public void RegexEncoding2()
{
var SJIS = RubyEncoding.KCodeSJIS.StrictEncoding;
// 1.9 encodings:
var invalidUtf8 = MutableString.CreateBinary(new byte[] { 0x80 }, RubyEncoding.UTF8);
AssertExceptionThrown <ArgumentException>(() => new RubyRegex(invalidUtf8, RubyRegexOptions.NONE));
// LastMatch
MatchData m;
var u = MutableString.CreateBinary(SJIS.GetBytes("あああ"), RubyEncoding.KCodeSJIS);
var p = MutableString.CreateBinary(SJIS.GetBytes("あ{2}"), RubyEncoding.KCodeSJIS);
var rs = new RubyRegex(p, RubyRegexOptions.SJIS);
// /あ{2}/ matches "あああ", the resulting index is in bytes:
m = rs.LastMatch(null, u);
Assert(m != null && m.Index == 2);
rs = new RubyRegex(MutableString.CreateBinary(SJIS.GetBytes("あ")), RubyRegexOptions.SJIS);
// "start at" in the middle of a character:
m = rs.LastMatch(null, u, 0);
Assert(m != null && m.Index == 0);
m = rs.LastMatch(null, u, 1);
Assert(m != null && m.Index == 0);
m = rs.LastMatch(null, u, 2);
Assert(m != null && m.Index == 2);
m = rs.LastMatch(null, u, 3);
Assert(m != null && m.Index == 2);
// Split
u = MutableString.CreateBinary(SJIS.GetBytes("あちあちあ"), RubyEncoding.UTF8);
rs = new RubyRegex(MutableString.CreateBinary(SJIS.GetBytes("ち")), RubyRegexOptions.SJIS);
var parts = rs.Split(null, u);
Assert(parts.Length == 3);
foreach (var part in parts)
{
Assert(part.Encoding == RubyEncoding.KCodeSJIS);
Assert(part.ToString() == "あ");
}
// groups
rs = new RubyRegex(MutableString.CreateBinary(SJIS.GetBytes("ち(a(あ+)(b+))+あ")), RubyRegexOptions.SJIS);
u = MutableString.CreateBinary(SJIS.GetBytes("ちaああbaあbbbあ"));
m = rs.Match(null, u);
Assert(m.GroupCount == 4);
int s, l;
Assert(m.GetGroupStart(0) == (s = 0));
Assert(m.GetGroupLength(0) == (l = u.GetByteCount()));
Assert(m.GetGroupEnd(0) == s + l);
// the group has 2 captures, the last one is its value:
Assert(m.GetGroupStart(1) == (s = SJIS.GetByteCount("ちaああb")));
Assert(m.GetGroupLength(1) == (l = SJIS.GetByteCount("aあbbb")));
Assert(m.GetGroupEnd(1) == s + l);
// the group has 2 captures, the last one is its value:
Assert(m.GetGroupStart(2) == (s = SJIS.GetByteCount("ちaああba")));
Assert(m.GetGroupLength(2) == (l = SJIS.GetByteCount("あ")));
Assert(m.GetGroupEnd(2) == s + l);
// the group has 2 captures, the last one is its value:
Assert(m.GetGroupStart(3) == (s = SJIS.GetByteCount("ちaああbaあ")));
Assert(m.GetGroupLength(3) == (l = SJIS.GetByteCount("bbb")));
Assert(m.GetGroupEnd(3) == s + l);
}
public void RegexEncoding2() {
var SJIS = RubyEncoding.KCodeSJIS.StrictEncoding;
// 1.9 encodings:
var invalidUtf8 = MutableString.CreateBinary(new byte[] { 0x80 }, RubyEncoding.UTF8);
AssertExceptionThrown<ArgumentException>(() => new RubyRegex(invalidUtf8, RubyRegexOptions.NONE));
// LastMatch
MatchData m;
var u = MutableString.CreateBinary(SJIS.GetBytes("あああ"), RubyEncoding.KCodeSJIS);
var p = MutableString.CreateBinary(SJIS.GetBytes("あ{2}"), RubyEncoding.KCodeSJIS);
var rs = new RubyRegex(p, RubyRegexOptions.SJIS);
// /あ{2}/ matches "あああ", the resulting index is in bytes:
m = rs.LastMatch(null, u);
Assert(m != null && m.Index == 2);
rs = new RubyRegex(MutableString.CreateBinary(SJIS.GetBytes("あ")), RubyRegexOptions.SJIS);
// "start at" in the middle of a character:
m = rs.LastMatch(null, u, 0);
Assert(m != null && m.Index == 0);
m = rs.LastMatch(null, u, 1);
Assert(m != null && m.Index == 0);
m = rs.LastMatch(null, u, 2);
Assert(m != null && m.Index == 2);
m = rs.LastMatch(null, u, 3);
Assert(m != null && m.Index == 2);
// Split
u = MutableString.CreateBinary(SJIS.GetBytes("あちあちあ"), RubyEncoding.UTF8);
rs = new RubyRegex(MutableString.CreateBinary(SJIS.GetBytes("ち")), RubyRegexOptions.SJIS);
var parts = rs.Split(null, u);
Assert(parts.Length == 3);
foreach (var part in parts) {
Assert(part.Encoding == RubyEncoding.KCodeSJIS);
Assert(part.ToString() == "あ");
}
// groups
rs = new RubyRegex(MutableString.CreateBinary(SJIS.GetBytes("ち(a(あ+)(b+))+あ")), RubyRegexOptions.SJIS);
u = MutableString.CreateBinary(SJIS.GetBytes("ちaああbaあbbbあ"));
m = rs.Match(null, u);
Assert(m.GroupCount == 4);
int s, l;
Assert(m.GetGroupStart(0) == (s = 0));
Assert(m.GetGroupLength(0) == (l = u.GetByteCount()));
Assert(m.GetGroupEnd(0) == s + l);
// the group has 2 captures, the last one is its value:
Assert(m.GetGroupStart(1) == (s = SJIS.GetByteCount("ちaああb")));
Assert(m.GetGroupLength(1) == (l = SJIS.GetByteCount("aあbbb")));
Assert(m.GetGroupEnd(1) == s + l);
// the group has 2 captures, the last one is its value:
Assert(m.GetGroupStart(2) == (s = SJIS.GetByteCount("ちaああba")));
Assert(m.GetGroupLength(2) == (l = SJIS.GetByteCount("あ")));
Assert(m.GetGroupEnd(2) == s + l);
// the group has 2 captures, the last one is its value:
Assert(m.GetGroupStart(3) == (s = SJIS.GetByteCount("ちaああbaあ")));
Assert(m.GetGroupLength(3) == (l = SJIS.GetByteCount("bbb")));
Assert(m.GetGroupEnd(3) == s + l);
}
public void RegexEncoding1()
{
MatchData m;
// the k-coding of the pattern string is irrelevant:
foreach (var pe in new[] { RubyEncoding.Binary })
{
var p = MutableString.CreateBinary(new byte[] { 0x82, 0xa0, (byte)'{', (byte)'2', (byte)'}' }, pe);
var r = new RubyRegex(p, RubyRegexOptions.NONE);
var rs = new RubyRegex(p, RubyRegexOptions.SJIS);
// the k-coding of the string is irrelevant:
foreach (var se in new[] { RubyEncoding.Binary })
{
var s = MutableString.CreateBinary(new byte[] { 0x82, 0xa0, 0xa0 }, se);
var t = MutableString.CreateBinary(new byte[] { 0x82, 0xa0, 0xa0, 0x82, 0xa0, 0xa0, 0xff }, se);
var u = MutableString.CreateBinary(new byte[] { 0x82, 0xa0, 0x82, 0xa0, 0x82, 0xa0 }, se);
// /あ{2}/ does not match "あ\xa0"
m = r.Match(RubyEncoding.KCodeSJIS, s);
Assert(m == null);
// /\x82\xa0{2}/ matches "[ \x82\xa0\xa0 ] \x82\xa0\xa0\xff"
m = r.Match(null, s);
Assert(m != null && m.Index == 0);
// /\x82\xa0{2}/ matches "\x82\xa0\xa0 [ \x82\xa0\xa0 ] \xff" starting from byte #1:
m = r.Match(null, t, 1, false);
Assert(m != null && m.Index == 3 && m.Length == 3);
// /あ{2}/s does not match "あ\xa0", current KCODE is ignored
m = rs.Match(null, s);
Assert(m == null);
// /あ{2}/s does not match "あ\xa0", current KCODE is ignored
m = rs.Match(RubyEncoding.KCodeUTF8, s);
Assert(m == null);
// /あ{2}/s matches "ああ\xff", current KCODE is ignored
m = rs.Match(RubyEncoding.KCodeUTF8, u, 2, false);
Assert(m != null && m.Index == 2 && m.Length == 4);
// /あ{2}/ does not match "あ\xa0あ\xa0"
m = r.LastMatch(RubyEncoding.KCodeSJIS, t);
Assert(m == null);
// /\x82\xa0{2}/ matches "\x82\xa0\xa0 [ \x82\xa0\xa0 ] \xff"
m = r.LastMatch(null, t);
Assert(m != null && m.Index == 3);
// /あ{2}/s does not match "あ\xa0あ\xa0", current KCODE is ignored
m = rs.LastMatch(null, t);
Assert(m == null);
// /あ{2}/s does not match "あ\xa0あ\xa0", current KCODE is ignored
m = rs.LastMatch(RubyEncoding.KCodeUTF8, t);
Assert(m == null);
}
}
}
public void RegexEncoding1() {
MatchData m;
// the k-coding of the pattern string is irrelevant:
foreach (var pe in new[] { RubyEncoding.KCodeSJIS, RubyEncoding.KCodeUTF8, RubyEncoding.Binary }) {
var p = MutableString.CreateBinary(new byte[] { 0x82, 0xa0, (byte)'{', (byte)'2', (byte)'}' }, pe);
var r = new RubyRegex(p, RubyRegexOptions.NONE);
var rs = new RubyRegex(p, RubyRegexOptions.SJIS);
// the k-coding of the string is irrelevant:
foreach (var se in new[] { RubyEncoding.KCodeSJIS, RubyEncoding.KCodeUTF8, RubyEncoding.Binary }) {
var s = MutableString.CreateBinary(new byte[] { 0x82, 0xa0, 0xa0 }, se);
var t = MutableString.CreateBinary(new byte[] { 0x82, 0xa0, 0xa0, 0x82, 0xa0, 0xa0, 0xff }, se);
var u = MutableString.CreateBinary(new byte[] { 0x82, 0xa0, 0x82, 0xa0, 0x82, 0xa0 }, se);
// /あ{2}/ does not match "あ\xa0"
m = r.Match(RubyEncoding.KCodeSJIS, s);
Assert(m == null);
// /\x82\xa0{2}/ matches "[ \x82\xa0\xa0 ] \x82\xa0\xa0\xff"
m = r.Match(null, s);
Assert(m != null && m.Index == 0);
// /\x82\xa0{2}/ matches "\x82\xa0\xa0 [ \x82\xa0\xa0 ] \xff" starting from byte #1:
m = r.Match(null, t, 1, false);
Assert(m != null && m.Index == 3 && m.Length == 3);
// /あ{2}/s does not match "あ\xa0", current KCODE is ignored
m = rs.Match(null, s);
Assert(m == null);
// /あ{2}/s does not match "あ\xa0", current KCODE is ignored
m = rs.Match(RubyEncoding.KCodeUTF8, s);
Assert(m == null);
// /あ{2}/s matches "ああ\xff", current KCODE is ignored
m = rs.Match(RubyEncoding.KCodeUTF8, u, 2, false);
Assert(m != null && m.Index == 2 && m.Length == 4);
// /あ{2}/ does not match "あ\xa0あ\xa0"
m = r.LastMatch(RubyEncoding.KCodeSJIS, t);
Assert(m == null);
// /\x82\xa0{2}/ matches "\x82\xa0\xa0 [ \x82\xa0\xa0 ] \xff"
m = r.LastMatch(null, t);
Assert(m != null && m.Index == 3);
// /あ{2}/s does not match "あ\xa0あ\xa0", current KCODE is ignored
m = rs.LastMatch(null, t);
Assert(m == null);
// /あ{2}/s does not match "あ\xa0あ\xa0", current KCODE is ignored
m = rs.LastMatch(RubyEncoding.KCodeUTF8, t);
Assert(m == null);
}
}
}