dnl PSPP - a program for statistical analysis.
dnl Copyright (C) 2017 Free Software Foundation, Inc.
dnl
dnl This program is free software: you can redistribute it and/or modify
dnl it under the terms of the GNU General Public License as published by
dnl the Free Software Foundation, either version 3 of the License, or
dnl (at your option) any later version.
dnl
dnl This program is distributed in the hope that it will be useful,
dnl but WITHOUT ANY WARRANTY; without even the implied warranty of
dnl MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
dnl GNU General Public License for more details.
dnl
dnl You should have received a copy of the GNU General Public License
dnl along with this program.  If not, see <http://www.gnu.org/licenses/>.
dnl
AT_BANNER([lexer])

AT_SETUP([lexer ignores initial @%:@! line])
AT_DATA([lexer.sps], [dnl
#! /usr/local/bin/pspp
DATA LIST LIST NOTABLE /a.
BEGIN DATA.
1
2
END DATA.
LIST.
])
AT_CHECK([pspp -o pspp.csv lexer.sps])
AT_CHECK([cat pspp.csv], [0], [dnl
Table: Data List
a
1.00
2.00
])
AT_CLEANUP

AT_SETUP([lexer properly reports scan errors])
AT_DATA([lexer.sps], [dnl
x'123'
x'1x'
u''
u'012345678'
u'd800'
u'110000'
'foo
'very long unterminated string that be ellipsized in its error message
1e .x
^
�
])
AT_CHECK([pspp -O format=csv lexer.sps], [1], [dnl
"lexer.sps:1.1-1.6: error: String of hex digits has 3 characters, which is not a multiple of 2.
    1 | x'123'
      | ^~~~~~"

"lexer.sps:2.1-2.5: error: `x' is not a valid hex digit.
    2 | x'1x'
      | ^~~~~"

"lexer.sps:3.1-3.3: error: Unicode string contains 0 bytes, which is not in the valid range of 1 to 8 bytes.
    3 | u''
      | ^~~"

"lexer.sps:4.1-4.12: error: Unicode string contains 9 bytes, which is not in the valid range of 1 to 8 bytes.
    4 | u'012345678'
      | ^~~~~~~~~~~~"

"lexer.sps:5.1-5.7: error: U+D800 is not a valid Unicode code point.
    5 | u'd800'
      | ^~~~~~~"

"lexer.sps:6.1-6.9: error: U+110000 is not a valid Unicode code point.
    6 | u'110000'
      | ^~~~~~~~~"

"lexer.sps:7.1-7.4: error: Unterminated string constant.
    7 | 'foo
      | ^~~~"

"lexer.sps:8.1-8.70: error: Unterminated string constant.
    8 | 'very long unterminated string that be ellipsized in its error message
      | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~"

"lexer.sps:9.1-9.2: error: Missing exponent following `1e'.
    9 | 1e .x
      | ^~"

"lexer.sps:9.4: error: Syntax error expecting command name.
    9 | 1e .x
      |    ^"

"lexer.sps:10.1: error: Bad character `^' in input.
   10 | ^
      | ^"

"lexer.sps:11.1-11.2: error: Bad character U+FFFD in input.
   11 | �
      | ^~"
])
AT_CLEANUP

# Bug #47602.
AT_SETUP([lexer crash due to null byte])
# Intentionally leave out the new-line and add a null byte:
printf "datA dist list notable file='input.txt'/a b c.
lis|.\0" > lexer.sps

AT_CHECK([pspp -O format=csv lexer.sps], [1], [dnl
"lexer.sps:1.1-1.9: error: Unknown command `datA dist'.
    1 | datA dist list notable file='input.txt'/a b c.
      | ^~~~~~~~~"

"lexer.sps:2.1-2.3: error: LIST: LIST is allowed only after the active dataset has been defined.
    2 | lis|."

"lexer.sps:2.6: error: LIST: Bad character U+0000 in input.
    2 | lis|."
])
AT_CLEANUP




# Bug #54684
AT_SETUP([lexer crash due to overflow])
printf "DATA LIST/5555555555555555." > lexer.sps

AT_CHECK([pspp -O format=csv lexer.sps], [1], [dnl
"lexer.sps:1.11-1.26: error: DATA LIST: Syntax error expecting integer between 1 and 2147483647.
    1 | DATA LIST/5555555555555555.
      |           ^~~~~~~~~~~~~~~~"
])

AT_CLEANUP