List of usage examples for org.apache.hadoop.conf Configuration getConfResourceAsReader
public Reader getConfResourceAsReader(String name)
name
. From source file:org.apache.nutch.urlfilter.automaton.AutomatonURLFilter.java
License:Apache License
/** * Rules specified as a config property will override rules specified as a * config file./*from w w w .j a v a2 s . co m*/ */ protected Reader getRulesReader(Configuration conf) throws IOException { String stringRules = conf.get(URLFILTER_AUTOMATON_RULES); if (stringRules != null) { return new StringReader(stringRules); } String fileRules = conf.get(URLFILTER_AUTOMATON_FILE); return conf.getConfResourceAsReader(fileRules); }
From source file:org.apache.nutch.urlfilter.domain.DomainURLFilter.java
License:Apache License
/** * Sets the configuration./*from ww w. j a v a2 s . co m*/ */ public void setConf(Configuration conf) { this.conf = conf; // get the extensions for domain urlfilter String pluginName = "urlfilter-domain"; Extension[] extensions = PluginRepository.get(conf).getExtensionPoint(URLFilter.class.getName()) .getExtensions(); for (int i = 0; i < extensions.length; i++) { Extension extension = extensions[i]; if (extension.getDescriptor().getPluginId().equals(pluginName)) { attributeFile = extension.getAttribute("file"); break; } } // handle blank non empty input if (attributeFile != null && attributeFile.trim().equals("")) { attributeFile = null; } if (attributeFile != null) { if (LOG.isInfoEnabled()) { LOG.info("Attribute \"file\" is defined for plugin " + pluginName + " as " + attributeFile); } } else { if (LOG.isWarnEnabled()) { LOG.warn("Attribute \"file\" is not defined in plugin.xml for plugin " + pluginName); } } // domain file and attribute "file" take precedence if defined String file = conf.get("urlfilter.domain.file"); String stringRules = conf.get("urlfilter.domain.rules"); if (domainFile != null) { file = domainFile; } else if (attributeFile != null) { file = attributeFile; } Reader reader = null; if (stringRules != null) { // takes precedence over files reader = new StringReader(stringRules); } else { reader = conf.getConfResourceAsReader(file); } try { if (reader == null) { reader = new FileReader(file); } readConfiguration(reader); } catch (IOException e) { LOG.error(org.apache.hadoop.util.StringUtils.stringifyException(e)); } }
From source file:org.apache.nutch.urlfilter.domainblacklist.DomainBlacklistURLFilter.java
License:Apache License
/** * Sets the configuration.//from ww w .jav a 2 s . co m */ public void setConf(Configuration conf) { this.conf = conf; // get the extensions for domain urlfilter String pluginName = "urlfilter-domainblacklist"; Extension[] extensions = PluginRepository.get(conf).getExtensionPoint(URLFilter.class.getName()) .getExtensions(); for (int i = 0; i < extensions.length; i++) { Extension extension = extensions[i]; if (extension.getDescriptor().getPluginId().equals(pluginName)) { attributeFile = extension.getAttribute("file"); break; } } // handle blank non empty input if (attributeFile != null && attributeFile.trim().equals("")) { attributeFile = null; } if (attributeFile != null) { if (LOG.isInfoEnabled()) { LOG.info("Attribute \"file\" is defined for plugin " + pluginName + " as " + attributeFile); } } else { if (LOG.isWarnEnabled()) { LOG.warn("Attribute \"file\" is not defined in plugin.xml for plugin " + pluginName); } } // domain file and attribute "file" take precedence if defined String file = conf.get("urlfilter.domainblacklist.file"); String stringRules = conf.get("urlfilter.domainblacklist.rules"); if (domainFile != null) { file = domainFile; } else if (attributeFile != null) { file = attributeFile; } Reader reader = null; if (stringRules != null) { // takes precedence over files reader = new StringReader(stringRules); } else { reader = conf.getConfResourceAsReader(file); } try { if (reader == null) { reader = new FileReader(file); } readConfiguration(reader); } catch (IOException e) { LOG.error(org.apache.hadoop.util.StringUtils.stringifyException(e)); } }
From source file:org.apache.nutch.urlfilter.exactduplicate.ExactDuplicateURLFilter.java
License:Apache License
public void setConf(Configuration conf) { this.conf = conf; String pluginName = "urlfilter-suffix"; Extension[] extensions = PluginRepository.get(conf).getExtensionPoint(URLFilter.class.getName()) .getExtensions();/* w w w . j a v a 2 s.c o m*/ for (int i = 0; i < extensions.length; i++) { Extension extension = extensions[i]; if (extension.getDescriptor().getPluginId().equals(pluginName)) { attributeFile = extension.getAttribute("file"); break; } } if (attributeFile != null && attributeFile.trim().equals("")) attributeFile = null; if (attributeFile != null) { if (LOG.isInfoEnabled()) { LOG.info("Attribute \"file\" is defined for plugin " + pluginName + " as " + attributeFile); } } else { // if (LOG.isWarnEnabled()) { // LOG.warn("Attribute \"file\" is not defined in plugin.xml for // plugin "+pluginName); // } } String file = conf.get("urlfilter.suffix.file"); String stringRules = conf.get("urlfilter.suffix.rules"); // attribute "file" takes precedence if defined if (attributeFile != null) file = attributeFile; Reader reader = null; if (stringRules != null) { // takes precedence over files reader = new StringReader(stringRules); } else { reader = conf.getConfResourceAsReader(file); } /*try { //readConfiguration(reader); } catch (IOException e) { if (LOG.isErrorEnabled()) { LOG.error(e.getMessage()); } throw new RuntimeException(e.getMessage(), e); }*/ }
From source file:org.apache.nutch.urlfilter.forbidden.ForbiddenURLFilter.java
License:Apache License
public void setConf(Configuration conf) { this.conf = conf; String pluginName = "urlfilter-forbidden"; Extension[] extensions = PluginRepository.get(conf).getExtensionPoint(URLFilter.class.getName()) .getExtensions();//from w ww . ja v a 2s .co m for (int i = 0; i < extensions.length; i++) { Extension extension = extensions[i]; if (extension.getDescriptor().getPluginId().equals(pluginName)) { attributeFile = extension.getAttribute("file"); break; } } if (attributeFile != null && attributeFile.trim().equals("")) attributeFile = null; if (attributeFile != null) { if (LOG.isInfoEnabled()) { LOG.info("Attribute \"file\" is defined for plugin " + pluginName + " as " + attributeFile); } } else { // if (LOG.isWarnEnabled()) { // LOG.warn("Attribute \"file\" is not defined in plugin.xml for // plugin "+pluginName); // } } String file = conf.get("urlfilter.forbidden.file"); // attribute "file" takes precedence if defined if (attributeFile != null) file = attributeFile; Reader reader = conf.getConfResourceAsReader(file); try { readConfiguration(reader); } catch (IOException e) { if (LOG.isErrorEnabled()) { LOG.error(e.getMessage()); } throw new RuntimeException(e.getMessage(), e); } }
From source file:org.apache.nutch.urlfilter.ignoreexempt.ExemptionUrlFilter.java
License:Apache License
/** * Gets reader for regex rules/*from w ww. j ava 2 s. c om*/ */ protected Reader getRulesReader(Configuration conf) throws IOException { String fileRules = conf.get(DB_IGNORE_EXTERNAL_EXEMPTIONS_FILE); return conf.getConfResourceAsReader(fileRules); }
From source file:org.apache.nutch.urlfilter.prefix.PrefixURLFilter.java
License:Apache License
public void setConf(Configuration conf) { this.conf = conf; String pluginName = "urlfilter-prefix"; Extension[] extensions = PluginRepository.get(conf).getExtensionPoint(URLFilter.class.getName()) .getExtensions();/*from ww w . ja va 2 s.com*/ for (int i = 0; i < extensions.length; i++) { Extension extension = extensions[i]; if (extension.getDescriptor().getPluginId().equals(pluginName)) { attributeFile = extension.getAttribute("file"); break; } } if (attributeFile != null && attributeFile.trim().equals("")) attributeFile = null; if (attributeFile != null) { if (LOG.isInfoEnabled()) { LOG.info("Attribute \"file\" is defined for plugin " + pluginName + " as " + attributeFile); } } else { // if (LOG.isWarnEnabled()) { // LOG.warn("Attribute \"file\" is not defined in plugin.xml for // plugin "+pluginName); // } } String file = conf.get("urlfilter.prefix.file"); String stringRules = conf.get("urlfilter.prefix.rules"); // attribute "file" takes precedence if defined if (attributeFile != null) file = attributeFile; Reader reader = null; if (stringRules != null) { // takes precedence over files reader = new StringReader(stringRules); } else { reader = conf.getConfResourceAsReader(file); } if (reader == null) { trie = new PrefixStringMatcher(new String[0]); } else { try { trie = readConfiguration(reader); } catch (IOException e) { if (LOG.isErrorEnabled()) { LOG.error(e.getMessage()); } // TODO mb@media-style.com: throw Exception? Because broken api. throw new RuntimeException(e.getMessage(), e); } } }
From source file:org.apache.nutch.urlfilter.suffix.SuffixURLFilter.java
License:Apache License
public void setConf(Configuration conf) { this.conf = conf; String pluginName = "urlfilter-suffix"; Extension[] extensions = PluginRepository.get(conf).getExtensionPoint(URLFilter.class.getName()) .getExtensions();//from w ww .j a v a 2s .c o m for (int i = 0; i < extensions.length; i++) { Extension extension = extensions[i]; if (extension.getDescriptor().getPluginId().equals(pluginName)) { attributeFile = extension.getAttribute("file"); break; } } if (attributeFile != null && attributeFile.trim().equals("")) attributeFile = null; if (attributeFile != null) { if (LOG.isInfoEnabled()) { LOG.info("Attribute \"file\" is defined for plugin " + pluginName + " as " + attributeFile); } } else { // if (LOG.isWarnEnabled()) { // LOG.warn("Attribute \"file\" is not defined in plugin.xml for // plugin "+pluginName); // } } String file = conf.get("urlfilter.suffix.file"); String stringRules = conf.get("urlfilter.suffix.rules"); // attribute "file" takes precedence if defined if (attributeFile != null) file = attributeFile; Reader reader = null; if (stringRules != null) { // takes precedence over files reader = new StringReader(stringRules); } else { reader = conf.getConfResourceAsReader(file); } try { readConfiguration(reader); } catch (IOException e) { if (LOG.isErrorEnabled()) { LOG.error(e.getMessage()); } throw new RuntimeException(e.getMessage(), e); } }