Best way to get all issues for large project

[Anonymouslemming]

Hi all,

As part of trying to resolve https://answers.atlassian.com/questions/28870803, I am trying to get a list of all issues in a project in a plugin. This will then be used by the client to delete all issues.

The first step is to get all issues, but I see that getProjectIssues is deprecated at https://developer.atlassian.com/static/javadoc/jira/6.4.1/reference/com/atlassian/jira/issue/IssueManager.html#getProjectIssues(org.ofbiz.core.entity.GenericValue)

That page mentions that I should be using a search provider instead, but does not link to what that is or an example of that.

What I'm trying to do is get all issues for a project, identify which are subtasks and delete those, then delete all remaining tasks, then delete the project. 

Any input into the best way to get all issues for large projects (20,000+ issues) would be greatly appreciated!

[Tomasz Stec · Accepted solution]

Hi,

Try JQL... process all results at once or implement your own way of handling numerous results using for instance PageFilter:

SearchProvider searchProvider = ComponentAccessor.getComponentOfType(SearchProvider.class);
List<Issue> fetchedIssues;
JqlQueryBuilder builder = JqlQueryBuilder.newBuilder();   builder.where().issueType().eq("Sub-Task").and().project(projectId);        
        try {
            final SearchResults results = searchProvider.searchOverrideSecurity(builder.buildQuery(), user, PagerFilter.getUnlimitedFilter() , null);
// if you have huge amount of issues try assemble something using PageFilter.newPageAlignedFilter(0, 100)
// then process it page by page.
            fetchedIssues = results.getIssues();
        } catch (SearchException e1) {
            e1.printStackTrace();
        }

[Anonymouslemming]

Thanks - looks good. That'll be a lot of iterations for 2,000 issues, but at least it keeps me under the barrier for large search results in JQL. Thanks again,

[Jeff Louwerse]

I do the same thing since I have had the internal delete project fail on me way too many times and have then had to go though and clean it up. 

If you want a pure REST solution, use JQL.

/rest/api/2/search?jql=project=ZAC&maxResults=2000&fields=key

Personally I use good old SQL (oracle example below) so I am not limited to JIRA timeouts, max results or anything jira. 

SELECT ID 
FROM JIRAISSUE JI, PROJECT P 
WHERE JI.PROJECT = P.ID and P.PKEY = '<YOUR KEY HERE>'

Also note that you don't have to delete subtasks before deleting an issue; Just add "?deleteSubtasks=true" to your REST URL

[Bob Swift OSS (Bob Swift Atlassian Apps)]

You can use JIRA Command Line Interface (CLI).

1. runFromIssueList using a project or a JQL selection - see How to use runFromIssueList
2. deleteIssue will handle the delete

[Anonymouslemming]

Unfortunately, Jira CLI is not suitable in our environment for a number of reasons.