DuyTa's picture
Upload folder using huggingface_hub
bc20498 verified
raw
history blame
8.03 kB
import {ExternalTokenizer, ContextTracker} from "@lezer/lr"
import {
newline as newlineToken, eof, newlineBracketed, blankLineStart, indent, dedent, printKeyword,
ParenthesizedExpression, TupleExpression, ComprehensionExpression,
PatternArgList, SequencePattern, MappingPattern, TypeParamList,
ArrayExpression, ArrayComprehensionExpression, ArgList, ParamList, importList, subscript,
DictionaryExpression, DictionaryComprehensionExpression, SetExpression, SetComprehensionExpression,
String as StringTerm, FormatString, FormatReplacement, nestedFormatReplacement,
stringStart, stringStartD, stringStartL, stringStartLD,
stringStartR, stringStartRD, stringStartRL, stringStartRLD,
stringStartF, stringStartFD, stringStartFL, stringStartFLD,
stringStartFR, stringStartFRD, stringStartFRL, stringStartFRLD,
stringContent, Escape, replacementStart, stringEnd,
ParenL, BraceL, BracketL
} from "./parser.terms.js"
const newline = 10, carriageReturn = 13, space = 32, tab = 9, hash = 35, parenOpen = 40, dot = 46,
braceOpen = 123, braceClose = 125, singleQuote = 39, doubleQuote = 34, backslash = 92,
letter_o = 111, letter_x = 120, letter_N = 78, letter_u = 117, letter_U = 85
const bracketed = new Set([
ParenthesizedExpression, TupleExpression, ComprehensionExpression, importList, ArgList, ParamList,
ArrayExpression, ArrayComprehensionExpression, subscript,
SetExpression, SetComprehensionExpression, FormatString, FormatReplacement, nestedFormatReplacement,
DictionaryExpression, DictionaryComprehensionExpression,
SequencePattern, MappingPattern, PatternArgList, TypeParamList
])
function isLineBreak(ch) {
return ch == newline || ch == carriageReturn
}
function isHex(ch) {
return ch >= 48 && ch <= 57 || ch >= 65 && ch <= 70 || ch >= 97 && ch <= 102
}
export const newlines = new ExternalTokenizer((input, stack) => {
let prev
if (input.next < 0) {
input.acceptToken(eof)
} else if (stack.context.flags & cx_Bracketed) {
if (isLineBreak(input.next)) input.acceptToken(newlineBracketed, 1)
} else if (((prev = input.peek(-1)) < 0 || isLineBreak(prev)) &&
stack.canShift(blankLineStart)) {
let spaces = 0
while (input.next == space || input.next == tab) { input.advance(); spaces++ }
if (input.next == newline || input.next == carriageReturn || input.next == hash)
input.acceptToken(blankLineStart, -spaces)
} else if (isLineBreak(input.next)) {
input.acceptToken(newlineToken, 1)
}
}, {contextual: true})
export const indentation = new ExternalTokenizer((input, stack) => {
let context = stack.context
if (context.flags) return
let prev = input.peek(-1), depth
if (prev == newline || prev == carriageReturn) {
let depth = 0, chars = 0
for (;;) {
if (input.next == space) depth++
else if (input.next == tab) depth += 8 - (depth % 8)
else break
input.advance()
chars++
}
if (depth != context.indent &&
input.next != newline && input.next != carriageReturn && input.next != hash) {
if (depth < context.indent) input.acceptToken(dedent, -chars)
else input.acceptToken(indent)
}
}
})
// Flags used in Context objects
const cx_Bracketed = 1, cx_String = 2, cx_DoubleQuote = 4, cx_Long = 8, cx_Raw = 16, cx_Format = 32
function Context(parent, indent, flags) {
this.parent = parent
this.indent = indent
this.flags = flags
this.hash = (parent ? parent.hash + parent.hash << 8 : 0) + indent + (indent << 4) + flags + (flags << 6)
}
const topIndent = new Context(null, 0, 0)
function countIndent(space) {
let depth = 0
for (let i = 0; i < space.length; i++)
depth += space.charCodeAt(i) == tab ? 8 - (depth % 8) : 1
return depth
}
const stringFlags = new Map([
[stringStart, 0],
[stringStartD, cx_DoubleQuote],
[stringStartL, cx_Long],
[stringStartLD, cx_Long | cx_DoubleQuote],
[stringStartR, cx_Raw],
[stringStartRD, cx_Raw | cx_DoubleQuote],
[stringStartRL, cx_Raw | cx_Long],
[stringStartRLD, cx_Raw | cx_Long | cx_DoubleQuote],
[stringStartF, cx_Format],
[stringStartFD, cx_Format | cx_DoubleQuote],
[stringStartFL, cx_Format | cx_Long],
[stringStartFLD, cx_Format | cx_Long | cx_DoubleQuote],
[stringStartFR, cx_Format | cx_Raw],
[stringStartFRD, cx_Format | cx_Raw | cx_DoubleQuote],
[stringStartFRL, cx_Format | cx_Raw | cx_Long],
[stringStartFRLD, cx_Format | cx_Raw | cx_Long | cx_DoubleQuote]
].map(([term, flags]) => [term, flags | cx_String]))
export const trackIndent = new ContextTracker({
start: topIndent,
reduce(context, term, _, input) {
if ((context.flags & cx_Bracketed) && bracketed.has(term) ||
(term == StringTerm || term == FormatString) && (context.flags & cx_String))
return context.parent
return context
},
shift(context, term, stack, input) {
if (term == indent)
return new Context(context, countIndent(input.read(input.pos, stack.pos)), 0)
if (term == dedent)
return context.parent
if (term == ParenL || term == BracketL || term == BraceL || term == replacementStart)
return new Context(context, 0, cx_Bracketed)
if (stringFlags.has(term))
return new Context(context, 0, stringFlags.get(term) | (context.flags & cx_Bracketed))
return context
},
hash(context) { return context.hash }
})
export const legacyPrint = new ExternalTokenizer(input => {
for (let i = 0; i < 5; i++) {
if (input.next != "print".charCodeAt(i)) return
input.advance()
}
if (/\w/.test(String.fromCharCode(input.next))) return
for (let off = 0;; off++) {
let next = input.peek(off)
if (next == space || next == tab) continue
if (next != parenOpen && next != dot && next != newline && next != carriageReturn && next != hash)
input.acceptToken(printKeyword)
return
}
})
export const strings = new ExternalTokenizer((input, stack) => {
let {flags} = stack.context
let quote = (flags & cx_DoubleQuote) ? doubleQuote : singleQuote
let long = (flags & cx_Long) > 0
let escapes = !(flags & cx_Raw)
let format = (flags & cx_Format) > 0
let start = input.pos
for (;;) {
if (input.next < 0) {
break
} else if (format && input.next == braceOpen) {
if (input.peek(1) == braceOpen) {
input.advance(2)
} else {
if (input.pos == start) {
input.acceptToken(replacementStart, 1)
return
}
break
}
} else if (escapes && input.next == backslash) {
if (input.pos == start) {
input.advance()
let escaped = input.next
if (escaped >= 0) {
input.advance()
skipEscape(input, escaped)
}
input.acceptToken(Escape)
return
}
break
} else if (input.next == quote && (!long || input.peek(1) == quote && input.peek(2) == quote)) {
if (input.pos == start) {
input.acceptToken(stringEnd, long ? 3 : 1)
return
}
break
} else if (input.next == newline) {
if (long) {
input.advance()
} else if (input.pos == start) {
input.acceptToken(stringEnd)
return
}
break
} else {
input.advance()
}
}
if (input.pos > start) input.acceptToken(stringContent)
})
function skipEscape(input, ch) {
if (ch == letter_o) {
for (let i = 0; i < 2 && input.next >= 48 && input.next <= 55; i++) input.advance()
} else if (ch == letter_x) {
for (let i = 0; i < 2 && isHex(input.next); i++) input.advance()
} else if (ch == letter_u) {
for (let i = 0; i < 4 && isHex(input.next); i++) input.advance()
} else if (ch == letter_U) {
for (let i = 0; i < 8 && isHex(input.next); i++) input.advance()
} else if (ch == letter_N) {
if (input.next == braceOpen) {
input.advance()
while (input.next >= 0 && input.next != braceClose && input.next != singleQuote &&
input.next != doubleQuote && input.next != newline) input.advance()
if (input.next == braceClose) input.advance()
}
}
}