| { |
| "jobs":[ |
| { |
| "name":"indexUpdate", |
| "workflow":"indexUpdate", |
| "parameters":{ |
| "tempStore":"temp" |
| } |
| }, |
| { |
| "name":"indexUpdateFeed", |
| "workflow":"indexUpdateFeed", |
| "parameters":{ |
| "tempStore":"temp" |
| } |
| }, |
| { |
| "name":"indexUpdateXml", |
| "workflow":"indexUpdateXml", |
| "parameters":{ |
| "tempStore":"temp" |
| } |
| }, |
| { |
| "name":"crawlFilesystem", |
| "workflow":"fileCrawling", |
| "parameters":{ |
| "tempStore":"temp", |
| "dataSource":"file", |
| "rootFolder":"< EDIT: directory to crawl >", |
| "jobToPushTo":"indexUpdate", |
| "mapping":{ |
| "fileContent":"Content", |
| "filePath":"Path", |
| "fileName":"Filename", |
| "fileSize":"Size", |
| "fileExtension":"Extension", |
| "fileLastModified":"LastModifiedDate" |
| } |
| } |
| }, |
| { |
| "name":"crawlSmilaWiki", |
| "workflow":"webCrawling", |
| "parameters":{ |
| "tempStore":"temp", |
| "dataSource":"web", |
| "jobToPushTo":"indexUpdate", |
| "startUrl":"http://wiki.eclipse.org/SMILA", |
| "linksPerBulk": 100, |
| "filters":{ |
| "urlPatterns": { |
| "include": ["http://wiki\\.eclipse\\.org/SMILA.*", |
| "http://wiki\\.eclipse\\.org/Image:.*", |
| "http://wiki\\.eclipse\\.org/images/.*"], |
| "exclude": [".*\\?.*", |
| "http://wiki\\.eclipse\\.org/images/archive/.*", |
| ".*\\.java"] |
| } |
| }, |
| "mapping": { |
| "httpCharset": "Charset", |
| "httpContenttype": "ContentType", |
| "httpLastModified": "LastModifiedDate", |
| "httpMimetype": "MimeType", |
| "httpSize": "Size", |
| "httpUrl": "Url", |
| "httpContent": "Content" |
| } |
| } |
| }, |
| { |
| "name":"crawlFeed", |
| "workflow":"feedCrawling", |
| "parameters":{ |
| "tempStore":"temp", |
| "dataSource":"feed", |
| "jobToPushTo":"indexUpdateFeed", |
| "feedUrls":["https://www.eclipse.org/forums/feed.php?mode=m&l=1&basic=1&frm=36&n=100", |
| "https://dev.eclipse.org/mhonarc/lists/smila-user/maillist.rss", |
| "https://dev.eclipse.org/mhonarc/lists/smila-dev/maillist.rss"], |
| "deltaProperties": ["itemPublishDate", "itemUpdateDate"], |
| "deltaImportStrategy":"additive", |
| "mapping": { |
| "itemUri":"Url", |
| "itemTitle":"Title", |
| "itemUpdateDate":"LastModifiedDate", |
| "itemContents": "Contents", |
| "itemDescription": "Description", |
| "itemAuthors": "Authors", |
| "itemLinks": "Links" |
| } |
| } |
| } |
| ] |
| } |