[jsinterp] Add support for quoted names and indexers (closes #13123, closes #13130)

youtube_dl/jsinterp.py

@@ -6,6 +6,7 @@ import re
 
 from .utils import (
     ExtractorError,
+    remove_quotes,
 )
 
 _OPERATORS = [
@@ -57,7 +58,6 @@ class JSInterpreter(object):
 
     def interpret_expression(self, expr, local_vars, allow_recursion):
         expr = expr.strip()
-
         if expr == '':  # Empty expression
             return None
 
@@ -121,11 +121,19 @@ class JSInterpreter(object):
             pass
 
         m = re.match(
-            r'(?P<var>%s)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$' % _NAME_RE,
+            r'(?P<in>%s)\[(?P<idx>.+)\]$' % _NAME_RE, expr)
+        if m:
+            val = local_vars[m.group('in')]
+            idx = self.interpret_expression(
+                m.group('idx'), local_vars, allow_recursion - 1)
+            return val[idx]
+
+        m = re.match(
+            r'(?P<var>%s)(?:\.(?P<member>[^(]+)|\[(?P<member2>[^]]+)\])\s*(?:\(+(?P<args>[^()]*)\))?$' % _NAME_RE,
             expr)
         if m:
             variable = m.group('var')
-            member = m.group('member')
+            member = remove_quotes(m.group('member') or m.group('member2'))
             arg_str = m.group('args')
 
             if variable in local_vars:
@@ -173,14 +181,6 @@ class JSInterpreter(object):
 
             return obj[member](argvals)
 
-        m = re.match(
-            r'(?P<in>%s)\[(?P<idx>.+)\]$' % _NAME_RE, expr)
-        if m:
-            val = local_vars[m.group('in')]
-            idx = self.interpret_expression(
-                m.group('idx'), local_vars, allow_recursion - 1)
-            return val[idx]
-
         for op, opfunc in _OPERATORS:
             m = re.match(r'(?P<x>.+?)%s(?P<y>.+)' % re.escape(op), expr)
             if not m:
@@ -211,21 +211,22 @@ class JSInterpreter(object):
         raise ExtractorError('Unsupported JS expression %r' % expr)
 
     def extract_object(self, objname):
+        _FUNC_NAME_RE = r'''(?:[a-zA-Z$0-9]+|"[a-zA-Z$0-9]+"|'[a-zA-Z$0-9]+')'''
         obj = {}
         obj_m = re.search(
             (r'(?<!this\.)%s\s*=\s*\{' % re.escape(objname)) +
-            r'\s*(?P<fields>([a-zA-Z$0-9]+\s*:\s*function\(.*?\)\s*\{.*?\}(?:,\s*)?)*)' +
-            r'\}\s*;',
+            r'\s*(?P<fields>(%s\s*:\s*function\(.*?\)\s*\{.*?\}(?:,\s*)?)*)' +
+            r'\}\s*;' % _FUNC_NAME_RE,
             self.code)
         fields = obj_m.group('fields')
         # Currently, it only supports function definitions
         fields_m = re.finditer(
-            r'(?P<key>[a-zA-Z$0-9]+)\s*:\s*function'
-            r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}',
+            r'(?P<key>%s)\s*:\s*function'
+            r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}' % _FUNC_NAME_RE,
             fields)
         for f in fields_m:
             argnames = f.group('args').split(',')
-            obj[f.group('key')] = self.build_function(argnames, f.group('code'))
+            obj[remove_quotes(f.group('key'))] = self.build_function(argnames, f.group('code'))
 
         return obj