This is a follow-up from bug 551670, see bug 551670 comment 59. Quoting Jonas about nsAttrValue.h and nsAttrValue.cpp: " >+ const EnumTable* table = sEnumTableArray-> >+ ElementAt(allEnumBits & NS_ATTRVALUE_ENUMTABLEINDEX_MASK); Hmm.. This looks wierd. This argument passed to ElementAt will always be a multiple of 16384 or 16 or something like that. I think. This bit twiddling is too complex and too spread out. I take full responsibility for this fact. Would you mind centralizing and cleaning it up? Something like the following defines: #define NS_ATTRVALUE_ENUM_TABLEINDEX_BITS (32 - 16 - 1 - NS_ATTRVALUE_INTEGERTYPE_BITS) #define NS_ATTRVALUE_ENUM_TABLEINDEX_OFFSET NS_ATTRVALUE_INTEGERTYPE_BITS #define NS_ATTRVALUE_ENUM_TABLEINDEX_MASK \ (PR_BITMASK(NS_ATTRVALUE_ENUM_TABLEINDEX_BITS) << NS_ATTRVALUE_ENUM_TABLEINDEX_OFFSET) #define NS_ATTRVALUE_ENUM_GET_TABLEINDEX(_bits) \ (((_bits) & NS_ATTRVALUE_ENUM_TABLEINDEX_MASK) >> NS_ATTRVALUE_ENUM_TABLEINDEX_OFFSET) #define NS_ATTRVALUE_ENUM_VALUE_NEEDS_TO_UPPER_BITS 1 #define NS_ATTRVALUE_ENUM_VALUE_NEEDS_TO_UPPER_OFFSET \ (NS_ATTRVALUE_ENUM_TABLEINDEX_OFFSET + NS_ATTRVALUE_ENUM_TABLEINDEX_BITS) #define NS_ATTRVALUE_ENUM_VALUE_NEEDS_TO_UPPER_MASK \ (PR_BITMASK(NS_ATTRVALUE_ENUM_VALUE_NEEDS_TO_UPPER_BITS) << NS_ATTRVALUE_ENUM_VALUE_NEEDS_TO_UPPER_OFFSET) #define NS_ATTRVALUE_ENUM_VALUE_NEEDS_TO_UPPER(_bits) \ ((_bits) & NS_ATTRVALUE_ENUM_VALUE_NEEDS_TO_UPPER_MASK) #define NS_ATTRVALUE_ENUM_VALUE_BITS 16 #define NS_ATTRVALUE_ENUM_VALUE_OFFSET \ (NS_ATTRVALUE_ENUM_TABLEINDEX_OFFSET + NS_ATTRVALUE_ENUM_TABLEINDEX_BITS) #define NS_ATTRVALUE_ENUM_VALUE_MASK \ (PR_BITMASK(NS_ATTRVALUE_ENUM_VALUE_BITS) << NS_ATTRVALUE_ENUM_VALUE_OFFSET) #define NS_ATTRVALUE_ENUM_GET_VALUE(_bits) \ (((_bits) & NS_ATTRVALUE_ENUM_VALUE_MASK) >> NS_ATTRVALUE_ENUM_VALUE_OFFSET) And then use these defines throughout. Feel free to simplify these macros. Might actually be more readable to calculate the values by hand and #define to those values directly. "

You should use functions (marked as inline, not that it should make a difference) in preference to macros, for what it's worth -- easier to debug in/through.

https://bugzilla.mozilla.org/show_bug.cgi?id=1472046 Move all DOM bugs that haven’t been updated in more than 3 years and has no one currently assigned to P5. If you have questions, please contact :mdaly.