[claude] fix(benchmarks-server): include commit_sha in measurement_id hash (#7642)

Without commit_sha in the hash input, every (dim tuple) collapses to one
row across commits via INSERT ... ON CONFLICT DO UPDATE, so the chart
pages render at most one point per series. Adding commit_sha to the
per-table hashers makes each (commit, dim) pair its own row, which is
the time series the UI is built around. Re-emission of the same (commit,
dim) is still the upsert case.

The web-ui chart_page_query snapshot now correctly shows three commits
with three points per series, matching the test fixture.

No public API change; measurement_id is server-internal.

<!--
Thank you for submitting a pull request! We appreciate your time and
effort.

Please make sure to provide enough information so that we can review
your pull
request. The Summary and Testing sections below contain guidance on what
to
include.
-->

## Summary

<!--
If this PR is related to a tracked effort, please link to the relevant
issue
here (e.g., `Closes: #123`). Otherwise, feel free to ignore / delete
this.

In this section, please:

1. Explain the rationale for this change.
2. Summarize the changes included in this PR.

A general rule of thumb is that larger PRs should have larger summaries.
If
there are a lot of changes, please help us review the code by explaining
what
was changed and why.

If there is an issue or discussion attached, there is no need to
duplicate all
the details, but clarity is always preferred over brevity.
-->

Closes: #000

<!--
## API Changes

Uncomment this section if there are any user-facing changes.

Consider whether the change affects users in one of the following ways:

1. Breaks public APIs in some way.
2. Changes the underlying behavior of one of the engine integrations.
3. Should some documentation be updated to reflect this change?

If a public API is changed in a breaking manner, make sure to add the
appropriate label. You can run `./scripts/public-api.sh` locally to see
if there
are any public API changes (and this also runs in our CI).
-->

## Testing

<!--
Please describe how this change was tested. Here are some common
categories for
testing in Vortex:

1. Verifying existing behavior is maintained.
2. Verifying new behavior and functionality works correctly.
3. Serialization compatibility (backwards and forwards) should be
maintained or
   explicitly broken.
-->

Signed-off-by: Claude <noreply@anthropic.com>
Co-authored-by: Claude <noreply@anthropic.com>
Signed-off-by: Connor Tsui <connor.tsui20@gmail.com>

Author: connortsui20

benchmarks-website/planning/01-schema.md

@@ -21,8 +21,11 @@ migration framework; those are deferred (see
    columns. Adding a nullable column is cheap; the readability win
    is worth it.
 4. **Hashed primary key per table.** Each fact table has a
-   `measurement_id` that is a deterministic 64-bit hash of that
-   table's dimensional tuple. Server-internal; not on the wire.
+   `measurement_id` that is a deterministic 64-bit hash of
+   `commit_sha` plus that table's dimensional tuple. Including
+   `commit_sha` makes every (commit, dim) pair a distinct row -
+   that's what the chart pages render as a time series.
+   Server-internal; not on the wire.
 5. **`commits` is the only dim table.** Engine, format, dataset,
    etc. stay as inline strings; DuckDB's dictionary encoding makes
    a lookup table pointless.
@@ -184,12 +187,15 @@ on read.
 
 ## `measurement_id` hash
 
-Per-table xxhash64 over each table's dimensional tuple. The hash is
-**server-internal** - the wire never carries it. The server's INSERT
-path computes it before each `INSERT ... ON CONFLICT DO UPDATE`,
-which gives idempotent upsert on re-emission of the same dim tuple.
-Encoding details (input order, NULL handling, byte layout) are the
-server's call, since the value never crosses a process boundary.
+Per-table xxhash64 over `commit_sha` plus that table's dimensional
+tuple. Including `commit_sha` makes every (commit, dim) pair a
+distinct row, which is what the chart pages render as a time
+series. The hash is **server-internal** - the wire never carries
+it. The server's INSERT path computes it before each
+`INSERT ... ON CONFLICT DO UPDATE`, which gives idempotent upsert
+on re-emission of the same (commit, dim) pair. Encoding details
+(input order, NULL handling, byte layout) are the server's call,
+since the value never crosses a process boundary.
 
 When the historical migrator lands (deferred), it reuses the
 server's hash function via a shared crate.

benchmarks-website/server/src/db.rs

@@ -7,9 +7,12 @@
 //! [`tokio::sync::Mutex`]. All DB work runs inside `spawn_blocking` so the
 //! Tokio runtime is never blocked on synchronous DuckDB calls.
 //!
-//! `measurement_id` is a server-internal xxhash64 over each table's
-//! dimensional tuple. The hash never crosses a process boundary, so the
-//! exact byte layout below is private to this server.
+//! `measurement_id` is a server-internal xxhash64 over `commit_sha` plus
+//! each table's dimensional tuple. Including `commit_sha` makes every
+//! (commit, dim) pair a distinct row, which is what the chart pages render
+//! as a time series; re-ingest of the same pair is the upsert case. The
+//! hash never crosses a process boundary, so the exact byte layout below
+//! is private to this server.
 
 use std::path::Path;
 use std::sync::Arc;
@@ -99,10 +102,12 @@ fn hasher_for(tag: &'static str) -> XxHash64 {
     h
 }
 
-/// Hash for `query_measurements` rows. Matches the dim columns marked NOT NULL
-/// in `01-schema.md` plus the optional dimensions that distinguish rows.
+/// Hash for `query_measurements` rows. Includes `commit_sha` so each
+/// (commit, dim tuple) pair gets a distinct row; re-emission of the same
+/// pair is the upsert case.
 pub fn measurement_id_query(r: &QueryMeasurement) -> i64 {
     let mut h = hasher_for("query_measurements");
+    write_str(&mut h, &r.commit_sha);
     write_str(&mut h, &r.dataset);
     write_opt_str(&mut h, r.dataset_variant.as_deref());
     write_opt_str(&mut h, r.scale_factor.as_deref());
@@ -116,6 +121,7 @@ pub fn measurement_id_query(r: &QueryMeasurement) -> i64 {
 /// Hash for `compression_times` rows.
 pub fn measurement_id_compression_time(r: &CompressionTime) -> i64 {
     let mut h = hasher_for("compression_times");
+    write_str(&mut h, &r.commit_sha);
     write_str(&mut h, &r.dataset);
     write_opt_str(&mut h, r.dataset_variant.as_deref());
     write_str(&mut h, &r.format);
@@ -126,6 +132,7 @@ pub fn measurement_id_compression_time(r: &CompressionTime) -> i64 {
 /// Hash for `compression_sizes` rows.
 pub fn measurement_id_compression_size(r: &CompressionSize) -> i64 {
     let mut h = hasher_for("compression_sizes");
+    write_str(&mut h, &r.commit_sha);
     write_str(&mut h, &r.dataset);
     write_opt_str(&mut h, r.dataset_variant.as_deref());
     write_str(&mut h, &r.format);
@@ -135,6 +142,7 @@ pub fn measurement_id_compression_size(r: &CompressionSize) -> i64 {
 /// Hash for `random_access_times` rows.
 pub fn measurement_id_random_access(r: &RandomAccessTime) -> i64 {
     let mut h = hasher_for("random_access_times");
+    write_str(&mut h, &r.commit_sha);
     write_str(&mut h, &r.dataset);
     write_str(&mut h, &r.format);
     finish(h)
@@ -144,6 +152,7 @@ pub fn measurement_id_random_access(r: &RandomAccessTime) -> i64 {
 /// of the dim tuple per `01-schema.md`.
 pub fn measurement_id_vector_search(r: &VectorSearchRun) -> i64 {
     let mut h = hasher_for("vector_search_runs");
+    write_str(&mut h, &r.commit_sha);
     write_str(&mut h, &r.dataset);
     write_str(&mut h, &r.layout);
     write_str(&mut h, &r.flavor);

benchmarks-website/server/tests/snapshots/chart_page_query.snap

@@ -1,5 +1,6 @@
 ---
 source: benchmarks-website/server/tests/web_ui.rs
+assertion_line: 267
 expression: body
 ---
-<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width, initial-scale=1"><title>tpch sf=1 Q1 [nvme] — bench.vortex.dev</title><link rel="stylesheet" href="/static/style.css"></head><body><header class="page-header"><h1><a href="/">bench.vortex.dev</a></h1><p class="subtitle">tpch sf=1 Q1 [nvme]</p></header><main><p class="chart-meta">unit: <code>ns</code> · 2 series · 1 commit</p><div class="chart-wrap"><canvas id="chart"></canvas></div><script id="chart-data" type="application/json">{"display_name":"tpch sf=1 Q1 [nvme]","unit":"ns","commits":[{"sha":"3333333333333333333333333333333333333333","timestamp":"2026-04-25 12:00:00+00","message":"third commit","url":"https://github.com/vortex-data/vortex/commit/3333333333333333333333333333333333333333"}],"series":{"datafusion:vortex-file-compressed":[1100000.0],"duckdb:parquet":[900000.0]}}</script><noscript><p class="no-script">JavaScript is required to render the chart.</p></noscript></main><script src="/static/chart.umd.js" defer></script><script src="/static/chart-init.js" defer></script></body></html>
+<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width, initial-scale=1"><title>tpch sf=1 Q1 [nvme] — bench.vortex.dev</title><link rel="stylesheet" href="/static/style.css"></head><body><header class="page-header"><h1><a href="/">bench.vortex.dev</a></h1><p class="subtitle">tpch sf=1 Q1 [nvme]</p></header><main><p class="chart-meta">unit: <code>ns</code> · 2 series · 3 commits</p><div class="chart-wrap"><canvas id="chart"></canvas></div><script id="chart-data" type="application/json">{"display_name":"tpch sf=1 Q1 [nvme]","unit":"ns","commits":[{"sha":"1111111111111111111111111111111111111111","timestamp":"2026-04-23 12:00:00+00","message":"first commit","url":"https://github.com/vortex-data/vortex/commit/1111111111111111111111111111111111111111"},{"sha":"2222222222222222222222222222222222222222","timestamp":"2026-04-24 12:00:00+00","message":"second commit","url":"https://github.com/vortex-data/vortex/commit/2222222222222222222222222222222222222222"},{"sha":"3333333333333333333333333333333333333333","timestamp":"2026-04-25 12:00:00+00","message":"third commit","url":"https://github.com/vortex-data/vortex/commit/3333333333333333333333333333333333333333"}],"series":{"datafusion:vortex-file-compressed":[1000000.0,1050000.0,1100000.0],"duckdb:parquet":[800000.0,850000.0,900000.0]}}</script><noscript><p class="no-script">JavaScript is required to render the chart.</p></noscript></main><script src="/static/chart.umd.js" defer></script><script src="/static/chart-init.js" defer></script></body></html>