How could I tell different rules with the same prefix?

[shuizaiku7]

struct integer : pegtl::plus<pegtl::digit> {};
struct lexFloat : pegtl::seq<pegtl::star<pegtl::digit>, pegtl::one<'.'>, pegtl::plus<pegtl::digit>> {};
struct grammar : pegtl::must<pegtl::sor<integer, lexFloat>> {};

if the input is 1.2, it will match integer rule, namely just match 1

I've got 2 ways to address this.
01.Carefully select the order in sor
02.Match interger and lexFloat along with its following space

Is there any simpler way to solve this problem?

[ColinH]

Given the deterministic semantics of PEGs, which is a fancy way of saying that sor<> goes from left to right and stops at the first sub-rule that matched (even if a later rule would be a "better" match), it is normal and expected to carefully order the sub-rules to avoid this prefix problem you encountered.

Now in general it's a good idea to factor out common prefixes, so instead of sor< seq< A, B >, seq< A, C > you'd write seq< A, sor< B, C > >, however with your rule it is not quite that simple and while you could rewrite it as

if_then_else< one< '.' >, plus< digit >, seq< plus< digit >, opt< one< '.' >, plus< digit > > > >

you might prefer sticking to your version, though with a couple of "renamed" intermediate rules you can make it clearer which role each set of digits is playing and simplify attaching your actions:

struct dot : one< '.' >;
struct int_digits : plus< digit >{};
struct frac_digits : plus< digit >{};

struct number : if_then_else< dot, frac_digits, seq< int_digits, opt< dot, frac_digits > > > {};

https://github.com/taocpp/PEGTL/blob/main/doc/Rule-Reference.md#if_then_else-r-s-t-

[shuizaiku7]

Yeah, I think the work of tell different tokens that token A may be the prefix of token B in hand-writen parser has been done by lexer.
Cause pegtl will handle lexer/parser together, so we need to handle it on our own.
01.Just distinguish at every sor, put the longer rule before others.
02.Or provide more intermediate rules so that it's not possible for token A to be prefix of token B.

Thanks.