Here is a patch that enables exporting csv with actual names of linked items instead of their ids. It adds a new action, "export_csv_names" to the "extensions" directory in your tracker.

The action works like export_csv except on columns with the Link type. The action tries to find a column called name in the linked class, or username if the class is user. If it is found, it prints this column instead of the id. Simply change the lines in the desired html pages from export_csv to export_csv_names, copy this file to the extensions directory, and that's it.

This action has been evaluated on Roundup 0.8.2 only, and was updated from the original version that ran under 0.7.

[BEWARE: get the patch from the wiki page source, not from the html display, since the wiki software removes quotes. Any suggestions of a better way to put this patch online are welcome.]

(cut here)----ExportCSVNamesAction.py--------------------
from roundup.cgi.actions import Action
from roundup.cgi import templating
from roundup import hyperdb

import csv
class ExportCSVNamesAction(Action):
    name = export
    permissionType = View
    def handle(self):
        ''' Export the specified search query as CSV. '''
        # figure the request
        request = templating.HTMLRequest(self.client)
        filterspec = request.filterspec
        sort = request.sort
        group = request.group
        columns = request.columns
        klass = self.db.getclass(request.classname)
        # full-text search
        if request.search_text:
            matches = self.db.indexer.search(
                re.findall(r'\b\w{2,25}\b', request.search_text), klass)
        else:
            matches = None
        h = self.client.additional_headers
        h['Content-Type'] = text/csv
        # some browsers will honor the filename here...
        h['Content-Disposition'] = inline; filename=query.csv
        self.client.header()
        if self.client.env['REQUEST_METHOD'] == 'HEAD':
            # all done, return a dummy string
            return dummy
        writer = csv.writer(self.client.request.wfile)
        writer.writerow(columns)
        # Figure out Link columns
        represent = {}
        def repr_link(cls,col):
            def f(x):
                if x==None:
                    return ""
                else:
                    return str(cls.get(x,col))
            return f
        props = klass.getprops()
        for col in columns:
            represent[col] = str
            if isinstance(props[col], hyperdb.Link):
                cn = props[col].classname
                cl = self.db.getclass(cn)
                if cl.getprops().has_key(name):
                    represent[col] = repr_link(cl, name)
                elif cn == 'user':
                    represent[col] = repr_link(cl, username)
        # and search
        for itemid in klass.filter(matches, filterspec, sort, group):
            writer.writerow([represent[col](klass.get(itemid, col)) for col in columns])
        return \n


def init(instance):
          instance.registerAction(export_csv_names, ExportCSVNamesAction)
# vim: set filetype=python sts=4 sw=4 et si
(cut here)----ExportCSVNamesAction.py--------------------