[Moved to DR at the October, 2015 meeting.]

Proposed resolution (May, 2015):

1. Change 5.4 [lex.pptoken] bullet 3.3 as follows:

• Otherwise, the next preprocessing token is the longest sequence of characters that could constitute a preprocessing token, even if that would cause further lexical analysis to fail, except that a header-name (5.8 [lex.header]) is only formed within a #include directive (15.3 [cpp.include]).

2. Change 5.8 [lex.header] paragraph 1 as follows:

[Note: Header name preprocessing tokens shall only appear within a #include preprocessing directive (15.3 [cpp.include] see 5.4 [lex.pptoken]). —end note] The sequences in both forms...

The “max munch” rule could be read to require the characters <int> in vector<int> to be parsed as a header-name rather than as three distinct tokens. 5.8 [lex.header] paragraph 1 says,

Header name preprocessing tokens shall only appear within a #include preprocessing directive (15.3 [cpp.include]).

However, that is not sufficiently clear that header-names are only to be recognized in that context.