wireshark/epan/dissectors/packet-http-urlencoded.c

247 lines
6.3 KiB
C

/* packet-http-urlencoded.c
* Routines for dissection of HTTP urlecncoded form, based on packet-text-media.c (C) Olivier Biot, 2004.
*
* Wireshark - Network traffic analyzer
* By Gerald Combs <gerald@wireshark.org>
* Copyright 1998 Gerald Combs
*
* SPDX-License-Identifier: GPL-2.0-or-later
*/
#include "config.h"
#include <epan/packet.h>
#include <epan/charsets.h>
#include <epan/strutil.h>
#include <wsutil/str_util.h>
#include "packet-http.h"
void proto_register_http_urlencoded(void);
void proto_reg_handoff_http_urlencoded(void);
static dissector_handle_t form_urlencoded_handle;
static int proto_urlencoded = -1;
static int hf_form_key = -1;
static int hf_form_value = -1;
static gint ett_form_urlencoded = -1;
static gint ett_form_keyvalue = -1;
static ws_mempbrk_pattern pbrk_key;
static ws_mempbrk_pattern pbrk_value;
static int
get_form_key_value(wmem_allocator_t *pool, tvbuff_t *tvb, char **ptr, int offset, const ws_mempbrk_pattern *pbrk)
{
const int orig_offset = offset;
int found_offset;
guint8 ch;
char *tmp;
int len;
len = 0;
while (tvb_reported_length_remaining(tvb, offset) > 0) {
found_offset = tvb_ws_mempbrk_pattern_guint8(tvb, offset, -1, pbrk, &ch);
if (found_offset == -1) {
len += tvb_reported_length_remaining(tvb, offset);
break;
}
len += (found_offset - offset);
offset = found_offset;
if (ch == '%') {
if (tvb_reported_length_remaining(tvb, offset) < 2) {
return -1;
}
offset++;
ch = tvb_get_guint8(tvb, offset);
if (ws_xton(ch) == -1)
return -1;
offset++;
ch = tvb_get_guint8(tvb, offset);
if (ws_xton(ch) == -1)
return -1;
} else if (ch != '+') {
/* Key, matched '=', stop. */
break;
}
len++;
offset++;
}
*ptr = tmp = (char*)wmem_alloc(pool, len + 1);
tmp[len] = '\0';
len = 0;
offset = orig_offset;
while (tvb_reported_length_remaining(tvb, offset)) {
found_offset = tvb_ws_mempbrk_pattern_guint8(tvb, offset, -1, pbrk, &ch);
if (found_offset == -1) {
tvb_memcpy(tvb, &tmp[len], offset, tvb_reported_length_remaining(tvb, offset));
offset = tvb_reported_length(tvb);
break;
}
tvb_memcpy(tvb, &tmp[len], offset, found_offset - offset);
len += (found_offset - offset);
offset = found_offset;
if (ch == '%') {
guint8 ch1, ch2;
offset++;
ch1 = tvb_get_guint8(tvb, offset);
offset++;
ch2 = tvb_get_guint8(tvb, offset);
tmp[len] = ws_xton(ch1) << 4 | ws_xton(ch2);
} else if (ch == '+') {
tmp[len] = ' ';
} else {
/* Key, matched '=', stop */
break;
}
len++;
offset++;
}
return offset;
}
static int
dissect_form_urlencoded(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void* data)
{
proto_tree *url_tree;
proto_tree *sub;
proto_item *ti;
gint offset = 0, next_offset, end_offset;
const char *data_name;
http_message_info_t *message_info;
tvbuff_t *sequence_tvb;
data_name = pinfo->match_string;
if (! (data_name && data_name[0])) {
/*
* No information from "match_string"
*/
message_info = (http_message_info_t *)data;
if (message_info == NULL) {
/*
* No information from dissector data
*/
data_name = NULL;
} else {
data_name = message_info->media_str;
if (! (data_name && data_name[0])) {
/*
* No information from dissector data
*/
data_name = NULL;
}
}
}
if (data_name)
col_append_sep_fstr(pinfo->cinfo, COL_INFO, " ", "(%s)", data_name);
ti = proto_tree_add_item(tree, proto_urlencoded, tvb, 0, -1, ENC_NA);
if (data_name)
proto_item_append_text(ti, ": %s", data_name);
url_tree = proto_item_add_subtree(ti, ett_form_urlencoded);
while (tvb_reported_length_remaining(tvb, offset) > 0) {
char *key, *value;
char *key_decoded, *value_decoded;
end_offset = tvb_find_guint8(tvb, offset, -1, '&');
if (end_offset == -1) {
end_offset = (int)tvb_reported_length(tvb);
}
sub = proto_tree_add_subtree(url_tree, tvb, offset, end_offset - offset, ett_form_keyvalue, NULL, "Form item");
sequence_tvb = tvb_new_subset_length(tvb, 0, end_offset);
next_offset = get_form_key_value(pinfo->pool, sequence_tvb, &key, offset, &pbrk_key);
if (next_offset == -1)
break;
/* XXX: Only UTF-8 is conforming according to WHATWG, though we
* ought to look for a "charset" parameter in media_str
* to handle other encodings.
* Our charset functions should probably return a boolean
* indicating that replacement characters had to be used,
* and that the string was not the expected encoding.
*/
key_decoded = get_utf_8_string(pinfo->pool, key, (int)strlen(key));
proto_tree_add_string(sub, hf_form_key, tvb, offset, next_offset - offset, key_decoded);
proto_item_append_text(sub, ": \"%s\"", format_text(pinfo->pool, key, strlen(key)));
offset = next_offset+1;
next_offset = get_form_key_value(pinfo->pool, sequence_tvb, &value, offset, &pbrk_value);
if (next_offset == -1)
break;
value_decoded = get_utf_8_string(pinfo->pool, value, (int)strlen(value));
proto_tree_add_string(sub, hf_form_value, tvb, offset, next_offset - offset, value_decoded);
proto_item_append_text(sub, " = \"%s\"", format_text(pinfo->pool, value, strlen(value)));
offset = next_offset+1;
}
return tvb_captured_length(tvb);
}
void
proto_register_http_urlencoded(void)
{
static hf_register_info hf[] = {
{ &hf_form_key,
{ "Key", "urlencoded-form.key",
FT_STRINGZ, BASE_NONE, NULL, 0x0,
NULL, HFILL }
},
{ &hf_form_value,
{ "Value", "urlencoded-form.value",
FT_STRINGZ, BASE_NONE, NULL, 0x0,
NULL, HFILL }
},
};
static gint *ett[] = {
&ett_form_urlencoded,
&ett_form_keyvalue
};
proto_urlencoded = proto_register_protocol("HTML Form URL Encoded", "URL Encoded Form Data", "urlencoded-form");
form_urlencoded_handle = register_dissector("urlencoded-form", dissect_form_urlencoded, proto_urlencoded);
proto_register_field_array(proto_urlencoded, hf, array_length(hf));
proto_register_subtree_array(ett, array_length(ett));
ws_mempbrk_compile(&pbrk_key, "%+=");
ws_mempbrk_compile(&pbrk_value, "%+");
}
void
proto_reg_handoff_http_urlencoded(void)
{
dissector_add_string("media_type", "application/x-www-form-urlencoded", form_urlencoded_handle);
}
/*
* Editor modelines - https://www.wireshark.org/tools/modelines.html
*
* Local variables:
* c-basic-offset: 8
* tab-width: 8
* indent-tabs-mode: t
* End:
*
* vi: set shiftwidth=8 tabstop=8 noexpandtab:
* :indentSize=8:tabSize=8:noTabs=false:
*/