From aee5b108e59dab0bfd8412d1b58af7881fc4000d Mon Sep 17 00:00:00 2001 From: Michael Peter Christen Date: Mon, 7 Jul 2014 13:37:17 +0200 Subject: [PATCH] added linkScraperParser, a parser which ignores the text like the generic parser but extracts links like the htmlParser. This should be used for ASCII documents without known text format annotation like source code files or json documents. Probably also good for xml files without known schema. --- source/net/yacy/document/Document.java | 4 +- source/net/yacy/document/TextParser.java | 2 + .../net/yacy/document/parser/htmlParser.java | 6 -- .../document/parser/linkScraperParser.java | 91 +++++++++++++++++++ 4 files changed, 95 insertions(+), 8 deletions(-) create mode 100644 source/net/yacy/document/parser/linkScraperParser.java diff --git a/source/net/yacy/document/Document.java b/source/net/yacy/document/Document.java index d687b64d2..040227428 100644 --- a/source/net/yacy/document/Document.java +++ b/source/net/yacy/document/Document.java @@ -433,7 +433,7 @@ dc_rights return this.anchors; } - public Map getRSS() { + public LinkedHashMap getRSS() { // returns all links embedded as anchors (clickeable entities) // this is a url(String)/text(String) map return this.rss; @@ -458,7 +458,7 @@ dc_rights return this.videolinks; } - public Map getImages() { + public LinkedHashMap getImages() { // returns all links enbedded as pictures (visible in document) // this resturns a htmlFilterImageEntry collection if (!this.resorted) resortLinks(); diff --git a/source/net/yacy/document/TextParser.java b/source/net/yacy/document/TextParser.java index 023299ad9..9aa0ccb20 100644 --- a/source/net/yacy/document/TextParser.java +++ b/source/net/yacy/document/TextParser.java @@ -43,6 +43,7 @@ import net.yacy.document.parser.docParser; import net.yacy.document.parser.genericParser; import net.yacy.document.parser.gzipParser; import net.yacy.document.parser.htmlParser; +import net.yacy.document.parser.linkScraperParser; import net.yacy.document.parser.mmParser; import net.yacy.document.parser.odtParser; import net.yacy.document.parser.ooxmlParser; @@ -91,6 +92,7 @@ public final class TextParser { if (Switchboard.getSwitchboard().getConfigBool("parserAugmentation.RDFa", true)) initParser(new RDFaParser()); initParser(new htmlParser()); // called within rdfa parser initParser(new genericImageParser()); + initParser(new linkScraperParser()); initParser(new mmParser()); initParser(new odtParser()); initParser(new ooxmlParser()); diff --git a/source/net/yacy/document/parser/htmlParser.java b/source/net/yacy/document/parser/htmlParser.java index e9e6cf244..021080aff 100644 --- a/source/net/yacy/document/parser/htmlParser.java +++ b/source/net/yacy/document/parser/htmlParser.java @@ -72,18 +72,12 @@ public class htmlParser extends AbstractParser implements Parser { this.SUPPORTED_EXTENSIONS.add("aspx"); this.SUPPORTED_EXTENSIONS.add("tex"); this.SUPPORTED_EXTENSIONS.add("txt"); - //SUPPORTED_EXTENSIONS.add("js"); - this.SUPPORTED_EXTENSIONS.add("jsp"); - this.SUPPORTED_EXTENSIONS.add("mf"); - this.SUPPORTED_EXTENSIONS.add("pl"); - this.SUPPORTED_EXTENSIONS.add("py"); this.SUPPORTED_MIME_TYPES.add("text/html"); this.SUPPORTED_MIME_TYPES.add("text/xhtml+xml"); this.SUPPORTED_MIME_TYPES.add("application/xhtml+xml"); this.SUPPORTED_MIME_TYPES.add("application/x-httpd-php"); this.SUPPORTED_MIME_TYPES.add("application/x-tex"); this.SUPPORTED_MIME_TYPES.add("text/plain"); - this.SUPPORTED_MIME_TYPES.add("text/sgml"); this.SUPPORTED_MIME_TYPES.add("text/csv"); } diff --git a/source/net/yacy/document/parser/linkScraperParser.java b/source/net/yacy/document/parser/linkScraperParser.java new file mode 100644 index 000000000..a7a586361 --- /dev/null +++ b/source/net/yacy/document/parser/linkScraperParser.java @@ -0,0 +1,91 @@ +/** + * linkScraperParser + * Copyright 2014 by Michael Peter Christen, mc@yacy.net, Frankfurt a. M., Germany + * First released 07.07.2014 at http://yacy.net + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public License + * along with this program in the file lgpl21.txt + * If not, see . + */ + +package net.yacy.document.parser; + +import java.io.InputStream; +import java.util.Date; + +import net.yacy.cora.document.id.AnchorURL; +import net.yacy.cora.document.id.MultiProtocolURL; +import net.yacy.document.AbstractParser; +import net.yacy.document.Document; +import net.yacy.document.Parser; + +/** + * This parser is used if we know that the content is text but the exact format is unknown. + * We just expect that the content has some html links and we use it to discover those links. + * Furthermore the URL is tokenized and added as the text part, just as the generic parser does. + * This parser can i.e. be used for application source code and json files. + */ +public class linkScraperParser extends AbstractParser implements Parser { + + public linkScraperParser() { + super("Link Scraper Parser"); + this.SUPPORTED_EXTENSIONS.add("js"); + this.SUPPORTED_EXTENSIONS.add("jsp"); + this.SUPPORTED_EXTENSIONS.add("json"); + this.SUPPORTED_EXTENSIONS.add("jsonp"); + this.SUPPORTED_EXTENSIONS.add("mf"); // metafont + this.SUPPORTED_EXTENSIONS.add("pl"); // prolog + this.SUPPORTED_EXTENSIONS.add("py"); // python + this.SUPPORTED_EXTENSIONS.add("c"); // c + this.SUPPORTED_EXTENSIONS.add("cpp"); // c++ + this.SUPPORTED_EXTENSIONS.add("h"); // header file (whatever) + this.SUPPORTED_MIME_TYPES.add("application/json"); // correct for json + this.SUPPORTED_MIME_TYPES.add("application/x-javascript"); // wrong, but used + this.SUPPORTED_MIME_TYPES.add("text/javascript"); // correct for jsonp + this.SUPPORTED_MIME_TYPES.add("text/x-javascript"); // wrong, but used + this.SUPPORTED_MIME_TYPES.add("text/x-json"); // wrong, but used + this.SUPPORTED_MIME_TYPES.add("text/sgml"); + this.SUPPORTED_MIME_TYPES.add("text/sgml"); + } + @Override + public Document[] parse(final AnchorURL location, final String mimeType, + final String charset, final InputStream source) + throws Parser.Failure, InterruptedException { + + Document[] htmlParserDocs = new htmlParser().parse(location, mimeType, charset, source); + Document htmlParserDoc = htmlParserDocs == null ? null : Document.mergeDocuments(location, mimeType, htmlParserDocs); + + + String filename = location.getFileName(); + final Document[] docs = new Document[]{new Document( + location, + mimeType, + charset, + this, + null, + null, + singleList(filename.isEmpty() ? location.toTokens() : MultiProtocolURL.unescape(filename)), // title + "", // author + location.getHost(), + null, + null, + 0.0f, 0.0f, + location.toTokens(), + htmlParserDoc == null ? null : htmlParserDoc.getAnchors(), + htmlParserDoc == null ? null : htmlParserDoc.getRSS(), + htmlParserDoc == null ? null : htmlParserDoc.getImages(), + false, + new Date())}; + return docs; + } +}