Added spec tests.

3 files changed, 358 insertions, 0 deletions

diff --git a/test/cmark.py b/test/cmark.py
new file mode 100644
index 0000000..253e3a8
--- /dev/null
+++ b/test/cmark.py
@@ -0,0 +1,40 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+from ctypes import CDLL, c_char_p, c_long
+from subprocess import *
+import platform
+
+def pipe_through_prog(prog, text):
+    p1 = Popen(prog.split(), stdout=PIPE, stdin=PIPE, stderr=PIPE)
+    [result, err] = p1.communicate(input=text.encode('utf-8'))
+    return [p1.returncode, result.decode('utf-8'), err]
+
+def use_library(lib, text):
+    textbytes = text.encode('utf-8')
+    textlen = len(textbytes)
+    return [0, lib(textbytes, textlen).decode('utf-8'), '']
+
+class CMark:
+    def __init__(self, prog=None, library_dir=None):
+        self.prog = prog
+        if prog:
+            self.to_html = lambda x: pipe_through_prog(prog, x)
+        else:
+            sysname = platform.system()
+            libname = "libcmark"
+            if sysname == 'Darwin':
+                libname += ".dylib"
+            elif sysname == 'Windows':
+                libname = "cmark.dll"
+            else:
+                libname += ".so"
+            if library_dir:
+                libpath = library_dir + "/" + libname
+            else:
+                libpath = "build/src/" + libname
+            cmark = CDLL(libpath)
+            markdown = cmark.cmark_markdown_to_html
+            markdown.restype = c_char_p
+            markdown.argtypes = [c_char_p, c_long]
+            self.to_html = lambda x: use_library(markdown, x)
diff --git a/test/normalize.py b/test/normalize.py
new file mode 100644
index 0000000..03d958e
--- /dev/null
+++ b/test/normalize.py
@@ -0,0 +1,179 @@
+# -*- coding: utf-8 -*-
+from html.parser import HTMLParser, HTMLParseError
+from html.entities import name2codepoint
+import sys
+import re
+import cgi
+
+# Normalization code, adapted from
+# https://github.com/karlcow/markdown-testsuite/
+significant_attrs = ["alt", "href", "src", "title"]
+whitespace_re = re.compile('\s+')
+class MyHTMLParser(HTMLParser):
+    def __init__(self):
+        HTMLParser.__init__(self)
+        self.last = "starttag"
+        self.in_pre = False
+        self.output = ""
+        self.last_tag = ""
+    def handle_data(self, data):
+        after_tag = self.last == "endtag" or self.last == "starttag"
+        after_block_tag = after_tag and self.is_block_tag(self.last_tag)
+        if after_tag and self.last_tag == "br":
+            data = data.lstrip('\n')
+        data = whitespace_re.sub(' ', data)
+        if after_block_tag and not self.in_pre:
+            if self.last == "starttag":
+                data = data.lstrip()
+            elif self.last == "endtag":
+                data = data.strip()
+        self.output += data
+        self.last = "data"
+    def handle_endtag(self, tag):
+        if tag == "pre":
+            self.in_pre = False
+        elif self.is_block_tag(tag):
+            self.output = self.output.rstrip()
+        self.output += "</" + tag + ">"
+        self.last_tag = tag
+        self.last = "endtag"
+    def handle_starttag(self, tag, attrs):
+        if tag == "pre":
+            self.in_pre = True
+        if self.is_block_tag(tag):
+            self.output = self.output.rstrip()
+        self.output += "<" + tag
+        # For now we don't strip out 'extra' attributes, because of
+        # raw HTML test cases.
+        # attrs = filter(lambda attr: attr[0] in significant_attrs, attrs)
+        if attrs:
+            attrs.sort()
+            for (k,v) in attrs:
+                self.output += " " + k
+                if v != None:
+                    self.output += ("=" + '"' + cgi.escape(v,quote=True) + '"')
+        self.output += ">"
+        self.last_tag = tag
+        self.last = "starttag"
+    def handle_startendtag(self, tag, attrs):
+        """Ignore closing tag for self-closing """
+        self.handle_starttag(tag, attrs)
+        self.last_tag = tag
+        self.last = "endtag"
+    def handle_comment(self, data):
+        self.output += '<!--' + data + '-->'
+        self.last = "comment"
+    def handle_decl(self, data):
+        self.output += '<!' + data + '>'
+        self.last = "decl"
+    def unknown_decl(self, data):
+        self.output += '<!' + data + '>'
+        self.last = "decl"
+    def handle_pi(self,data):
+        self.output += '<?' + data + '>'
+        self.last = "pi"
+    def handle_entityref(self, name):
+        try:
+            c = chr(name2codepoint[name])
+        except KeyError:
+            c = None
+        self.output_char(c, '&' + name + ';')
+        self.last = "ref"
+    def handle_charref(self, name):
+        try:
+            if name.startswith("x"):
+                c = chr(int(name[1:], 16))
+            else:
+                c = chr(int(name))
+        except ValueError:
+                c = None
+        self.output_char(c, '&' + name + ';')
+        self.last = "ref"
+    # Helpers.
+    def output_char(self, c, fallback):
+        if c == '<':
+            self.output += "&lt;"
+        elif c == '>':
+            self.output += "&gt;"
+        elif c == '&':
+            self.output += "&amp;"
+        elif c == '"':
+            self.output += "&quot;"
+        elif c == None:
+            self.output += fallback
+        else:
+            self.output += c
+
+    def is_block_tag(self,tag):
+        return (tag in ['article', 'header', 'aside', 'hgroup', 'blockquote',
+            'hr', 'iframe', 'body', 'li', 'map', 'button', 'object', 'canvas',
+            'ol', 'caption', 'output', 'col', 'p', 'colgroup', 'pre', 'dd',
+            'progress', 'div', 'section', 'dl', 'table', 'td', 'dt',
+            'tbody', 'embed', 'textarea', 'fieldset', 'tfoot', 'figcaption',
+            'th', 'figure', 'thead', 'footer', 'tr', 'form', 'ul',
+            'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'video', 'script', 'style'])
+
+def normalize_html(html):
+    r"""
+    Return normalized form of HTML which ignores insignificant output
+    differences:
+
+    Multiple inner whitespaces are collapsed to a single space (except
+    in pre tags):
+
+        >>> normalize_html("<p>a  \t b</p>")
+        '<p>a b</p>'
+
+        >>> normalize_html("<p>a  \t\nb</p>")
+        '<p>a b</p>'
+
+    * Whitespace surrounding block-level tags is removed.
+
+        >>> normalize_html("<p>a  b</p>")
+        '<p>a b</p>'
+
+        >>> normalize_html(" <p>a  b</p>")
+        '<p>a b</p>'
+
+        >>> normalize_html("<p>a  b</p> ")
+        '<p>a b</p>'
+
+        >>> normalize_html("\n\t<p>\n\t\ta  b\t\t</p>\n\t")
+        '<p>a b</p>'
+
+        >>> normalize_html("<i>a  b</i> ")
+        '<i>a b</i> '
+
+    * Self-closing tags are converted to open tags.
+
+        >>> normalize_html("<br />")
+        '<br>'
+
+    * Attributes are sorted and lowercased.
+
+        >>> normalize_html('<a title="bar" HREF="foo">x</a>')
+        '<a href="foo" title="bar">x</a>'
+
+    * References are converted to unicode, except that '<', '>', '&', and
+      '"' are rendered using entities.
+
+        >>> normalize_html("&forall;&amp;&gt;&lt;&quot;")
+        '\u2200&amp;&gt;&lt;&quot;'
+
+    """
+    html_chunk_re = re.compile("(\<!\[CDATA\[.*?\]\]\>|\<[^>]*\>|[^<]+)")
+    try:
+        parser = MyHTMLParser()
+        # We work around HTMLParser's limitations parsing CDATA
+        # by breaking the input into chunks and passing CDATA chunks
+        # through verbatim.
+        for chunk in re.finditer(html_chunk_re, html):
+            if chunk.group(0)[:8] == "<![CDATA":
+                parser.output += chunk.group(0)
+            else:
+                parser.feed(chunk.group(0))
+        parser.close()
+        return parser.output
+    except HTMLParseError as e:
+        sys.stderr.write("Normalization error: " + e.msg + "\n")
+        return html  # on error, return unnormalized HTML
diff --git a/test/spec_tests.py b/test/spec_tests.py
new file mode 100755
index 0000000..6c276ca
--- /dev/null
+++ b/test/spec_tests.py
@@ -0,0 +1,139 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+import sys
+from difflib import unified_diff
+import argparse
+import re
+import json
+from normalize import normalize_html
+from cmark import CMark
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description='Run cmark tests.')
+    parser.add_argument('-p', '--program', dest='program', nargs='?', default=None,
+            help='program to test')
+    parser.add_argument('-s', '--spec', dest='spec', nargs='?', default='spec.txt',
+            help='path to spec')
+    parser.add_argument('-P', '--pattern', dest='pattern', nargs='?',
+            default=None, help='limit to sections matching regex pattern')
+    parser.add_argument('--library-dir', dest='library_dir', nargs='?',
+            default=None, help='directory containing dynamic library')
+    parser.add_argument('--no-normalize', dest='normalize',
+            action='store_const', const=False, default=True,
+            help='do not normalize HTML')
+    parser.add_argument('-d', '--dump-tests', dest='dump_tests',
+            action='store_const', const=True, default=False,
+            help='dump tests in JSON format')
+    parser.add_argument('--debug-normalization', dest='debug_normalization',
+            action='store_const', const=True,
+            default=False, help='filter stdin through normalizer for testing')
+    parser.add_argument('-n', '--number', type=int, default=None,
+            help='only consider the test with the given number')
+    args = parser.parse_args(sys.argv[1:])
+
+def print_test_header(headertext, example_number, start_line, end_line):
+    print("Example %d (lines %d-%d) %s" % (example_number,start_line,end_line,headertext))
+
+def do_test(test, normalize, result_counts):
+    [retcode, actual_html, err] = cmark.to_html(test['markdown'])
+    if retcode == 0:
+        expected_html = test['html']
+        unicode_error = None
+        if normalize:
+            try:
+                passed = normalize_html(actual_html) == normalize_html(expected_html)
+            except UnicodeDecodeError as e:
+                unicode_error = e
+                passed = False
+        else:
+            passed = actual_html == expected_html
+        if passed:
+            result_counts['pass'] += 1
+        else:
+            print_test_header(test['section'], test['example'], test['start_line'], test['end_line'])
+            sys.stdout.write(test['markdown'])
+            if unicode_error:
+                print("Unicode error: " + str(unicode_error))
+                print("Expected: " + repr(expected_html))
+                print("Got:      " + repr(actual_html))
+            else:
+                expected_html_lines = expected_html.splitlines(True)
+                actual_html_lines = actual_html.splitlines(True)
+                for diffline in unified_diff(expected_html_lines, actual_html_lines,
+                                "expected HTML", "actual HTML"):
+                    sys.stdout.write(diffline)
+            sys.stdout.write('\n')
+            result_counts['fail'] += 1
+    else:
+        print_test_header(test['section'], test['example'], test['start_line'], test['end_line'])
+        print("program returned error code %d" % retcode)
+        print(err)
+        result_counts['error'] += 1
+
+def get_tests(specfile):
+    line_number = 0
+    start_line = 0
+    end_line = 0
+    example_number = 0
+    markdown_lines = []
+    html_lines = []
+    state = 0  # 0 regular text, 1 markdown example, 2 html output
+    headertext = ''
+    tests = []
+
+    header_re = re.compile('#+ ')
+
+    with open(specfile, 'r', encoding='utf-8') as specf:
+        for line in specf:
+            line_number = line_number + 1
+            if state == 0 and re.match(header_re, line):
+                headertext = header_re.sub('', line).strip()
+            if line.strip() == ".":
+                state = (state + 1) % 3
+                if state == 0:
+                    example_number = example_number + 1
+                    end_line = line_number
+                    tests.append({
+                        "markdown":''.join(markdown_lines).replace('→',"\t"),
+                        "html":''.join(html_lines),
+                        "example": example_number,
+                        "start_line": start_line,
+                        "end_line": end_line,
+                        "section": headertext})
+                    start_line = 0
+                    markdown_lines = []
+                    html_lines = []
+            elif state == 1:
+                if start_line == 0:
+                    start_line = line_number - 1
+                markdown_lines.append(line)
+            elif state == 2:
+                html_lines.append(line)
+    return tests
+
+if __name__ == "__main__":
+    if args.debug_normalization:
+        print(normalize_html(sys.stdin.read()))
+        exit(0)
+
+    all_tests = get_tests(args.spec)
+    if args.pattern:
+        pattern_re = re.compile(args.pattern, re.IGNORECASE)
+    else:
+        pattern_re = re.compile('.')
+    tests = [ test for test in all_tests if re.search(pattern_re, test['section']) and (not args.number or test['example'] == args.number) ]
+    if args.dump_tests:
+        print(json.dumps(tests, ensure_ascii=False, indent=2))
+        exit(0)
+    else:
+        skipped = len(all_tests) - len(tests)
+        cmark = CMark(prog=args.program, library_dir=args.library_dir)
+        result_counts = {'pass': 0, 'fail': 0, 'error': 0, 'skip': skipped}
+        for test in tests:
+            do_test(test, args.normalize, result_counts)
+        print("{pass} passed, {fail} failed, {error} errored, {skip} skipped".format(**result_counts))
+        if result_counts['fail'] == 0 and result_counts['error'] == 0:
+            exit(0)
+        else:
+            exit(1)