RFC Errata


Errata Search
 
Source of RFC  
Summary Table Full Records

RFC 3986, "Uniform Resource Identifier (URI): Generic Syntax", January 2005

Note: This RFC has been updated by RFC 6874, RFC 7320, RFC 8820

Source of RFC: IETF - NON WORKING GROUP
Area Assignment: app

Errata ID: 4393
Status: Rejected
Type: Technical
Publication Format(s) : TEXT
Reported By: Steven Liekens
Date Reported: 2015-06-15
Rejected by: Barry Leiba
Date Rejected: 2015-06-15

Section 3.2.2 says: 

dec-octet   = DIGIT                 ; 0-9
            / %x31-39 DIGIT         ; 10-99
            / "1" 2DIGIT            ; 100-199
            / "2" %x30-34 DIGIT     ; 200-249
            / "25" %x30-35          ; 250-255

It should say:

dec-octet = "25" %x30-35          ; 250-255
          / "2" %x30-34 DIGIT     ; 200-249
          / "1" 2DIGIT            ; 100-199
          / %x31-39 DIGIT         ; 10-99
          / DIGIT                 ; 0-9

Notes:

The 'dec-octet' rule requires more than 1 lookahead symbol.

Example value: 127

Parsers that implement a first-match-wins strategy will erroneously match 127 as ( DIGIT ), followed by two unexpected symbols.

Parsers that implement a first-match-wins strategy with the corrected grammar will correctly match 127 as ( "1" 2DIGIT ).
--VERIFIER NOTES--
Yes, except that ABNF defined in RFC 5234 is designed to describe what's valid to produce -- it's not designed as the definitive source for building a perfect parser. In particular, the order of the alternatives is not significant in ABNF, so reordering them is irrelevant.

In fact, the ABNF here is more specific than it often is. In other RFCs it will say things like
xyz = 0 / %x31-39 *2DIGIT ; valid values are 0-255
...and just let the comment restrict the maximum value.

Report New Errata



Advanced Search