Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.
# decimal integer
(r'[-+]?\d+', Number.Integer),
# hex integer
(r'#x[0-9a-f]+', Number.Hex),
# Macro parameters
(r'(\?' + valid_name + ')(:)'
r'(token|name|variable|expression|body|case-body|\*)',
bygroups(Name.Tag, Operator, Name.Builtin)),
(r'(\?)(:)(token|name|variable|expression|body|case-body|\*)',
bygroups(Name.Tag, Operator, Name.Builtin)),
(r'\?' + valid_name, Name.Tag),
# Punctuation
(r'(=>|::|#\(|#\[|##|\?\?|\?=|\?|[(){}\[\],.;])', Punctuation),
# Most operators are picked up as names and then re-flagged.
# This one isn't valid in a name though, so we pick it up now.
(r':=', Operator),
# Pick up #t / #f before we match other stuff with #.
(r'#[tf]', Literal),
# #"foo" style keywords
(r'#"', String.Symbol, 'keyword'),
# #rest, #key, #all-keys, etc.
(r'#[a-z0-9-]+', Keyword),
# required-init-keyword: style keywords.
(valid_name + ':', Keyword),
(r'/(\\\n)?[*][\w\W]*?[*](\\\n)?/', Comment.Multiline),
# Open until EOF, so no ending delimeter
(r'/(\\\n)?[*][\w\W]*', Comment.Multiline),
],
'statements': [
(r'(L?)(")', bygroups(String.Affix, String), 'string'),
(r"(L?)(')(\\.|\\[0-7]{1,3}|\\x[a-fA-F0-9]{1,2}|[^\\\'\n])(')",
bygroups(String.Affix, String.Char, String.Char, String.Char)),
(r'(\d+\.\d*|\.\d+|\d+)[eE][+-]?\d+[LlUu]*', Number.Float),
(r'(\d+\.\d*|\.\d+|\d+[fF])[fF]?', Number.Float),
(r'0x[0-9a-fA-F]+[LlUu]*', Number.Hex),
(r'0[0-7]+[LlUu]*', Number.Oct),
(r'\d+[LlUu]*', Number.Integer),
(r'\*/', Error),
(r'[~!%^&*+=|?:<>/-]', Operator),
(r'[()\[\],.]', Punctuation),
(words(('asm', 'auto', 'break', 'case', 'const', 'continue',
'default', 'do', 'else', 'enum', 'extern', 'for', 'goto',
'if', 'register', 'restricted', 'return', 'sizeof',
'static', 'struct', 'switch', 'typedef', 'union',
'volatile', 'while'),
suffix=r'\b'), Keyword),
(r'(bool|int|long|float|short|double|char|unsigned|signed|void)\b',
Keyword.Type),
(words(('inline', '_inline', '__inline', 'naked', 'restrict',
'thread', 'typename'), suffix=r'\b'), Keyword.Reserved),
# Vector intrinsics
(r'(__m(128i|128d|128|64))\b', Keyword.Reserved),
# Microsoft-isms
(words((
'asm', 'int8', 'based', 'except', 'int16', 'stdcall', 'cdecl',
'fastcall', 'int32', 'declspec', 'finally', 'int64', 'try',
(r'[a-zA-Z_]\w*', Name),
(r'.', Text),
],
'string': [
(r"[^']*'", String, '#pop'),
],
'blockcomment': [
(r'^\s*%\}', Comment.Multiline, '#pop'),
(r'^.*\n', Comment.Multiline),
(r'.', Comment.Multiline),
],
'deffunc': [
(r'(\s*)(?:(.+)(\s*)(=)(\s*))?(.+)(\()(.*)(\))(\s*)',
bygroups(Whitespace, Text, Whitespace, Punctuation,
Whitespace, Name.Function, Punctuation, Text,
Punctuation, Whitespace), '#pop'),
# function with no args
(r'(\s*)([a-zA-Z_]\w*)', bygroups(Text, Name.Function), '#pop'),
],
r'allow_archetype|use_archetype|use_node)\W', Keyword.Type),
(r'(and|or|not|there_exists|xor|implies|for_all)\W', Keyword.Type),
(r'(after|before|closed)\W', Keyword.Type),
(r'(not)\W', Operator),
(r'(matches|is_in)\W', Operator),
# is_in / not is_in char
('(\u2208|\u2209)', Operator),
# there_exists / not there_exists / for_all / and / or
('(\u2203|\u2204|\u2200|\u2227|\u2228|\u22BB|\223C)',
Operator),
# regex in slot or as string constraint
(r'(\{)(\s*/[^}]+/\s*)(\})',
bygroups(Punctuation, String.Regex, Punctuation)),
# regex in slot or as string constraint
(r'(\{)(\s*\^[^}]+\^\s*)(\})',
bygroups(Punctuation, String.Regex, Punctuation)),
(r'/', Punctuation, 'path'),
# for cardinality etc
(r'(\{)((?:\d+\.\.)?(?:\d+|\*))'
r'((?:\s*;\s*(?:ordered|unordered|unique)){,2})(\})',
bygroups(Punctuation, Number, Number, Punctuation)),
# [{ is start of a tuple value
(r'\[\{', Punctuation),
(r'\}\]', Punctuation),
(r'\{', Punctuation),
(r'\}', Punctuation),
include('constraint_values'),
# type name
(r'[A-Z]\w+(<[A-Z]\w+([A-Za-z_<>]*)>)?', Name.Class),
# attribute name
(r'[a-z_]\w*', Name.Class),
(r'\[', Punctuation, 'any_code'),
(r'\$\w+:[^$\n]*\$', Comment.Multiline),
# directives: begin, end
(r'(\$)(begin|end)(\{(%)?)(.*?)((?(4)%)\})',
bygroups(Punctuation, Name.Builtin, Punctuation, None,
String, Punctuation, None)),
# directives: evoque, overlay
# see doc for handling first name arg: /directives/evoque/
#+ minor inconsistency: the "name" in e.g. $overlay{name=site_base}
# should be using(PythonLexer), not passed out as String
(r'(\$)(evoque|overlay)(\{(%)?)(\s*[#\w\-"\'.]+[^=,%}]+?)?'
r'(.*?)((?(4)%)\})',
bygroups(Punctuation, Name.Builtin, Punctuation, None,
String, using(PythonLexer), Punctuation, None)),
# directives: if, for, prefer, test
(r'(\$)(\w+)(\{(%)?)(.*?)((?(4)%)\})',
bygroups(Punctuation, Name.Builtin, Punctuation, None,
using(PythonLexer), Punctuation, None)),
# directive clauses (no {} expression)
(r'(\$)(else|rof|fi)', bygroups(Punctuation, Name.Builtin)),
# expressions
(r'(\$\{(%)?)(.*?)((!)(.*?))?((?(2)%)\})',
bygroups(Punctuation, None, using(PythonLexer),
Name.Builtin, None, None, Punctuation, None)),
(r'#', Other),
],
'comment': [
(r'[^\]#]', Comment.Multiline),
(r'#\[', Comment.Multiline, '#push'),
(r'\]#', Comment.Multiline, '#pop'),
(r'[\]#]', Comment.Multiline)
],
}
mimetypes = ['application/x-pypylog']
tokens = {
"root": [
(r"\[\w+\] \{jit-log-.*?$", Keyword, "jit-log"),
(r"\[\w+\] \{jit-backend-counts$", Keyword, "jit-backend-counts"),
include("extra-stuff"),
],
"jit-log": [
(r"\[\w+\] jit-log-.*?}$", Keyword, "#pop"),
(r"^\+\d+: ", Comment),
(r"--end of the loop--", Comment),
(r"[ifp]\d+", Name),
(r"ptr\d+", Name),
(r"(\()(\w+(?:\.\w+)?)(\))",
bygroups(Punctuation, Name.Builtin, Punctuation)),
(r"[\[\]=,()]", Punctuation),
(r"(\d+\.\d+|inf|-inf)", Number.Float),
(r"-?\d+", Number.Integer),
(r"'.*'", String),
(r"(None|descr|ConstClass|ConstPtr|TargetToken)", Name),
(r"<.*?>+", Name.Builtin),
(r"(label|debug_merge_point|jump|finish)", Name.Class),
(r"(int_add_ovf|int_add|int_sub_ovf|int_sub|int_mul_ovf|int_mul|"
r"int_floordiv|int_mod|int_lshift|int_rshift|int_and|int_or|"
r"int_xor|int_eq|int_ne|int_ge|int_gt|int_le|int_lt|int_is_zero|"
r"int_is_true|"
r"uint_floordiv|uint_ge|uint_lt|"
r"float_add|float_sub|float_mul|float_truediv|float_neg|"
r"float_eq|float_ne|float_ge|float_gt|float_le|float_lt|float_abs|"
r"ptr_eq|ptr_ne|instance_ptr_eq|instance_ptr_ne|"
r"cast_int_to_float|cast_float_to_int|"
bygroups(Punctuation, Name.Builtin, Punctuation, None,
String, Punctuation)),
# directives: evoque, overlay
# see doc for handling first name arg: /directives/evoque/
# + minor inconsistency: the "name" in e.g. $overlay{name=site_base}
# should be using(PythonLexer), not passed out as String
(r'(\$)(evoque|overlay)(\{(%)?)(\s*[#\w\-"\'.]+[^=,%}]+?)?'
r'(.*?)((?(4)%)\})',
bygroups(Punctuation, Name.Builtin, Punctuation, None,
String, using(PythonLexer), Punctuation)),
# directives: if, for, prefer, test
(r'(\$)(\w+)(\{(%)?)(.*?)((?(4)%)\})',
bygroups(Punctuation, Name.Builtin, Punctuation, None,
using(PythonLexer), Punctuation)),
# directive clauses (no {} expression)
(r'(\$)(else|rof|fi)', bygroups(Punctuation, Name.Builtin)),
# expressions
(r'(\$\{(%)?)(.*?)((!)(.*?))?((?(2)%)\})',
bygroups(Punctuation, None, using(PythonLexer),
Name.Builtin, None, None, Punctuation)),
(r'#', Other),
],
'comment': [
(r'[^\]#]', Comment.Multiline),
(r'#\[', Comment.Multiline, '#push'),
(r'\]#', Comment.Multiline, '#pop'),
(r'[\]#]', Comment.Multiline)
],
}
class EvoqueHtmlLexer(DelegatingLexer):
"""
Common objdump lexer tokens to wrap an ASM lexer.
"""
hex_re = r'[0-9A-Za-z]'
return {
'root': [
# File name & format:
('(.*?)(:)( +file format )(.*?)$',
bygroups(Name.Label, Punctuation, Text, String)),
# Section header
('(Disassembly of section )(.*?)(:)$',
bygroups(Text, Name.Label, Punctuation)),
# Function labels
# (With offset)
('('+hex_re+'+)( )(<)(.*?)([-+])(0[xX][A-Za-z0-9]+)(>:)$',
bygroups(Number.Hex, Text, Punctuation, Name.Function,
Punctuation, Number.Hex, Punctuation)),
# (Without offset)
('('+hex_re+'+)( )(<)(.*?)(>:)$',
bygroups(Number.Hex, Text, Punctuation, Name.Function,
Punctuation)),
# Code line with disassembled instructions
('( *)('+hex_re+r'+:)(\t)((?:'+hex_re+hex_re+' )+)( *\t)([a-zA-Z].*?)$',
bygroups(Text, Name.Label, Text, Number.Hex, Text,
using(asm_lexer))),
# Code line with ascii
('( *)('+hex_re+r'+:)(\t)((?:'+hex_re+hex_re+' )+)( *)(.*?)$',
bygroups(Text, Name.Label, Text, Number.Hex, Text, String)),
# Continued code line, only raw opcodes without disassembled
# instruction
('( *)('+hex_re+r'+:)(\t)((?:'+hex_re+hex_re+' )+)$',
bygroups(Text, Name.Label, Text, Number.Hex)),
def _objdump_lexer_tokens(asm_lexer):
"""
Common objdump lexer tokens to wrap an ASM lexer.
"""
hex_re = r'[0-9A-Za-z]'
return {
'root': [
# File name & format:
('(.*?)(:)( +file format )(.*?)$',
bygroups(Name.Label, Punctuation, Text, String)),
# Section header
('(Disassembly of section )(.*?)(:)$',
bygroups(Text, Name.Label, Punctuation)),
# Function labels
# (With offset)
('('+hex_re+'+)( )(<)(.*?)([-+])(0[xX][A-Za-z0-9]+)(>:)$',
bygroups(Number.Hex, Text, Punctuation, Name.Function,
Punctuation, Number.Hex, Punctuation)),
# (Without offset)
('('+hex_re+'+)( )(<)(.*?)(>:)$',
bygroups(Number.Hex, Text, Punctuation, Name.Function,
Punctuation)),
# Code line with disassembled instructions
('( *)('+hex_re+r'+:)(\t)((?:'+hex_re+hex_re+' )+)( *\t)([a-zA-Z].*?)$',
bygroups(Text, Name.Label, Text, Number.Hex, Text,
using(asm_lexer))),
include('comments'),
(r'\{', Punctuation),
(r'(' + _TOKEN_REF + r')(\s*)(=)?(\s*)(' + _STRING_LITERAL
+ r')?(\s*)(;)',
bygroups(Name.Label, Whitespace, Punctuation, Whitespace,
String, Whitespace, Punctuation)),
(r'\}', Punctuation, '#pop'),
],
'options': [
include('whitespace'),
include('comments'),
(r'\{', Punctuation),
(r'(' + _id + r')(\s*)(=)(\s*)(' +
'|'.join((_id, _STRING_LITERAL, _INT, r'\*')) + r')(\s*)(;)',
bygroups(Name.Variable, Whitespace, Punctuation, Whitespace,
Text, Whitespace, Punctuation)),
(r'\}', Punctuation, '#pop'),
],
'action': [
(r'(' + r'|'.join(( # keep host code in largest possible chunks
r'[^${}\'"/\\]+', # exclude unsafe characters
# strings and comments may safely contain unsafe characters
r'"(\\\\|\\"|[^"])*"', # double quote string
r"'(\\\\|\\'|[^'])*'", # single quote string
r'//.*$\n?', # single line comment
r'/\*(.|\n)*?\*/', # multi-line javadoc-style comment
# regular expression: There's no reason for it to start
# with a * and this stops confusion with comments.
r'/(?!\*)(\\\\|\\/|[^/])*/',