Log evals as JSON

.github/workflows/ci.yml

@@ -60,8 +60,31 @@ jobs:
       - name: Get dependencies
         run: go mod download
 
+      - name: Install evals
+        run: go install maragu.dev/evals
+
       - name: Evaluate
-        run: go test -json -run TestEval ./... | jq 'select(.Test != null and .Action == "output" and (.Output | contains("result"))) | del(.Action)'
+        run: go test -json -run TestEval ./... | evals | tee evals.txt >> $GITHUB_STEP_SUMMARY
+
+      - uses: actions/upload-artifact@v4
+        id: evalsdb
+        with:
+          name: evals.db
+          path: evals.db
+          if-no-files-found: error
+
+      - name: Add evals comment to PR
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const fs = require('fs')
+            const table = fs.readFileSync('evals.txt', 'utf8')
+            github.rest.issues.createComment({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              issue_number: context.issue.number,
+              body: `# Evals\n\n${table}\n\n[Download evals.db](${{ steps.evalsdb.outputs.artifact-url }})`
+            })
 
   lint:
     name: Lint

.gitignore

@@ -1,2 +1,3 @@
-/cover.out
 /.env*.local
+/cover.out
+evals.db

Makefile

@@ -8,7 +8,7 @@ cover:
 
 .PHONY: evaluate
 evaluate:
-	go test -json -run TestEval ./... | jq 'select(.Test != null and .Action == "output" and (.Output | contains("result"))) | del(.Action)'
+	@go test -json -run TestEval ./... | evals
 
 .PHONY: lint
 lint:

eval/run.go

@@ -1,6 +1,7 @@
 package eval
 
 import (
+	"encoding/json"
 	"os"
 	"strings"
 	"testing"
@@ -70,10 +71,32 @@ func (e *E) Score(s Sample, scorer Scorer) Result {
 	return r
 }
 
+type logLine struct {
+	Sample   Sample
+	Result   Result
+	Duration time.Duration
+}
+
 // Log a [Sample] and [Result].
 // This effectively logs the eval name, sample, and result, along with timing information.
 // TODO include token information?
 func (e *E) Log(s Sample, r Result) {
 	e.T.Helper()
-	e.T.Logf("sample=%+v result=%+v duration=%v", s, r, time.Since(e.start))
+
+	l := logLine{
+		Sample:   s,
+		Result:   r,
+		Duration: time.Since(e.start),
+	}
+
+	e.T.Log(mustJSON(l))
+}
+
+func mustJSON(l logLine) string {
+	b, err := json.Marshal(l)
+	if err != nil {
+		panic(err)
+	}
+
+	return string(b)
 }

go.mod

@@ -9,6 +9,7 @@ require (
 	github.com/openai/openai-go v0.1.0-alpha.43
 	google.golang.org/api v0.214.0
 	maragu.dev/env v0.2.0
+	maragu.dev/evals v0.0.0-20250110091058-dfafc4ece961
 	maragu.dev/is v0.2.0
 )
 
@@ -26,6 +27,8 @@ require (
 	github.com/google/uuid v1.6.0 // indirect
 	github.com/googleapis/enterprise-certificate-proxy v0.3.4 // indirect
 	github.com/googleapis/gax-go/v2 v2.14.0 // indirect
+	github.com/jmoiron/sqlx v1.4.0 // indirect
+	github.com/mattn/go-sqlite3 v1.14.24 // indirect
 	github.com/tidwall/gjson v1.14.4 // indirect
 	github.com/tidwall/match v1.1.1 // indirect
 	github.com/tidwall/pretty v1.2.1 // indirect
@@ -46,4 +49,6 @@ require (
 	google.golang.org/genproto/googleapis/rpc v0.0.0-20241209162323-e6fa225c2576 // indirect
 	google.golang.org/grpc v1.67.1 // indirect
 	google.golang.org/protobuf v1.35.2 // indirect
+	maragu.dev/errors v0.3.0 // indirect
+	maragu.dev/migrate v0.6.0 // indirect
 )

go.sum

@@ -10,6 +10,8 @@ cloud.google.com/go/compute/metadata v0.6.0 h1:A6hENjEsCDtC1k8byVsgwvVcioamEHvZ4
 cloud.google.com/go/compute/metadata v0.6.0/go.mod h1:FjyFAW1MW0C203CEOMDTu3Dk1FlqW3Rga40jzHL4hfg=
 cloud.google.com/go/longrunning v0.5.7 h1:WLbHekDbjK1fVFD3ibpFFVoyizlLRl73I7YKuAKilhU=
 cloud.google.com/go/longrunning v0.5.7/go.mod h1:8GClkudohy1Fxm3owmBGid8W0pSgodEMwEAztp38Xng=
+filippo.io/edwards25519 v1.1.0 h1:FNf4tywRC1HmFuKW5xopWpigGjJKiJSV0Cqo0cJWDaA=
+filippo.io/edwards25519 v1.1.0/go.mod h1:BxyFTGdWcka3PhytdK4V28tE5sGfRvvvRV7EaN4VDT4=
 github.com/agnivade/levenshtein v1.2.0 h1:U9L4IOT0Y3i0TIlUIDJ7rVUziKi/zPbrJGaFrtYH3SY=
 github.com/agnivade/levenshtein v1.2.0/go.mod h1:QVVI16kDrtSuwcpd0p1+xMC6Z/VfhtCyDIjcwga4/DU=
 github.com/anthropics/anthropic-sdk-go v0.2.0-alpha.8 h1:ss/c/eeyILgoK2sMsTJdcdLdhY3wZSt//+nanM41B9w=
@@ -27,6 +29,8 @@ github.com/go-logr/logr v1.4.2 h1:6pFjapn8bFcIbiKo3XT4j/BhANplGihG6tvd+8rYgrY=
 github.com/go-logr/logr v1.4.2/go.mod h1:9T104GzyrTigFIr8wt5mBrctHMim0Nb2HLGrmQ40KvY=
 github.com/go-logr/stdr v1.2.2 h1:hSWxHoqTgW2S2qGc0LTAI563KZ5YKYRhT3MFKZMbjag=
 github.com/go-logr/stdr v1.2.2/go.mod h1:mMo/vtBO5dYbehREoey6XUKy/eSumjCCveDpRre4VKE=
+github.com/go-sql-driver/mysql v1.8.1 h1:LedoTUt/eveggdHS9qUFC1EFSa8bU2+1pZjSRpvNJ1Y=
+github.com/go-sql-driver/mysql v1.8.1/go.mod h1:wEBSXgmK//2ZFJyE+qWnIsVGmvmEKlqwuVSjsCm7DZg=
 github.com/google/generative-ai-go v0.19.0 h1:R71szggh8wHMCUlEMsW2A/3T+5LdEIkiaHSYgSpUgdg=
 github.com/google/generative-ai-go v0.19.0/go.mod h1:JYolL13VG7j79kM5BtHz4qwONHkeJQzOCkKXnpqtS/E=
 github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
@@ -39,6 +43,29 @@ github.com/googleapis/enterprise-certificate-proxy v0.3.4 h1:XYIDZApgAnrN1c855gT
 github.com/googleapis/enterprise-certificate-proxy v0.3.4/go.mod h1:YKe7cfqYXjKGpGvmSg28/fFvhNzinZQm8DGnaburhGA=
 github.com/googleapis/gax-go/v2 v2.14.0 h1:f+jMrjBPl+DL9nI4IQzLUxMq7XrAqFYB7hBPqMNIe8o=
 github.com/googleapis/gax-go/v2 v2.14.0/go.mod h1:lhBCnjdLrWRaPvLWhmc8IS24m9mr07qSYnHncrgo+zk=
+github.com/jackc/chunkreader/v2 v2.0.1 h1:i+RDz65UE+mmpjTfyz0MoVTnzeYxroil2G82ki7MGG8=
+github.com/jackc/chunkreader/v2 v2.0.1/go.mod h1:odVSm741yZoC3dpHEUXIqA9tQRhFrgOHwnPIn9lDKlk=
+github.com/jackc/pgconn v1.14.3 h1:bVoTr12EGANZz66nZPkMInAV/KHD2TxH9npjXXgiB3w=
+github.com/jackc/pgconn v1.14.3/go.mod h1:RZbme4uasqzybK2RK5c65VsHxoyaml09lx3tXOcO/VM=
+github.com/jackc/pgio v1.0.0 h1:g12B9UwVnzGhueNavwioyEEpAmqMe1E/BN9ES+8ovkE=
+github.com/jackc/pgio v1.0.0/go.mod h1:oP+2QK2wFfUWgr+gxjoBH9KGBb31Eio69xUb0w5bYf8=
+github.com/jackc/pgpassfile v1.0.0 h1:/6Hmqy13Ss2zCq62VdNG8tM1wchn8zjSGOBJ6icpsIM=
+github.com/jackc/pgpassfile v1.0.0/go.mod h1:CEx0iS5ambNFdcRtxPj5JhEz+xB6uRky5eyVu/W2HEg=
+github.com/jackc/pgproto3/v2 v2.3.3 h1:1HLSx5H+tXR9pW3in3zaztoEwQYRC9SQaYUHjTSUOag=
+github.com/jackc/pgproto3/v2 v2.3.3/go.mod h1:WfJCnwN3HIg9Ish/j3sgWXnAfK8A9Y0bwXYU5xKaEdA=
+github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 h1:iCEnooe7UlwOQYpKFhBabPMi4aNAfoODPEFNiAnClxo=
+github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761/go.mod h1:5TJZWKEWniPve33vlWYSoGYefn3gLQRzjfDlhSJ9ZKM=
+github.com/jackc/pgtype v1.14.0 h1:y+xUdabmyMkJLyApYuPj38mW+aAIqCe5uuBB51rH3Vw=
+github.com/jackc/pgtype v1.14.0/go.mod h1:LUMuVrfsFfdKGLw+AFFVv6KtHOFMwRgDDzBt76IqCA4=
+github.com/jackc/pgx/v4 v4.18.2 h1:xVpYkNR5pk5bMCZGfClbO962UIqVABcAGt7ha1s/FeU=
+github.com/jackc/pgx/v4 v4.18.2/go.mod h1:Ey4Oru5tH5sB6tV7hDmfWFahwF15Eb7DNXlRKx2CkVw=
+github.com/jmoiron/sqlx v1.4.0 h1:1PLqN7S1UYp5t4SrVVnt4nUVNemrDAtxlulVe+Qgm3o=
+github.com/jmoiron/sqlx v1.4.0/go.mod h1:ZrZ7UsYB/weZdl2Bxg6jCRO9c3YHl8r3ahlKmRT4JLY=
+github.com/lib/pq v1.10.9 h1:YXG7RB+JIjhP29X+OtkiDnYaXQwpS4JEWq7dtCCRUEw=
+github.com/lib/pq v1.10.9/go.mod h1:AlVN5x4E4T544tWzH6hKfbfQvm3HdbOxrmggDNAPY9o=
+github.com/mattn/go-sqlite3 v1.14.22/go.mod h1:Uh1q+B4BYcTPb+yiD3kU8Ct7aC0hY9fxUwlHK0RXw+Y=
+github.com/mattn/go-sqlite3 v1.14.24 h1:tpSp2G2KyMnnQu99ngJ47EIkWVmliIizyZBfPrBWDRM=
+github.com/mattn/go-sqlite3 v1.14.24/go.mod h1:Uh1q+B4BYcTPb+yiD3kU8Ct7aC0hY9fxUwlHK0RXw+Y=
 github.com/openai/openai-go v0.1.0-alpha.43 h1:6XWGUsrHSaPyh8U6ocs/XJGb/UX7jhQRK2bYefvTuAg=
 github.com/openai/openai-go v0.1.0-alpha.43/go.mod h1:3SdE6BffOX9HPEQv8IL/fi3LYZ5TUpRYaqGQZbyk11A=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
@@ -93,5 +120,13 @@ gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
 gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
 maragu.dev/env v0.2.0 h1:nQKitDEB65ArZsh6E7vxzodOqY9bxEVFdBg+tskS1ys=
 maragu.dev/env v0.2.0/go.mod h1:t5CCbaEnjCM5mewiAVVzTS4N+oXTus2+SRnzKQbQVME=
+maragu.dev/errors v0.3.0 h1:huI+n+ddMfVgQFD+cEqIPaozUlfz3TkfgpkssNip5G0=
+maragu.dev/errors v0.3.0/go.mod h1:cygLiyNnq4ofF3whYscilo2ecUADCaUQXwvwFrMOhmM=
+maragu.dev/evals v0.0.0-20250110090454-18cd820491c8 h1:04jcfKTCUUW0iU1v6leNl3nWmE/w3l/es1niEjX8JRs=
+maragu.dev/evals v0.0.0-20250110090454-18cd820491c8/go.mod h1:Qyj56mYe9ApUvas0pVm5M7BbxH5KSyMmifSDFyaUSco=
+maragu.dev/evals v0.0.0-20250110091058-dfafc4ece961 h1:H7iZNzGQ7RZ+Se/OqiKsj4bCFUPPWhauk8iY8HtjwhQ=
+maragu.dev/evals v0.0.0-20250110091058-dfafc4ece961/go.mod h1:Qyj56mYe9ApUvas0pVm5M7BbxH5KSyMmifSDFyaUSco=
 maragu.dev/is v0.2.0 h1:poeuVEA5GG3vrDpGmzo2KjWtIMZmqUyvGnOB0/pemig=
 maragu.dev/is v0.2.0/go.mod h1:bviaM5S0fBshCw7wuumFGTju/izopZ/Yvq4g7Klc7y8=
+maragu.dev/migrate v0.6.0 h1:gJLAIVaRh9z9sN55Q2sWwScpEH+JsT6N0L1DnzedXFE=
+maragu.dev/migrate v0.6.0/go.mod h1:TdZBD5wRvBbzLocsSV08kyvLiLCn0Q6DvgYHmyygWVQ=

tools.go

@@ -0,0 +1,3 @@
+package llm
+
+import _ "maragu.dev/evals/tool"