Nmap\nselib/http.lua:77: attempt to call field 'parse' (a nil
value) [Jah]
Also added Win2K dependencies section to nmap-install (I should have
done a separate checkin for that)
the parsing of the standard nmap data files (e.g. nmap-services,
nmap-protocols, etc.) now uses those generic routines. NSE scripts
and libraries may find them useful for dealing with their own data
files, such as password lists. [Jah]
when I adapted some libdnet 1.11 code to convert a dnet name to a pcap name by
matching hardware addresses. I removed an earlier comment that this one
supersedes.
using sortas in an index entry. Previously I had things like
<indexterm><primary sortas="nmap"><filename>.nmap</filename> filename extension</primary></indexterm>
<indexterm><primary sortas="nmap"><filename>.nmap</filename> directory</primary></indexterm>
which, because they have the same sort key, would result in one index entires
under either ".nmap filename extension" or ".nmap directory", with the page
numbers for both. Now I just spell out the full text: sortas="nmap filename
extension".
This makes the rule to generate a module general so all you should have to do
in the common case is add <module>.so as a dependency of the "all" target.
Attached corrects answerFetcher[types.TXT]() in dns.lua which inserts
v.data into answers only if v.domain is present - I believe the test
should be for v.data. This was preventing TXT answers from being
returned from dns.query() with the retAll option.
o Nmap sometimes sent packets with incorrect IP checksums,
particularly when sending the UDP probes in OS detection. This has
been fixed. Thanks to Gisle Vanem for reporting and investigating the
bug. [David]
o In the interest of forward compatibility, the xmloutputversion
attribute in Nmap XML output is no longer constrained to be a
certain string ("1.02"). The xmloutputversion should be taken as
merely advisory by authors of parsers.
The partial checksum for the TCP/UDP pseudo-header is calculated and then it is
added to the checksum for the rest of the packet. I started to write the
functions for such incremental checksum calculation but then I saw they are
already implemented in libdnet.
