expose vnu options

.gitignore

@@ -7,3 +7,4 @@ pypy*
 dist
 
 .DS_Store
+.vscode

html5validator/cli.py

@@ -24,8 +24,16 @@ def main():
                         help='match file pattern (default: *.html)')
     parser.add_argument('--blacklist', type=str, nargs='*',
                         help='directory names to skip', default=[])
-    parser.add_argument('--show-warnings', dest='error_only',
-                        action='store_false', default=True)
+
+    parser.add_argument('--show-warnings', dest='errors_only',
+                        action='store_false', default=True,
+                        help='show warnings')
+    parser.add_argument('--no-langdetect',
+                        action='store_true', default=False,
+                        help='disable language detection')
+    parser.add_argument('--format', choices=['gnu', 'xml', 'json', 'text'],
+                        help='output format', default=None)
+
     parser.add_argument('--ignore', nargs='*', default=None,
                         type=lambda s: (s.decode('utf-8')
                                         if isinstance(s, bytes) else s),
@@ -35,42 +43,54 @@ def main():
                                         if isinstance(s, bytes) else s),
                         dest='ignore_re',
                         help='regular expression of messages to ignore')
+
     parser.add_argument('-l', action='store_const', const=2048,
                         dest='stack_size',
                         help=('run on larger files: sets Java '
-                              'stack size to 2048k')
-                        )
+                              'stack size to 2048k'))
     parser.add_argument('-ll', action='store_const', const=8192,
                         dest='stack_size',
                         help=('run on larger files: sets Java '
-                              'stack size to 8192k')
-                        )
+                              'stack size to 8192k'))
     parser.add_argument('-lll', action='store_const', const=32768,
                         dest='stack_size',
                         help=('run on larger files: sets Java '
-                              'stack size to 32768k')
-                        )
+                              'stack size to 32768k'))
+
     parser.add_argument('--log', default='WARNING',
                         help=('log level: DEBUG, INFO or WARNING '
                               '(default: WARNING)'))
     parser.add_argument('--version', action='version',
                         version='%(prog)s ' + VERSION)
     args = parser.parse_args()
 
+    vnu_options = []
+    if args.errors_only:
+        vnu_options.append('--errors_only')
+    if args.no_langdetect:
+        vnu_options.append('--no-langdetect')
+    if args.format:
+        vnu_options.append('--format')
+        vnu_options.append(args.format)
+
+    java_options = []
+    if args.stack_size:
+        java_options.append('-Xss{}k'.format(args.stack_size))
+
     logging.basicConfig(level=getattr(logging, args.log))
 
-    v = Validator(directory=args.root, match=args.match,
-                  blacklist=args.blacklist,
-                  ignore=args.ignore, ignore_re=args.ignore_re)
-    files = v.all_files()
+    validator = Validator(directory=args.root,
+                          match=args.match,
+                          blacklist=args.blacklist,
+                          ignore=args.ignore,
+                          ignore_re=args.ignore_re,
+                          java_options=java_options,
+                          vnu_options=vnu_options)
+    files = validator.all_files()
     LOGGER.info('Files to validate: \n  {0}'.format('\n  '.join(files)))
     LOGGER.info('Number of files: {0}'.format(len(files)))
 
-    error_count = v.validate(
-        files,
-        errors_only=args.error_only,
-        stack_size=args.stack_size,
-    )
+    error_count = validator.validate(files)
     sys.exit(min(error_count, 255))
 
 

html5validator/validator.py

@@ -25,12 +25,15 @@ def __str__(self):
 class Validator(object):
 
     def __init__(self, directory='.', match='*.html', blacklist=None,
-                 ignore=None, ignore_re=None):
+                 ignore=None, ignore_re=None,
+                 java_options=None, vnu_options=None):
         self.directory = directory
         self.match = match
         self.blacklist = blacklist if blacklist else []
         self.ignore = ignore if ignore else []
         self.ignore_re = ignore_re if ignore_re else []
+        self.java_options = java_options if java_options is not None else []
+        self.vnu_options = vnu_options if vnu_options is not None else []
 
         # add default ignore_re
         self.ignore_re += DEFAULT_IGNORE_RE
@@ -77,12 +80,7 @@ def all_files(self, skip_invisible=True):
                 files.append(os.path.join(root, filename))
         return files
 
-    def validate(self, files=None, errors_only=True, stack_size=None):
-        vnu_opts, java_opts = [], []
-        if errors_only:
-            vnu_opts.append('--errors-only')
-        if stack_size:
-            java_opts.append('-Xss{}k'.format(stack_size))
+    def validate(self, files=None):
         if not files:
             files = self.all_files()
 
@@ -95,9 +93,10 @@ def validate(self, files=None, errors_only=True, stack_size=None):
                 raise JavaNotFoundException()
 
         try:
+            cmd = (['java'] + self.java_options +
+                   ['-jar', self.vnu_jar_location] + self.vnu_options + files)
             o = subprocess.check_output(
-                ['java'] + java_opts +
-                ['-jar', self.vnu_jar_location] + vnu_opts + files,
+                cmd,
                 stderr=subprocess.STDOUT,
             ).decode('utf-8')
         except subprocess.CalledProcessError as e:

vnujar/CHANGELOG.md

@@ -2,11 +2,26 @@ With a few exceptions, this is a record of mainly just user-facing
 changes—that is, either changes to the actual behavior of the checker, or
 changes to any options/interfaces the checker exposes for developers.
 
+# 17.11.1
+07 October 2017
+  - Fix bug that made the vnu.jar `--Werror` option not work as expected
+  - Make vnu.jar exit 0 if all errors have been filtered out
+
+# 17.11.0
+06 October 2017
+  - Allow DPUB roles on more elements (per ARIA in HTML spec updates)
+  - Add `--Werror` option to the vnu.jar command-line checker. The option
+    causes the checker to exit non-zero if any warnings are encountered
+    (even if there are no errors).
+  - Fix mismatch that caused message-filtering failures
+  - Fix memory leak in language detector (patch from @tgyurci)
+  - Stop reporting HTML4-specific parse errors for HTML4-doctype docs
+
 # 17.9.0
 20 August 2017
-  - Allow script[nomodule]
-  - Allow the hover, any-hover, pointer, and any-pointer media features
-  - Allow @scope, @updateviacache, @workertype for link[rel=serviceworker]
+  - Allow `script[nomodule]`
+  - Allow `hover`, `any-hover`, `pointer`, and `any-pointer` media features
+  - Allow `@scope`, `@updateviacache`, `@workertype` for `link[rel=serviceworker]`
   - Allow `&;` (don’t report it as “`&` did not start a character reference”)
   - Add `acceptlanguage` query parameter, to specify an Accept-Language
     request-header value for checker to send when fetching remote documents
@@ -21,21 +36,21 @@ changes to any options/interfaces the checker exposes for developers.
     context**” wording in error message with “Attribute "foo" not allowed on
     element "param" **at this point**” (for consistent working for that error
     between the command-line checker and the web-based checker).
-  - Disallow the "contextmenu" attribute and type=contextmenu and type=toolbar
+  - Disallow the `contextmenu` attribute and `type=contextmenu` and `type=toolbar`
     for the `menu` element.
-  - Allow link[rel=serviceworker]
+  - Allow `link[rel=serviceworker]`
   - Allow floating-point numbers in attribute values to start with decimal point
-  - Allow a[href] in SVG wherever a[xlink:href] is allowed
-  - Allow the "focusable" and "tabindex" attributes on SVG elements
-  - Fix bug that disallowed `progress` & `meter` as label[for] targets
+  - Allow `a[href]` in SVG wherever `a[xlink:href]` is allowed
+  - Allow the `focusable` and `tabindex` attributes on SVG elements
+  - Fix bug that disallowed `progress` & `meter` as `label[for]` targets
   - Default to text/html for checking file uploads
-  - Emit warnings for use of rel={copyright,previous}
+  - Emit warnings for use of `rel={copyright,previous}`
   - Prevent Bulgarian ➡ Russian misidentifications in language detector
   - Skip figcaption elements when running the language detector
 
 # 17.3.0
 26 March 2017
-  - Allow “color” attribute with link[rel="mask-icon"]
+  - Allow `color` attribute with `link[rel="mask-icon"]`
   - Allow `meta[name]` to have `itemref`/`itemscope`/`itemtype`/`itemid`
   - Allow `allow-top-navigation-by-user-activation` in `iframe[sandbox]`
   - Stop hiding “sectioning roots” headings in “Heading-level outline”

vnujar/README.md

@@ -62,7 +62,7 @@ frontend such as:
 Use the `vnu.jar` HTML checker as an executable for command-line checking of
 documents by invoking it like this:
 
-      java -jar ~/vnu.jar [--errors-only] [--exit-zero-always]
+      java -jar ~/vnu.jar [--errors-only] [--Werror] [--exit-zero-always]
            [--asciiquotes] [--no-stream] [--format gnu|xml|json|text]
            [--filterfile FILENAME] [--filterpattern PATTERN] [--html]
            [--skip-non-html] [--no-langdetect] [--help] [--verbose] [--version]
@@ -111,9 +111,7 @@ executable provides the following options:
     Specifies whether ASCII quotation marks are substituted for Unicode smart
     quotation marks in messages.
 
-    default: "no"
-
-    possible values: "yes" or "no"
+    default: [unset; Unicode smart quotation marks are used in messages]
 
 #### --errors-only
 
@@ -122,6 +120,13 @@ executable provides the following options:
 
     default: [unset; all message reported, including warnings & info messages]
 
+#### --Werror
+
+    Makes the checker exit non-zero if any warnings are encountered (even if
+    there are no errors).
+
+    default: [unset; checker exits zero if only warnings are encountered]
+
 #### --exit-zero-always
 
     Makes the checker exit zero even if errors are reported for any documents.

vnujar/index.html

@@ -193,7 +193,7 @@ <h2>Usage</h2>
 <p>Use the <code>vnu.jar</code> HTML checker as an executable for
 command-line checking of documents by invoking it like this:
 <pre>
-  java -jar ~/vnu.jar [--errors-only] [--exit-zero-always]
+  java -jar ~/vnu.jar [--errors-only] [--Werror] [--exit-zero-always]
        [--asciiquotes] [--no-stream] [--format gnu|xml|json|text]
        [--filterfile FILENAME] [--filterpattern PATTERN]
        [--html] [--skip-non-html] [--no-langdetect]
@@ -249,14 +249,12 @@ <h3 id="options">Options</h3>
 <p>When used from the command line as described in this section, the
 <code>vnu.jar</code> executable provides the following options:
 
-<h4 id="errors-only">--asciiquotes</h4>
+<h4 id="asciiquotes">--asciiquotes</h4>
 <pre>
 Specifies whether ASCII quotation marks are substituted for Unicode
 smart quotation marks in messages.
 
-default: "no"
-
-possible values: "yes" or "no"
+default: [unset; Unicode smart quotation marks are used in messages]
 </pre>
 
 <h4 id="errors-only">--errors-only</h4>
@@ -267,6 +265,14 @@ <h4 id="errors-only">--errors-only</h4>
 default: [unset; all message reported, including warnings & info messages]
 </pre>
 
+<h4 id="Werror">--Werror</h4>
+<pre>
+Makes the checker exit non-zero if any warnings are encountered (even if
+there are no errors).
+
+default: [unset; checker exits zero if only warnings are encountered]
+</pre>
+
 <h4 id="exit-zero-always">--exit-zero-always</h4>
 <pre>
 Makes the checker exit zero even if errors are reported for any documents.