Skip to content

Commit

Permalink
Feat Supporting string escaping sequences.
Browse files Browse the repository at this point in the history
  • Loading branch information
@deleterium
deleterium committed Feb 15, 2024
1 parent 5c540f5 commit 0d073ec
Show file tree
Hide file tree
Showing 2 changed files with 120 additions and 30 deletions.
2 changes: 1 addition & 1 deletion src/parser/parser.ts
View file
Original file line number Diff line number Diff line change
Expand Up @@ -98,7 +98,7 @@ export default function parser (preTokens: PRE_TOKEN[]): TOKEN[] {
sequence: ['string'],
action (tokenID): TOKEN {
const PreTkn = preTokens[tokenID]
let val = stringToHexstring(PreTkn.value)
let val = stringToHexstring(PreTkn.value, PreTkn.line)
const parts = /^(BURST-|S-|TS-)([0-9A-Z]{4}-[0-9A-Z]{4}-[0-9A-Z]{4}-[0-9A-Z]{5})/.exec(PreTkn.value)
if (parts !== null) {
val = ReedSalomonAddressDecode(parts[2], PreTkn.line)
Expand Down
148 changes: 119 additions & 29 deletions src/repository/repository.ts
View file
Original file line number Diff line number Diff line change
Expand Up @@ -73,48 +73,138 @@ export function deepCopy<T1> (source: T1): T1 {
}

/**
* Converts a utf-16 string to utf-8 hexstring
* Converts a utf-16 string to utf-8 hexstring. Also process escape chars
* \n \r \t \\ \' \" \xHH \uHHHH
*
* @param inStr Input string
* @returns Same string converted, padded and reversed.
* (multiple of 8 bytes)
*/
export function stringToHexstring (inStr: string) : string {
export function stringToHexstring (inStr: string, line: string) : string {
const byarr : number [] = []
const hexDigits = [ // charCodes of /[0-9A-Fa-f]/ are ones here
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0,
0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1,
1, 1, 1
]
const hexValues = [ // charCodes of /[0-9A-Fa-f]/ and their value in HEX
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 0,
0, 0, 0, 0, 0, 10, 11, 12, 13, 14, 15, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 11, 12,
13, 14, 15

]
if (inStr.length === 0) byarr.push(0)
let charCode, nextCharCode: number
let helperUnicode: number[]
let i = 0
while (i < inStr.length) {
const charCode = inStr.charCodeAt(i)
switch (true) {
case (charCode < 0x80):
byarr.push(charCode)
break
case (charCode < 0x800):
byarr.push(0xc0 | (charCode >> 6))
byarr.push(0x80 | (charCode & 0x3f))
break
case (charCode < 0xd800):
case (charCode > 0xdfff):
byarr.push(0xe0 | (0x3f & (charCode >> 12)))
byarr.push(0x80 | (0x3f & (charCode >> 6)))
byarr.push(0x80 | (0x3f & charCode))
let state = 0
for (i = 0; !Number.isNaN(charCode = inStr.charCodeAt(i)); i++) {
switch (state) {
case 0: // Regular chars
switch (true) {
case (charCode < 0x80):
if (charCode === 92) { // Starting escape sequence with \
state = 1
break
}
byarr.push(charCode)
break
case (charCode < 0x800):
byarr.push(0xc0 | (charCode >> 6))
byarr.push(0x80 | (charCode & 0x3f))
break
case (charCode < 0xd800):
case (charCode > 0xdfff):
byarr.push(0xe0 | (0x3f & (charCode >> 12)))
byarr.push(0x80 | (0x3f & (charCode >> 6)))
byarr.push(0x80 | (0x3f & charCode))
break
default:
nextCharCode = inStr.charCodeAt(i + 1)
if (Number.isNaN(nextCharCode)) {
throw new Error(`At line: ${line}. String with invalid character at index ${i}.`)
}
if ((charCode & 0xfc00) === 0xd800 && (nextCharCode & 0xfc00) === 0xdc00) {
i++
const newCharCode = ((charCode & 0x3ff) << 10) + (nextCharCode & 0x3ff) + 0x10000
byarr.push(0xf0 | (0x3f & (newCharCode >> 18)))
byarr.push(0x80 | (0x3f & (newCharCode >> 12)))
byarr.push(0x80 | (0x3f & (newCharCode >> 6)))
byarr.push(0x80 | (0x3f & newCharCode))
break
}
throw new Error(`At line: ${line}. String with invalid character at index ${i}.`)
}
break
default: {
const nextCharCode = inStr.charCodeAt(i + 1)
if (isNaN(nextCharCode)) {
case 1: // Possible escape char
state = 0
switch (charCode) {
case 110: // 'n'
byarr.push('\n'.charCodeAt(0))
break
case 114: // 'r'
byarr.push('\r'.charCodeAt(0))
break
case 116: // 't'
byarr.push('\t'.charCodeAt(0))
break
case 92: // '\\'
byarr.push('\\'.charCodeAt(0))
break
case 39: // '\''
byarr.push('\''.charCodeAt(0))
break
case 34: // '"'
byarr.push('"'.charCodeAt(0))
break
case 120: // 'x'
state = 2
break
case 117: // 'u'
state = 3
break
default:
// not escaping char
throw new Error(`At line: ${line}. String with invalid escaping character '\\${String.fromCharCode(charCode)}.`)
}
if ((charCode & 0xfc00) === 0xd800 && (nextCharCode & 0xfc00) === 0xdc00) {
i++
const newCharCode = ((charCode & 0x3ff) << 10) + (nextCharCode & 0x3ff) + 0x10000
byarr.push(0xf0 | (0x3f & (newCharCode >> 18)))
byarr.push(0x80 | (0x3f & (newCharCode >> 12)))
byarr.push(0x80 | (0x3f & (newCharCode >> 6)))
byarr.push(0x80 | (0x3f & newCharCode))
break
case 2: // Read escaped \xHH
nextCharCode = inStr.charCodeAt(++i)
state = 0
if (!(hexDigits[charCode] && hexDigits[nextCharCode])) {
throw new Error(`At line: ${line}. String with malformed escaping '\\x' character. ` +
'Expected 2 hexadecimal chars, but found invalid char or end of string.')
}
byarr.push(hexValues[charCode] * 16 + hexValues[nextCharCode])
break
case 3: // Read escaped \uHHHH (unicode)
helperUnicode = [
charCode,
inStr.charCodeAt(++i),
inStr.charCodeAt(++i),
inStr.charCodeAt(++i)
]
state = 0
if (!helperUnicode.every(code => hexDigits[code] === 1)) {
throw new Error(`At line: ${line}. String with malformed escaping '\\u' (unicode) character. ` +
'Expected 4 hexadecimal chars, but found invalid char or end of string.')
}
// Change input string to include
inStr = inStr.slice(0, i - 5) +
String.fromCharCode(
hexValues[helperUnicode[0]] * 4096 + hexValues[helperUnicode[1]] * 256 +
hexValues[helperUnicode[2]] * 16 + hexValues[helperUnicode[3]]
) +
inStr.slice(i + 1)
i -= 6
break
}
}
i++
}
const byteSize = (Math.floor((byarr.length - 1) / 8) + 1) * 8
byarr.reverse()
Expand Down

0 comments on commit 0d073ec

Please sign in to comment.