pingcap/tidb

Prefer upstream unless you specifically need a 2019 TiDB snapshot. This fork is far behind and looks unsuitable for production adoption except as a legacy compatibility or debugging baseline.

Prefer this fork only if you want a near-upstream TiDB mirror. If you need new capabilities or the newest fixes, upstream is the better choice.

Prefer this fork only if you need Shopify-specific TiDB behavior, test tooling, or operational controls and are prepared to maintain a large upstream gap. If you want the broadest compatibility, fastest access to bug fixes, and lowest long-term maintenance cost, upstream TiDB is the safer default.

Choose this fork only if you specifically need its custom testing and internal workflow additions and are prepared to own a large, stale divergence from upstream. For general production use, upstream TiDB is the safer default.

Prefer upstream unless you specifically need this fork's testing/tooling customizations and can tolerate major drift. For most adopters, the staleness and divergence outweigh the added workflows.

Prefer this fork only if you need its custom local changes and can absorb long-term merge debt. For most adopters, upstream TiDB is the better choice because this fork is far behind and operationally risky.

Prefer the upstream TiDB unless you specifically need the fork’s custom parser/binlog/metrics/workflow changes and are willing to own a stale, highly divergent codebase.

Prefer upstream TiDB unless you specifically need this fork’s old local experiments or test scaffolding. The fork is materially stale and highly divergent, so adoption is only sensible for users willing to own long-term maintenance and restore missing upstream capabilities.

Prefer this fork only if you need its specific 2019-era customizations and testing workflow. For most adopters, upstream TiDB is the better choice because this fork is stale and substantially behind.