diff options
author | ph10 <ph10@2f5784b3-3f2a-0410-8824-cb99058d5e15> | 2012-09-10 11:02:48 +0000 |
---|---|---|
committer | ph10 <ph10@2f5784b3-3f2a-0410-8824-cb99058d5e15> | 2012-09-10 11:02:48 +0000 |
commit | 7a3e01c442328741902c3c4b974d0e3024f645d1 (patch) | |
tree | e61b4534e91f97df907571a92db57623422e71d7 /testdata | |
parent | 640564bfeb0860032d98439dfa9f5585af59a09e (diff) | |
download | pcre-7a3e01c442328741902c3c4b974d0e3024f645d1.tar.gz |
General spring-clean of EBCDIC-related issues in the code, which had decayed
over time. Also the documentation. Added one test that can be run in an ASCII
world to do a little testing of EBCDIC-related things.
git-svn-id: svn://vcs.exim.org/pcre/code/trunk@1033 2f5784b3-3f2a-0410-8824-cb99058d5e15
Diffstat (limited to 'testdata')
-rw-r--r-- | testdata/testinputEBC | 121 | ||||
-rw-r--r-- | testdata/testoutputEBC | 182 |
2 files changed, 303 insertions, 0 deletions
diff --git a/testdata/testinputEBC b/testdata/testinputEBC new file mode 100644 index 0000000..56efcd0 --- /dev/null +++ b/testdata/testinputEBC @@ -0,0 +1,121 @@ +/-- This is a specialized test for checking, when PCRE is compiled with the +EBCDIC option but in an ASCII environment, that newline and white space +functionality is working. It catches cases where explicit values such as 0x0a +have been used instead of names like CHAR_LF. Needless to say, it is not a +genuine EBCDIC test! In patterns, alphabetic characters that follow a backslash +must be in EBCDIC code. In data, newlines and other spacing characters must be +in EBCDIC, but can be specified as escapes. --/ + +/-- Test default newline and variations --/ + +/^A/m + ABC + 12\x15ABC + +/^A/m<any> + 12\x15ABC + 12\x0dABC + 12\x0d\x15ABC + 12\x25ABC + +/^A/m<anycrlf> + 12\x15ABC + 12\x0dABC + 12\x0d\x15ABC + ** Fail + 12\x25ABC + +/-- Test \h --/ + +/^A\ˆ/ + A B + +/-- Test \H --/ + +/^A\È/ + AB + ** Fail + A B + +/-- Test \R --/ + +/^A\Ù/ + A\x15B + A\x0dB + A\x25B + A\x0bB + A\x0cB + ** Fail + A B + +/-- Test \v --/ + +/^A\¥/ + A\x15B + A\x0dB + A\x25B + A\x0bB + A\x0cB + ** Fail + A B + +/-- Test \V --/ + +/^A\å/ + A B + ** Fail + A\x15B + A\x0dB + A\x25B + A\x0bB + A\x0cB + +/-- For repeated items, use an atomic group so that the output is the same +for DFA matching (otherwise it may show multiple matches). --/ + +/-- Test \h+ --/ + +/^A(?>\ˆ+)/ + A B + +/-- Test \H+ --/ + +/^A(?>\È+)/ + AB + ** Fail + A B + +/-- Test \R+ --/ + +/^A(?>\Ù+)/ + A\x15B + A\x0dB + A\x25B + A\x0bB + A\x0cB + ** Fail + A B + +/-- Test \v+ --/ + +/^A(?>\¥+)/ + A\x15B + A\x0dB + A\x25B + A\x0bB + A\x0cB + ** Fail + A B + +/-- Test \V+ --/ + +/^A(?>\å+)/ + A B + ** Fail + A\x15B + A\x0dB + A\x25B + A\x0bB + A\x0cB + +/-- End --/ diff --git a/testdata/testoutputEBC b/testdata/testoutputEBC new file mode 100644 index 0000000..abbfdc4 --- /dev/null +++ b/testdata/testoutputEBC @@ -0,0 +1,182 @@ +/-- This is a specialized test for checking, when PCRE is compiled with the +EBCDIC option but in an ASCII environment, that newline and white space +functionality is working. It catches cases where explicit values such as 0x0a +have been used instead of names like CHAR_LF. Needless to say, it is not a +genuine EBCDIC test! In patterns, alphabetic characters that follow a backslash +must be in EBCDIC code. In data, newlines and other spacing characters must be +in EBCDIC, but can be specified as escapes. --/ + +/-- Test default newline and variations --/ + +/^A/m + ABC + 0: A + 12\x15ABC + 0: A + +/^A/m<any> + 12\x15ABC + 0: A + 12\x0dABC + 0: A + 12\x0d\x15ABC + 0: A + 12\x25ABC + 0: A + +/^A/m<anycrlf> + 12\x15ABC + 0: A + 12\x0dABC + 0: A + 12\x0d\x15ABC + 0: A + ** Fail +No match + 12\x25ABC +No match + +/-- Test \h --/ + +/^A\ˆ/ + A B + 0: A\x20 + +/-- Test \H --/ + +/^A\È/ + AB + 0: AB + ** Fail +No match + A B +No match + +/-- Test \R --/ + +/^A\Ù/ + A\x15B + 0: A\x15 + A\x0dB + 0: A\x0d + A\x25B + 0: A\x25 + A\x0bB + 0: A\x0b + A\x0cB + 0: A\x0c + ** Fail +No match + A B +No match + +/-- Test \v --/ + +/^A\¥/ + A\x15B + 0: A\x15 + A\x0dB + 0: A\x0d + A\x25B + 0: A\x25 + A\x0bB + 0: A\x0b + A\x0cB + 0: A\x0c + ** Fail +No match + A B +No match + +/-- Test \V --/ + +/^A\å/ + A B + 0: A\x20 + ** Fail +No match + A\x15B +No match + A\x0dB +No match + A\x25B +No match + A\x0bB +No match + A\x0cB +No match + +/-- For repeated items, use an atomic group so that the output is the same +for DFA matching (otherwise it may show multiple matches). --/ + +/-- Test \h+ --/ + +/^A(?>\ˆ+)/ + A B + 0: A\x20 + +/-- Test \H+ --/ + +/^A(?>\È+)/ + AB + 0: AB + ** Fail +No match + A B +No match + +/-- Test \R+ --/ + +/^A(?>\Ù+)/ + A\x15B + 0: A\x15 + A\x0dB + 0: A\x0d + A\x25B + 0: A\x25 + A\x0bB + 0: A\x0b + A\x0cB + 0: A\x0c + ** Fail +No match + A B +No match + +/-- Test \v+ --/ + +/^A(?>\¥+)/ + A\x15B + 0: A\x15 + A\x0dB + 0: A\x0d + A\x25B + 0: A\x25 + A\x0bB + 0: A\x0b + A\x0cB + 0: A\x0c + ** Fail +No match + A B +No match + +/-- Test \V+ --/ + +/^A(?>\å+)/ + A B + 0: A\x20B + ** Fail +No match + A\x15B +No match + A\x0dB +No match + A\x25B +No match + A\x0bB +No match + A\x0cB +No match + +/-- End --/ |