Changeset 126

Timestamp:

May 24, 2010, 4:07:06 PM (15 years ago)

Author:

shunfa

Message:

commit test/install_func.sh

File:

• nutchez-0.2/src/test/install_func.sh (modified) (2 diffs)

nutchez-0.2/src/test/install_func.sh

@@ -114 +114 @@
 }
 
-###################
-# undo... sed or cp
+
 set_crawl-urlfilter () {
   echo "set NutchEZ_HOME/conf/set_crawl-urlfilter.txt"
+  Line_NO=`cat '$NutchEZ'/conf/crawl-urlfilter.txt | grep -n 'skip URLs containing certain characters as probable queries, etc.' | sed 's/:.*//g'`
+  sed -i ''$((Line_NO+1))'d' $NutchEZ/conf/crawl-urlfilter.txt
+  sed -i ''$Line_NO'a -[*!@]' $NutchEZ/conf/crawl-urlfilter.txt
+  Line_NO=`cat crawl-urlfilter.txt | grep -n 'accept hosts in MY.DOMAIN.NAME' | sed 's/:.*//g'`
+  sed -i ''$((Line_NO+1))'d' $NutchEZ/conf/crawl-urlfilter.txt
+  sed -i ''$Line_NO'a +^http://([a-z0-9]*\.)*.*/' $NutchEZ/conf/crawl-urlfilter.txt
 }
 
@@ -145 +150 @@
 }
 
-####################
-# undo...sed or cp
 set_server () {
   echo "$NutchEZ_HOME/tomcat/conf/server.xml"
+  Line_NO=`cat '$NutchEZ'/tomcat/conf/server.xml | grep -n '<Connector port="8080" protocol="HTTP/1.1"' | sed 's/:.*//g'`
+  sed -i ''$Line_NO','$(($Line_NO+2))'d' server.xml
+  sed -i ''$Line_NO'a <Connector port="8080" protocol="HTTP/1.1"\
+               connectionTimeout="20000"\
+               redirectPort="8443" URIEncoding="UTF-8"/>\
+' $NutchEZ/tomcat/conf/server.xml
 }
 
-####################
-# undo...sed or cp
+
 set_nutch-site () {
   echo "$NutchEZ_HOME/tomcat/webapps/ROOT/WEB-INF/classes/nutch-site.xml"
+  
+  # 搜尋加入設定的行號位址
+  line_NO=`cat '$NutchEZ'/conf/nutch-site.xml | grep -n '<'configuration'>' | sed 's/:.*//g'`
+  
+  # 加入設定檔
+  sed -i ''$line_NO'a <property>\
+  <name>http.agent.name</name>\
+  <value>waue</value>\
+  <description>HTTP 'User-Agent' request header. </description>\
+</property>\
+<property>\
+  <name>http.agent.description</name>\
+  <value>MyTest</value>\
+  <description>Further description</description>\
+</property>\
+<property>\
+  <name>http.agent.url</name>\
+  <value>'$MasterDNS'</value>\
+  <description>A URL to advertise in the User-Agent header. </description>\
+</property>\
+<property>\
+  <name>http.agent.email</name>\
+  <value>'$Admin_email'</value>\
+  <description>An email address\
+  </description>\
+</property>\
+' $NutchEZ/conf/nutch-site.xml
 }