You can translate the question and the replies:

How do I tell aracne to index a location on the local hard drive?

Hello! I'm trying to figure out how to use Denodo Aracne to index some files, but the Aracne Administration Guide doesn't seem to have the answer. How do I configure Denodo Aracne to index a bunch of files so I can use Denodo VDP to then run queries to search for documents that contain my keywords? Thank you~!
01-06-2015 18:10:32 -0400

4 Answers

Hi, First, you have to create an index from the aracne web administration tool. Once you have created an index, you have to use the scheduler web administration tool to create a new ARN job. In the 'Extraction section' you can specify the URL to be indexed. This can be a location in your system or a website. You can then create a new Aracne datasource using the VDP Administration tool to run queries for the documents that contain the keywords. Check out the section 'Aracne Sources' in the 'Virtual DataPort Administration Guide' ( to know more about creating Aracne datasource in VDP. Hope this helps.
Denodo Team
02-06-2015 07:33:58 -0400
This does help... here is what I tried to do: 1) Created a new index in aracne called "wiki_index" and used the default schema and default analyzer 2) Created a new job in scheduler called "wiki_index_job" and told it to index a web site with depth of 1 Ran the job and it completed successfully but didn't save any of the data 3) I modified the wiki_index_job and added an ARN-Index exporter using: filter_sequence: arn_default data source: arnIndex index name: wiki_index clear index: TRUE 4) I re-ran the wiki_index_job and it completed successfully and stored 38 tuples in the index 5) I went back to the aracne administration tool > AracneIndexer and did a search in my index using the "Search in index" option and I can search the index and I see the "INDEXSCORE" seems to be dynamically populated based on the search keywords and the relevance for the found items in the index (I'm having to guess what INDEXSCORE is because I cannot find this in any of your documentation). 6) I went to Denodo VDP and created a new data source, connecting it to my index server (NOT the crawler)... port 9000... and then I created a new base view. 7) I read the information about creating MAIN TERMS for a document in the Denodo Virtual Data Port administration guide... this was interesting, but I'm not sure I would use the main terms when trying to search for data. I decided to create main terms for the "content" field. 8) Now I have a base view pointing at my "wiki_index" and a content_MAIN_TERM field with an array of registers. I decide to execute a query against my "wiki_index" base view: SELECT * FROM wiki_index WHERE content contains 'middleware' I get EXTREMELY similar results to #5 above, however, the index score is different. Why is the index score different? How can I simulate the "Search in index" functionality that exists via /webadmin/denodo-aracne-admin, but do it from VDP? I'd love to have a simple way to MIMIC the simpleness of the search keywords (I can't seem to make the "contains" operator work exactly the same, especially when searching for multiple keywords). Finally, I can see how to index a web page, but how do I index WORD, EXCEL, and POWERPOINT documents? I see that I can upload my documents to a web server and then point the crawler (via scheduler job) at the web server. Is this the only way? THANK YOU~!
02-06-2015 13:39:14 -0400
Update: The Denodo Scheduler Administration guide has answers about where files can be stored: has information on "WebBot" and how you can use file:///path/to/documents It seems my only remaining question is how to use Virtual Data Port to mimic the "Search in Index" functionality.
02-06-2015 13:57:57 -0400
I'm going to post a new Q&A question since the original question to this one has been answered. Thank you!!!
03-06-2015 13:09:35 -0400
You must sign in to add an answer. If you do not have an account, you can register here