Have separate packet_provider_data structures and packet_provider_funcs
structures; the latter holds a table of functions that libwireshark can
call for information about packets, the latter holds the data that those
functions use.
This means we no longer need to expose the structure of an epan_t
outside epan/epan.c; get rid of epan/epan-int.h.
Change-Id: I381b88993aa19e55720ce02c42ad33738e3f51f4
Reviewed-on: https://code.wireshark.org/review/24732
Reviewed-by: Guy Harris <guy@alum.mit.edu>
libwireshark now expects an epan_t to be created with a pointer to a
"packet provider" structure; that structure is opaque within
libwireshark, and a pointer to it is passed to the callbacks that
provide interface names, interface, descriptions, user comments, and
packet time stamps, and that set user comments. The code that calls
epan_new() is expected to provide those callbacks, and to define the
structure, which can be used by the providers. If none of the callbacks
need that extra information, the "packet provider" structure can be
null.
Have a "file" packet provider for all the programs that provide packets
from a file.
Change-Id: I4b5709a3dd7b098ebd7d2a7d95bcdd7b5903c1a0
Reviewed-on: https://code.wireshark.org/review/24731
Reviewed-by: Guy Harris <guy@alum.mit.edu>
The split isn't necessary now that epan no longer uses the capture_file
structure.
Change-Id: Ia232712a2fb5db511865805518e8d03509b2167f
Reviewed-on: https://code.wireshark.org/review/24693
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Embed one of those structures in a capture_file, and have a struct
epan_session point to that structure rather than to a capture_file.
Pass that structure to the routines that fetch data that libwireshark
uses when dissecting.
That separates the stuff that libwireshark expects from the stuff that
it doesn't look at.
Change-Id: Ia3cd28efb9622476437a2ce32204597fae720877
Reviewed-on: https://code.wireshark.org/review/24692
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Have cfile-int.h declare the structure, and use it in files that
directly access the structure.
Have cfile.h just incompletely declare the structure and include it
rather than explicitly declaring it in source files or other header
files.
Never directly refer to struct _capture_file except when typedeffing
capture_file.
Add #includes as necessary, now that cfile.h doesn't drag in a ton of
Change-Id: I7931c8039d75ff7c980b0f2a6e221f20e602a556
Reviewed-on: https://code.wireshark.org/review/24686
Reviewed-by: Guy Harris <guy@alum.mit.edu>
A while back Graham pointed out the SPDX project (spdx.org), which is
working on standardizing license specifications:
https://www.wireshark.org/lists/wireshark-dev/201509/msg00119.html
Appendix V of the specification describes a short identifier
(SPDX-License-Identifier) that you can use in place of boilerplate in
your source files:
https://spdx.org/spdx-specification-21-web-version#h.twlc0ztnng3b
Start the conversion process with our top-level C and C++ files.
Change-Id: Iba1d835776714deb6285e2181e8ca17f95221878
Reviewed-on: https://code.wireshark.org/review/24302
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Reviewed-by: Balint Reczey <balint@balintreczey.hu>
Tested-by: Petri Dish Buildbot
Reviewed-by: Michael Mann <mmann78@netscape.net>
Change-Id: Ia46903586219ee79210a980a04024af02acb0db0
Reviewed-on: https://code.wireshark.org/review/24189
Petri-Dish: João Valverde <j@v6e.pt>
Tested-by: Petri Dish Buildbot
Reviewed-by: Michael Mann <mmann78@netscape.net>
If we're building on Windows we're going to have windows.h and
winsock2.h. Don't bother checking for them.
Change-Id: I0004c44d7364ab3f41682f34b8c84cd8617c9603
Reviewed-on: https://code.wireshark.org/review/24068
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Tested-by: Petri Dish Buildbot
Reviewed-by: Anders Broman <a.broman58@gmail.com>
In preparation for possibly using AUTOUIC in CMake which treats "ui_*.h"
files specially, rename ui_util.h. No other changes.
Change-Id: Id026572c000b713ff0e9388dc7fff8d81d4df73e
Reviewed-on: https://code.wireshark.org/review/23916
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
It's not installed so like most other files it doesn't need or benefit
from the prefix.
Change-Id: I01517e06f12b3101fee21b68cba3bc6842bbef5c
Reviewed-on: https://code.wireshark.org/review/23751
Petri-Dish: João Valverde <j@v6e.pt>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: João Valverde <j@v6e.pt>
Rename cf_read_shb_comment() to cf_read_section_comment(); an SHB is a
record type in a particular capture file format (pcapng), and not all
files that have per-file or per-file-section comments have something
called a Section Header Block.
Rename cf_update_capture_comment() to cf_update_section_comment();
pcapng, at least, supports multiple sections, although we don't curently
support that.
This also gives them matching names.
Change-Id: Idd8cb0f0fd9125b9626411274aebfb1ec0097665
Reviewed-on: https://code.wireshark.org/review/22659
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Support per-packet comments in ERF_TYPE_META through a new Anchor ID
extension header with per-Host unique 48-bit Anchor ID which links an
ERF_TYPE_META record with a packet record. There may be more than one
Anchor ID associated with a packet, where they are grouped by Host ID
extension header in the extension header list. Like other ERF_TYPE_META
existing comments should not be overwritten and instead a new record
generated. See erf_write_anchor_meta_update_phdr() for detailed comments
on the extension header stack required.
As Wireshark only supports one comment currently, use the one one with
the latest metadata generation time (gen_time). Do this for capture
comment too.
Write various wtap metadata in periodic per-second ERF_TYPE_META records
if non-WTAP_ENCAP_ERF or we have an updated capture comment.
Refactor erf_dump to create fake ERF header first then follow common
pseudoheadr and payload write code rather than two separate code paths.
Support an ERF_HOST_ID environment variable to define Wireshark's Host
ID when writing. Defaults to 0 for now.
ERF dissector updates to support Anchor ID extension header with basic
frame linking.
Update ERF_TYPE_META naming and descriptions to official name
(Provenance)
Core changes:
Add has_comment_changed to wtap_pkthdr, TRUE when a packet
opt_comment has unsaved changes by the user.
Add needs_reload to wtap_dumper which forces a full reload of the file
on save, otherwise wireshark gets confused by additional packets being
written.
Change-Id: I0bb04411548c7bcd2d6ed82af689fbeed104546c
Ping-Bug: 12303
Reviewed-on: https://code.wireshark.org/review/21873
Petri-Dish: Anders Broman <a.broman58@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Stephen Donnelly <stephen.donnelly@endace.com>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Adds the --no-duplicate-keys option to tshark. If -T json is specified,
this option can be specified in order to transform the duplicate keys
produced by -T json into single keys with as value a json array of all
separate values.
Specifying --no-duplicate-keys changes the function which groups node
children that is passed to write_json_proto_tree. Instead of a function
that puts each node in a separate group (proto_node_group_children_by_unique)
a function is passed that groups children that have the same json key
together (proto_node_group_children_by_json_key). This will lead to
some groups having multiple values. Groups with multiple values are
written to the output as a json array. This includes normal json keys
but also keys with the "_raw" and "_tree" suffix.
If --no-duplicate-keys is specified with an option other than "-T json"
or "-T jsonraw" or without -T an error is shown and tshark will exit.
"Export Packet Dissections -> As JSON" in the GUI is hardcoded to use
the duplicated keys format.
Fixes one regression in the output where a filtered json key (-j) with
both a value and children would not have the "_tree" suffix added to the
json key containing the children.
Includes a little code cleanup (removes one instance of code
duplication and simplifies a while loop).
Fixes a memory leak (I thought this fix was already included in the
previous refactor patch but something must have gone wrong when updating
the patch so I'm including it again in this patch).
Bug: 12958
Change-Id: I401f8fc877b5c590686567c3c44cdb832e9e7dfe
Reviewed-on: https://code.wireshark.org/review/22166
Petri-Dish: Jaap Keuter <jaap.keuter@xs4all.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
The Qt UI doesn't have a popup that tells you how long your file has been
loading. So let's set the load time each time we update the packets bar.
(Obviously this is only useful when you're waiting a long time for a file to
load...)
Change-Id: I9da372800a12454888439e2baf3d2a848c611501
Reviewed-on: https://code.wireshark.org/review/22234
Petri-Dish: Anders Broman <a.broman58@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
testCaptureFileClose can also be invoked while reading an existing
capture file (the original comment only applied to GTK+, not Qt). When
the user quits Wireshark while reading an offline pcap, this could
result in a confusing "Unsaved packets" dialog. Fix this by checking the
actual capture session state.
After fixing this, the next issue is that cf_close trips on an assertion
("cf->state != FILE_READ_IN_PROGRESS"). To address this problem, do not
close the capture file immediately, but signal to the reader (cf_read)
that this should be done (similar to the quit logic in GTK+).
Bug: 13563
Change-Id: I12d4b813557bf354199320df2ed8609070fdc58a
Reviewed-on: https://code.wireshark.org/review/22096
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Bug: 6682
Change-Id: I19330d06aa3d5692503c61369c3c650d595971f5
Reviewed-on: https://code.wireshark.org/review/22077
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Stephen Donnelly <stephen.donnelly@endace.com>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Use WTAP_MAX_PACKET_SIZE_STANDARD, set to 256KB, for everything except
for D-Bus captures. Use WTAP_MAX_PACKET_SIZE_DBUS, set to 128MB, for
them, because that's the largest possible D-Bus message size. See
https://bugs.freedesktop.org/show_bug.cgi?id=100220
for an example of the problems caused by limiting the snapshot length to
256KB for D-Bus.
Have a snapshot length of 0 in a capture_file structure mean "there is
no snapshot length for the file"; we don't need the has_snap field in
that case, a value of 0 mean "no, we don't have a snapshot length".
In dumpcap, start out with a pipe buffer size of 2KB, and grow it as
necessary. When checking for a too-big packet from a pipe, check
against the appropriate maximum - 128MB for DLT_DBUS, 256KB for
everything else.
Change-Id: Ib2ce7a0cf37b971fbc0318024fd011e18add8b20
Reviewed-on: https://code.wireshark.org/review/21952
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Change-Id: I2ace2a6847ca01435651a48934acf9c1c6d2f23d
Reviewed-on: https://code.wireshark.org/review/21444
Petri-Dish: Martin Kaiser <wireshark@kaiser.cx>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Martin Kaiser <wireshark@kaiser.cx>
The cfile_ error-reporting routines free err_info; the caller doesn't
have to and, in fact, mustn't do so themselves.
While we're at it, make sure wtap_seek_read() always zeroes out *err and
nulls out *err_info, so the latter either points to a freshly-allocated
string or is null.
Change-Id: Idfe05a3ba2fbf2647ba14e483187617ee53e3c69
Reviewed-on: https://code.wireshark.org/review/21407
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Packet ranges are used only in the UI; move the packet range stuff into
libui.
Don't pass a print_args_t structure to libwireshark packet-printing
routines, just pass the few parameters they need. Move the declaration
of print_args_t into file.h.
Change-Id: Icff5991eea7d7d56f33b4716105895263d275bcf
Reviewed-on: https://code.wireshark.org/review/21308
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Have them just return the information needed for the caller to produce
an error message, and have the callers use the new cfile_ routines for
reporting errors.
This requires that the "write failure alert box" routine take the
*input* file name as an argument, so that, on a merge, if the problem is
that a record from a given input file can't be written out to the type
of output file we're generating, the input file name can be given, along
with the record number in that file.
Change-Id: If5a5e00539e7e652008a523dec92c0b359a48e71
Reviewed-on: https://code.wireshark.org/review/21257
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Expand comments while we're at it.
Change-Id: I6dcc791eab1c9e323a9572f3d54720d223bdd64b
Reviewed-on: https://code.wireshark.org/review/21252
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Take the code from save_record() to pop up an alert box and put it into
libui, with the name cfile_write_failure_alert_box().
While we're at it, clean up some variable names and comments.
Change-Id: If56e649e5afe50e7a20b39720ec96e8fe2983b24
Reviewed-on: https://code.wireshark.org/review/21233
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Take the code from cf_read() to pop up an alert box and put it into
libui, with the name cfile_read_failure_alert_box(). Use it in a couple
of places where we pop up such an error dialog.
While we're at it, get rid of the "err" argument to rescan_file();
nobody uses what it returns.
Change-Id: Iba7099b95de24309359d94eb96f606020e2ff2c3
Reviewed-on: https://code.wireshark.org/review/21232
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Take cf_close_failure_alert_box() and put it into libui, with the name
cfile_close_failure_alert_box(). Use it not only in file.c but also
in ui/export_pdu_ui_utils.c, ui/gtk/file_import_dlg.c, and
ui/qt/import_text_dialog.cpp where the error we get back isn't
necessarily an errno.
Have ui/gtk/file_import_dlg.c and ui/qt/import_text_dialog.cpp also use
cfile_open_failure_alert_box() on open errors.
Change-Id: I987f339a23ea58609390306a319923e7f92d5c07
Reviewed-on: https://code.wireshark.org/review/21203
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Take cf_open_failure_alert_box() and put it into libui, with the name
cfile_open_failure_alert_box(). Use it not only in file.c but also in
ui/export_pdu_ui_utils.c, where the error we get back isn't necessarily
an errno.
Change-Id: Ia053f3e403ba464d446bd9530778d5ed302796d2
Reviewed-on: https://code.wireshark.org/review/21174
Reviewed-by: Guy Harris <guy@alum.mit.edu>
If fdata->flags.visited *isn't* set, this is the first pass over the
packets since opening a file or starting a redissection.
Change-Id: Ib72d122c9291ce5ac9c3a3df1b6dfb4db163698e
Reviewed-on: https://code.wireshark.org/review/21155
Reviewed-by: Guy Harris <guy@alum.mit.edu>
They deal with sets of hfids, which can belong to protocols as well as
fields (I guess you could argue that a protocol is a field, but...).
Change-Id: Ibd103cfa26427ead4ef54be89f1251908004cfae
Reviewed-on: https://code.wireshark.org/review/21154
Reviewed-by: Guy Harris <guy@alum.mit.edu>
This makes sure that postdissectors that indicate that they need certain
fields in the first pass will get them.
While we're at it:
Fix the field-fetching code in TRANSUM not to assume it got any
instances of the field being fetched.
Rename process_packet_first_pass() in sharkd to process_packet(), as
it's the only routine in sharkd that processes packets.
Rename process_packet() in tshark and tfshark to
process_packet_single_pass(), as it's what's used if we're only doing
one-pass analysis.
Clean up comments and whitespace.
Change-Id: I3769af952c66f5ca4b68002ad6213858ab9cab9b
Reviewed-on: https://code.wireshark.org/review/21063
Reviewed-by: Guy Harris <guy@alum.mit.edu>
XXX_prime_with_YYY makes it a bit clearer than does XXX_prime_YYY that
we're not priming YYY, we're priming XXX *using* YYY.
Change-Id: I1686b8b5469bc0f0bd6db8551fb6301776a1b133
Reviewed-on: https://code.wireshark.org/review/21031
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Currently, this is only used to determine whether a protocol tree needs
to be built on the first pass or not - if there are postdissectors that
need fields, it does - but eventually we should be able to use it to
prime the dissection to deliver those fields in cases where we don't
need the *entire* protocol tree (rather than using a hack such as
cooking up a fake tap with a fake filter to do that).
Update MATE and TRANSUM to use it.
Clean up code to check whether we need a protocol tree, and add comments
before that code indicating, in each case, what the criteria are.
The array of postdissectors includes a length, so we don't need to
separately keep track of the number of postdissectors.
Clean up indentation while we're at it.
Change-Id: I71d4025848206d144bc54cc82941089a50e80ab7
Reviewed-on: https://code.wireshark.org/review/21029
Reviewed-by: Guy Harris <guy@alum.mit.edu>
This patch introduces the frame.interface_description field.
While testing this new functionality it became obvious that we have
a non-optimal interaction between the existing cfile.c's
cap_file_get_interface_name(), the recently added frame.interface_name
field and this new frame.interface_description field.
The string returned from cap_file_get_interface_name() may in fact
come from one of three different sources: the idb's interface name
(if it exists) or the idb's interface description (if that exists)
or a default text of "unknown". The string ultimately becomes the
rame.interface_name whether or not the idb had an interface name
option to begin with. This behavior does not allow one to test for
the simple presence of frame.interface_name. The new peer function
cap_file_get_interface_description() added by this patch returns
NULL instead of "unknown" if the idb does not have an interface
description. Should cap_file_get_interface_name() be similarly
modified to return NULL if the idb does not have an interface name?
Bug: 9781
Change-Id: Ie479f373c5080c004dd22bd88919838feca71e95
Reviewed-on: https://code.wireshark.org/review/19861
Petri-Dish: Stig Bjørlykke <stig@bjorlykke.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Add "file merge" callback plumbing. Use it to display "Merging files" in
the main statusbar.
Make sure we have a usable window pointer when we merge files.
Change-Id: I236b6edb30685f0b06703ab8304bc88ae592f83c
Reviewed-on: https://code.wireshark.org/review/19716
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Just directly call the packet_list_select_ routine from the GTK+ code.
Change-Id: I9146fb968c407d6186b146a86aa34678765f7352
Reviewed-on: https://code.wireshark.org/review/19225
Reviewed-by: Guy Harris <guy@alum.mit.edu>
This is similar to what we have for opening a dump file - one API that
uses the file name as specified, one that creates a temporary file and
provides the file name, and one that uses the standard output.
All of those APIs handle closing the output file.
Change-Id: I56beea7be347402773460b9148ab31a8f8bc51e1
Reviewed-on: https://code.wireshark.org/review/19059
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Setting our compiler warning flags in CMAKE_C_FLAGS does not allow
using different flags per target.
Allow for that possibility by setting the internal WS_WARNINGS_{C,CXX}_FLAGS
and using the COMPILE_OPTIONS property to set them.
This change is just setting mechanism and there should be no difference
in generated warnings.
The check_X_compiler_flag cmake test is changed to test each flag individually.
We need a list, not a space separated string, and the aggregate test is not
significant.
Change-Id: I59fc5cd7e130c7a5e001c598e3df3e13f83a6a25
Reviewed-on: https://code.wireshark.org/review/17150
Petri-Dish: João Valverde <j@v6e.pt>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: João Valverde <j@v6e.pt>
A string option, if present, always has a value; it might be a null
*string*, but you won't get a null pointer (if the option isn't present,
it simply isn't present).
Fix some comments while we're at it.
Change-Id: I9c1420f56998a7d04de5c5cc2e92631b181f303a
Reviewed-on: https://code.wireshark.org/review/16564
Reviewed-by: Guy Harris <guy@alum.mit.edu>
That information can, and will, get saved in some file formats, so mark
the file as changed so it can and will get saved by "Save".
XXX - we need to treat it as a type of data that can be discarded when
saving in some file formats, just like comments.
Bug: 12629
Change-Id: I1fd69b95f4f7345c339961b4c53c28b98b364e4e
Reviewed-on: https://code.wireshark.org/review/16538
Reviewed-by: Guy Harris <guy@alum.mit.edu>
In file.c, update our UI based on the time elapsed since we last
finished updating the UI instead of on the number of packets we've
processed. Remove the timer code from progress_frame.cpp since it's now
redundant.
This makes the UI more responsive here for captures with large numbers
of packets and it should mean that packets_bar_update spends less time
repainting, at least on Windows.
Change-Id: I9edfa944c44192350bef75b8c0c3ad63bae9c131
Reviewed-on: https://code.wireshark.org/review/16476
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
A block can have zero or more instances of a given option. We
distinguish between "one instance only" options, where a block can have
zero or one instance, and "multiple instances allowed" options, where a
block can have zero or more instances.
For "one instance only" options:
"add" routines add an instance if there isn't one already
and fail if there is;
"set" routines add an instance if there isn't one already
and change the value of the existing instance if there is one;
"set nth" routines fail;
"get" routines return the value of the instance if there is one
and fail if there isn't;
"get nth" routines fail.
For "multiple instances allowed" options:
"add" routines add an instance;
"set" routines fail;
"set nth" routines set the value of the nth instance if there is
one and fail otherwise;
"get" routines fail;
"get nth" routines get the value if the nth instance if there is
one and fail otherwise.
Rename "optionblock" to just "block"; it describes the contents of a
block, including both mandatory items and options.
Add some support for NRB options, including IPv4 and IPv6 option types.
Change-Id: Iad184f668626c3d1498b2ed00c7f1672e4abf52e
Reviewed-on: https://code.wireshark.org/review/16444
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Fixed json and ek escape function
Fixed -j protocol filter to do exact match
Fixed -T json to correctly close json
Added -j protocol filter also to pdml output
Bug: 11754
Change-Id: I02f274e4a5a02346922b37bbe946c10340c242ea
Reviewed-on: https://code.wireshark.org/review/16034
Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Guy Harris