textmachine/backend/cmd/tmctl/export_cli_test.go

73 lines
2.8 KiB
Go
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

package main
import (
"bytes"
"encoding/json"
"strings"
"testing"
"textmachine/backend/internal/pipeline"
)
// export_cli_test.go pins the tmctl surface of the read-only export projection (D39 слой 6): the
// parse of the `export` command + `--plaintext`, the default stable-JSON render, and the human
// plaintext render (banner + text, flagged chunks marked, dropped chunks empty).
func TestParseExportPlaintext(t *testing.T) {
inv, err := parseInvocation([]string{"export", "--config", "b.yaml", "--plaintext"}, &bytes.Buffer{})
if err != nil {
t.Fatalf("export --plaintext must parse: %v", err)
}
if inv.cmd != "export" || inv.cfgPath != "b.yaml" || !inv.asPlaintext {
t.Fatalf("inv = %+v", inv)
}
// Default (no flag) is JSON, not plaintext.
inv2, err := parseInvocation([]string{"export", "--config", "b.yaml"}, &bytes.Buffer{})
if err != nil || inv2.asPlaintext {
t.Fatalf("export default must be JSON (asPlaintext=false): inv=%+v err=%v", inv2, err)
}
}
func sampleExport() *pipeline.BookExport {
return &pipeline.BookExport{BookID: "b1", Chunks: []pipeline.ChunkExport{
{Chapter: 1, ChunkIdx: 0, SnapshotID: "snap", Disposition: "ok", FinalText: "Чистый текст."},
{Chapter: 2, ChunkIdx: 0, SnapshotID: "snap", Disposition: "flagged", FlagReason: "sanitizer_stripped", FinalText: "Очищенный текст."},
{Chapter: 3, ChunkIdx: 0, SnapshotID: "snap", Disposition: "flagged", FlagReason: "sanitizer_defect", FinalText: ""},
}}
}
func TestRenderExportJSONIsStable(t *testing.T) {
var b bytes.Buffer
if err := renderExport(&b, sampleExport(), false); err != nil {
t.Fatal(err)
}
// The default surface is valid JSON the polygon extractor can consume: it round-trips and carries
// every chunk's export bytes and metadata.
var got pipeline.BookExport
if err := json.Unmarshal(b.Bytes(), &got); err != nil {
t.Fatalf("export JSON must round-trip: %v\n%s", err, b.String())
}
if got.BookID != "b1" || len(got.Chunks) != 3 {
t.Fatalf("round-tripped export = %+v", got)
}
if got.Chunks[1].FinalText != "Очищенный текст." || got.Chunks[1].FlagReason != "sanitizer_stripped" {
t.Fatalf("stripped chunk lost fidelity: %+v", got.Chunks[1])
}
}
func TestRenderExportPlaintextBanners(t *testing.T) {
var b bytes.Buffer
if err := renderExport(&b, sampleExport(), true); err != nil {
t.Fatal(err)
}
out := b.String()
for _, want := range []string{
"=== ГЛАВА 1 ЧАНК 0 ===\nЧистый текст.",
"=== ГЛАВА 2 ЧАНК 0 — flagged (sanitizer_stripped) (утечка вычищена, проверь) ===\nОчищенный текст.",
"=== ГЛАВА 3 ЧАНК 0 — flagged (sanitizer_defect) (не переведён, помечен для человека) ===",
} {
if !strings.Contains(out, want) {
t.Fatalf("plaintext export must contain %q, got:\n%s", want, out)
}
}
}