bug处理

2026-10-05 04:56:37 +08:00 · 2026-10-02 15:14:36 +08:00
parent 83ec494a36
commit 2c6646beea
23 changed files with 1900 additions and 108 deletions
@@ -0,0 +1,159 @@
+package reader
+
+import (
+	"fmt"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/truewhile/MeBox/internal/model"
+)
+
+// TestGetTocFollowsNextTocUrl 长书目录按 offset/limit 分页时必须翻页取全。
+// 这是「目录少章」的根因：nextTocUrl 没实现时只会拿到第一页
+// （实测某漫画站点 total=399/limit=100，MeBox 只得到 100 章）。
+func TestGetTocFollowsNextTocUrl(t *testing.T) {
+	pages := map[string]string{
+		"0": `{"count":4,"list":[{"name":"第1话","url":"/c/1"},{"name":"第2话","url":"/c/2"}]}`,
+		"2": `{"count":4,"list":[{"name":"第3话","url":"/c/3"},{"name":"第4话","url":"/c/4"}]}`,
+	}
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		offset := r.URL.Query().Get("offset")
+		body, ok := pages[offset]
+		if !ok {
+			http.NotFound(w, r)
+			return
+		}
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = fmt.Fprint(w, body)
+	}))
+	defer srv.Close()
+
+	// 与真实书源（拷贝漫画）一致的写法：用 baseUrl 拼下一页地址
+	nextTocURL := `<js>baseUrl.indexOf('offset=0') >= 0 ? baseUrl.replace('offset=0','offset=2') : ''</js>`
+	header := ""
+	src := &model.ReaderBookSource{Name: "分页目录测试源", SourceURL: srv.URL}
+	bs := &BookSource{
+		BookSourceURL: srv.URL,
+		Header:        &header,
+		RuleToc: &TocRule{
+			ChapterList: strPtr("$.list[*]"),
+			ChapterName: strPtr("$.name"),
+			ChapterURL:  strPtr("$.url"),
+			NextTocURL:  strPtr(nextTocURL),
+		},
+	}
+
+	svc, _ := newLoginTestService(t)
+	tocURL := srv.URL + "/toc?limit=2&offset=0"
+	chapters, _, err := svc.getTocFrom(t.Context(), src, bs, srv.URL+"/book/1", tocURL)
+	if err != nil {
+		t.Fatalf("取目录失败: %v", err)
+	}
+	if len(chapters) != 4 {
+		t.Fatalf("章节数 = %d，期望 4（第二页没有被抓取）", len(chapters))
+	}
+	for i, want := range []string{"第1话", "第2话", "第3话", "第4话"} {
+		if chapters[i].Title != want {
+			t.Errorf("章节[%d] = %q，期望 %q", i, chapters[i].Title, want)
+		}
+		if chapters[i].Index != i {
+			t.Errorf("章节[%d] 的 Index = %d，应按翻页顺序重排", i, chapters[i].Index)
+		}
+	}
+}
+
+// TestGetTocNextTocUrlListsAllPagesFromFirstPage 书源只在第一页能算出完整后续页列表
+// （拷贝漫画的写法：规则锚定 offset=0，第二页上算出来还是同一批地址）。
+// 这种源必须把所有后续页一次性入队，否则只翻一页就停。
+func TestGetTocNextTocUrlListsAllPagesFromFirstPage(t *testing.T) {
+	pages := map[string]string{
+		"0":   `{"total":4,"list":[{"name":"第1话","url":"/c/1"},{"name":"第2话","url":"/c/2"}]}`,
+		"100": `{"total":4,"list":[{"name":"第3话","url":"/c/3"},{"name":"第4话","url":"/c/4"}]}`,
+	}
+	var hits []string
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		offset := r.URL.Query().Get("offset")
+		hits = append(hits, offset)
+		body, ok := pages[offset]
+		if !ok {
+			w.Header().Set("Content-Type", "application/json")
+			_, _ = fmt.Fprint(w, `{"total":4,"list":[]}`)
+			return
+		}
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = fmt.Fprint(w, body)
+	}))
+	defer srv.Close()
+
+	// 与真实书源一致：一次列出全部后续页，且锚点写死 offset=0（在第二页上等于原地踏步）
+	nextTocURL := `<js>
+list=[];
+for(i=0;i<=1;i++){list.push(baseUrl.replace('offset=0','offset='+100*i))}
+list
+</js>`
+	header := ""
+	src := &model.ReaderBookSource{Name: "全量下一页测试源", SourceURL: srv.URL}
+	bs := &BookSource{
+		BookSourceURL: srv.URL,
+		Header:        &header,
+		RuleToc: &TocRule{
+			ChapterList: strPtr("$.list[*]"),
+			ChapterName: strPtr("$.name"),
+			ChapterURL:  strPtr("$.url"),
+			NextTocURL:  strPtr(nextTocURL),
+		},
+	}
+
+	svc, _ := newLoginTestService(t)
+	chapters, _, err := svc.getTocFrom(t.Context(), src, bs, srv.URL+"/book/1", srv.URL+"/toc?limit=2&offset=0")
+	if err != nil {
+		t.Fatalf("取目录失败: %v", err)
+	}
+	if len(chapters) != 4 {
+		t.Fatalf("章节数 = %d，期望 4（第一页列出的后续页没有被全部抓取）", len(chapters))
+	}
+	// 每页只抓一次（队列去重）
+	if len(hits) != 2 {
+		t.Fatalf("上游请求次数 = %d（%v），期望 2", len(hits), hits)
+	}
+}
+
+// TestGetTocNextTocUrlSelfReferenceStops nextTocUrl 指回当前页时必须停下，
+// 不能靠「规则永远给下一页」把自己转死。
+func TestGetTocNextTocUrlSelfReferenceStops(t *testing.T) {
+	var hits int
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		hits++
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = fmt.Fprint(w, `{"list":[{"name":"第1话","url":"/c/1"}]}`)
+	}))
+	defer srv.Close()
+
+	// 永远返回同一页地址（用 baseUrl 原样返回）
+	nextTocURL := `<js>baseUrl</js>`
+	header := ""
+	src := &model.ReaderBookSource{Name: "自引用目录测试源", SourceURL: srv.URL}
+	bs := &BookSource{
+		BookSourceURL: srv.URL,
+		Header:        &header,
+		RuleToc: &TocRule{
+			ChapterList: strPtr("$.list[*]"),
+			ChapterName: strPtr("$.name"),
+			ChapterURL:  strPtr("$.url"),
+			NextTocURL:  strPtr(nextTocURL),
+		},
+	}
+
+	svc, _ := newLoginTestService(t)
+	chapters, _, err := svc.getTocFrom(t.Context(), src, bs, srv.URL+"/book/1", srv.URL+"/toc")
+	if err != nil {
+		t.Fatalf("取目录失败: %v", err)
+	}
+	if len(chapters) != 1 {
+		t.Fatalf("章节数 = %d，期望 1", len(chapters))
+	}
+	if hits != 1 {
+		t.Fatalf("上游被请求 %d 次，自引用时应只请求 1 次", hits)
+	}
+}