v2_indent.go

     1  // Copyright 2010 The Go Authors. All rights reserved.
     2  // Use of this source code is governed by a BSD-style
     3  // license that can be found in the LICENSE file.
     4  
     5  //go:build goexperiment.jsonv2
     6  
     7  package json
     8  
     9  import (
    10  	"bytes"
    11  	"strings"
    12  
    13  	"encoding/json/jsontext"
    14  )
    15  
    16  // HTMLEscape appends to dst the JSON-encoded src with <, >, &, U+2028 and U+2029
    17  // characters inside string literals changed to \u003c, \u003e, \u0026, \u2028, \u2029
    18  // so that the JSON will be safe to embed inside HTML <script> tags.
    19  // For historical reasons, web browsers don't honor standard HTML
    20  // escaping within <script> tags, so an alternative JSON encoding must be used.
    21  func HTMLEscape(dst *bytes.Buffer, src []byte) {
    22  	dst.Grow(len(src))
    23  	dst.Write(appendHTMLEscape(dst.AvailableBuffer(), src))
    24  }
    25  
    26  func appendHTMLEscape(dst, src []byte) []byte {
    27  	const hex = "0123456789abcdef"
    28  	// The characters can only appear in string literals,
    29  	// so just scan the string one byte at a time.
    30  	start := 0
    31  	for i, c := range src {
    32  		if c == '<' || c == '>' || c == '&' {
    33  			dst = append(dst, src[start:i]...)
    34  			dst = append(dst, '\\', 'u', '0', '0', hex[c>>4], hex[c&0xF])
    35  			start = i + 1
    36  		}
    37  		// Convert U+2028 and U+2029 (E2 80 A8 and E2 80 A9).
    38  		if c == 0xE2 && i+2 < len(src) && src[i+1] == 0x80 && src[i+2]&^1 == 0xA8 {
    39  			dst = append(dst, src[start:i]...)
    40  			dst = append(dst, '\\', 'u', '2', '0', '2', hex[src[i+2]&0xF])
    41  			start = i + len("\u2029")
    42  		}
    43  	}
    44  	return append(dst, src[start:]...)
    45  }
    46  
    47  // Compact appends to dst the JSON-encoded src with
    48  // insignificant space characters elided.
    49  func Compact(dst *bytes.Buffer, src []byte) error {
    50  	dst.Grow(len(src))
    51  	b := dst.AvailableBuffer()
    52  	b, err := jsontext.AppendFormat(b, src,
    53  		ReportErrorsWithLegacySemantics(true),
    54  		jsontext.AllowDuplicateNames(true),
    55  		jsontext.AllowInvalidUTF8(true),
    56  		jsontext.PreserveRawStrings(true))
    57  	if err != nil {
    58  		return transformSyntacticError(err)
    59  	}
    60  	dst.Write(b)
    61  	return nil
    62  }
    63  
    64  // indentGrowthFactor specifies the growth factor of indenting JSON input.
    65  // Empirically, the growth factor was measured to be between 1.4x to 1.8x
    66  // for some set of compacted JSON with the indent being a single tab.
    67  // Specify a growth factor slightly larger than what is observed
    68  // to reduce probability of allocation in appendIndent.
    69  // A factor no higher than 2 ensures that wasted space never exceeds 50%.
    70  const indentGrowthFactor = 2
    71  
    72  // Indent appends to dst an indented form of the JSON-encoded src.
    73  // Each element in a JSON object or array begins on a new,
    74  // indented line beginning with prefix followed by one or more
    75  // copies of indent according to the indentation nesting.
    76  // The data appended to dst does not begin with the prefix nor
    77  // any indentation, to make it easier to embed inside other formatted JSON data.
    78  // Although leading space characters (space, tab, carriage return, newline)
    79  // at the beginning of src are dropped, trailing space characters
    80  // at the end of src are preserved and copied to dst.
    81  // For example, if src has no trailing spaces, neither will dst;
    82  // if src ends in a trailing newline, so will dst.
    83  func Indent(dst *bytes.Buffer, src []byte, prefix, indent string) error {
    84  	dst.Grow(indentGrowthFactor * len(src))
    85  	b := dst.AvailableBuffer()
    86  	b, err := appendIndent(b, src, prefix, indent)
    87  	dst.Write(b)
    88  	return err
    89  }
    90  
    91  func appendIndent(dst, src []byte, prefix, indent string) ([]byte, error) {
    92  	// In v2, only spaces and tabs are allowed, while v1 allowed any character.
    93  	dstLen := len(dst)
    94  	if len(strings.Trim(prefix, " \t"))+len(strings.Trim(indent, " \t")) > 0 {
    95  		// Use placeholder spaces of correct length, and replace afterwards.
    96  		invalidPrefix, invalidIndent := prefix, indent
    97  		prefix = strings.Repeat(" ", len(prefix))
    98  		indent = strings.Repeat(" ", len(indent))
    99  		defer func() {
   100  			b := dst[dstLen:]
   101  			for i := bytes.IndexByte(b, '\n'); i >= 0; i = bytes.IndexByte(b, '\n') {
   102  				b = b[i+len("\n"):]
   103  				n := len(b) - len(bytes.TrimLeft(b, " ")) // len(prefix)+n*len(indent)
   104  				spaces := b[:n]
   105  				spaces = spaces[copy(spaces, invalidPrefix):]
   106  				for len(spaces) > 0 {
   107  					spaces = spaces[copy(spaces, invalidIndent):]
   108  				}
   109  				b = b[n:]
   110  			}
   111  		}()
   112  	}
   113  
   114  	dst, err := jsontext.AppendFormat(dst, src,
   115  		ReportErrorsWithLegacySemantics(true),
   116  		jsontext.AllowDuplicateNames(true),
   117  		jsontext.AllowInvalidUTF8(true),
   118  		jsontext.PreserveRawStrings(true),
   119  		jsontext.Multiline(true),
   120  		jsontext.WithIndentPrefix(prefix),
   121  		jsontext.WithIndent(indent))
   122  	if err != nil {
   123  		return dst[:dstLen], transformSyntacticError(err)
   124  	}
   125  
   126  	// In v2, trailing whitespace is discarded, while v1 preserved it.
   127  	if n := len(src) - len(bytes.TrimRight(src, " \n\r\t")); n > 0 {
   128  		dst = append(dst, src[len(src)-n:]...)
   129  	}
   130  	return dst, nil
   131  }
   132
View as plain text