package docpipe
import (
"os"
"path/filepath"
"strings"
"testing"
)
func TestHOCRToMarkdown_Fixture(t *testing.T) {
// Minimal hOCR 1.2 snippet
hocrXML := `
tesseract
`
dir := t.TempDir()
p := filepath.Join(dir, "sample.hocr")
if err := os.WriteFile(p, []byte(hocrXML), 0o644); err != nil {
t.Fatal(err)
}
md, yml, err := HOCRToMarkdown(p, 0)
if err != nil {
t.Fatal(err)
}
if !strings.Contains(md, "Hello") || !strings.Contains(md, "world") {
t.Fatalf("md=%q", md)
}
if !strings.Contains(yml, "Hello") {
t.Fatalf("yaml missing word: %s", yml)
}
}