1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
|
#! /usr/bin/env perl
#
# %CopyrightBegin%
#
# Copyright Ericsson AB 2009-2011. All Rights Reserved.
#
# The contents of this file are subject to the Erlang Public License,
# Version 1.1, (the "License"); you may not use this file except in
# compliance with the License. You should have received a copy of the
# Erlang Public License along with this software. If not, it can be
# retrieved online at http://www.erlang.org/.
#
# Software distributed under the License is distributed on an "AS IS"
# basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See
# the License for the specific language governing rights and limitations
# under the License.
#
# %CopyrightEnd%
#
use strict;
# Generate ADT (Abstract Data Type) access and generation functions
# for internal records.
#
# The following defines which ADT function sets that will be generated
# and which record fields that will be exponated.
#
# (FunctionBaseName => [RecordName, FieldName ...], ...)
my %Names = ('msg' => ['dns_rec', 'header', 'qdlist',
'anlist', 'nslist', 'arlist'],
'dns_rr' => ['dns_rr', 'domain', 'type', 'class', 'ttl', 'data'],
'dns_rr_opt' => ['dns_rr_opt', 'domain', 'type',
'udp_payload_size', 'ext_rcode', 'version',
'z', 'data'],
'dns_query' => ['dns_query', 'domain', 'type', 'class'],
'header' => ['dns_header', 'id', 'qr', 'opcode', 'aa', 'tc',
'rd', 'ra', 'pr', 'rcode']);
# The functions are defined in the __DATA__ section at the end.
# Read in __DATA__ and merge lines.
my $line = '';
my @DATA;
my @INDEX;
while(<DATA>) {
chomp;
$line .= $_;
unless ($line =~ s/\\$//) {
if ($line =~ s/^[+]//) {
push(@INDEX, $line);
} else {
push(@DATA, $line);
}
$line = '';
}
}
$" = ',';
$\ = "\n";
while( my ($Name, $r) = each(%Names)) {
# Create substitutions for this Name
my ($Record, @Fields) = @{ $r };
my @FieldMatchValues;
my @FieldValueTuples;
my @Values;
my $n = $#{ $r };
for my $i ( 1 .. $n ) {
push(@FieldMatchValues, "$Fields[$i-1]=V$i");
push(@FieldValueTuples, "{$Fields[$i-1],V$i}");
push(@Values, "V$i");
}
# "@FieldMatchValues" = "field1=V1,field2=V2"...",fieldN=VN"
# "@FieldMatchTuples" = "{field1,V1},{field2,V2}"...",{fieldN,VN}"
# "@Values" = "V1,V2"...",VN"
my @D = @DATA;
foreach my $line (@D) {
# Ignore !name lines
if ($line =~ s/^\!(\S+)\s+//) {
next if $1 eq $Name;
}
my $m = 1;
# For leading * iterate $n times, otherwise once
$line =~ s/^\s*[*]// and $m = $n;
for my $i ( 1 .. $m ) {
# For this iteration - substitute and print
my $Value = "V$i";
my $SemicolonDot = ";";
$SemicolonDot = "." if $i == $m;
my @ValuesIgnoreValue = @Values;
$ValuesIgnoreValue[$i-1] = '_';
# "$Value" = "V1" or "V2" or ... "VN"
# "@ValuesIgnoreValue" = "_,V2"...",VN"
# or "V1,_"...",VN"
# or ... "V1,V2"...",_"
$_ = $line;
s/FieldMatchValues\b/@FieldMatchValues/g;
s/FieldValueTuples\b/@FieldValueTuples/g;
s/Field\b/$Fields[$i-1]/g;
s/Name\b/$Name/g;
s/Record\b/$Record/g;
s/ValuesIgnoreValue\b/@ValuesIgnoreValue/g;
s/Values\b/@Values/g;
s/Value\b/$Value/g;
s/[;][.]/$SemicolonDot/g;
s/->\s*/->\n /;
print;
}
}
}
for my $i ( 0 .. $#INDEX ) {
my $line = $INDEX[$i];
if ($line =~ s/^[*]//) {
while( my ($Name, $r) = each(%Names)) {
my ($Record) = @{ $r };
$_ = $line;
s/Name\b/$Name/g;
s/Record\b/$Record/g;
s/->\s*/->\n /;
print;
}
} else {
print $line;
}
}
# Trailing \ will merge line with the following.
# Leading * will iterate the (merged) line over all field names.
# Sub-words in the loop above are substituted.
__DATA__
%%
%% Abstract Data Type functions for #Record{}
%%
%% -export([Name/1, Name/2,
%% make_Name/0, make_Name/1, make_Name/2, make_Name/3]).
%% Split #Record{} into property list
%%
Name(#Record{FieldMatchValues}) -> \
[FieldValueTuples].
%% Get one field value from #Record{}
%%
*Name(#Record{Field=Value}, Field) -> \
Value;
%% Map field name list to value list from #Record{}
%%
Name(#Record{}, []) -> \
[];
*Name(#Record{Field=Value}=R, [Field|L]) -> \
[Value|Name(R, L)];.
%% Generate default #Record{}
%%
make_Name() -> \
#Record{}.
%% Generate #Record{} from property list
%%
make_Name(L) when is_list(L) -> \
make_Name(#Record{}, L).
!dns_rr_opt %% Generate #Record{} with one updated field
!dns_rr_opt %%
!dns_rr_opt *make_Name(Field, Value) -> \
#Record{Field=Value};
%%
%% Update #Record{} from property list
%%
make_Name(#Record{FieldMatchValues}, L) when is_list(L) -> \
do_make_Name(L, Values).
do_make_Name([], Values) -> \
#Record{FieldMatchValues};
*do_make_Name([{Field,Value}|L], ValuesIgnoreValue) -> \
do_make_Name(L, Values);.
%% Update one field of #Record{}
%%
*make_Name(#Record{}=R, Field, Value) -> \
R#Record{Field=Value};.
+%% Record type index
+%%
+*record_adts(#Record{}) -> Name;
+record_adts(_) -> undefined.
|