def node2json(ast, data):
assert (ast['step']['type'] == 'CALL')
if ast['step']['data'] == 'Object':
return object2json(ast, data)
elif ast['step']['data'] == 'Array':
return array2json(ast, data)
elif ast['step']['data'] == 'String':
str = peggy.astdata(ast, data)
assert (str[0] == '"' and str[-1] == '"')
return str[1:-1].encode('utf-8').decode('unicode_escape')
elif ast['step']['data'] == 'Number':
num = peggy.astdata(ast, data)
try:
return int(num)
except:
return float(num)
elif ast['step']['data'] == 'Literal':
literal = peggy.astdata(ast, data)
if literal == 'true':
return True
elif literal == 'false':
return False
elif literal == 'null':
return None
else:
raise RuntimeError('Invalid literal {}'.format(literal))
else:
raise RuntimeError('Invalid node type {}'.format(ast['step']['data']))
def node2json(ast, data):
assert (ast['step']['type'] == 'CALL')
if ast['step']['data'] == 'Object':
return object2json(ast, data)
elif ast['step']['data'] == 'Array':
return array2json(ast, data)
elif ast['step']['data'] == 'String':
str = peggy.astdata(ast, data)
assert (str[0] == '"' and str[-1] == '"')
return str[1:-1].encode('utf-8').decode('unicode_escape')
elif ast['step']['data'] == 'Number':
num = peggy.astdata(ast, data)
try:
return int(num)
except:
return float(num)
elif ast['step']['data'] == 'Literal':
literal = peggy.astdata(ast, data)
if literal == 'true':
return True
elif literal == 'false':
return False
elif literal == 'null':
return None
else:
raise RuntimeError('Invalid literal {}'.format(literal))
else:
raise RuntimeError('Invalid node type {}'.format(ast['step']['data']))
def grammar2json(ast, data):
g = {}
assert (ast['step']['type'] == 'CALL' and ast['step']['data'] == 'Grammar')
for definition in ast['children']:
assert (definition['step']['type'] == 'CALL')
assert (definition['step']['data'] == 'Definition')
assert (len(definition['children']) == 3)
identifier = definition['children'][0]
assert (
identifier['step']['type'] == 'CALL'
and identifier['step']['data'] == 'Identifier'
)
arrow = definition['children'][1]
#print(peggy.astdata(identifier,data),peggy.astdata(arrow,data))
assert (arrow['step']['type'] == 'CALL')
rule = peggy.astdata(identifier, data).strip().split()[0]
g[rule] = {
'type': 'RULE',
'ast': {
'LeftArrow': 'BUILD',
'VoidArrow': 'VOID',
'PruneArrow': 'PRUNE',
}[arrow['step']['data']],
'data': node2json(definition['children'][2], data),
}
return g
def grammar2json(ast, data):
g = {}
filter_node(ast)
assert (ast['step']['type'] == 'CALL' and ast['step']['data'] == 'Grammar')
if ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'Code':
ast['children'].pop(0)
for definition in ast['children']:
assert (definition['step']['type'] == 'CALL')
assert (definition['step']['data'] == 'Definition')
identifier = definition['children'][0]
assert (
identifier['step']['type'] == 'CALL'
and identifier['step']['data'] == 'Identifier'
)
definition['children'].pop(0)
if definition['children'][0]['step']['type'] == 'CALL' and definition[
'children'][0]['step']['data'] == 'CaseLiteral':
definition['children'].pop(0)
assert (len(definition['children']) == 1)
rule = peggy.astdata(identifier, data).strip().split()[0]
g[rule] = {
'type': 'RULE',
'ast': 'BUILD',
'data': node2json(definition['children'][0], data),
}
return g
def item2json(ast, data):
s = peggy.astdata(ast, data)
assert (s)
assert (ast['step']['type'] == 'CALL')
if ast['step']['data'] == 'Char':
if s[0] == '\\':
return ord(EscapeChar[s[1:]])
return ord(s)
elif ast['step']['data'] == 'Hex':
assert (s[0] == '\\' and s[1] == '<' and s[-1] == '>')
return ord(bytes.fromhex(s[2:-1]).decode('utf-8'))
else:
raise RuntimeError('Invalid range item {}'.format(s))
def item2json(ast, data):
s = peggy.astdata(ast, data)
assert (s)
assert (ast['step']['type'] == 'CALL')
if ast['step']['data'] == 'Char':
if s[0] == '\\':
return ord(EscapeChar[s[1:]])
return ord(s)
elif ast['step']['data'] == 'Hex':
assert (s[0] == '\\' and s[1] == '<' and s[-1] == '>')
return ord(bytes.fromhex(s[2:-1]).decode('utf-8'))
elif ast['step']['data'] == 'Unicode':
assert (len(s) == 6 and s[0] == '\\' and s[1] == 'u')
return ord(s.encode('utf-8').decode('unicode_escape'))
else:
raise RuntimeError('Invalid range item {}'.format(s))
def unit2json(ast, data):
unit = {'ast': 'BUILD'}
prefix = ''
if ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'Prefix':
prefix = peggy.astdata(ast['children'][0], data)
ast['children'].pop(0)
if ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'Literal':
unit['type'] = 'STRING'
unit['data'] = peggy.astdata(
ast['children'][0], data
).strip()[1:-1].encode('utf-8').decode('unicode_escape')
unit['ast'] = 'VOID'
elif ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'CaseLiteral':
unit['type'] = 'STRINGI'
unit['data'] = peggy.astdata(
ast['children'][0], data
).strip()[1:-1].encode('utf-8').decode('unicode_escape')
unit['ast'] = 'VOID'
elif ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'Identifier':
unit['type'] = 'CALL'
unit['data'] = peggy.astdata(ast['children'][0],
data).strip().split()[0]
elif ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'CharClass':
unit['type'] = 'RANGE'
unit['data'] = peggy.Range()
unit['ast'] = 'VOID'
for child in ast['children'][0]['children']:
range2json(child, data, unit)
elif ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'WildCard':
unit['type'] = 'WILDCARD'
unit['ast'] = 'VOID'
else:
unit = node2json(ast['children'][0], data)
if prefix:
if prefix == '*':
unit['min'] = 0
unit['max'] = sys.maxsize
elif prefix == '+':
unit['min'] = 1
unit['max'] = sys.maxsize
elif prefix == '?':
unit['min'] = 0
unit['max'] = 1
elif prefix == '&':
unit['predicate'] = True
unit['ast'] = 'VOID'
elif prefix == '!':
unit['predicate'] = False
unit['ast'] = 'VOID'
elif prefix == ':':
unit['ast'] = 'VOID'
else:
raise NotImplementedError(
'prefix {} not implemented'.format(prefix)
)
return unit
def unit2json(ast, data):
unit = {'ast': 'BUILD'}
prefix = ''
suffix = ''
if ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'Prefix':
prefix = peggy.astdata(ast['children'][0], data)
ast['children'].pop(0)
if ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'Literal':
unit['type'] = 'STRING'
unit['data'] = peggy.astdata(
ast['children'][0], data
).strip()[1:-1].encode('utf-8').decode('unicode_escape')
unit['ast'] = 'VOID'
elif ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'CaseLiteral':
unit['type'] = 'STRING'
unit['data'] = peggy.astdata(
ast['children'][0], data
).strip()[1:-1].encode('utf-8').decode('unicode_escape')
unit['ast'] = 'VOID'
elif ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'Identifier':
unit['type'] = 'CALL'
unit['data'] = peggy.astdata(ast['children'][0],
data).strip().split()[0]
elif ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'CharClass':
unit['type'] = 'RANGE'
unit['data'] = peggy.Range()
unit['ast'] = 'VOID'
for child in ast['children'][0]['children']:
range2json(child, data, unit)
elif ast['children'][0]['step']['type'] == 'CALL' and ast['children'][0][
'step']['data'] == 'WildCard':
unit['type'] = 'WILDCARD'
unit['ast'] = 'VOID'
else:
unit = node2json(ast['children'][0], data)
ast['children'].pop(0)
if ast['children'] and ast['children'][0]['step'][
'type'] == 'CALL' and ast['children'][0]['step']['data'] == 'Suffix':
suffix = peggy.astdata(ast['children'][0], data)
ast['children'].pop(0)
if prefix:
if prefix == '&':
unit['predicate'] = True
unit['ast'] = 'VOID'
elif prefix == '!':
unit['predicate'] = False
unit['ast'] = 'VOID'
elif prefix == '$':
pass
else:
raise NotImplementedError(
'prefix {} not implemented'.format(prefix)
)
if suffix:
if suffix == '*':
unit['min'] = 0
unit['max'] = sys.maxsize
elif suffix == '+':
unit['min'] = 1
unit['max'] = sys.maxsize
elif suffix == '?':
unit['min'] = 0
unit['max'] = 1
elif suffix == 'i':
assert (unit['type'] == 'STRING')
unit['type'] = 'STRINGI'
else:
raise NotImplementedError(
'suffix {} not implemented'.format(suffix)
)
return unit