plan 10: a format with no text is no match, not unreadable

2 files changed, 50 insertions, 1 deletions

diff --git a/internal/engine/exclude_test.go b/internal/engine/exclude_test.go
index 87fa16c..4bb1512 100644
--- a/internal/engine/exclude_test.go
+++ b/internal/engine/exclude_test.go
@@ -260,3 +260,45 @@ func TestLoadChecksMainExcludesWithoutDirectories(t *testing.T) {
 		t.Error("a broken krino.conf exclude was not reported")
 	}
 }
+
+// TestNoTextFormatIsNoMatch: a file whose format has no text cannot contain
+// a keyword, so a content exclude without a type does not set it aside, and
+// no "content unreadable" warning is raised - while a real read failure (over
+// max-read) still fails closed (re-review N2, Łukasz's decision).
+func TestNoTextFormatIsNoMatch(t *testing.T) {
+	big := "confidential " + strings.Repeat("x", 2048)
+	h, dl := excludeTree(t, map[string]string{"photo.jpg": "\xff\xd8\xff\x00\x01binary", "big.txt": big})
+	main := writeConfig(t, h, `(include "dl")`, map[string]string{"dl": `
+(path "~/dl")
+(max-read 1K)
+(exclude (content "confidential"))
+(rule "pics" (when (type jpg)) (move "Pictures"))
+`})
+	e, errs := Load(main)
+	if len(errs) > 0 {
+		t.Fatal(errs)
+	}
+	r, err := e.Match(context.Background(), e.Dirs[0])
+	if err != nil {
+		t.Fatal(err)
+	}
+	for _, fm := range r.Matched {
+		switch fm.File.Rel {
+		case "photo.jpg":
+			if fm.Excluded != "" || len(fm.Rules) != 1 || len(fm.Warnings) != 0 {
+				t.Errorf("photo.jpg: Excluded %q, rules %d, warnings %v; want the pics rule and no warning", fm.Excluded, len(fm.Rules), fm.Warnings)
+			}
+		case "big.txt":
+			if !strings.HasSuffix(fm.Excluded, "(content unreadable)") {
+				t.Errorf("big.txt: Excluded %q; a real read failure must still fail closed", fm.Excluded)
+			}
+		}
+	}
+	x, err := e.Explain(context.Background(), filepath.Join(dl, "photo.jpg"))
+	if err != nil {
+		t.Fatal(err)
+	}
+	if x.Excluded != "" {
+		t.Errorf("explain: Excluded %q, want none", x.Excluded)
+	}
+}
diff --git a/internal/engine/facts.go b/internal/engine/facts.go
index 487604b..ccdf18d 100644
--- a/internal/engine/facts.go
+++ b/internal/engine/facts.go
@@ -4,6 +4,7 @@ package engine
 
 import (
 	"context"
+	"errors"
 	"path/filepath"
 	"sort"
 	"strings"
@@ -178,7 +179,13 @@ func (f *facts) ContentContains(opt cond.Options, keywords []string) (int, error
 func (f *facts) extract(opt cond.Options, keywords []string) {
 	f.contentDone = true
 	text, err := f.run.e.Extract.Text(f.run.ctx, f.file.Path, f.file.Size, f.run.d.Settings.MaxRead)
-	if err != nil {
+	switch {
+	case errors.Is(err, extract.ErrUnsupported):
+		// A format with no text cannot contain a keyword: every answer is
+		// no, with no warning, and the answers are cached like any other
+		// (re-review N2). Only a real read failure is "unreadable".
+		text = ""
+	case err != nil:
 		f.contentErr = err
 		return
 	}