How to map fields in JIRA during a transition?

[Jacek Bregulla]

I have created a workflow in JIRA, where in the first step users can fill out a screen and then assign it to someone. Now I would like the values inserted by the users to be mapped, when he sends them to the assignee. Desired outcome would be: the user fills out a screen with certain values and the assignee receives the same screen, but with other mapped values. Background is, that users are familiar with logical names, but the assignees need cryptical technical names for their analysis. I assume in order to accomplish that some kind of script will need to be written. It would be hugely helpful, if someone could tell me which plugin to use and maybe how to write it.

I agree, it would be easier to have only one set of names. However this convention has grown historically and I won't be able to change that, therefore I have to deal with what is.

Currently we have an Excel with two columns, one with the logical names and the other one with the technical names. The create issue screen in JIRA consists of fields, where users can choose a value (only the logical names are displayed) via a dropbox for each field. Now these fields should be mapped as in the Excel, i.e. IF value of field X is Y, THEN field X receives Z and so on.

The language pack solution won't work, because I had to create custom fields. I have also thought about enhancing the name of the fields with the technical name, but that is only a backup solution, I would prefer to solve it differently. Therefore what do I need to do to write a plugin, are there any examples of the code I could use? Secondly regarding the duplication of data. What exactly would be duplicated and how can I do that?

[CL]

My assumption: If you talk about "values" it seems to be "predefined values" (not terms within free text). I.e. you would use kind of select fields for the data rather than text fields.

What about combining both terms in each option of the selection? "user term - technical term"
With Nic's example: "Penguin - MrFlibble"

By that both sides would also learn about the terms of the other side. And may be in future you can reduce to 1 terminology.

If you talk about "terms within free text" you will probably need more than just a "mapping".

[Jacek Bregulla]

I agree, it would be easier to have only one set of names. However this convention has grown historically and I won't be able to change that, therefore I have to deal with what is.

Currently we have an Excel with two columns, one with the logical names and the other one with the technical names. The create issue screen in JIRA consists of fields, where users can choose a value (only the logical names are displayed) via a dropbox for each field. Now these fields should be mapped as in the Excel, i.e. IF value of field X is Y, THEN field X receives Z and so on.

The language pack solution won't work, because I had to create custom fields. I have also thought about enhancing the name of the fields with the technical name, but that is only a backup solution, I would prefer to solve it differently. Therefore what do I need to do to write a plugin, are there any examples of the code I could use? Secondly regarding the duplication of data. What exactly would be duplicated and how can I do that?

[Nic Brough -Adaptavist-]

I'm not sure how you expect this "mapping" to work, unless you're writing an external screen, or a new "create" screen for Jira as a plugin. Could you explain how you are expecting this to work?

On the human side though, this sounds like a bad idea - two sets of users using different language to talk about the same thing is asking for trouble. I can't count the number of times I've been in situations where people using different words for the same thing has caused problems. I'm tempted to tell your assignees to stop using "cryptical technical names" for their analysis and tell them to do the mapping themselves. The worst thing about this is that your user who created the issue is going to be confounded by it - "I entered a field called Penguin, and I can't find it and your analyst keeps talking about MrFlibble" - it's an instant break in human terms.

I can't work out an easy solution either. You could think about language packs - if real users are happy with English, have a "translation" for your analysis - but that only works for system fields at the moment (you can't translate custom field names). You could add a description to every English field with the crypic name, that would at least tell your analysts where the fields are when they update it. You might think about duplicating the data, but that's ugly because updating it means having to write listeners that can capture edits, and you'll have two sets of data to work with all the time.