Import user-defined keywords

[Taxi]

According to the Announcement of Version 3.0.3, one of the Enhancements included was

* Support CSV import to user-defined keywords.

I have been unable to find any additional information about this facility in the online manual or these forums.

Can anyone provide some such additional information to explain how a list of user-defined keywords can be imported into a URD file?

Please!

[quant]

hmmm, interesting, I gave it a try, there is "Keyword (user-defined)" attribute in the import section, but no explanation how the user keywords should be delimited, so I was able to import only one user keyword per item ...

I tried to reverse-engineer, ie to export to CSV, but as opposed to XML export, there is no option to choose either auto- or user-defined keywords for CSV exporting ...

[ashwken]

In the CSV file you are Importing you can have a field that contains a series of delimited values. I don't think the function is particular about the delimiter, but my original feature request was based on an example that used the semi-colon:

Banking;Finance;Mortgage;Asset Managment

might be a good idea to stay away from using the comma as a delimiter in case text fields are not enclosed in quotes by the program generating the CSV.

As quant mentioned, when doing the Import to UR, map this field from the CSV to the UR field - Keyword (user-defined). It really doesn't matter what the field is called in the CSV since the Import mapping handles the assignment of the incoming values.

Later,
KenA

[quant]

Originally posted by ashwken
... , but my original feature request was based on an example that used the semi-colon:

Banking;Finance;Mortgage;Asset Managment

heh, so one has to use <TAB> or , to delimit field names, because ; is restricted to delimit user keywords. That works! This is funny (I mean crap), user should read user requests to find the way to use UR ...

First, there should definitely be a note about this and second there should be an option to choose the way to delimit user keywords. Without both of these, this feature is good just for the user who requested it! Disappointed

[ashwken]

Originally posted by quant
heh, so one has to use <TAB> or , to delimit field names, because ; is restricted to delimit user keywords. That works! This is funny (I mean crap), user should read user requests to find the way to use UR ...

First, there should definitely be a note about this and second there should be an option to choose the way to delimit user keywords. Without both of these, this feature is good just for the user who requested it! Disappointed

From the Help File:
The data in the delimited file simply needs to be separated with a consistent character (by default a comma [,] but any symbol such as a TAB or ~ could be specified and used).


Actually, I don't know that the keyword delimiter is restricted to the semi-colon, I suspect that a comma could be used if the field data were enclosed in quotes for a Comma Separated Value import file. If the import file were Tab delimited, I suspect that any delimiter could be used within the imported Keyword text string.

It's unfortunate that the Help File has not been updated to reflect this Import function; my original feature request was made during an email exchange to the support address.

Later,
KenA

[quant]

KenA,

when I tried to find out how this works, I had ; delimited data and tried space and , to delimit keywords. Didn't work. Then you replied that actually you requested ; to be used to delimit keywords and that worked. So I was unlucky that I have chosen ; to delimit fields cause it seems that it is hard-coded to use to delimit keywords.

I'm disappointed at the approach Kinook used in adding new feature. It shouldn't be that hard to the mature company and software like UR:

- add new feature
- test it enough to prevent obvious errors
- update help file (and it would be a good practice if any new feature has a simple example how to use it)
- update version list
- release it ...

[ashwken]

Originally posted by quant
when I tried to find out how this works, I had ; delimited data and tried space and , to delimit keywords. Didn't work. Then you replied that actually you requested ; to be used to delimit keywords and that worked. So I was unlucky that I have chosen ; to delimit fields cause it seems that it is hard-coded to use to delimit keywords.

Uhmm, this is interesting and disappointing.

In my original feature request I only used the semi-colon as an example to describe the delimited text string that should be importable, I never dreamed that it would be taken so literally.

Once the function became available I was pleased that it met my needs, but assumed that steps had been taken to make it more universally applicable.

As you mentioned in a post upstream, it would appear that an option needs to be available so that the user can declare what Keyword Delimiter has been used - just as we declare what delimiter is used for the import file.

Thank you for the correction and the insight,
KenA

[kinook]

The delimiter for keywords when importing is the semi-colon. We'll update the docs and consider offering other delimiter characters within this particular field.