.

go/zodb/storage/fs1/filestorage.go

@@ -462,7 +462,7 @@ func (fs *FileStorage) watcher(w *fsnotify.Watcher) {
 		}
 	}
 
-	// XXX fs.watchErr = err  (-> fail other operations)
+	// XXX fs.watchErr = err  (-> fail other operations)?
 	if err != nil {
 		log.Print(err)
 	}
@@ -479,7 +479,7 @@ func (fs *FileStorage) _watcher(w *fsnotify.Watcher) (err error) {
 
 	// loop checking f.size vs topPos
 	//
-	// besides relying on notify we also check file periodically to avoid
+	// besides relying on fsnotify we also check file periodically to avoid
 	// stalls due to e.g. OS notification errors.
 	tick := time.NewTicker(1*time.Second)
 	defer tick.Stop()
@@ -488,14 +488,12 @@ func (fs *FileStorage) _watcher(w *fsnotify.Watcher) (err error) {
 mainloop:
 	for {
 		if !first {
-			//tracef("select ...")
 			select {
 			case <-fs.down:
 				// closed
 				return nil
 
 			case err := <-w.Errors:
-				//tracef("error: %s", err)
 				if err != fsnotify.ErrEventOverflow {
 					return err
 				}
@@ -504,12 +502,12 @@ mainloop:
 			case <-w.Events:
 				// we got some kind of "file was modified" event (e.g.
 				// write, truncate, chown ...) -> it is time to check the file again.
-				//tracef("event: %s", e)
 
 			case <-tick.C:
 				// recheck the file periodically.
-				//tracef("tick")
 			}
+
+			// XXX dequeue everything from w?
 		}
 		first = false
 
@@ -519,7 +517,6 @@ mainloop:
 			return err
 		}
 		fsize := fi.Size()
-		//tracef("toppos: %d\tfsize: %d\n", idx.TopPos, fsize)
 		switch {
 		case fsize == idx.TopPos:
 			continue // same as before
@@ -529,15 +526,14 @@ mainloop:
 		}
 
 		// there is some data after toppos - try to advance as much as we can.
-		// start iterating afresh with empty buffer.
-		//tracef("scanning ...")
+		// start iterating afresh with new empty buffer.
 		it := Iterate(seqReadAt(f), idx.TopPos, IterForward)
 		for {
 			err = it.NextTxn(LoadNoStrings)
 			if err != nil {
 				// transaction header could not be fully read.
 				//
-				// Even though FileStorage code always calls write with full txn
+				// even though FileStorage code always calls write with full txn
 				// header, the kernel could do the write in parts, e.g. if written
 				// region overlaps page boundary.
 				//
@@ -568,19 +564,17 @@ mainloop:
 			// read ok - reset t₀(partial)
 			t0partial = time.Time{}
 
-			//tracef("@%d tid=%s st=%q", it.Txnh.Pos, it.Txnh.Tid, it.Txnh.Status)
-
 			// XXX dup wrt Index.Update
 
 			// we could successfully read the transaction header. Try to see now,
 			// whether it is finished transaction or not.
 			if it.Txnh.Status == zodb.TxnInprogress {
-				// not yet. we have to resync because transaction finish writes
+				// not yet. We have to resync because transaction finish writes
 				// to what we have already buffered.
 				continue mainloop
 			}
 
-			// it is fully-committed transaction. scan its data records to update
+			// it is fully-committed transaction. Scan its data records to update
 			// our index & notify watchers. There is no expected errors here.
 			var oidv []zodb.Oid
 			update := map[zodb.Oid]int64{}
@@ -610,8 +604,7 @@ mainloop:
 			}
 			fs.mu.Unlock()
 
-			//tracef("-> tid=%s  δoidv=%v", it.Txnh.Tid, oidv)
-
+			// notify client
 			if fs.watchq != nil {
 				select {
 				case <-fs.down: