diff options
author | 2024-04-29 10:45:17 +0000 | |
---|---|---|
committer | 2024-04-29 10:45:17 +0000 | |
commit | bfc21e4850a893a7bebd566e46ba94bd1912a98b (patch) | |
tree | 33678bc0024f150bd80c09b42e6d76c58d4e19fe /vendor/github.com/tdewolff/parse/v2/common.go | |
parent | [chore]: Bump github.com/minio/minio-go/v7 from 7.0.69 to 7.0.70 (#2877) (diff) | |
download | gotosocial-bfc21e4850a893a7bebd566e46ba94bd1912a98b.tar.xz |
[chore]: Bump github.com/tdewolff/minify/v2 from 2.20.19 to 2.20.20 (#2875)
Diffstat (limited to 'vendor/github.com/tdewolff/parse/v2/common.go')
-rw-r--r-- | vendor/github.com/tdewolff/parse/v2/common.go | 305 |
1 files changed, 305 insertions, 0 deletions
diff --git a/vendor/github.com/tdewolff/parse/v2/common.go b/vendor/github.com/tdewolff/parse/v2/common.go index da46cc3df..e0795304c 100644 --- a/vendor/github.com/tdewolff/parse/v2/common.go +++ b/vendor/github.com/tdewolff/parse/v2/common.go @@ -5,6 +5,7 @@ import ( "bytes" "encoding/base64" "errors" + "strconv" ) var ( @@ -235,3 +236,307 @@ func QuoteEntity(b []byte) (quote byte, n int) { } return 0, 0 } + +// ReplaceMultipleWhitespace replaces character series of space, \n, \t, \f, \r into a single space or newline (when the serie contained a \n or \r). +func ReplaceMultipleWhitespace(b []byte) []byte { + j, k := 0, 0 // j is write position, k is start of next text section + for i := 0; i < len(b); i++ { + if IsWhitespace(b[i]) { + start := i + newline := IsNewline(b[i]) + i++ + for ; i < len(b) && IsWhitespace(b[i]); i++ { + if IsNewline(b[i]) { + newline = true + } + } + if newline { + b[start] = '\n' + } else { + b[start] = ' ' + } + if 1 < i-start { // more than one whitespace + if j == 0 { + j = start + 1 + } else { + j += copy(b[j:], b[k:start+1]) + } + k = i + } + } + } + if j == 0 { + return b + } else if j == 1 { // only if starts with whitespace + b[k-1] = b[0] + return b[k-1:] + } else if k < len(b) { + j += copy(b[j:], b[k:]) + } + return b[:j] +} + +// replaceEntities will replace in b at index i, assuming that b[i] == '&' and that i+3<len(b). The returned int will be the last character of the entity, so that the next iteration can safely do i++ to continue and not miss any entitites. +func replaceEntities(b []byte, i int, entitiesMap map[string][]byte, revEntitiesMap map[byte][]byte) ([]byte, int) { + const MaxEntityLength = 31 // longest HTML entity: CounterClockwiseContourIntegral + var r []byte + j := i + 1 + if b[j] == '#' { + j++ + if b[j] == 'x' { + j++ + c := 0 + for ; j < len(b) && (b[j] >= '0' && b[j] <= '9' || b[j] >= 'a' && b[j] <= 'f' || b[j] >= 'A' && b[j] <= 'F'); j++ { + if b[j] <= '9' { + c = c<<4 + int(b[j]-'0') + } else if b[j] <= 'F' { + c = c<<4 + int(b[j]-'A') + 10 + } else if b[j] <= 'f' { + c = c<<4 + int(b[j]-'a') + 10 + } + } + if j <= i+3 || 10000 <= c { + return b, j - 1 + } + if c < 128 { + r = []byte{byte(c)} + } else { + r = append(r, '&', '#') + r = strconv.AppendInt(r, int64(c), 10) + r = append(r, ';') + } + } else { + c := 0 + for ; j < len(b) && c < 128 && b[j] >= '0' && b[j] <= '9'; j++ { + c = c*10 + int(b[j]-'0') + } + if j <= i+2 || 128 <= c { + return b, j - 1 + } + r = []byte{byte(c)} + } + } else { + for ; j < len(b) && j-i-1 <= MaxEntityLength && b[j] != ';'; j++ { + } + if j <= i+1 || len(b) <= j { + return b, j - 1 + } + + var ok bool + r, ok = entitiesMap[string(b[i+1:j])] + if !ok { + return b, j + } + } + + // j is at semicolon + n := j + 1 - i + if j < len(b) && b[j] == ';' && 2 < n { + if len(r) == 1 { + if q, ok := revEntitiesMap[r[0]]; ok { + if len(q) == len(b[i:j+1]) && bytes.Equal(q, b[i:j+1]) { + return b, j + } + r = q + } else if r[0] == '&' { + // check if for example & is followed by something that could potentially be an entity + k := j + 1 + if k < len(b) && (b[k] >= '0' && b[k] <= '9' || b[k] >= 'a' && b[k] <= 'z' || b[k] >= 'A' && b[k] <= 'Z' || b[k] == '#') { + return b, k + } + } + } + + copy(b[i:], r) + copy(b[i+len(r):], b[j+1:]) + b = b[:len(b)-n+len(r)] + return b, i + len(r) - 1 + } + return b, i +} + +// ReplaceEntities replaces all occurrences of entites (such as ") to their respective unencoded bytes. +func ReplaceEntities(b []byte, entitiesMap map[string][]byte, revEntitiesMap map[byte][]byte) []byte { + for i := 0; i < len(b); i++ { + if b[i] == '&' && i+3 < len(b) { + b, i = replaceEntities(b, i, entitiesMap, revEntitiesMap) + } + } + return b +} + +// ReplaceMultipleWhitespaceAndEntities is a combination of ReplaceMultipleWhitespace and ReplaceEntities. It is faster than executing both sequentially. +func ReplaceMultipleWhitespaceAndEntities(b []byte, entitiesMap map[string][]byte, revEntitiesMap map[byte][]byte) []byte { + j, k := 0, 0 // j is write position, k is start of next text section + for i := 0; i < len(b); i++ { + if IsWhitespace(b[i]) { + start := i + newline := IsNewline(b[i]) + i++ + for ; i < len(b) && IsWhitespace(b[i]); i++ { + if IsNewline(b[i]) { + newline = true + } + } + if newline { + b[start] = '\n' + } else { + b[start] = ' ' + } + if 1 < i-start { // more than one whitespace + if j == 0 { + j = start + 1 + } else { + j += copy(b[j:], b[k:start+1]) + } + k = i + } + } + if i+3 < len(b) && b[i] == '&' { + b, i = replaceEntities(b, i, entitiesMap, revEntitiesMap) + } + } + if j == 0 { + return b + } else if j == 1 { // only if starts with whitespace + b[k-1] = b[0] + return b[k-1:] + } else if k < len(b) { + j += copy(b[j:], b[k:]) + } + return b[:j] +} + +// URLEncodingTable is a charmap for which characters need escaping in the URL encoding scheme +var URLEncodingTable = [256]bool{ + // ASCII + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + + true, false, true, true, true, true, true, false, // space, ", #, $, %, & + false, false, false, true, true, false, false, true, // +, comma, / + false, false, false, false, false, false, false, false, + false, false, true, true, true, true, true, true, // :, ;, <, =, >, ? + + true, false, false, false, false, false, false, false, // @ + false, false, false, false, false, false, false, false, + false, false, false, false, false, false, false, false, + false, false, false, true, true, true, true, false, // [, \, ], ^ + + true, false, false, false, false, false, false, false, // ` + false, false, false, false, false, false, false, false, + false, false, false, false, false, false, false, false, + false, false, false, true, true, true, false, true, // {, |, }, DEL + + // non-ASCII + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, +} + +// DataURIEncodingTable is a charmap for which characters need escaping in the Data URI encoding scheme +// Escape only non-printable characters, unicode and %, #, &. +// IE11 additionally requires encoding of \, [, ], ", <, >, `, {, }, |, ^ which is not required by Chrome, Firefox, Opera, Edge, Safari, Yandex +// To pass the HTML validator, restricted URL characters must be escaped: non-printable characters, space, <, >, #, %, " +var DataURIEncodingTable = [256]bool{ + // ASCII + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + + true, false, true, true, false, true, true, false, // space, ", #, %, & + false, false, false, false, false, false, false, false, + false, false, false, false, false, false, false, false, + false, false, false, false, true, false, true, false, // <, > + + false, false, false, false, false, false, false, false, + false, false, false, false, false, false, false, false, + false, false, false, false, false, false, false, false, + false, false, false, true, true, true, true, false, // [, \, ], ^ + + true, false, false, false, false, false, false, false, // ` + false, false, false, false, false, false, false, false, + false, false, false, false, false, false, false, false, + false, false, false, true, true, true, false, true, // {, |, }, DEL + + // non-ASCII + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, + true, true, true, true, true, true, true, true, +} + +// EncodeURL encodes bytes using the URL encoding scheme +func EncodeURL(b []byte, table [256]bool) []byte { + for i := 0; i < len(b); i++ { + c := b[i] + if table[c] { + b = append(b, 0, 0) + copy(b[i+3:], b[i+1:]) + b[i+0] = '%' + b[i+1] = "0123456789ABCDEF"[c>>4] + b[i+2] = "0123456789ABCDEF"[c&15] + } + } + return b +} + +// DecodeURL decodes an URL encoded using the URL encoding scheme +func DecodeURL(b []byte) []byte { + for i := 0; i < len(b); i++ { + if b[i] == '%' && i+2 < len(b) { + j := i + 1 + c := 0 + for ; j < i+3 && (b[j] >= '0' && b[j] <= '9' || b[j] >= 'a' && b[j] <= 'f' || b[j] >= 'A' && b[j] <= 'F'); j++ { + if b[j] <= '9' { + c = c<<4 + int(b[j]-'0') + } else if b[j] <= 'F' { + c = c<<4 + int(b[j]-'A') + 10 + } else if b[j] <= 'f' { + c = c<<4 + int(b[j]-'a') + 10 + } + } + if j == i+3 && c < 128 { + b[i] = byte(c) + b = append(b[:i+1], b[i+3:]...) + } + } else if b[i] == '+' { + b[i] = ' ' + } + } + return b +} |