diff options
| author | Luke Shumaker <lukeshu@lukeshu.com> | 2023-02-16 19:06:46 -0700 |
|---|---|---|
| committer | Luke Shumaker <lukeshu@lukeshu.com> | 2023-02-18 22:45:54 -0700 |
| commit | 00187950437a10952b82353405e5ba4b4515fb29 (patch) | |
| tree | 826c4ff76310bf6f58e79f37f2107c329810aaa8 /encode_escape.go | |
| parent | a87d6cbbb51a19071c5c742ef3c91bbb90a727c6 (diff) | |
reencode: Don't normalize the capitalization of \uXXXX hex escapes
Diffstat (limited to 'encode_escape.go')
| -rw-r--r-- | encode_escape.go | 39 |
1 files changed, 36 insertions, 3 deletions
diff --git a/encode_escape.go b/encode_escape.go index c9e2bc9..664c762 100644 --- a/encode_escape.go +++ b/encode_escape.go @@ -36,7 +36,8 @@ const ( // - as a short "well-known" `\X` backslash sequence (where `X` is a // single-character) // -// - as a long Unicode `\uXXXX` backslash sequence +// - as a long Unicode `\uXXXX` backslash sequence (with 16 +// permutations of capitalization) // // - as a raw byte; this allows you to emit invalid JSON; JSON must // be valid UTF-8, but this allows you to emit arbitrary binary @@ -47,8 +48,29 @@ type BackslashEscapeMode = jsonstring.BackslashEscapeMode const ( BackslashEscapeNone = jsonstring.BackslashEscapeNone BackslashEscapeShort = jsonstring.BackslashEscapeShort - BackslashEscapeUnicode = jsonstring.BackslashEscapeUnicode BackslashEscapeRawByte = jsonstring.BackslashEscapeRawByte + + BackslashEscapeUnicodeXXXX = jsonstring.BackslashEscapeUnicodeXXXX + BackslashEscapeUnicodeXXXx = jsonstring.BackslashEscapeUnicodeXXXx + BackslashEscapeUnicodeXXxX = jsonstring.BackslashEscapeUnicodeXXxX + BackslashEscapeUnicodeXXxx = jsonstring.BackslashEscapeUnicodeXXxx + BackslashEscapeUnicodeXxXX = jsonstring.BackslashEscapeUnicodeXxXX + BackslashEscapeUnicodeXxXx = jsonstring.BackslashEscapeUnicodeXxXx + BackslashEscapeUnicodeXxxX = jsonstring.BackslashEscapeUnicodeXxxX + BackslashEscapeUnicodeXxxx = jsonstring.BackslashEscapeUnicodeXxxx + BackslashEscapeUnicodexXXX = jsonstring.BackslashEscapeUnicodexXXX + BackslashEscapeUnicodexXXx = jsonstring.BackslashEscapeUnicodexXXx + BackslashEscapeUnicodexXxX = jsonstring.BackslashEscapeUnicodexXxX + BackslashEscapeUnicodexXxx = jsonstring.BackslashEscapeUnicodexXxx + BackslashEscapeUnicodexxXX = jsonstring.BackslashEscapeUnicodexxXX + BackslashEscapeUnicodexxXx = jsonstring.BackslashEscapeUnicodexxXx + BackslashEscapeUnicodexxxX = jsonstring.BackslashEscapeUnicodexxxX + BackslashEscapeUnicodexxxx = jsonstring.BackslashEscapeUnicodexxxx + + BackslashEscapeUnicodeMin = jsonstring.BackslashEscapeUnicodeMin + BackslashEscapeUnicodeMax = jsonstring.BackslashEscapeUnicodeMax + + BackslashEscapeUnicode = jsonstring.BackslashEscapeUnicode // back-compat ) func hexToInt(c byte) rune { @@ -72,13 +94,24 @@ func hexToRune(a, b, c, d byte) rune { hexToInt(d)<<0 } +func hexToMode(a, b, c, d byte) BackslashEscapeMode { + // The 0b0010_0000 bit is the ASCII "lowercase bit". + return BackslashEscapeUnicodeMin + BackslashEscapeMode(0| + ((a&0b0010_0000)>>2)| + ((b&0b0010_0000)>>3)| + ((c&0b0010_0000)>>4)| + ((d&0b0010_0000)>>5)) +} + // A BackslashEscaper controls how a ReEncoder emits a character in a // JSON string. The `rune` argument is the character being // considered, and the `BackslashEscapeMode` argument is how it was // originally encoded in the input. // // The ReEncoder will panic if a BackslashEscaper returns an unknown -// BackslashEscapeMode. +// BackslashEscapeMode. However, a BackslashEscaper should be +// permissive of BackslashEscapeModes it doesn't recognize; it is safe +// to just return them unmodified. type BackslashEscaper = func(rune, BackslashEscapeMode) BackslashEscapeMode // EscapePreserve is a BackslashEscaper that preserves the original |