md4c

C Markdown parser. Fast. SAX-like interface. Compliant to CommonMark specification.
git clone https://noulin.net/git/md4c.git
Log | Files | Refs | README | LICENSE

commit 26f14899ed326b123dd8f7accef6e0422c59fe12
parent ad4f28bb85136cde06eb01b5cb4255b28870f145
Author: Martin Mitas <mity@morous.org>
Date:   Mon, 24 Jul 2017 19:27:27 +0200

Add pathological_tests.py from cmark.

Diffstat:
Mscripts/run-tests.sh | 3+++
Atest/pathological_tests.py | 89+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 92 insertions(+), 0 deletions(-)

diff --git a/scripts/run-tests.sh b/scripts/run-tests.sh @@ -37,3 +37,6 @@ $PYTHON "$TEST_DIR/spec_tests.py" -s "$TEST_DIR/permissive-url-autolinks.txt" -p $PYTHON "$TEST_DIR/spec_tests.py" -s "$TEST_DIR/permissive-www-autolinks.txt" -p "$PROGRAM --fpermissive-www-autolinks" $PYTHON "$TEST_DIR/spec_tests.py" -s "$TEST_DIR/tables.txt" -p "$PROGRAM --ftables" $PYTHON "$TEST_DIR/spec_tests.py" -s "$TEST_DIR/strikethrough.txt" -p "$PROGRAM --fstrikethrough" + +# Run pathological tests: +$PYTHON "$TEST_DIR/pathological_tests.py" -p "$PROGRAM" diff --git a/test/pathological_tests.py b/test/pathological_tests.py @@ -0,0 +1,89 @@ +#!/usr/bin/env python3 +# -*- coding: utf-8 -*- + +import re +import argparse +import sys +import platform +from cmark import CMark + +if __name__ == "__main__": + parser = argparse.ArgumentParser(description='Run cmark tests.') + parser.add_argument('--program', dest='program', nargs='?', default=None, + help='program to test') + parser.add_argument('--library-dir', dest='library_dir', nargs='?', + default=None, help='directory containing dynamic library') + args = parser.parse_args(sys.argv[1:]) + +cmark = CMark(prog=args.program, library_dir=args.library_dir) + +# list of pairs consisting of input and a regex that must match the output. +pathological = { + # note - some pythons have limit of 65535 for {num-matches} in re. + "nested strong emph": + (("*a **a " * 65000) + "b" + (" a** a*" * 65000), + re.compile("(<em>a <strong>a ){65000}b( a</strong> a</em>){65000}")), + "many emph closers with no openers": + (("a_ " * 65000), + re.compile("(a[_] ){64999}a_")), + "many emph openers with no closers": + (("_a " * 65000), + re.compile("(_a ){64999}_a")), + "many link closers with no openers": + (("a]" * 65000), + re.compile("(a\]){65000}")), + "many link openers with no closers": + (("[a" * 65000), + re.compile("(\[a){65000}")), + "mismatched openers and closers": + (("*a_ " * 50000), + re.compile("([*]a[_] ){49999}[*]a_")), + "openers and closers multiple of 3": + (("a**b" + ("c* " * 50000)), + re.compile("a[*][*]b(c[*] ){49999}c[*]")), + "link openers and emph closers": + (("[ a_" * 50000), + re.compile("(\[ a_){50000}")), + "hard link/emph case": + ("**x [a*b**c*](d)", + re.compile("\\*\\*x <a href=\"d\">a<em>b\\*\\*c</em></a>")), + "nested brackets": + (("[" * 50000) + "a" + ("]" * 50000), + re.compile("\[{50000}a\]{50000}")), + "nested block quotes": + ((("> " * 50000) + "a"), + re.compile("(<blockquote>\n){50000}")), + "U+0000 in input": + ("abc\u0000de\u0000", + re.compile("abc\ufffd?de\ufffd?")), + "backticks": + ("".join(map(lambda x: ("e" + "`" * x), range(1,10000))), + re.compile("^<p>[e`]*</p>\n$")) + } + +whitespace_re = re.compile('/s+/') +passed = 0 +errored = 0 +failed = 0 + +print("Testing pathological cases:") +for description in pathological: + (inp, regex) = pathological[description] + [rc, actual, err] = cmark.to_html(inp) + if rc != 0: + errored += 1 + print(description, '[ERRORED (return code %d)]' %rc) + print(err) + elif regex.search(actual): + print(description, '[PASSED]') + passed += 1 + else: + print(description, '[FAILED]') + print(repr(actual)) + failed += 1 + +print("%d passed, %d failed, %d errored" % (passed, failed, errored)) +if (failed == 0 and errored == 0): + exit(0) +else: + exit(1)