parasitepool/para

Track user's participated blockcount #306

Merged

parabitdev merged 8 commits from parabit/track-user-participated-blocks into master

2025-12-11 22:09:35 +00:00

parabitdev commented

2025-12-11 19:32:40 +00:00

(Migrated from github.com)

closes #302
- update account migration to recalculate participated blocks
- update share sync to increment blockcount
- come up with plan to handle deduplication
- add tests

- [x] closes #302 - [x] update account migration to recalculate participated blocks - [x] update share sync to increment blockcount - [x] come up with plan to handle deduplication - [x] add tests

parabitdev (Migrated from github.com) reviewed

2025-12-11 20:45:27 +00:00

bin/postgres-init

					
				@ -387,3 +354,1 @@

				                                 ELSE accounts.lnurl_updated_at

				              END,

				          updated_at       = NOW();

				      INSERT INTO accounts (username, lnurl, past_lnurls, total_diff, created_at, updated_at)

Only the final part (referencing account_metadata) is changed, rest just had indentation fixed.

Only the final part (referencing account_metadata) is changed, rest just had indentation fixed.

parabitdev (Migrated from github.com) reviewed

2025-12-11 20:46:53 +00:00

src/subcommand/server.rs

					
				@ -804,0 +802,4 @@

				                if share.result == Some(true) {

				                    if let Some(diff) = share.diff {

				                        entry.total_diff += diff;

				                    }

Typically, this will only be a single block, but there is an edge-case (one of the two causes of duplicate counts) where we send records from more than one block.

Typically, this will only be a single block, but there is an edge-case (one of the two causes of duplicate counts) where we send records from more than one block.

parabitdev (Migrated from github.com) reviewed

2025-12-11 21:57:45 +00:00

src/subcommand/server.rs

Two is better as we get an explicit per-query error. If we were performance constrained then this may be an area for some improvement.

Two is better as we get an explicit per-query error. If we were performance constrained then this may be an area for some improvement.

parabitdev (Migrated from github.com) reviewed

2025-12-11 22:00:14 +00:00

src/subcommand/server.rs

The sending side sends blocks in order. It is possible for blocks to come in out of order, but would require a server to fall behing on sync while a reorg was happening for it to impact this. The effect would be that all users would be missing one point (which can be corrected by running the migration/refresh).

The sending side sends blocks in order. It is possible for blocks to come in out of order, but would require a server to fall behing on sync while a reorg was happening for it to impact this. The effect would be that all users would be missing one point (which can be corrected by running the migration/refresh).

parabitdev (Migrated from github.com) reviewed

2025-12-11 22:06:07 +00:00

src/subcommand/server.rs

					
				@ -844,6 +849,39 @@ impl Server {

				                    .execute(&mut *tx)

Yes we can 😂

Yes we can :joy:

Sign in to join this conversation.

No reviewers

No labels

good first issue

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

parasitepool/para!306

No description provided.