08-09-2012 06:58 AM - edited 08-09-2012 06:59 AM
We have configured the Knowlegde management and we are able to add and update the data manually.
We are trying to update the knowledge management by uploading the ".csv" file.
But no records are getting updated. Kindly find the attached file for the error we are getting.
Thanks in advance.
Solved! Go to Solution.
08-09-2012 07:14 AM
It seems that you are getting error because you are not providing the id of Knowledge document in you csv file or ids are not unique.
It is mandatory to have unique key for each document/record to be added in to knowledge table.
So just generate the id first according to documents already uploaded in table then add the id in your csv and importer descriptor.
08-09-2012 07:34 AM
Firstly, thank you for the response piku!
Could you please tell me how to define the unique key in the csv file?
We have tried adding random numbers which would act as unique keys, but that did not help.
08-10-2012 07:31 AM
To upload the data from csv file you have to use the import descriptor and have to map the field in both of these.
To add the id first of all , add id column in your importer descriptor then according to that position add the id field in your csv file.To add the value in your .csv file for this id column check what is last id of knowledge doc in your environment and generate next ids accordingly by increasing the number for each record in .csv file.
Also to check the id you may use the number file ,type number in command line then put kmdocument in class field now enter ,you will get the one record note last number value and prefix and combined them like KM+list number = KM0023
So according to above generate and mention your id in .csv file.
Other way is to use the FC in import descriptor ,create a FC and mention it's name in Format Control field of import descriptor,
and in FC mentioned as follow
Subroutines tab ::
RAD Application: getnumb.fc
Parameter Names: record, name, prompt, and text
Parameter Values: $file, kmdocument, id, and string
08-13-2012 06:17 AM
We are now able to add the csv file. Also, we are getting the message that records have got updated.
But after full reindexing, the record is not being displayed on searching in search knowlegdebases.
Thanks in advance.
10-11-2012 12:35 PM
The supported method for importing records into the kmdocument table is using the KMImport Utility and importing xml files of the documents. Now, that is not to say that you cannot try to use an import descriptor record with a .csv file. However, what you might run across by doing that is importing special characters that the indexer or search engine might reject.
In your last update you said that records needed to be indexed. Anytime you create, import or update knowledge a record is created in kmknowledgebaseupdates for that document. If you ran a full index how many records were submitted for indexing and how many were actually indexed? Are the ones that were imported missing from the results set while the manually added/created documents are in the results set?
Brett Alan Christlieb
Level 3 Technical Support Engineer
HP Software Support Team
If you find that this or any post resolves your issue please be sure to mark is as an accepted solution.