package gutenberg

import (
	"strings"
	"testing"
)

func TestStripBoilerplate(t *testing.T) {
	input := `The Project Gutenberg eBook of Test Book

*** START OF THE PROJECT GUTENBERG EBOOK TEST BOOK ***

This is the actual content of the book.
It has multiple lines.
And some more content here.

*** END OF THE PROJECT GUTENBERG EBOOK TEST BOOK ***

End of the Project Gutenberg EBook of Test Book
Blah blah license info.`

	result := StripBoilerplate(input)

	if strings.Contains(result, "Project Gutenberg") {
		t.Error("boilerplate not fully stripped")
	}
	if !strings.Contains(result, "actual content") {
		t.Error("body content was stripped")
	}
	if !strings.Contains(result, "multiple lines") {
		t.Error("body content was stripped")
	}
	if strings.Contains(result, "license info") {
		t.Error("footer not stripped")
	}
}

func TestStripBoilerplateNoMarkers(t *testing.T) {
	input := "This is plain text without any Gutenberg markers."
	result := StripBoilerplate(input)
	if result != input {
		t.Error("text without markers should be returned unchanged")
	}
}

func TestStripBoilerplateCaseInsensitive(t *testing.T) {
	input := `header stuff
*** start of THE PROJECT GUTENBERG EBOOK ***
body content
*** END of THE PROJECT GUTENBERG EBOOK ***
footer`

	result := StripBoilerplate(input)
	if !strings.Contains(result, "body content") {
		t.Error("case-insensitive matching failed")
	}
}

func TestBookIDsFromCatalog(t *testing.T) {
	input := strings.NewReader(`# Comment
1
2
3
# Another comment

5
not-a-number
10
`)

	ids, err := BookIDsFromCatalog(input)
	if err != nil {
		t.Fatal(err)
	}
	if len(ids) != 5 {
		t.Errorf("got %d IDs, want 5", len(ids))
	}
	expected := []int{1, 2, 3, 5, 10}
	for i, want := range expected {
		if ids[i] != want {
			t.Errorf("ids[%d] = %d, want %d", i, ids[i], want)
		}
	}
}