[utils] Improve comments processing in js_to_json (closes #11947)

This commit is contained in:
Sergey M․ 2017-02-03 02:55:06 +07:00
parent 0bbcc8a10a
commit 4195096ea8
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
2 changed files with 35 additions and 9 deletions

View File

@ -785,12 +785,24 @@ def test_js_to_json_edgecases(self):
on = js_to_json('["abc", "def",]') on = js_to_json('["abc", "def",]')
self.assertEqual(json.loads(on), ['abc', 'def']) self.assertEqual(json.loads(on), ['abc', 'def'])
on = js_to_json('[/*comment\n*/"abc"/*comment\n*/,/*comment\n*/"def",/*comment\n*/]')
self.assertEqual(json.loads(on), ['abc', 'def'])
on = js_to_json('[//comment\n"abc" //comment\n,//comment\n"def",//comment\n]')
self.assertEqual(json.loads(on), ['abc', 'def'])
on = js_to_json('{"abc": "def",}') on = js_to_json('{"abc": "def",}')
self.assertEqual(json.loads(on), {'abc': 'def'}) self.assertEqual(json.loads(on), {'abc': 'def'})
on = js_to_json('{/*comment\n*/"abc"/*comment\n*/:/*comment\n*/"def"/*comment\n*/,/*comment\n*/}')
self.assertEqual(json.loads(on), {'abc': 'def'})
on = js_to_json('{ 0: /* " \n */ ",]" , }') on = js_to_json('{ 0: /* " \n */ ",]" , }')
self.assertEqual(json.loads(on), {'0': ',]'}) self.assertEqual(json.loads(on), {'0': ',]'})
on = js_to_json('{ /*comment\n*/0/*comment\n*/: /* " \n */ ",]" , }')
self.assertEqual(json.loads(on), {'0': ',]'})
on = js_to_json('{ 0: // comment\n1 }') on = js_to_json('{ 0: // comment\n1 }')
self.assertEqual(json.loads(on), {'0': 1}) self.assertEqual(json.loads(on), {'0': 1})
@ -803,15 +815,27 @@ def test_js_to_json_edgecases(self):
on = js_to_json("['a\\\nb']") on = js_to_json("['a\\\nb']")
self.assertEqual(json.loads(on), ['ab']) self.assertEqual(json.loads(on), ['ab'])
on = js_to_json("/*comment\n*/[/*comment\n*/'a\\\nb'/*comment\n*/]/*comment\n*/")
self.assertEqual(json.loads(on), ['ab'])
on = js_to_json('{0xff:0xff}') on = js_to_json('{0xff:0xff}')
self.assertEqual(json.loads(on), {'255': 255}) self.assertEqual(json.loads(on), {'255': 255})
on = js_to_json('{/*comment\n*/0xff/*comment\n*/:/*comment\n*/0xff/*comment\n*/}')
self.assertEqual(json.loads(on), {'255': 255})
on = js_to_json('{077:077}') on = js_to_json('{077:077}')
self.assertEqual(json.loads(on), {'63': 63}) self.assertEqual(json.loads(on), {'63': 63})
on = js_to_json('{/*comment\n*/077/*comment\n*/:/*comment\n*/077/*comment\n*/}')
self.assertEqual(json.loads(on), {'63': 63})
on = js_to_json('{42:42}') on = js_to_json('{42:42}')
self.assertEqual(json.loads(on), {'42': 42}) self.assertEqual(json.loads(on), {'42': 42})
on = js_to_json('{/*comment\n*/42/*comment\n*/:/*comment\n*/42/*comment\n*/}')
self.assertEqual(json.loads(on), {'42': 42})
def test_extract_attributes(self): def test_extract_attributes(self):
self.assertEqual(extract_attributes('<e x="y">'), {'x': 'y'}) self.assertEqual(extract_attributes('<e x="y">'), {'x': 'y'})
self.assertEqual(extract_attributes("<e x='y'>"), {'x': 'y'}) self.assertEqual(extract_attributes("<e x='y'>"), {'x': 'y'})

View File

@ -2103,6 +2103,13 @@ def strip_jsonp(code):
def js_to_json(code): def js_to_json(code):
COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*'
SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE)
INTEGER_TABLE = (
(r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16),
(r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8),
)
def fix_kv(m): def fix_kv(m):
v = m.group(0) v = m.group(0)
if v in ('true', 'false', 'null'): if v in ('true', 'false', 'null'):
@ -2118,11 +2125,6 @@ def fix_kv(m):
'\\x': '\\u00', '\\x': '\\u00',
}.get(m.group(0), m.group(0)), v[1:-1]) }.get(m.group(0), m.group(0)), v[1:-1])
INTEGER_TABLE = (
(r'^(0[xX][0-9a-fA-F]+)\s*:?$', 16),
(r'^(0+[0-7]+)\s*:?$', 8),
)
for regex, base in INTEGER_TABLE: for regex, base in INTEGER_TABLE:
im = re.match(regex, v) im = re.match(regex, v)
if im: if im:
@ -2134,11 +2136,11 @@ def fix_kv(m):
return re.sub(r'''(?sx) return re.sub(r'''(?sx)
"(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"| "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
'(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'| '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
/\*.*?\*/|//[^\n]*|,(?=\s*[\]}])| {comment}|,(?={skip}[\]}}])|
[a-zA-Z_][.a-zA-Z_0-9]*| [a-zA-Z_][.a-zA-Z_0-9]*|
\b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:\s*:)?| \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
[0-9]+(?=\s*:) [0-9]+(?={skip}:)
''', fix_kv, code) '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code)
def qualities(quality_ids): def qualities(quality_ids):