mirror of
https://github.com/lukaszraczylo/llm-testbench.git
synced 2026-10-12 00:45:19 +00:00
Add 9 tests to research/web, research/whitepapers, and research/codebase, bringing each subcategory from 1 test to 10. Web tests extract data from inline HTML, headers, and URLs, and answer protocol scenarios for HTTP status codes, CORS, DNS, and canonicalization. Every fixture stays inline; no test makes a live web call. Whitepaper tests each pair a 120 to 180 word, technically accurate excerpt with numeric or JSON extraction questions. Topics cover product quantization, Bloom filters, TF-IDF, recall@k, LSM-tree write amplification, Raft quorum sizing, B-tree capacity, the CAP theorem, and attention scaling. Every derived constant carries a ground-truth comment and a unit test that recomputes the value with an independent formula. Codebase tests trace a Python function and a TypeScript function to exact output, then reason over an inline diff, a data race, dead code, Big-O complexity, an import cycle, a generic return type, and a git-bisect scenario. The Python trace test runs the same source through python3 when available and skips otherwise; the TypeScript trace test cross-checks its hand-derivation against an equivalent Go function. New package-level helpers use a wp or code prefix to avoid name collisions with sibling category authors sharing this package. Test IDs keep the paper, web, and code prefixes already established by the existing catalog entries. New tests omit MaxTokens so the runner falls back to its default, per the catalog's authoring rules. catalog_test.go's exact-map assertion now fails as expected, since it still lists only the original 17-test catalog; that assertion is owned centrally and is not touched here.
147 lines
3.9 KiB
Go
147 lines
3.9 KiB
Go
package tests
|
|
|
|
import (
|
|
"bytes"
|
|
"context"
|
|
"os"
|
|
"os/exec"
|
|
"strings"
|
|
"testing"
|
|
)
|
|
|
|
func TestCodeExactAnswer(t *testing.T) {
|
|
ev := codeExactAnswer("efabcd")
|
|
|
|
tests := []struct {
|
|
name string
|
|
response string
|
|
want float64
|
|
}{
|
|
{"bare correct", "efabcd", 1},
|
|
{"double-quoted correct", `"efabcd"`, 1},
|
|
{"different case", "EFABCD", 1},
|
|
{"trailing period", "efabcd.", 1},
|
|
{"wrong string", "abcdef", 0},
|
|
{"correct but truncated", "efabc", 0},
|
|
}
|
|
for _, tt := range tests {
|
|
t.Run(tt.name, func(t *testing.T) {
|
|
got := ev.Evaluate(context.Background(), tt.response)
|
|
if got.Value != tt.want {
|
|
t.Errorf("Evaluate(%q) = %v, want %v (detail: %s)", tt.response, got.Value, tt.want, got.Detail)
|
|
}
|
|
})
|
|
}
|
|
}
|
|
|
|
func TestCodeTracePythonWant_GroundTruth(t *testing.T) {
|
|
// Hand-derivation, double-checked: len("abcdef")=6 is even at every
|
|
// step, so every call takes the rotate branch (s[1:]+s[0]) rather than
|
|
// the reverse branch.
|
|
s := "abcdef"
|
|
for i := 0; i < 4; i++ {
|
|
if len(s)%2 == 0 {
|
|
s = s[1:] + string(s[0])
|
|
} else {
|
|
for l, r := 0, len(s)-1; l < r; l, r = l+1, r-1 {
|
|
b := []byte(s)
|
|
b[l], b[r] = b[r], b[l]
|
|
s = string(b)
|
|
}
|
|
}
|
|
}
|
|
if s != codeTracePythonWant {
|
|
t.Errorf("hand-derived trace = %q, want %q", s, codeTracePythonWant)
|
|
}
|
|
|
|
if _, err := exec.LookPath("python3"); err != nil {
|
|
t.Skip("python3 not found on PATH, skipping exec ground-truth cross-check")
|
|
}
|
|
|
|
script := codeTracePythonSource + "\nprint(transform(\"abcdef\", 4))\n"
|
|
// #nosec G204 -- script is a fixed, hardcoded string built from the
|
|
// same codeTracePythonSource constant embedded in the prompt, not
|
|
// external or response-controlled input.
|
|
cmd := exec.Command("python3", "-c", script) //nolint:gosec // see comment above
|
|
var stdout bytes.Buffer
|
|
cmd.Stdout = &stdout
|
|
cmd.Stderr = os.Stderr
|
|
if err := cmd.Run(); err != nil {
|
|
t.Fatalf("running python3 ground-truth script failed: %v", err)
|
|
}
|
|
got := strings.TrimSpace(stdout.String())
|
|
if got != codeTracePythonWant {
|
|
t.Errorf("python3-computed transform(\"abcdef\", 4) = %q, want %q", got, codeTracePythonWant)
|
|
}
|
|
}
|
|
|
|
func TestCodeTracePythonTest_Eval(t *testing.T) {
|
|
tc := codeTracePythonTest()
|
|
|
|
tests := []struct {
|
|
name string
|
|
response string
|
|
want float64
|
|
}{
|
|
{"correct", "efabcd", 1},
|
|
{"correct quoted", `"efabcd"`, 1},
|
|
{"wrong: forgot final rotation", "defabc", 0},
|
|
{"wrong: reversed instead of rotated", "fedcba", 0},
|
|
}
|
|
for _, tt := range tests {
|
|
t.Run(tt.name, func(t *testing.T) {
|
|
got := tc.Eval.Evaluate(context.Background(), tt.response)
|
|
if got.Value != tt.want {
|
|
t.Errorf("Eval.Evaluate(%q) = %v, want %v (detail: %s)", tt.response, got.Value, tt.want, got.Detail)
|
|
}
|
|
})
|
|
}
|
|
}
|
|
|
|
func TestCodeTraceTSWant_GroundTruth(t *testing.T) {
|
|
// Hand-derivation, double-checked against codeTSPackTrace: acc=0;
|
|
// i=0,x=3 (even index) -> 3; i=1,x=4 (odd) -> 3+8=11; i=2,x=5 (even) ->
|
|
// 16; i=3,x=6 (odd) -> 16+12=28; i=4,x=7 (even) -> 35. 35 = 32+2+1 in
|
|
// binary is 100011.
|
|
acc := 0
|
|
items := []int{3, 4, 5, 6, 7}
|
|
for i, x := range items {
|
|
if i%2 == 0 {
|
|
acc += x
|
|
} else {
|
|
acc += x * 2
|
|
}
|
|
}
|
|
if acc != 35 {
|
|
t.Fatalf("hand-derived acc = %d, want 35", acc)
|
|
}
|
|
|
|
got := codeTSPackTrace(items)
|
|
if got != codeTraceTSWant {
|
|
t.Errorf("codeTSPackTrace(%v) = %q, want %q", items, got, codeTraceTSWant)
|
|
}
|
|
}
|
|
|
|
func TestCodeTraceTSTest_Eval(t *testing.T) {
|
|
tc := codeTraceTSTest()
|
|
|
|
tests := []struct {
|
|
name string
|
|
response string
|
|
want float64
|
|
}{
|
|
{"correct", "100011", 1},
|
|
{"correct quoted", `"100011"`, 1},
|
|
{"wrong: doubled every element instead of alternating", "1000110", 0},
|
|
{"wrong: decimal instead of binary", "35", 0},
|
|
}
|
|
for _, tt := range tests {
|
|
t.Run(tt.name, func(t *testing.T) {
|
|
got := tc.Eval.Evaluate(context.Background(), tt.response)
|
|
if got.Value != tt.want {
|
|
t.Errorf("Eval.Evaluate(%q) = %v, want %v (detail: %s)", tt.response, got.Value, tt.want, got.Detail)
|
|
}
|
|
})
|
|
}
|
|
}
|