mirror of
https://github.com/hanzoai/s3.git
synced 2026-11-03 03:07:41 +00:00
zooqueen
dd74d3e6d6
hanzoai/s3 was an AGPL-3.0 MinIO fork whose network-copyleft blocks embedding
and reselling as a managed service. Replace it wholesale with a fork of
SeaweedFS (github.com/seaweedfs/seaweedfs, Apache-2.0, Go) which ships a
built-in S3 gateway + Filer.
- License: AGPL-3.0 -> Apache-2.0 (upstream LICENSE kept verbatim, incl.
"Copyright 2025 Chris Lu"); NOTICE added crediting SeaweedFS. No AGPL/BUSL/
SSPL/GPL or enterprise ("ee") files remain.
- Module: go.mod path set to github.com/hanzoai/s3 with a self-replace
(replace github.com/seaweedfs/seaweedfs => ./) so the ~1800 internal Go files
keep their upstream import paths with zero churn. Internal identifiers and
import paths are intentionally NOT renamed.
- CLI rebrand: the binary is built as `s3` (was `weed`); root usage, per-command
help/usage/flag strings, shell autocomplete program name, and the product name
in CLI output now read "Hanzo S3" / `s3`. Users never see `weed`.
(e.g. `s3 server`, `s3 filer`, `s3 master`, `s3 volume`, `s3 s3`, `s3 mount`.)
- Build: `make` (or `go build -o s3 ./weed`) builds the `s3` binary green;
`cmd/weed-db` and `cmd/weed-sql` also build; `go vet ./weed/command/` clean.
- README/LLM.md rewritten for Hanzo S3; AGENTS.md/CLAUDE.md symlinks preserved.
- Dropped a 33MB prebuilt Mach-O `stress` binary upstream had committed; `.gitignore`
updated to ignore the `s3` build output.
51 lines
2.1 KiB
Go
51 lines
2.1 KiB
Go
package logstore
|
|
|
|
import (
|
|
"github.com/seaweedfs/seaweedfs/weed/mq/topic"
|
|
"github.com/seaweedfs/seaweedfs/weed/pb/filer_pb"
|
|
"github.com/seaweedfs/seaweedfs/weed/util/log_buffer"
|
|
)
|
|
|
|
func GenMergedReadFunc(filerClient filer_pb.FilerClient, t topic.Topic, p topic.Partition) log_buffer.LogReadFromDiskFuncType {
|
|
fromParquetFn := GenParquetReadFunc(filerClient, t, p)
|
|
readLogDirectFn := GenLogOnDiskReadFunc(filerClient, t, p)
|
|
// Reversed order: live logs first (recent), then Parquet files (historical)
|
|
// This provides better performance for real-time analytics queries
|
|
return mergeReadFuncs(readLogDirectFn, fromParquetFn)
|
|
}
|
|
|
|
func mergeReadFuncs(readLogDirectFn, fromParquetFn log_buffer.LogReadFromDiskFuncType) log_buffer.LogReadFromDiskFuncType {
|
|
// CRITICAL FIX: Removed stateful closure variables (exhaustedLiveLogs, lastProcessedPosition)
|
|
// These caused the function to skip disk reads on subsequent calls, leading to
|
|
// Schema Registry timeout when data was flushed after the first read attempt.
|
|
// The function must be stateless and check for data on EVERY call.
|
|
return func(startPosition log_buffer.MessagePosition, stopTsNs int64, eachLogEntryFn log_buffer.EachLogEntryFuncType) (lastReadPosition log_buffer.MessagePosition, isDone bool, err error) {
|
|
// Always try reading from live logs first (recent data)
|
|
lastReadPosition, isDone, err = readLogDirectFn(startPosition, stopTsNs, eachLogEntryFn)
|
|
if isDone {
|
|
// For very early timestamps (like timestamp=1 for RESET_TO_EARLIEST),
|
|
// we want to continue to read from in-memory data
|
|
isDone = false
|
|
}
|
|
if err != nil {
|
|
return
|
|
}
|
|
|
|
// If live logs returned data, update startPosition for parquet read
|
|
if lastReadPosition.Offset > startPosition.Offset || lastReadPosition.Time.After(startPosition.Time) {
|
|
startPosition = lastReadPosition
|
|
}
|
|
|
|
// Then try reading from Parquet files (historical data)
|
|
lastReadPosition, isDone, err = fromParquetFn(startPosition, stopTsNs, eachLogEntryFn)
|
|
|
|
if isDone {
|
|
// For very early timestamps (like timestamp=1 for RESET_TO_EARLIEST),
|
|
// parquet files won't exist, but we want to continue to in-memory data reading
|
|
isDone = false
|
|
}
|
|
|
|
return
|
|
}
|
|
}
|