How to get the rank with Jira API

[JORDAN JONES]

I am looking to get the rank(The order of items in the backlog for example) from my jira board, and read it from the API. We used to be able to do this with a customfield but now we can not. Is there any way to get this custom field back? Or is there another way to do it?

[bschmuhl]

I found that value in customfield_10019 and appears to be sortable alphanumerically. You won't be able to derive a human-readable number unless you can compare it to the context of what you're speaking to. If you have that set of issues, you should be able to rank them, and then number them. If I understand correctly, the rank is global—meaning it still applies after an issue has been resolved, and whether or not its in a sprint. This makes a single rank number pretty meaningless as discussed here: https://community.atlassian.com/t5/Jira-questions/How-to-see-numeric-rank-order-for-an-issue/qaq-p/611802

[MattS]

Rank is a string now, not an obvious number. Will that be useful?

[JORDAN JONES]

I found the string, but it is really weird. I get strings like x|xxxxx: how do I decode these strings to usable numbers?

[Mikhail Kurskiy]

@JORDAN JONES, how did you done it?
Which method? - I could not find it in API documentation :(

[Jared Kells]

The string is sortable, so while it isn't a number if you lexicographically sort based on the string you will get the correct order.

The problem with using numbers for a global rank is that when one issue is changed every other issue ranked below it would need its rank updated. You can imagine how expensive this would be when you have hundreds of thousands of issues.

Jira uses the LexoRank algorithm, you can learn a bit about it here: https://confluence.atlassian.com/adminjiraserver/managing-lexorank-938847803.html

It's a cool algorithm because given any two strings you can always find a new string that's ranked between them so long as you're happy for the string length to increase.

For example given the strings "a" and "b" you can add "aa "ab" "ac" and they will all slot in between "a" and "b"!

Jira then periodically re-balances the field as a background job when the strings get too long.