📄 runtimelimitenforcer.html
字号:
<a name="79" href="#79">79</a> OP_PAUSE, OP_TERMINATE, OP_BLOCK_URIS};<a name="80" href="#80">80</a> <a name="81" href="#81">81</a> <strong>public</strong> <a href="../../../../org/archive/crawler/prefetch/RuntimeLimitEnforcer.html">RuntimeLimitEnforcer</a>(String name) {<a name="82" href="#82">82</a> <strong>super</strong>(name, <span class="string">"A processor that halts further progress once a fixed "</span> +<a name="83" href="#83">83</a> <span class="string">"amount of time has elapsed since the start of a crawl. "</span> +<a name="84" href="#84">84</a> <span class="string">"It is possible to configure this processor per host, but "</span> +<a name="85" href="#85">85</a> <span class="string">"it should be noted that Heritrix does not track runtime "</span> +<a name="86" href="#86">86</a> <span class="string">"per host seperately. Especially when using facilities "</span> +<a name="87" href="#87">87</a> <span class="string">"like the BdbFrontier's hold-queues, the actual amount of "</span> +<a name="88" href="#88">88</a> <span class="string">"time spent crawling a host may have little relevance to "</span> +<a name="89" href="#89">89</a> <span class="string">"total elapsed time. Note however that using overrides "</span> +<a name="90" href="#90">90</a> <span class="string">"and/or refinements only makes sense when using the "</span> +<a name="91" href="#91">91</a> <span class="string">"'Block URIs' end operation. The pause and terminate "</span> +<a name="92" href="#92">92</a> <span class="string">"operations have global impact once encountered."</span>);<a name="93" href="#93">93</a> <a href="../../../../org/archive/crawler/settings/Type.html">Type</a> t = <strong>new</strong> SimpleType(<a name="94" href="#94">94</a> ATTR_RUNTIME_SECONDS,<a name="95" href="#95">95</a> <span class="string">"The amount of time, in seconds, that the crawl will be "</span> +<a name="96" href="#96">96</a> <span class="string">"allowed to run before this processor performs it's 'end "</span> +<a name="97" href="#97">97</a> <span class="string">"operation.'"</span>,<a name="98" href="#98">98</a> DEFAULT_RUNTIME_SECONDS);<a name="99" href="#99">99</a> addElementToDefinition(t);<a name="100" href="#100">100</a> t = <strong>new</strong> <a href="../../../../org/archive/crawler/settings/SimpleType.html">SimpleType</a>(<a name="101" href="#101">101</a> ATTR_END_OPERATION,<a name="102" href="#102">102</a> <span class="string">"The action that the processor takes once the runtime has "</span> +<a name="103" href="#103">103</a> <span class="string">"elapsed.\n "</span> +<a name="104" href="#104">104</a> <span class="string">"Operation: Pause job - Pauses the crawl. A change "</span> +<a name="105" href="#105">105</a> <span class="string">"(increase) to the runtime duration will "</span> +<a name="106" href="#106">106</a> <span class="string">"make it pausible to resume the crawl. Attempts to resume "</span> +<a name="107" href="#107">107</a> <span class="string">"the crawl without modifying the run time will cause it to "</span> +<a name="108" href="#108">108</a> <span class="string">"be immediately paused again.\n "</span> +<a name="109" href="#109">109</a> <span class="string">"Operation: Terminate job - Terminates the job. Equivalent "</span> +<a name="110" href="#110">110</a> <span class="string">"to using the max-time setting on the CrawlController.\n "</span> +<a name="111" href="#111">111</a> <span class="string">"Operation: Block URIs - Blocks each URI with an -5002 "</span> +<a name="112" href="#112">112</a> <span class="string">"(blocked by custom processor) fetch status code. This will "</span> +<a name="113" href="#113">113</a> <span class="string">"cause all the URIs queued to wind up in the crawl.log."</span>,<a name="114" href="#114">114</a> DEFAULT_END_OPERATION, <a name="115" href="#115">115</a> AVAILABLE_END_OPERATIONS);<a name="116" href="#116">116</a> addElementToDefinition(t);<a name="117" href="#117">117</a> }<a name="118" href="#118">118</a> <a name="119" href="#119">119</a> <strong>protected</strong> <strong>void</strong> innerProcess(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi) throws InterruptedException {<a name="120" href="#120">120</a> <strong>long</strong> allowedRuntime = getRuntime(curi);<a name="121" href="#121">121</a> <strong>long</strong> currentRuntime = getController().getStatistics().crawlDuration();<a name="122" href="#122">122</a> <strong>if</strong>(currentRuntime > allowedRuntime){<a name="123" href="#123">123</a> String op = (String)getUncheckedAttribute(curi,ATTR_END_OPERATION);<a name="124" href="#124">124</a> <strong>if</strong>(op != <strong>null</strong>){<a name="125" href="#125">125</a> <strong>if</strong>(op.equals(OP_PAUSE)){<a name="126" href="#126">126</a> getController().requestCrawlPause();<a name="127" href="#127">127</a> } <strong>else</strong> <strong>if</strong>(op.equals(OP_TERMINATE)){<a name="128" href="#128">128</a> getController().requestCrawlStop(<a name="129" href="#129">129</a> CrawlJob.STATUS_FINISHED_TIME_LIMIT);<a name="130" href="#130">130</a> } <strong>else</strong> <strong>if</strong>(op.equals(OP_BLOCK_URIS)){<a name="131" href="#131">131</a> curi.setFetchStatus(S_BLOCKED_BY_RUNTIME_LIMIT);<a name="132" href="#132">132</a> curi.addAnnotation(<span class="string">"Runtime exceeded "</span> + allowedRuntime + <a name="133" href="#133">133</a> <span class="string">"ms"</span>);<a name="134" href="#134">134</a> curi.skipToProcessorChain(<a name="135" href="#135">135</a> getController().getPostprocessorChain());<a name="136" href="#136">136</a> }<a name="137" href="#137">137</a> } <strong>else</strong> {<a name="138" href="#138">138</a> logger.log(Level.SEVERE,<span class="string">"Null value for "</span> + ATTR_END_OPERATION + <a name="139" href="#139">139</a> <span class="string">" when processing "</span> + curi.toString());<a name="140" href="#140">140</a> }<a name="141" href="#141">141</a> }<a name="142" href="#142">142</a> }<a name="143" href="#143">143</a> <a name="144" href="#144">144</a> <em>/**<em>*</em></em><a name="145" href="#145">145</a> <em> * Returns the amount of time to allow the crawl to run before this </em><a name="146" href="#146">146</a> <em> * processor interrupts.</em><a name="147" href="#147">147</a> <em> * @return the amount of time in milliseconds.</em><a name="148" href="#148">148</a> <em> */</em><a name="149" href="#149">149</a> <strong>protected</strong> <strong>long</strong> getRuntime(<a href="../../../../org/archive/crawler/datamodel/CrawlURI.html">CrawlURI</a> curi){<a name="150" href="#150">150</a> Object o = getUncheckedAttribute(curi,ATTR_RUNTIME_SECONDS);<a name="151" href="#151">151</a> <strong>if</strong>(o == <strong>null</strong>){<a name="152" href="#152">152</a> logger.log(Level.SEVERE,<span class="string">"Null value for "</span> + ATTR_RUNTIME_SECONDS + <a name="153" href="#153">153</a> <span class="string">" when processing "</span> + curi.toString());<a name="154" href="#154">154</a> <strong>return</strong> Long.MAX_VALUE;<a name="155" href="#155">155</a> }<a name="156" href="#156">156</a> <strong>return</strong> ((Long)o).longValue()*1000; <em class="comment">//extract value and convert to ms.</em><a name="157" href="#157">157</a> }<a name="158" href="#158">158</a> <a name="159" href="#159">159</a> }</pre><hr/><div id="footer">This page was automatically generated by <a href="http://maven.apache.org/">Maven</a></div></body></html>
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -