Last modified: 2012-04-12 13:55:53 UTC
Hello, Please try the following query: http://en.wikipedia.org/w/api.php?action=query&list=usercontribs&ucuser=Catrope&uclimit=3&ucdir=newer The results is: <?xml version="1.0"?> <api> <query> <usercontribs> <item user="Catrope" pageid="11650099" revid="136629050" ns="3" title="User talk:Catrope" timestamp="2007-06-07T16:45:30Z" minor="" comment="Creation; directing to BW" size="119" /> <item user="Catrope" pageid="11650100" revid="136629120" ns="2" title="User:Catrope" timestamp="2007-06-07T16:45:49Z" minor="" comment="Creation; directing to BW" size="109" /> <item user="Catrope" pageid="3383874" revid="136629407" ns="3" title="User talk:Ruud Koot" timestamp="2007-06-07T16:47:09Z" comment="Nested collapsible tables fix" size="70939" /> </usercontribs> </query> <query-continue> <usercontribs ucstart="2007-06-11T09:23:57Z" /> </query-continue> </api> Three results are returned (as asked) but the API indicates that there is a continuation, which should not happen because all 3 results that where asked for were returned. Thanks, Nakor
(In reply to comment #0) > Three results are returned (as asked) but the API indicates that there is a > continuation, which should not happen because all 3 results that where asked > for were returned. > You're misunderstanding what query-continue means. It means that there are more results, regardless of whether you asked for them.
It used to work that way. A recent change broke the algorithm tat was in place in pywikipedia. Are we then guaranteed that if we ask for any uclimit, all those entries will be back in the first query?
(In reply to comment #2) > It used to work that way. A recent change broke the algorithm tat was in place > in pywikipedia. > > Are we then guaranteed that if we ask for any uclimit, all those entries will > be back in the first query? > No. You have to count the results yourself to see if you're missing anything.