问题
I am looking for pseudo code or the actual code in the source repository (https://github.com/apache/couchdb) defining how CouchDB calculates the seq value.
Reviewing the technical overview in the docs: http://docs.couchdb.org/en/2.2.0/intro/overview.html it says:
Each update to a database instance generates a new sequential number.
But it doesn't go into how. I think that it may be implementation dependent, looking at the replication protocol docs it states:
Sequence ID: An ID provided by the Changes Feed. It MUST be incremental, but MAY NOT always be an integer.
1-g1AAAAIreJyVkEsKwjAURZ-toI5cgq5A0sQ0OrI70XyppcaRY92J7kR3ojupaSPUUgotgRd4yTlwbw4A0zRUMLdnpaMkwmyF3Ily9xBwEIuiKLI05KOTW0wkV4rruP29UyGWbordzwKVxWBNOGMKZhertDlarbr5pOT3DV4gudUC9-MPJX9tpEAYx4TQASns2E24ucuJ7rXJSL1BbEgf3vTwpmedCZkYa7Pulck7Xt7x_usFU2aIHOD4eEfVTVA5KMGUkqhNZV-8_o5i
This questions is somewhat related in that it's asking how it calculates another component of a document, the rev:
How does CouchDB calculate the Revision number
回答1:
You can find here some information about the change sequence internals
The sequence number reflects the cluster state for a document update sequence, this information is encoded into the sequence number.
The numeric part of the sequence number is the sum of the update sequences of a document in each of the cluster nodes.
The second part of the sequence number is a base64 encoded string with the update sequence information in each of the cluster nodes.
来源：https://stackoverflow.com/questions/53962975/how-does-couchdb-calculate-the-sequence-id-seq