Return-Path: X-Original-To: apmail-any23-dev-archive@www.apache.org Delivered-To: apmail-any23-dev-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id A74D318030 for ; Fri, 25 Mar 2016 22:24:27 +0000 (UTC) Received: (qmail 60902 invoked by uid 500); 25 Mar 2016 22:24:26 -0000 Delivered-To: apmail-any23-dev-archive@any23.apache.org Received: (qmail 60858 invoked by uid 500); 25 Mar 2016 22:24:26 -0000 Mailing-List: contact dev-help@any23.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@any23.apache.org Delivered-To: mailing list dev@any23.apache.org Received: (qmail 60705 invoked by uid 99); 25 Mar 2016 22:24:26 -0000 Received: from arcas.apache.org (HELO arcas) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 25 Mar 2016 22:24:26 +0000 Received: from arcas.apache.org (localhost [127.0.0.1]) by arcas (Postfix) with ESMTP id E9D162C1F62 for ; Fri, 25 Mar 2016 22:24:25 +0000 (UTC) Date: Fri, 25 Mar 2016 22:24:25 +0000 (UTC) From: "ASF GitHub Bot (JIRA)" To: dev@any23.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (ANY23-247) FIX Attribute name "itemscope" associated with an element type "html" must be followed by the ' = ' character. MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/ANY23-247?page=3Dcom.atlassian.= jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D15212= 500#comment-15212500 ]=20 ASF GitHub Bot commented on ANY23-247: -------------------------------------- Github user lewismc commented on the pull request: https://github.com/apache/any23/pull/17#issuecomment-201550510 =20 I agree. Jumping through this in the debugged made me think the same. I think it is different if Any23 is to be a PURE implementation... But = that is clearly not the case. Any23 fits in best when it can be used to extr= act semantics from any old crap input that it is fed. Parsers and extractor= s *should not* fail when there is a piece of crap input HTML. Currently, that's exactly what happens and it is extremely limiting. =20 I would like to propose that this PR is committed to master as is, we t= hen open a brand new issue which acts exactly your comments refactoring out content extractor and reusing the input stream which has been fixed, et= c. =20 Any thoughts Peter? Thanks fr quick response. =20 On Friday, March 25, 2016, Peter Ansell wrot= e: =20 > The system does seem a little too complex for our purposes and isn't > usable because of that. > > Removing generics would be the first step IMO as there are too many > rawtypes definitions which indicate generics are being used badly. > > ContentExtractor may be able to be completely removed instead of bein= g > refitted into the process after that and the parser should always be = set to > parse as far as practical for our purposes. > > It is a little strange that there isn't a buffered, markable, InputSt= ream > provided for all of the steps to reuse as necessary rather than pushi= ng a > raw InputStream or other source into different extractors. > > =E2=80=94 > You are receiving this because you authored the thread. > Reply to this email directly or view it on GitHub > > =20 =20 --=20 *Lewis* > FIX Attribute name "itemscope" associated with an element type "html" mus= t be followed by the ' =3D ' character. > -------------------------------------------------------------------------= ------------------------------------- > > Key: ANY23-247 > URL: https://issues.apache.org/jira/browse/ANY23-247 > Project: Apache Any23 > Issue Type: Improvement > Affects Versions: 1.1 > Reporter: Lewis John McGibbney > Assignee: Lewis John McGibbney > Fix For: 1.2 > > > In the following markup > {code} > > > > > > > ... > {code} > Due to the absence of any subsequent value for *itemscope*, we get the fo= llowing error in our web server logs > {code} > [Fatal Error] :2:185: Attribute name "itemscope" associated with an eleme= nt type "html" must be followed by the ' =3D ' character. > {code} > Although the markup semantics are incorrect, Any23 should simply perform = a check for the itemscope value being null, if this is the case then add *= =3D""*, there is a precedent for us doing something like this before, I jus= t cant find the ticket right now! > The code we need to add is present within either=20 > core/src/main/java/org/apache/any23/extractor/microdata/ItemScope.java > core/src/main/java/org/apache/any23/extractor/microdata/MicrodataParser.j= ava -- This message was sent by Atlassian JIRA (v6.3.4#6332)