blib/lib/Encode/CN/HZ.pm

filter: Statementstmt 63.4% 73 / 115 Branchbran 54.5% 36 / 66 Conditioncond 100% 2 / 2 MC/DCmcdc 100% 1 / 1 Subroutinesub 91.6% 11 / 12 Podpod 100% 4 / 4 totaltotal 63.5% 127 / 200 CC 35 SCAR 38.7
CC
35
Cov
61%
cat_decode
156.0
encode
14.3
decode
11.8
CRAP 48.1
« blib/lib/Encode/MIME/Header.pm ↑ Summary blib/lib/Encode/Guess.pm »
1 package Encode::CN::HZ;
2  
3 9 ▶use strict;
Subroutine BEGIN: called CC=1 SCAR=0.0
4 9 ▶use warnings;
Subroutine BEGIN: called CC=1 SCAR=0.0
5 9 ▶use utf8 ();
Subroutine BEGIN: called CC=1 SCAR=0.0
6  
7 9 ▶use vars qw($VERSION);
Subroutine BEGIN: called CC=1 SCAR=0.0
8 9 $VERSION = '2.11';
9  
10 9 ▶use Encode qw(:fallbacks);
Subroutine BEGIN: called CC=1 SCAR=0.0
11  
12 9 ▶use parent qw(Encode::Encoding);
Subroutine BEGIN: called CC=1 SCAR=0.0
13 9 __PACKAGE__->Define('hz');
14  
15 # HZ is a combination of ASCII and escaped GB, so we implement it
16 # with the GB2312(raw) encoding here. Cf. RFCs 1842 & 1843.
17  
18 # not ported for EBCDIC. Which should be used, "~" or "\x7E"?
19  
20 4 ▶sub needs_lines { 1 }
Subroutine needs_lines: called CC=1 SCAR=0.0
Pod documented
21  
22 13 ▶sub decode ($$;$) {
Subroutine decode: called CC=11 SCAR=24.7
Pod documented
23 13 my ( $obj, $str, $chk ) = @_;
24 13 ▶ return undef unless defined $str;
Branch: 1 branch
BranchTrueFalseTotal
unless defined $str 2 11 13
25  
26 11 my $GB = Encode::find_encoding('gb2312-raw');
27 11 my $ret = substr($str, 0, 0); # to propagate taintedness
28 11 my $in_ascii = 1; # default mode is ASCII.
29  
30 11 ▶ while ( length $str ) {
Branch: 1 branch
BranchTrueFalseTotal
if (length $str) 3011 11 3022
31 3011 ▶ if ($in_ascii) { # ASCII mode
Branch: 1 branch
BranchTrueFalseTotal
if ($in_ascii) { } 1509 1502 3011
32 1509 ▶ if ( $str =~ s/^([\x00-\x7D\x7F]+)// ) { # no '~' => ASCII
Branch: 4 branches
BranchTrueFalseTotal
if ($str =~ s/^([\x00-\x7D\x7F]+)//) { } 758 751 1509
elsif ($str =~ s/^\x7E\x7E//) { } 0 751 751
elsif ($str =~ s/^\x7E\cJ//) { } 0 751 751
elsif ($str =~ s/^\x7E\x7B//) { } 751 0 751
33 758 $ret .= $1;
34  
35 # EBCDIC should need ascii2native, but not ported.
36 }
37 elsif ( $str =~ s/^\x7E\x7E// ) { # escaped tilde
38 0 $ret .= '~';
39 }
40 elsif ( $str =~ s/^\x7E\cJ// ) { # '\cJ' == LF in ASCII
41 0 1; # no-op
42 }
43 elsif ( $str =~ s/^\x7E\x7B// ) { # '~{'
44 751 $in_ascii = 0; # to GB
45 }
46 else { # encounters an invalid escape, \x80 or greater
47 0 last;
48 }
49 }
50 else { # GB mode; the byte ranges are as in RFC 1843.
51 9 ▶ no warnings 'uninitialized';
Subroutine BEGIN: called CC=1 SCAR=0.0
52 1502 ▶ if ( $str =~ s/^((?:[\x21-\x77][\x21-\x7E])+)// ) {
Branch: 2 branches
BranchTrueFalseTotal
if ($str =~ s/^((?:[\x21-\x77][\x21-\x7E])+)//) { } 751 751 1502
elsif ($str =~ s/^\x7E\x7D//) { } 751 0 751
53 751 my $prefix = $1;
54 751 $ret .= $GB->decode( $prefix, $chk );
55 }
56 elsif ( $str =~ s/^\x7E\x7D// ) { # '~}'
57 751 $in_ascii = 1;
58 }
59 else { # invalid
60 0 last;
61 }
62 }
63 }
64 11 ▶ $_[1] = '' if $chk; # needs_lines guarantees no partial character
Branch: 1 branch
BranchTrueFalseTotal
if $chk 6 5 11
65 11 return $ret;
66 }
67  
68 0 ▶sub cat_decode {
Subroutine cat_decode: not called CC=12 SCAR=50.5
Pod documented
69 0 my ( $obj, undef, $src, $pos, $trm, $chk ) = @_;
70 0 my ( $rdst, $rsrc, $rpos ) = \@_[ 1 .. 3 ];
71  
72 0 my $GB = Encode::find_encoding('gb2312-raw');
73 0 my $ret = '';
74 0 my $in_ascii = 1; # default mode is ASCII.
75  
76 0 my $ini_pos = pos($$rsrc);
77  
78 0 substr( $src, 0, $pos ) = '';
79  
80 0 my $ini_len = bytes::length($src);
81  
82 # $trm is the first of the pair '~~', then 2nd tilde is to be removed.
83 # XXX: Is better C<$src =~ s/^\x7E// or die if ...>?
84 0 ▶ $src =~ s/^\x7E// if $trm eq "\x7E";
Branch: 1 branch
BranchTrueFalseTotal
if $trm eq "~" 0 0 0
85  
86 0 ▶ while ( length $src ) {
Branch: 1 branch
BranchTrueFalseTotal
if (length $src) 0 0 0
87 0 my $now;
88 0 ▶ if ($in_ascii) { # ASCII mode
Branch: 1 branch
BranchTrueFalseTotal
if ($in_ascii) { } 0 0 0
89 0 ▶ if ( $src =~ s/^([\x00-\x7D\x7F])// ) { # no '~' => ASCII
Branch: 4 branches
BranchTrueFalseTotal
if ($src =~ s/^([\x00-\x7D\x7F])//) { } 0 0 0
elsif ($src =~ s/^\x7E\x7E//) { } 0 0 0
elsif ($src =~ s/^\x7E\cJ//) { } 0 0 0
elsif ($src =~ s/^\x7E\x7B//) { } 0 0 0
90 0 $now = $1;
91 }
92 elsif ( $src =~ s/^\x7E\x7E// ) { # escaped tilde
93 0 $now = '~';
94 }
95 elsif ( $src =~ s/^\x7E\cJ// ) { # '\cJ' == LF in ASCII
96 0 next;
97 }
98 elsif ( $src =~ s/^\x7E\x7B// ) { # '~{'
99 0 $in_ascii = 0; # to GB
100 0 next;
101 }
102 else { # encounters an invalid escape, \x80 or greater
103 0 last;
104 }
105 }
106 else { # GB mode; the byte ranges are as in RFC 1843.
107 0 ▶ if ( $src =~ s/^((?:[\x21-\x77][\x21-\x7F])+)// ) {
Branch: 2 branches
BranchTrueFalseTotal
if ($src =~ s/^((?:[\x21-\x77][\x21-\x7F])+)//) { } 0 0 0
elsif ($src =~ s/^\x7E\x7D//) { } 0 0 0
108 0 $now = $GB->decode( $1, $chk );
109 }
110 elsif ( $src =~ s/^\x7E\x7D// ) { # '~}'
111 0 $in_ascii = 1;
112 0 next;
113 }
114 else { # invalid
115 0 last;
116 }
117 }
118  
119 0 ▶ next if !defined $now;
Branch: 1 branch
BranchTrueFalseTotal
unless defined $now 0 0 0
120  
121 0 $ret .= $now;
122  
123 0 ▶ if ( $now eq $trm ) {
Branch: 1 branch
BranchTrueFalseTotal
if ($now eq $trm) 0 0 0
124 0 $$rdst .= $ret;
125 0 $$rpos = $ini_pos + $pos + $ini_len - bytes::length($src);
126 0 pos($$rsrc) = $ini_pos;
127 0 return 1;
128 }
129 }
130  
131 0 $$rdst .= $ret;
132 0 $$rpos = $ini_pos + $pos + $ini_len - bytes::length($src);
133 0 pos($$rsrc) = $ini_pos;
134 0 return ''; # terminator not found
135 }
136  
137 492 ▶sub encode($$;$) {
Subroutine encode: called CC=14 SCAR=26.6
Pod documented
138 492 my ( $obj, $str, $chk ) = @_;
139 492 ▶ return undef unless defined $str;
Branch: 1 branch
BranchTrueFalseTotal
unless defined $str 2 490 492
140  
141 490 my $GB = Encode::find_encoding('gb2312-raw');
142 490 my $ret = substr($str, 0, 0); # to propagate taintedness;
143 490 my $in_ascii = 1; # default mode is ASCII.
144  
145 9 ▶ no warnings 'utf8'; # $str may be malformed UTF8 at the end of a chunk.
Subroutine BEGIN: called CC=1 SCAR=0.0
146  
147 490 ▶ while ( length $str ) {
Branch: 1 branch
BranchTrueFalseTotal
if (length $str) 31438 490 31928
148 31438 ▶ if ( $str =~ s/^([[:ascii:]]+)// ) {
Branch: 2 branches
BranchTrueFalseTotal
if ($str =~ s/^([[:ascii:]]+)//) { } 1482 29956 31438
elsif ($str =~ s/(.)//) { } 29956 0 29956
149 1482 my $tmp = $1;
150 1482 $tmp =~ s/~/~~/g; # escapes tildes
151 1482 ▶ if ( !$in_ascii ) {
Branch: 1 branch
BranchTrueFalseTotal
unless ($in_ascii) 994 488 1482
152 994 $ret .= "\x7E\x7D"; # '~}'
153 994 $in_ascii = 1;
154 }
155 1482 $ret .= pack 'a*', $tmp; # remove UTF8 flag.
156 }
157 elsif ( $str =~ s/(.)// ) {
158 29956 my $s = $1;
159 29956 ▶ my $tmp = $GB->encode( $s, $chk || 0 );
Condition100%2/2
$chk || 0
l!l
1489015066
MC/DC100%1/1
$chk || 0
$chk
Truth table100%2/2
A
A= $chk
Aresult
11
00
160 29956 ▶ last if !defined $tmp;
Branch: 1 branch
BranchTrueFalseTotal
unless defined $tmp 0 29956 29956
161 29956 ▶ if ( length $tmp == 2 ) { # maybe a valid GB char (XXX)
Branch: 2 branches
BranchTrueFalseTotal
if (length $tmp == 2) { } 29954 2 29956
elsif (length $tmp) { } 2 0 2
162 29954 ▶ if ($in_ascii) {
Branch: 1 branch
BranchTrueFalseTotal
if ($in_ascii) 996 28958 29954
163 996 $ret .= "\x7E\x7B"; # '~{'
164 996 $in_ascii = 0;
165 }
166 29954 $ret .= $tmp;
167 }
168 elsif ( length $tmp ) { # maybe FALLBACK in ASCII (XXX)
169 2 ▶ if ( !$in_ascii ) {
Branch: 1 branch
BranchTrueFalseTotal
unless ($in_ascii) 0 2 2
170 0 $ret .= "\x7E\x7D"; # '~}'
171 0 $in_ascii = 1;
172 }
173 2 $ret .= $tmp;
174 }
175 }
176 else { # if $str is malformed UTF8 *and* if length $str != 0.
177 0 last;
178 }
179 }
180 490 ▶ $_[1] = $str if $chk;
Branch: 1 branch
BranchTrueFalseTotal
if $chk 484 6 490
181  
182 # The state at the end of the chunk is discarded, even if in GB mode.
183 # That results in the combination of GB-OUT and GB-IN, i.e. "~}~{".
184 # Parhaps it is harmless, but further investigations may be required...
185  
186 490 ▶ if ( !$in_ascii ) {
Branch: 1 branch
BranchTrueFalseTotal
unless ($in_ascii) 2 488 490
187 2 $ret .= "\x7E\x7D"; # '~}'
188 2 $in_ascii = 1;
189 }
190 490 utf8::encode($ret); # https://rt.cpan.org/Ticket/Display.html?id=35120
191 490 return $ret;
192 }
193  
194 9 1;
195 __END__
« blib/lib/Encode/MIME/Header.pm ↑ Summary blib/lib/Encode/Guess.pm »