从php中的couchdb清除已删除的文件

Question

I am trying to write an automated script to run through a couchdb, find the deleted documents (About 100,000 doc deletetions a month) and purge them and their revisions.

I haven't found documentation explaining how to get all deleted documents, all their revisions, and construct them into the POST request shown here.

http://docs.couchdb.org/en/stable/api/database/misc.html

How do I construct the views, get the data, and create the POST to do this?

Thanks

Answer 1

So I ended up making a nodejs script for it that runs periodically. Some things are hard coded since this was for testing, and not production (yet). Anyone referencing this should watch for hard coded references. I used pouchdb to handle the replication, as it was able to run all 400,000 tombstones in a few minutes, whereas I never actually got couchdb to finish replicating after a full day of running. This is my first node script ever, so I apologize if it's awful.

var PouchDB = require('pouchdb');
var fs = require('fs');
var child = require('child_process');

var originalDB = "dbName";
var tempDB = "dbName_clean";
var serviceName = "Apache CouchDB";
var couchDBDataPath = "C:\\bin\\couchdb\\couchdb.2.1.1\\data\\";

var db = new PouchDB('http://URL/' + originalDB);
var db2 = new PouchDB('http://URL/' + tempDB);

console.log("Compacting");
return db.compact().then(function (result) {
    console.log("Compacting done, replicating");
    var batch_size = 100; 

    return db2.destroy().then(function () {
        db2 = new PouchDB('http://URL/' + tempDB);
        return db.replicate.to(db2, {
            batch_size: batch_size,
            filter: function (doc, req) {
                return !doc._deleted;
            }
        }).on('change', function (info) {
            console.log("batch done");
        }).on('complete', function () {
            console.log('complete');
        }).on('paused', function (err) {
            // replication paused (e.g. replication up to date, user went offline)
            console.log("paused", err);
        }).on('active', function () {
            // replicate resumed (e.g. new changes replicating, user went back online)
            console.log("Active");
        }).on('denied', function (err) {
            // a document failed to replicate (e.g. due to permissions)
            console.log("Denied", err);
        }).on('error', function (err) {
            // handle error
            console.log("error", err);
            // reject(err);
        }).then(function () {
            //replicate done, rename everything
            var date = new Date();
            console.log("Stopping service");
            child.execSync('net stop "' + serviceName + '"');
            console.log("Service Stopped");
            var newName;
            var counter = 0;
            do {
                newName = "_" + date.getFullYear() + date.getMonth() + date.getDay() + "_" + counter;
                counter++;
            } while (fs.existsSync(couchDBDataPath + originalDB + newName + ".couch") || fs.existsSync(couchDBDataPath + "." + originalDB + "_design" + newName));


            console.log("Renaming original couch to backup labeled", originalDB + newName);
            fs.renameSync(couchDBDataPath + originalDB + ".couch", couchDBDataPath + originalDB + newName + ".couch");
            fs.renameSync(couchDBDataPath + "." + originalDB + "_design", couchDBDataPath + "." + originalDB + newName + "_design");

            console.log("Renaming clean couch back to original", tempDB);
            fs.renameSync(couchDBDataPath + tempDB + ".couch", couchDBDataPath + originalDB + ".couch");
            fs.mkdirSync(couchDBDataPath + "." + originalDB + "_design"); 

            console.log("Starting service");
            child.execSync('net start "' + serviceName + '"');
            console.log("Service started");
        });
    });
}).catch(function (err) {
    console.log(err);
});