Future/bison
1
0
Fork 0
mirror of https://git.savannah.gnu.org/git/bison.git synced 2026-03-09 04:13:03 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
e51e89856a5bb4462bf850048a355607dbe70cd5
bison/examples/c/bistromathic/bistromathic.test
Akim Demaille b7aab2dbad fix: crash when redefining the EOF token 
Reported by Agency for Defense Development.
https://lists.gnu.org/r/bug-bison/2020-08/msg00008.html

On an empty such as

    %token FOO
           BAR
           FOO 0
    %%
    input: %empty

we crash because when we find FOO 0, we decrement ntokens (since FOO
was discovered to be EOF, which is already known to be a token, so we
increment ntokens for it, and need to cancel this).  This "works well"
when EOF is properly defined in one go, but here it is first defined
and later only assign token code 0.  In the meanwhile BAR was given
the token number that we just decremented.

To fix this, assign symbol numbers after parsing, not during parsing,
so that we also saw all the explicit token codes.  To maintain the
current numbers (I'd like to keep no difference in the output, not
just equivalence), we need to make sure the symbols are numbered in
the same order: that of appearance in the source file.  So we need the
locations to be correct, which was almost the case, except for nterms
that appeared several times as LHS (i.e., several times as "foo:
...").  Fixing the use of location_of_lhs sufficed (it appears it was
intended for this use, but its implementation was unfinished: it was
always set to "false" only).

* src/symtab.c (symbol_location_as_lhs_set): Update location_of_lhs.
(symbol_code_set): Remove broken hack that decremented ntokens.
(symbol_class_set, dummy_symbol_get): Don't set number, ntokens and
nnterms.
(symbol_check_defined): Do it.
(symbols): Don't count nsyms here.
Actually, don't count nsyms at all: let it be done in...
* src/reader.c (check_and_convert_grammar): here.  Define nsyms from
ntokens and nnterms after parsing.
* tests/input.at (EOF redeclared): New.

* examples/c/bistromathic/bistromathic.test: Adjust the traces: in
"%nterm <double> exp %% input: ...", exp used to be numbered before
input.
2020-08-07 07:30:06 +02:00

362 lines
8.8 KiB
Bash
Executable File
Raw Blame History

This file contains invisible Unicode characters
This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
#! /bin/sh
# Copyright (C) 2020 Free Software Foundation, Inc.
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
# Users may customize the behavior of readline, which might break our
# expected results.
INPUTRC=/dev/null
export INPUTRC
# Beware of portability issues of readline when not feeding it from a
# terminal.
#
# With recent versions of GNU Readline, input "1+2*3\n" gives
# "> 1+2*3\n7\n> \n"
#
# macOS' version does not display the prompt and does not repeat stdin
# on stdout, so input "1+2*3\n" gives "7\n" as output. Let's try to
# cope with this.
#
# On OpenBSD 6.5 the prompt is displayed, but the input is not
# repeated (!). So input "1+2*3\n" gives "> 7\n> \n" as output.
#
# On AIX, you get some escaping sequence before the prompt:
# "<ESC>[?1034h> 1+2*3". It appears to pass the terminfo capability
# "smm", to put the terminal in "meta mode": as if the user had hit
# META.
echo >perfect '> 0
0
> '
echo >ok '0'
echo '0' | prog >effective
echo "checking for readline output..."
if diff perfect effective; then
# Alles ist gut.
strip_prompt=false
elif diff ok effective; then
strip_prompt=true
else
skip "this is not the GNU Readline we expect"
fi
cat >input <<EOF
1+2*3
EOF
run 0 '> 1+2*3
7
> '
cat >input <<EOF
(1+2) * 3
EOF
run 0 '> (1+2) * 3
9
> '
run -noerr 0 '> (1+2) * 3
9
> ' -p
cat >input <<EOF
a = 256
sqrt (a)
EOF
run 0 '> a = 256
256
> sqrt (a)
16
> '
cat >input <<EOF
a = .16
b = 10 ^ 2
sqrt (a * b)
EOF
run 0 '> a = .16
0.16
> b = 10 ^ 2
100
> sqrt (a * b)
4
> '
cat >input <<EOF
*
EOF
run 0 '> *
> ''
err: 1.1: syntax error: expected end of file or - or ( or exit or number or function etc., before *
err: 1 | *
err: | ^'
# Underline long errors.
cat >input <<EOF
123 123456
EOF
run 0 '> 123 123456
> ''
err: 1.5-10: syntax error: expected end of file or + or - or * or / or ^ before number
err: 1 | 123 123456
err: | ^~~~~~'
cat >input <<EOF
1 + 2 * * 3
EOF
run 0 '> 1 + 2 * * 3
> ''
err: 1.9: syntax error: expected - or ( or number or function or variable before *
err: 1 | 1 + 2 * * 3
err: | ^'
cat >input <<EOF
1 / 0
EOF
run 0 '> 1 / 0
> ''
err: 1.1-5: error: division by zero'
## ---------------- ##
## Error recovery. ##
## ---------------- ##
cat >input <<EOF
((1 ++ 2) ** 3)
(1 ++ 2) + (3 ** 4)
EOF
run 0 '> ((1 ++ 2) ** 3)
666
> (1 ++ 2) + (3 ** 4)
1332
> ''
err: 1.6: syntax error: expected - or ( or number or function or variable before +
err: 1 | ((1 ++ 2) ** 3)
err: | ^
err: 2.5: syntax error: expected - or ( or number or function or variable before +
err: 2 | (1 ++ 2) + (3 ** 4)
err: | ^
err: 2.16: syntax error: expected - or ( or number or function or variable before *
err: 2 | (1 ++ 2) + (3 ** 4)
err: | ^'
# The rule "( error )" should work even if there are no tokens between "(" and ")".
cat >input <<EOF
()
EOF
run 0 '> ()
666
> ''
err: 1.2: syntax error: expected - or ( or number or function or variable before )
err: 1 | ()
err: | ^'
cat >input <<EOF
100% + 10
EOF
run 0 '> 100% + 10
> ''
err: 1.4: syntax error: invalid character: %'
# Traces. This allows to check the location of the error. If we
# forget to map YYerror to YYUNDEF, error recovery enters an endless
# loop with this input.
cat >input <<EOF
(+_)
EOF
run 0 '> (+_)
666
> ''
err: Starting parse
err: Entering state 0
err: Stack now 0
err: Reading a token
err: Next token is token ( (1.1: )
err: Shifting token ( (1.1: )
err: Entering state 2
err: Stack now 0 2
err: Return for a new token:
err: Reading a token
err: Next token is token + (1.2: )
err: LAC: initial context established for +
err: LAC: checking lookahead +: Err
err: LAC: checking lookahead end of file: Err
err: LAC: checking lookahead +: Err
err: LAC: checking lookahead -: S1
err: LAC: checking lookahead *: Err
err: LAC: checking lookahead /: Err
err: LAC: checking lookahead ^: Err
err: LAC: checking lookahead (: S2
err: LAC: checking lookahead ): Err
err: LAC: checking lookahead =: Err
err: LAC: checking lookahead exit: Err
err: LAC: checking lookahead number: S4
err: LAC: checking lookahead function: S5
err: LAC: checking lookahead variable: S6
err: LAC: checking lookahead NEG: Err
err: 1.2: syntax error: expected - or ( or number or function or variable before +
err: 1 | (+_)
err: | ^
err: LAC: initial context discarded due to error recovery
err: Shifting token error (1.2: )
err: Entering state 10
err: Stack now 0 2 10
err: Next token is token + (1.2: )
err: LAC: initial context established for +
err: LAC: checking lookahead +: Err
err: Error: discarding token + (1.2: )
err: Error: popping token error (1.2: )
err: Stack now 0 2
err: LAC: initial context discarded due to error recovery
err: Shifting token error (1.2: )
err: Entering state 10
err: Stack now 0 2 10
err: Return for a new token:
err: 1.3: syntax error: invalid character: _
err: Reading a token
err: Error: popping token error (1.2: )
err: Stack now 0 2
err: Shifting token error (1.2-3: )
err: Entering state 10
err: Stack now 0 2 10
err: Next token is token invalid token (1.3: )
err: LAC: initial context established for invalid token
err: LAC: checking lookahead invalid token: Always Err
err: Error: discarding token invalid token (1.3: )
err: Error: popping token error (1.2-3: )
err: Stack now 0 2
err: LAC: initial context discarded due to error recovery
err: Shifting token error (1.2-3: )
err: Entering state 10
err: Stack now 0 2 10
err: Return for a new token:
err: Reading a token
err: Next token is token ) (1.4: )
err: Shifting token ) (1.4: )
err: Entering state 20
err: Stack now 0 2 10 20
err: Reducing stack by rule XX (line XXX):
err: $1 = token ( (1.1: )
err: $2 = token error (1.2-3: )
err: $3 = token ) (1.4: )
err: -> $$ = nterm exp (1.1-4: 666)
err: Entering state 8
err: Stack now 0 8
err: Return for a new token:
err: Reading a token
err: Now at end of input.
err: LAC: initial context established for end of file
err: LAC: checking lookahead end of file: R2 G7 S14
err: Reducing stack by rule XX (line XXX):
err: $1 = nterm exp (1.1-4: 666)
err: -> $$ = nterm input (1.1-4: )
err: Entering state 7
err: Stack now 0 7
err: Now at end of input.
err: Shifting token end of file (1.5: )
err: LAC: initial context discarded due to shift
err: Entering state 14
err: Stack now 0 7 14
err: Stack now 0 7 14
err: Cleanup: popping token end of file (1.5: )
err: Cleanup: popping nterm input (1.1-4: )' -p
## ------------ ##
## Completion. ##
## ------------ ##
# From now on, the differences between versions of GNU Readline are
# too painful to try to cope with.
if $strip_prompt; then
echo "SKIP: this is not the GNU Readline we expect"
exit $status
fi
# On Windows10/MSYS2 the ^G coming from <tab> completion is not
# emitted the same way
# (https://lists.gnu.org/r/bug-bison/2020-05/msg00076.html).
echo "checking for kernel name... $(uname -s)"
case $(uname -s) in
(MSYS*)
echo "SKIP: this is Windows/MSYS"
exit $status
;;
esac
# Check completion after an operator.
sed -e 's/\\t/ /g' >input <<EOF
(1+\t\t
EOF
run 0 '> (1+
( - atan cos exp ln number sin sqrt
> (1+
> ''
err: 1.4: syntax error: expected - or ( or number or function or variable before end of file
err: 1 | (1+
err: | ^'
# Check the completion of a word.
sed -e 's/\\t/ /g' >input <<EOF
(at\t\t
EOF
run 0 '> (atan ( ''
> ''
err: 1.9: syntax error: expected - or ( or number or function or variable before end of file
err: 1 | (atan ( ''
err: | ^'
# Check the completion at the very beginning.
sed -e 's/\\t/ /g' >input <<EOF
e\t\t
EOF
run -n 0 '> e
end of file exit exp ''
> e
0
> ''
err: '
# Check that completion prints valid locations even when there is an error.
sed -e 's/\\t/ /g' >input <<EOF
1++\t
EOF
run -n 0 '> 1++ ''
> ''
err: 1.3: syntax error: expected - or ( or number or function or variable before +
err: 1 | 1++ ''
err: | ^
'
# And even when the error was recovered from.
sed -e 's/\\t/ /g' >input <<EOF
(1++2) + 3 +\t\t
EOF
run -n 0 '> (1++2) + 3 + ''
> ''
err: 1.4: syntax error: expected - or ( or number or function or variable before +
err: 1 | (1++2) + 3 + ''
err: | ^
err: 1.15: syntax error: expected - or ( or number or function or variable before end of file
err: 1 | (1++2) + 3 + ''
err: | ^
'
Reference in New Issue View Git Blame Copy Permalink