date as YYYY/DDD, where DDD is a 1-origin day of year. Move the formats
to a "time_fmt.h" file, included by the headers that use it. Have
abs_time_to_str() and abs_time_secs_to_str() take the date format value,
rather than a Boolean "show this as UTC" flag, as an argument. Document
the ABSOLUTE_TIME_ formats a bit better. Use that format in the CCSDS
and VCDU dissectors, rather than having those dissectors do the
formatting themselves.
svn path=/trunk/; revision=32034
bit, so as not to imply that there's some form of global "mode"
Wireshark is in when it passes a null or non-null pointer (there isn't),
and to explicitly note that there is *no* guarantee about the value of
"tree" on the first call to the dissector. (I.e., please do not build a
mental model of how Wireshark works in that regard, and write your
dissector based on that mental model - you *will* be wrong.)
svn path=/trunk/; revision=31560
ABSOLUTE_TIME_LOCAL or ABSOLUTE_TIME_UTC, indicating whether to display
the date/time in local time or UTC. (int)ABSOLUTE_TIME_LOCAL ==
(int)BASE_NONE, so there's no source or binary compatiblity issue,
although we might want to eliminate BASE_NONE at some point and have the
BASE_ values used with integral types start at 0, so that you can't
specify BASE_NONE for an integral field.
svn path=/trunk/; revision=31319
itself a valid value for that field - it should be ORed with a value.
Indicate that it will never be possible to record in a header_field_info
a byte order for all fields, as some protocols do not specify the
endianness of fields (for example, DCE RPC uses "receiver makes it
right", with the sender sending data in its byte order, with an
indication in the packet of what that byte order is).
svn path=/trunk/; revision=31248
level) way to handle passing the result of strlen() to a routine
expecting a int-sized value, mark it as "OK", not "Compiler warning".
svn path=/trunk/; revision=30747
checks that really check whether the packet is valid; DISSECTOR_ASSERT()
should only be used for cases where the dissector is making an
assumption about its internal state.
svn path=/trunk/; revision=29006
tvb_get_seasonal_string();
tvb_get_seasonal_stringz();
.. which work the same as the ephemeral versions of the functions, but use
se_alloc() instead of ep_alloc().
svn path=/trunk/; revision=27868
This patch implements a function for dissecting bitfields with better control
over the resulting representation than the existing proto_tree_add_bitmask()
routine. This function will be used by reworked IPMI/ATCA dissector (bug 2048).
The function is described in README.developer. In short, the differences are as
follows:
- The new function does not require a hf_XXX field for the whole bitmask. When
the bitmask includes several unrelated fields, such hf_XXX field does not make
sense.
- The new function allows better control over the way the sub-item descriptions
are added to the top-level item. For example, proto_tree_add_bitmask() function
does not add non-enumerated integers, does not use true_false_string to display
boolean.
- The new function allows to specify "fallback" text for the top-level item
which is used if no items were added to the top-level item.
svn path=/trunk/; revision=25920
- Change ugly GLIB version checking statements to GLIB_CHECK_VERSION
- Remove ws_strsplit files because we no longer need to borrow GLIB2's
g_strsplit code for the no longer supported GLIB1 builds
svn path=/trunk/; revision=24829
- The "showHex" name cannot be the name paramter. Changing it to "show_hex" should be fine.
- There is also a missing ';' at the end of a line in the example.
This fixes bug 2092.
svn path=/trunk/; revision=23840
Fixed two typos in ReadMe.Developer documentation:
In the example code given, a comment is not properly closed and a semicolon was
missing in variable definition.
This fixes bug 2085.
svn path=/trunk/; revision=23824
http://library.gnome.org/devel/glib/unstable/glib-Miscellaneous-Macros.html#id2571572
G_INLINE_FUNC
#define G_INLINE_FUNC
This macro is used to export function prototypes so they can be linked with an external version when no inlining is performed. The file which implements the functions should define G_IMPLEMENTS_INLINES before including the headers which contain G_INLINE_FUNC declarations. Since inlining is very compiler-dependent using these macros correctly is very difficult. Their use is strongly discouraged.
This macro is often mistaken for a replacement for the inline keyword; inline is already declared in a portable manner in the glib headers and can be used normally.
svn path=/trunk/; revision=22980
case N ... M:
as that's not supported by all compilers.
Say so in the Portability section of README.developer, in the hopes of
discouraging others from using that GCCism.
svn path=/trunk/; revision=22976
* ptvcursor_push_subtree(), ptvcursor_pop_subtree() for pushing/popping
subtrees. Multiple levels of subtrees (256 max.), allocation per 8 levels.
* Two new functions creating an item in the tree and pushing a subtree at the
same time. These two functions accept an undefined length
(SUBTREE_UNDEFINED_LENGTH). The length of the item is set at the next pop.
1) ptvcursor_add_with_subtree
2) ptvcursor_add_text_with_subtree
- get rid of potential memory leaks with g_new in ptvcursor_new().
- Documentation of the new ptvcursor functions in README.developer
svn path=/trunk/; revision=21276
http://bugs.wireshark.org/bugzilla/show_bug.cgi?id=552
by enforcing that header fields have names of length > 0. This should fix
the display of those fields and also make them filterable (which was the
subject of the bug). Abbreviations are (still) optional: if they are empty
then the field is not filterable.
Update README.developer with this information.
Add header field names in several dissectors where they were missing.
In packet-arp.c give "packet-storm-detected" a name (as above) but also set it
as _GENERATED.
Also remove trailing white space from all the files checked in.
svn path=/trunk/; revision=21018
I've had a good look at the code in packet-tcp.c, and whilst it's
somewhat impenetrable, I've come to the conclusion that it just doesn't
support multiple pdus as described.
That's not entirely unreasonable in itself; my objection is solely to
the fact that README.developer is completely misleading. In fact, even
the example dissect_cstr won't work on the tcp dissector, because if you
set desegment_len=1 the tcp dissector believes that you know what you
are doing and doesn't let you change your mind later.
Furthermore, 2.7.2 says that you can set desegment_len=-1; that doesn't
work either, because the tcp dissector expects
DESEGMENT_ONE_MORE_SEGMENT, which is 0x0fffffff, which is nowhere near -1.
In short, I think the relevant section of README.developer needs a
rewrite. I attach a patch - comments welcome.
svn path=/trunk/; revision=20974
Here is an updated patch for proto_tree_add_item and the
range_string structure. The new macro RVALS() can be used as the macro
VALS() in the declaration of your hf_register_info with another
structure (range_string). Be aware that you *have to* ORed the value of
the field display with BASE_RANGE_STRING constant and it can 'only' be
used with FT_(U)INT* types in a header_field_info.
svn path=/trunk/; revision=20805
Create two new files (ws_strsplit.[ch]) that use GTK2 code to override
the buggy g_strsplit() function when compiling for GTK1. Include this
work-around function (ws_strsplit) in libwireshark.def. Add notes on usage
to README.developer. Include epan/ws_strsplit.h in all files that use
g_strsplit().
svn path=/trunk/; revision=20804
In the Developers Guide, Section 9.3, Example 9.17. Decompressing data
packets for dissection
The code calls tvb_set_free_cb() for the newly created next_tvb. This
is unnecessary as the call to tvb_set_child_real_data() adds next_tvb to
the chained list of tvb, thus ensuring that next_tvb is correctly
deleted. In fact when I had the call in, Visual Studio kept breaking
deep down in ntdll.dll, probably because of a double free every time the
main tvb was deleted.
In README.developer, para 2..2.7 The example conversation code doesn't
assign the result of conversation_new() back into the conversation variable.
svn path=/trunk/; revision=20569
I defined a range_string struct. It's like value_string
but stores range <-> string pairs.
Moreover I wrote rval_to_str(), match_strrval_idx()
match_strrval() which are behaving exactly as
val_to_str(), match_strval_idx() and match_strval().
svn path=/trunk/; revision=20061
tcp_dissect_pdus(), pinfo->desegment_len indicates whether your
dissector needs more data from TCP or not - the return value doesn't
indicate that.
Fix typo.
It appears that the Id keyword is one of the case-insensitive ones in
the svn:keywords property, so if you set it to "ID" it still expands
"$Id$"; it also appears not to expand "$ID$". We use Revision, Date,
and Author in the document to indicate the revision, and don't expand
Id, so that references to "$Id$" get left alone.
Rewrap paragraphs.
svn path=/trunk/; revision=19950
"I ran doc/README.developer through a spell checker and conservatively
changed misspelled words. Attached is a compressed patch with the
corrections."
svn path=/trunk/; revision=19070
IPv6 addresses. Use "tvb_get_ipv4()" in the WINS Replication dissector,
so that it gets the right answer on little-endian *AND* big-endian
machines.
svn path=/trunk/; revision=15753
This is very naughty and will cause problems when we have assigned a dissector to a dynamic port using conversation_set_dissector().
To make ethereal handle this case I have changed the try_conversation_dissector() to allow it to fail and return 0, meaning yes there is indeed a protocol registered for this conversation but that protocol rejected this packet.
(which only happens for "new" style dissectors, "old" style dissectors will never reject a packet that way)
When this happens the decode_udp_port() helper will still allow other dissectors to be tried, in the hope that the conversation is now used for some other protocol and thus someone else might be able to decode the packet.
Update SNMP and TFTP dissectors to check that even if there already is a conversation but that conversation does NOT have snmp/tftp registered as the dissector for it, then create a new conversation anyway and attach the proper dissector.
Since ethereal keeps track of which frame number a conversation started in, this actually works really well.
svn path=/trunk/; revision=14345
the platform for which we're building (and that both should be avoided
if possible, i.e. write your code so that it works on all platforms).
svn path=/trunk/; revision=11973
particularly, should disregard all the renaming they did of some
routines, as the old names work Just Fine in 2.x but the new names don't
work in 1.2[.x]).
svn path=/trunk/; revision=11936
integers.
Make FT_INT64 and FT_UINT64 add numerical values, rather than byte-array
values, to the protocol tree, and add routines to add specified 64-bit
integer values to the protocol tree.
Use those routines in the RSVP dissector.
svn path=/trunk/; revision=11796
before using its value, or must check for a null return value and handle
it specially, otherwise you put Ethereal at risk of crashing with bad
packet data.
svn path=/trunk/; revision=11475
they have LF at the end of the line on UN*X and CR/LF on Windows;
hopefully this means that if a CR/LF version is checked in on Windows,
the CRs will be stripped so that they show up only when checked out on
Windows, not on UN*X.
svn path=/trunk/; revision=11400
have to", indicating that if it's too much work to explicitly test for a
null protocol tree, you might want to avoid those tests and rely on the
protocol tree routines not to do much work if passed a null protocol
tree pointer.
svn path=/trunk/; revision=11346
use to format 64-bit integers.
Fix the RSVP dissector to use that rather than hardcoding "%ll" in.
Remove the "only if G_HAVE_GINT64 is defined" bit from the discussion of
64-bit integers - we're too dependent on having them to support
compilers that don't have a 64-bit integral data type. Do, however,
note that neither "long" nor "long long" are acceptable, and also note
that you shouldn't assume "%ll" does the trick for printing them.
svn path=/trunk/; revision=11182
convenient to put into a command line (no capital letters, no spaces to
require quotes), and one that's a detailed description for use in the
UI. Allow either of them in the preferences file or "-o" option; use
the detailed description in the UI, and also use it when writing the
preferences out, so that the preference will be readable by older
versions of Ethereal (assuming the preference existed in that version).
Update "README.developer" to give more detail about an enum_val_t (and
to put the _t in), and to give a more detailed description of the
"radio_buttons" argument to "prefs_register_enum_preference()".
svn path=/trunk/; revision=10982
"tvb_get_ptr()".
Add a section on roubustness, giving a number of potential problems that
aren't just portability problems.
Document "tvb_get_string()" and "tvb_get_stringz()", better document
"tvb_memcpy()" and "tvb_memdup()".
Fix a typo.
svn path=/trunk/; revision=10239
to export to other dissectors.
Describe the "if (tree)" construct and its sense by introducing 2 operation
modes of Ethereal:
(a) operational dissection (tree == NULL)
and
(b) detailed dissection (tree != NULL).
Fix some typos.
svn path=/trunk/; revision=9495
Guy Harris