1 Reply Latest reply on Apr 8, 2015 5:27 PM by Mussie Zemikael

    AIE Import***Error: Number of tokens exceed the number of columns***

      Share This:

      Hi Guys,
      We have a 7.604 environment, I am trying to import people records using AIE (AI is not installed) from a flatfile (CSV) to a staging form, then using some work-flows import it to CTM:People. For that I created a Data Exchange and AR System Mapping entry. Please have a look the screenshots I have attached.

      However, I am facing few issues:

      1. I am getting an error message “***Error: Number of tokens exceed the number of columns***” (check the attached debug log). In order to investigate the cause of this issue, I have checked everything I possibly can, field mappings etc… and I concluded that this is caused by records with blank values. And indeed when I removed these blank records the import went well. My understanding is since my primary is CorporateID (have a look at the screenshot), then apart from this field any other column can have blank values. Is my understanding correct? How can make the AIE import to run smoothly for the records with blank values?
      2. The second question I have is, every time I run the AIE, it updates all the records whether the records are changed or not and as a result it changes the modified date to the current date (which unnecessarily causes my work-flows to be fired for the records that haven’t been changed). In AR System mappings, since I selected “Update existing records” to Yes and “Create a record” to “If no duplicate exist”, why is it updating each and every record. My requirement is only the records which are changed need to be updated, how can I achieve that?

      Thank you,

      Mussie

      AR_Form_Mapping_1.png

      AR_Form_Mapping_2.png

      AR_Form_Mapping_3.png

      AR_Form_Mapping_6.png

      Data_Exchange_1.png

      Data_Exchange_2.png

      Data_Exchange_3.png

        • 1. Re: AIE Import***Error: Number of tokens exceed the number of columns***

          I discovered that this was the problem with the CSV format. Even though I took it from a spreadsheet, when I saved it as .csv for some reason, it didn't have comma at the end of each record. This was resolved by using another csv file.

          Also, I fixed the issue where all the records were getting modified instead of just the ones which are changed. I used a filter which sets a flag from No to Yes based on the below condition:

          ('TR.City' != 'DB.City') OR ('TR.Company' != 'DB.Company') OR ('TR.Country' != 'DB.Country') OR ('TR.Dept_Name_Long' != 'DB.Dept_Name_Long') OR ('TR.Email Address' != 'DB.Email Address') OR ('TR.First Name' != 'DB.First Name') OR ('TR.Job Title' != 'DB.Job Title') OR ('TR.Last Name' != 'DB.Last Name') OR ('TR.Mobile Number' != 'DB.Mobile Number') OR ('TR.Organization' != 'DB.Organization') OR ('TR.Phone Number' != 'DB.Phone Number') OR ('TR.Site_Name_Long' != 'DB.Site_Name_Long') OR ('TR.Username' != 'DB.Username')

           

          Regards,

          Mussie