[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH 1/2] gram: correct token numbering in precedence declarations
From: |
Akim Demaille |
Subject: |
Re: [PATCH 1/2] gram: correct token numbering in precedence declarations |
Date: |
Wed, 6 Mar 2013 10:33:39 +0100 |
Thanks! I have installed it as follows. Please note:
- the commit message should read
* file (feature): New.
not
* file (new): Added feature.
Also, use the imperative, not the past.
- no space before [;!?:] in English
- comma after 'e.g.' and 'i.e.' and so forth.
- 'a is inferior to b' is not correct, 'a is less than b'
I have also preferred to make a single unified item in NEWS,
so that we can explain the point of these changes.
commit 5202b6ac1d3662fe283240aab1b29f35607995cd
Author: Valentin Tolmer <address@hidden>
Date: Tue Mar 5 12:29:50 2013 +0100
gram: correct token numbering in precedence declarations
In a precedence declaration, when tokens are declared with a litteral
character (e.g., 'a') or with a identifier (e.g., B), Bison behaved
differently: the litteral tokens would be numbered first, and then the
other ones, leading to the following grammar:
%right A B 'c' 'd'
being numbered as such: 'c' 'd' A B.
* src/parse-gram.y (symbol.prec): Set the symbol number when reading the
symbols.
* tests/conflicts.at (Token declaration order: literals vs. identifiers):
New.
Signed-off-by: Akim Demaille <address@hidden>
diff --git a/NEWS b/NEWS
index ef9f4fa..3bfc1f9 100644
--- a/NEWS
+++ b/NEWS
@@ -284,11 +284,27 @@ GNU Bison NEWS
It used to be an error only if used in non GLR mode, _and_ if there are
reduce/reduce conflicts.
-** Token numbering has changed to preserve the user-defined order
+** Tokens are numbered in their order of appearance
- When declaring %token A B, the numbering for A is inferior to B. Up to now,
- when declaring associativity at the same time, with %left (or %right,
- %precedence, %nonassoc), B was inferior to A.
+ Contributed by Valentin Tolmer.
+
+ With '%token A B', A had a number less than the one of B. However,
+ precedence declarations used to generate a reversed order. This is now
+ fixed, and introducing tokens with any of %token, %left, %right,
+ %precedence, or %nonassoc yields the same result.
+
+ When mixing declarations of tokens with a litteral character (e.g., 'a')
+ or with an identifier (e.g., B) in a precedence declaration, Bison
+ numbered the litteral characters first. For example
+
+ %right A B 'c' 'd'
+
+ would lead to the tokens declared in this order: 'c' 'd' A B. Again, the
+ input order is now preserved.
+
+ These changes were made so that one can remove useless precedence and
+ associativity declarations (i.e., map %nonassoc, %left or %right to
+ %precedence, or to %token) and get exactly the same output.
** Useless precedence and associativity
diff --git a/src/parse-gram.y b/src/parse-gram.y
index 4e887e8..e4d36cc 100644
--- a/src/parse-gram.y
+++ b/src/parse-gram.y
@@ -503,8 +503,17 @@ symbols.prec:
;
symbol.prec:
- symbol { $$ = $1; }
-| symbol INT { $$ = $1; symbol_user_token_number_set ($1, $2, @2); }
+ symbol
+ {
+ $$ = $1;
+ symbol_class_set ($1, token_sym, @1, false);
+ }
+| symbol INT
+ {
+ $$ = $1;
+ symbol_user_token_number_set ($1, $2, @2);
+ symbol_class_set ($1, token_sym, @1, false);
+ }
;
/* One or more symbols to be %typed. */
diff --git a/tests/conflicts.at b/tests/conflicts.at
index 07ff178..2d67a35 100644
--- a/tests/conflicts.at
+++ b/tests/conflicts.at
@@ -87,6 +87,92 @@ AT_BISON_OPTION_POPDEFS
AT_CLEANUP
+## --------------------------------------------------- ##
+## Token declaration order: literals vs. identifiers. ##
+## --------------------------------------------------- ##
+
+# This test checks that when several tokens are declared by the same keyword,
+# some of them defined as a character ('a'), others as simple textual reference
+# (A), they are declared correctly left to right.
+# Previously, the following test would declare the states in the order 'o' 'p'
+# M N, instead of M N 'o' 'p'.
+
+AT_SETUP([Token declaration order: literals vs. identifiers])
+
+AT_DATA_GRAMMAR([[input.y]],
+[[%token 'a' 'b' C D
+%token E F 'g' 'h'
+%right 'i' 'j' K L
+%right M N 'o' 'p'
+%%
+exp: 'a'
+ | 'b'
+ | C
+ | D
+ | E
+ | F
+ | 'g'
+ | 'h'
+ | 'i'
+ | 'j'
+ | K
+ | L
+ | M
+ | N
+ | 'o'
+ | 'p'
+;
+%%
+]])
+
+AT_BISON_CHECK([[--report=all -o input.c input.y]], 0, [], [ignore])
+AT_CHECK([[cat input.output | sed -n '/^State 0$/,/^State 1$/p']], 0,
+[[State 0
+
+ 0 $accept: . exp $end
+ 1 exp: . 'a'
+ 2 | . 'b'
+ 3 | . C
+ 4 | . D
+ 5 | . E
+ 6 | . F
+ 7 | . 'g'
+ 8 | . 'h'
+ 9 | . 'i'
+ 10 | . 'j'
+ 11 | . K
+ 12 | . L
+ 13 | . M
+ 14 | . N
+ 15 | . 'o'
+ 16 | . 'p'
+
+ 'a' shift, and go to state 1
+ 'b' shift, and go to state 2
+ C shift, and go to state 3
+ D shift, and go to state 4
+ E shift, and go to state 5
+ F shift, and go to state 6
+ 'g' shift, and go to state 7
+ 'h' shift, and go to state 8
+ 'i' shift, and go to state 9
+ 'j' shift, and go to state 10
+ K shift, and go to state 11
+ L shift, and go to state 12
+ M shift, and go to state 13
+ N shift, and go to state 14
+ 'o' shift, and go to state 15
+ 'p' shift, and go to state 16
+
+ exp go to state 17
+
+
+State 1
+]])
+
+AT_CLEANUP
+
+
## ------------------------------- ##
## Useless associativity warning. ##
## ------------------------------- ##