CouchDB equivalent of Sql NOT IN?

半世苍凉 提交于 2019-12-04 06:52:44

You can use the view collation to join feeds and data sources in map:

function(doc) {
  if (!doc.type) return;
  if (doc.type == "feed") emit(doc._id, null);
  if (doc.type == "ds" && doc.feed) emit(doc.feed, null);
}

and reduce to filter those feed ids which have data source documents linking to them. Eg. use of the build-in _count and query with group_level:

http://127.0.0.1:5984/test/_design/join/_view/not_in?group_level=1

for database:

{"id":"1", "type":"feed"}
{"id":"2", "type":"feed"}
{"id":"3", "type":"ds", "feed":1}
{"id":"4", "type":"ds", "feed":1}}

will give you:

{"rows":[
{"key":"1","value":3},
{"key":"2","value":1}
]}

Values >1 are those feed docs which have reference from data sources. To get pure feed list w/o datasources you can omit records with value>1 in client or list function.

Edit: With list function:

function(head, req) {
  var row;
  while (row = getRow()) {
    if (row.value == 1)
      send(row.key + "\n");
  }
}

and query:

http://127.0.0.1:5984/test/_design/join/_list/not_ds/not_in?group_level=1

You will get the final result with feed documents with out reference from data sources. It is plaint text with list of ids, you can also format it for JSON array.

The straight answer to CouchDB equivalent of Sql NOT IN is: There is no equivalent. There is no querying option in the couchdb view API for NOT IN.

However, there may be alternative ways to achieve your objectives depending if you are interested in getting feeds associated to no datasource or not associated to a specific datasource.

I know this is an old post, but having been confronted with the same problem I think that Marcin's answer is correct but incomplete.

In my case I used the following list function to select get all rows except the ones with a certain excludeId:

function(head, req) {
start({ 'headers': { 'Content-Type': 'application/json' } });
 var rows = [];
 while (row = getRow()) {
    if (!req.query.exceptId || doc.documentId != req.query.exceptId) {
        rows.push(row);
    }
 }
 send(JSON.stringify({ total_rows: rows.length, rows: rows }));
}

you can read more here

易学教程内所有资源均来自网络或用户发布的内容,如有违反法律规定的内容欢迎反馈
该文章没有解决你所遇到的问题?点击提问,说说你的问题,让更多的人一起探讨吧!