diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..01ac3bc --- /dev/null +++ b/.gitignore @@ -0,0 +1,7 @@ +# Build output +/monkeyd-crawler +/monkeyd-crawler.exe + +# Cached pages and exported books +/.cache/ +*.pdf diff --git a/README.md b/README.md new file mode 100644 index 0000000..484e04e --- /dev/null +++ b/README.md @@ -0,0 +1,114 @@ +# monkeyd-crawler + +Downloads every chapter of a [monkeydd.com](https://monkeydd.com) novel and exports it as a +single PDF sized for reading on a phone. + +## Install + +Requires Go 1.22+. + +```sh +go build ./cmd/monkeyd-crawler +``` + +## Usage + +Pass the novel's landing page URL: + +```sh +./monkeyd-crawler -url https://monkeydd.com/tro-lai-nam-thang-cu.html +``` + +The PDF is named after the novel unless you pass `-out`. + +```sh +# Bigger type, A5 page for a tablet +./monkeyd-crawler -url https://monkeydd.com/tro-lai-nam-thang-cu.html \ + -page a5 -font-size 14 -out truyen.pdf + +# Try the layout on 3 chapters before fetching the whole book +./monkeyd-crawler -url https://monkeydd.com/tro-lai-nam-thang-cu.html -limit 3 +``` + +### Flags + +| Flag | Default | Purpose | +| --- | --- | --- | +| `-url` | *required* | Novel landing page URL | +| `-out` | novel title | Output PDF path | +| `-page` | `phone` | Page size: `phone`, `a5`, `a4` | +| `-font-size` | `12` | Body font size in points | +| `-line-spacing` | `1.55` | Line height as a multiple of font size | +| `-margin` | `6` | Page margin in mm | +| `-font` | auto | Path to a `.ttf`; defaults to a system font with Vietnamese coverage | +| `-workers` | `4` | Concurrent chapter fetches | +| `-delay` | `400ms` | Minimum delay between requests | +| `-retries` | `3` | Retries per request | +| `-limit` | `0` | Fetch only the first N chapters (0 = all) | +| `-cache` | `.cache` | Cache directory for raw pages (empty to disable) | + +## Why the default page is 90×160 mm + +Phone readability is governed by page *shape* more than by font size. A PDF viewer scales a +whole page to fit the screen, so a large font on an A4 page still ends up small: the page is +about three times wider than a phone screen and gets shrunk to match. The default page is cut +to a 9:16 ratio so it fills the screen at 100% zoom, where 12 pt renders at a comfortable +size with roughly 35–40 characters per line. + +Use `-page a5` or `-page a4` for a tablet or for printing. + +## How it works + +1. Fetch the landing page and read the chapter list from `div.list-chapters`. +2. Cross-check that list against the `#selected_chapter` dropdown embedded in the first + chapter page. When the dropdown is a superset it wins; a disagreement is reported. +3. Fetch each chapter concurrently and extract its text. +4. Render one PDF, each chapter starting on a new page. + +### Two site behaviours the extractor has to handle + +**Chapter text is partly served through CSS.** The markup contains empty elements, and the +stylesheet supplies the missing word: + +```html +Nghe trưởng tử +``` + +```css +.t-3e625e…:before { content: "vị"; } +``` + +Reading DOM text alone silently drops these — about 19% of the words in a sampled chapter. The +extractor parses the `:before` rules and substitutes each word back in. + +**Chapter URLs cannot be generated.** Numbering has gaps (the sample novel has no chapter 4) +and slugs are not uniform across novels (`/14.html` on one, `/chuong-12.html` on another), so +chapter links are always parsed from the page rather than constructed from a count. + +## Politeness and caching + +Requests are spaced by `-delay` globally, so raising `-workers` does not raise the request +rate. Raw pages are cached under `.cache/`, so re-exporting with different font or page +settings costs no requests. Delete the directory to refetch. + +## Tests + +```sh +go test ./... +``` + +Tests run against synthetic fixtures that reproduce the CSS-injected words, the newest-first +chapter ordering, and the numbering gap. No network access required. + +## Layout + +``` +cmd/monkeyd-crawler/ CLI +internal/monkeyd/ fetching, HTML/CSS parsing, crawl orchestration +internal/pdfout/ PDF rendering and font discovery +``` + +## Scope + +Downloaded text stays on your machine; only fetch content you are allowed to read offline, and +respect the site's terms. diff --git a/cmd/monkeyd-crawler/main.go b/cmd/monkeyd-crawler/main.go new file mode 100644 index 0000000..41c4b18 --- /dev/null +++ b/cmd/monkeyd-crawler/main.go @@ -0,0 +1,184 @@ +// Command monkeyd-crawler downloads every chapter of a monkeydd.com novel and +// exports it as a PDF sized for reading on a phone. +package main + +import ( + "context" + "flag" + "fmt" + "net/url" + "os" + "os/signal" + "path/filepath" + "regexp" + "strings" + "syscall" + "time" + + "github.com/tiennm99/monkeyd-crawler/internal/monkeyd" + "github.com/tiennm99/monkeyd-crawler/internal/pdfout" +) + +type config struct { + novelURL string + out string + page string + fontFile string + fontSize float64 + lineSpacing float64 + margin float64 + workers int + delay time.Duration + retries int + limit int + cacheDir string +} + +func main() { + if err := run(); err != nil { + fmt.Fprintln(os.Stderr, "error:", err) + os.Exit(1) + } +} + +func run() error { + cfg, err := parseFlags() + if err != nil { + return err + } + + // Ctrl-C cancels in-flight fetches instead of leaving a partial PDF. + ctx, stop := signal.NotifyContext(context.Background(), os.Interrupt, syscall.SIGTERM) + defer stop() + + crawler := &monkeyd.Crawler{ + Client: monkeyd.NewClient(cfg.delay, cfg.retries), + CacheDir: cfg.cacheDir, + Workers: cfg.workers, + Log: func(format string, args ...any) { + fmt.Fprintf(os.Stderr, format+"\n", args...) + }, + } + + novel, err := crawler.Novel(ctx, cfg.novelURL) + if err != nil { + return err + } + + if cfg.limit > 0 && cfg.limit < len(novel.Chapters) { + fmt.Fprintf(os.Stderr, "limiting to first %d of %d chapters\n", cfg.limit, len(novel.Chapters)) + novel.Chapters = novel.Chapters[:cfg.limit] + } + + chapters, err := crawler.Chapters(ctx, novel) + if err != nil { + return err + } + + fontFile := cfg.fontFile + if fontFile == "" { + if fontFile, err = pdfout.FindFont(); err != nil { + return err + } + } + + outPath := cfg.out + if outPath == "" { + outPath = safeFileName(novel.Title, novel.Slug) + ".pdf" + } + + opts := pdfout.Options{ + Page: pdfout.Presets[cfg.page], + Margin: cfg.margin, + FontFile: fontFile, + FontSize: cfg.fontSize, + LineSpacing: cfg.lineSpacing, + Title: novel.Title, + SourceURL: novel.URL, + } + if err := pdfout.Write(outPath, opts, toPDFChapters(chapters)); err != nil { + return err + } + + fmt.Fprintf(os.Stderr, "\n%s\n", monkeyd.Describe(novel, chapters)) + fmt.Fprintf(os.Stderr, "font: %s at %.0fpt on %s page (%.0f x %.0f mm)\n", + filepath.Base(fontFile), cfg.fontSize, opts.Page.Name, opts.Page.W, opts.Page.H) + fmt.Println(outPath) + return nil +} + +func parseFlags() (*config, error) { + cfg := &config{} + + flag.StringVar(&cfg.novelURL, "url", "", "novel page URL, e.g. https://monkeydd.com/tro-lai-nam-thang-cu.html") + flag.StringVar(&cfg.out, "out", "", "output PDF path (default: novel title)") + flag.StringVar(&cfg.page, "page", "phone", + "page size: "+strings.Join(pdfout.PresetNames(), ", ")) + flag.StringVar(&cfg.fontFile, "font", "", "path to a .ttf font (default: a Vietnamese-capable system font)") + flag.Float64Var(&cfg.fontSize, "font-size", 12, "body font size in points") + flag.Float64Var(&cfg.lineSpacing, "line-spacing", 1.55, "line height as a multiple of font size") + flag.Float64Var(&cfg.margin, "margin", 6, "page margin in millimetres") + flag.IntVar(&cfg.workers, "workers", 4, "concurrent chapter fetches") + flag.DurationVar(&cfg.delay, "delay", 400*time.Millisecond, "minimum delay between requests") + flag.IntVar(&cfg.retries, "retries", 3, "retries per request") + flag.IntVar(&cfg.limit, "limit", 0, "only fetch the first N chapters (0 = all)") + flag.StringVar(&cfg.cacheDir, "cache", ".cache", + "directory for cached pages, so re-exports need no requests (empty to disable)") + + flag.Usage = func() { + fmt.Fprintf(flag.CommandLine.Output(), + "Download a monkeydd.com novel and export it as a phone-friendly PDF.\n\n"+ + "Usage:\n monkeyd-crawler -url [flags]\n\nFlags:\n") + flag.PrintDefaults() + } + flag.Parse() + + if cfg.novelURL == "" { + flag.Usage() + return nil, fmt.Errorf("-url is required") + } + parsed, err := url.Parse(cfg.novelURL) + if err != nil { + return nil, fmt.Errorf("invalid -url: %w", err) + } + if parsed.Scheme != "http" && parsed.Scheme != "https" { + return nil, fmt.Errorf("invalid -url: want an http(s) URL, got %q", cfg.novelURL) + } + if _, ok := pdfout.Presets[cfg.page]; !ok { + return nil, fmt.Errorf("unknown -page %q: want one of %s", + cfg.page, strings.Join(pdfout.PresetNames(), ", ")) + } + if cfg.fontSize <= 0 { + return nil, fmt.Errorf("-font-size must be positive") + } + if cfg.lineSpacing <= 0 { + return nil, fmt.Errorf("-line-spacing must be positive") + } + if cfg.margin < 0 { + return nil, fmt.Errorf("-margin cannot be negative") + } + if cfg.workers < 1 { + return nil, fmt.Errorf("-workers must be at least 1") + } + return cfg, nil +} + +func toPDFChapters(chapters []*monkeyd.Chapter) []pdfout.Chapter { + out := make([]pdfout.Chapter, 0, len(chapters)) + for _, ch := range chapters { + out = append(out, pdfout.Chapter{Heading: ch.Heading(), Paragraphs: ch.Paragraphs}) + } + return out +} + +var unsafeNameChars = regexp.MustCompile(`[^\p{L}\p{N}]+`) + +// safeFileName builds a file name from the novel title, falling back to the +// slug when the title has no usable characters. +func safeFileName(title, fallback string) string { + name := strings.Trim(unsafeNameChars.ReplaceAllString(title, "-"), "-") + if name == "" { + return fallback + } + return name +} diff --git a/go.mod b/go.mod new file mode 100644 index 0000000..4c5fcb7 --- /dev/null +++ b/go.mod @@ -0,0 +1,9 @@ +module github.com/tiennm99/monkeyd-crawler + +go 1.26.4 + +require ( + github.com/go-pdf/fpdf v0.9.0 + golang.org/x/net v0.57.0 + golang.org/x/sync v0.22.0 +) diff --git a/go.sum b/go.sum new file mode 100644 index 0000000..5038e44 --- /dev/null +++ b/go.sum @@ -0,0 +1,6 @@ +github.com/go-pdf/fpdf v0.9.0 h1:PPvSaUuo1iMi9KkaAn90NuKi+P4gwMedWPHhj8YlJQw= +github.com/go-pdf/fpdf v0.9.0/go.mod h1:oO8N111TkmKb9D7VvWGLvLJlaZUQVPM+6V42pp3iV4Y= +golang.org/x/net v0.57.0 h1:K5+3DljvIuDG9/Jv9rvyMywYNFCQ9RSUY6OOTTkT+tE= +golang.org/x/net v0.57.0/go.mod h1:KpXc8iv+r3XplLAG/f7Jsf9RPszJzdR0f58q9vGOuEU= +golang.org/x/sync v0.22.0 h1:SZjpbeLmrCk4xhRSZFNZW5gFUeCeFgjekvI/+gfScek= +golang.org/x/sync v0.22.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0= diff --git a/internal/monkeyd/chapter.go b/internal/monkeyd/chapter.go new file mode 100644 index 0000000..40d0548 --- /dev/null +++ b/internal/monkeyd/chapter.go @@ -0,0 +1,153 @@ +package monkeyd + +import ( + "bytes" + "fmt" + "strconv" + "strings" + + "golang.org/x/net/html" +) + +// contentElementID is the container holding a chapter's body text. +const contentElementID = "chapter-content-render" + +// Chapter is one fetched chapter, reduced to plain paragraphs. +type Chapter struct { + Label string + URL string + Paragraphs []string +} + +// Heading is the chapter title to print. Labels are often a bare number, which +// reads poorly as a heading, so those get the Vietnamese word for "chapter". +func (c *Chapter) Heading() string { + label := strings.TrimSpace(c.Label) + if label == "" { + return "Chương" + } + if _, err := strconv.Atoi(label); err == nil { + return "Chương " + label + } + return label +} + +// WordCount is a rough word count, used to sanity-check extraction. +func (c *Chapter) WordCount() int { + n := 0 + for _, p := range c.Paragraphs { + n += len(strings.Fields(p)) + } + return n +} + +// blockTags end the current paragraph when opened or closed. +var blockTags = map[string]bool{ + "p": true, "div": true, "br": true, "hr": true, "blockquote": true, + "h1": true, "h2": true, "h3": true, "h4": true, "h5": true, "h6": true, + "li": true, "ul": true, "ol": true, "tr": true, +} + +// skipTags never contribute text: scripts, styles and the ad/report widgets +// the site injects inside the content container. +var skipTags = map[string]bool{ + "script": true, "style": true, "noscript": true, "iframe": true, + "ins": true, "form": true, "select": true, "button": true, "textarea": true, +} + +// ParseChapter extracts a chapter's paragraphs, restoring the words the site +// serves through CSS :before rules instead of markup. +func ParseChapter(page []byte, ref ChapterRef) (*Chapter, error) { + doc, err := html.Parse(bytes.NewReader(page)) + if err != nil { + return nil, fmt.Errorf("parse chapter %s: %w", ref.URL, err) + } + content := elementByID(doc, contentElementID) + if content == nil { + return nil, fmt.Errorf("chapter %s: no #%s container (page layout may have changed)", + ref.URL, contentElementID) + } + + ch := &Chapter{ + Label: ref.Label, + URL: ref.URL, + Paragraphs: extractParagraphs(content, ParseWordClasses(page)), + } + if len(ch.Paragraphs) == 0 { + return nil, fmt.Errorf("chapter %s: extracted no text", ref.URL) + } + ch.Paragraphs = dropRepeatedTitle(ch.Paragraphs, ref.Label) + return ch, nil +} + +// extractParagraphs walks the content subtree into plain paragraphs, replacing +// each word-carrying element with the word its CSS rule injects. +func extractParagraphs(content *html.Node, words map[string]string) []string { + var b strings.Builder + + var walk func(*html.Node) + walk = func(n *html.Node) { + switch n.Type { + case html.TextNode: + // The HTML parser has already decoded entities such as ư. + b.WriteString(n.Data) + return + case html.ElementNode: + if skipTags[n.Data] { + return + } + // These elements are empty in the markup; the CSS word replaces them. + if word, ok := injectedWord(n, words); ok { + b.WriteString(word) + return + } + if blockTags[n.Data] { + b.WriteByte('\n') + } + } + + for c := n.FirstChild; c != nil; c = c.NextSibling { + walk(c) + } + + if n.Type == html.ElementNode && blockTags[n.Data] { + b.WriteByte('\n') + } + } + walk(content) + + var paragraphs []string + for _, line := range strings.Split(b.String(), "\n") { + if p := collapseSpaces(line); p != "" { + paragraphs = append(paragraphs, p) + } + } + return paragraphs +} + +// injectedWord returns the word a node's class supplies via CSS, if any. +func injectedWord(n *html.Node, words map[string]string) (string, bool) { + class := attr(n, "class") + if class == "" { + return "", false + } + for _, tok := range strings.Fields(class) { + if word, ok := words[tok]; ok { + return word, true + } + } + return "", false +} + +// dropRepeatedTitle removes a leading paragraph that only repeats the chapter +// label, since the export prints its own heading. +func dropRepeatedTitle(paragraphs []string, label string) []string { + if len(paragraphs) < 2 { + return paragraphs + } + first := strings.TrimSpace(paragraphs[0]) + if strings.EqualFold(first, strings.TrimSpace(label)) { + return paragraphs[1:] + } + return paragraphs +} diff --git a/internal/monkeyd/chapter_test.go b/internal/monkeyd/chapter_test.go new file mode 100644 index 0000000..94e64a2 --- /dev/null +++ b/internal/monkeyd/chapter_test.go @@ -0,0 +1,129 @@ +package monkeyd + +import ( + "strings" + "testing" +) + +// chapterFixture mirrors the real page shape: words split between markup and +// CSS :before rules, HTML entities,   spacer paragraphs and an ad script +// inside the content container. +const chapterFixture = ` + +

TEN TRUYEN - 1

+
+

1

+

 

+

Nghe trưởng tử noi.

+

 

+

Một di.

+ +
` + +func TestParseWordClassesDecodesEscapes(t *testing.T) { + words := ParseWordClasses([]byte(chapterFixture)) + + for class, want := range map[string]string{ + "t-aaa": "vị", + "j-bbb": "trồ", + "z-ccc": "nàng", + } { + if got := words[class]; got != want { + t.Errorf("words[%q] = %q, want %q", class, got, want) + } + } + if len(words) != 4 { + t.Errorf("got %d rules, want 4", len(words)) + } +} + +func TestParseChapterRestoresCSSWords(t *testing.T) { + ch, err := ParseChapter([]byte(chapterFixture), ChapterRef{Label: "1", URL: "http://x/1.html"}) + if err != nil { + t.Fatalf("ParseChapter: %v", err) + } + + want := []string{ + "Nghe vị trưởng tử noi.", + "Một trồ và nàng di.", + } + if len(ch.Paragraphs) != len(want) { + t.Fatalf("got %d paragraphs %q, want %d", len(ch.Paragraphs), ch.Paragraphs, len(want)) + } + for i, w := range want { + if ch.Paragraphs[i] != w { + t.Errorf("paragraph %d = %q, want %q", i, ch.Paragraphs[i], w) + } + } +} + +// The CSS-injected words are the difference between real text and text with +// silent holes, so guard against a regression that drops them. +func TestParseChapterWithoutCSSWouldLoseWords(t *testing.T) { + withoutCSS := strings.Replace(chapterFixture, `.t-aaa:before { content: "v\1ecb "; }`, "", 1) + + ch, err := ParseChapter([]byte(withoutCSS), ChapterRef{Label: "1", URL: "http://x/1.html"}) + if err != nil { + t.Fatalf("ParseChapter: %v", err) + } + if strings.Contains(ch.Paragraphs[0], "vị") { + t.Fatal("word appeared without its CSS rule; fixture no longer proves anything") + } + if want := "Nghe trưởng tử noi."; ch.Paragraphs[0] != want { + t.Errorf("paragraph 0 = %q, want %q", ch.Paragraphs[0], want) + } +} + +func TestParseChapterDropsScriptsAndSpacers(t *testing.T) { + ch, err := ParseChapter([]byte(chapterFixture), ChapterRef{Label: "1", URL: "http://x/1.html"}) + if err != nil { + t.Fatalf("ParseChapter: %v", err) + } + for _, p := range ch.Paragraphs { + if strings.Contains(p, "ads()") { + t.Errorf("script text leaked into paragraph %q", p) + } + if strings.TrimSpace(p) == "" { + t.Error("empty spacer paragraph was kept") + } + if strings.Contains(p, " ") { + t.Errorf("non-breaking space survived in %q", p) + } + } +} + +// The leading "

1

" repeats the chapter label and would print twice. +func TestParseChapterDropsRepeatedTitle(t *testing.T) { + ch, err := ParseChapter([]byte(chapterFixture), ChapterRef{Label: "1", URL: "http://x/1.html"}) + if err != nil { + t.Fatalf("ParseChapter: %v", err) + } + if ch.Paragraphs[0] == "1" { + t.Error("repeated chapter label was kept as a paragraph") + } +} + +func TestParseChapterMissingContainer(t *testing.T) { + if _, err := ParseChapter([]byte(`

hi

`), + ChapterRef{URL: "http://x/1.html"}); err == nil { + t.Fatal("want an error when the content container is absent") + } +} + +func TestChapterHeading(t *testing.T) { + for _, tc := range []struct{ label, want string }{ + {"14", "Chương 14"}, + {"Chương 12", "Chương 12"}, + {"", "Chương"}, + } { + ch := &Chapter{Label: tc.label} + if got := ch.Heading(); got != tc.want { + t.Errorf("Heading(%q) = %q, want %q", tc.label, got, tc.want) + } + } +} diff --git a/internal/monkeyd/client.go b/internal/monkeyd/client.go new file mode 100644 index 0000000..5dc7cab --- /dev/null +++ b/internal/monkeyd/client.go @@ -0,0 +1,135 @@ +package monkeyd + +import ( + "context" + "errors" + "fmt" + "io" + "net/http" + "sync" + "time" +) + +// The site rejects requests without a browser-like User-Agent. +const defaultUserAgent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 " + + "(KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36" + +// maxPageSize caps how much of a response we buffer; chapter pages are ~130 KB. +const maxPageSize = 8 << 20 + +// statusError reports an unexpected HTTP status so Get can decide whether +// retrying is worthwhile. +type statusError struct { + code int + status string +} + +func (e *statusError) Error() string { return "unexpected status " + e.status } + +// retryable is true for transient failures. A 404 or 403 will not fix itself, +// so those fail immediately instead of burning the retry budget. +func (e *statusError) retryable() bool { + return e.code == http.StatusTooManyRequests || e.code >= 500 +} + +// Client fetches pages from monkeydd.com. It spaces requests out by a fixed +// delay no matter how many goroutines call Get, so raising the worker count +// never raises the request rate, and it retries transient failures with +// exponential backoff. +type Client struct { + http *http.Client + ua string + delay time.Duration + retries int + + mu sync.Mutex + nextSlot time.Time +} + +func NewClient(delay time.Duration, retries int) *Client { + return &Client{ + http: &http.Client{Timeout: 45 * time.Second}, + ua: defaultUserAgent, + delay: delay, + retries: retries, + } +} + +// reserve claims the next request slot and blocks until it comes due, holding +// the global rate at one request per delay across all callers. +func (c *Client) reserve(ctx context.Context) error { + c.mu.Lock() + slot := c.nextSlot + if now := time.Now(); slot.Before(now) { + slot = now + } + c.nextSlot = slot.Add(c.delay) + c.mu.Unlock() + + wait := time.Until(slot) + if wait <= 0 { + return nil + } + timer := time.NewTimer(wait) + defer timer.Stop() + select { + case <-ctx.Done(): + return ctx.Err() + case <-timer.C: + return nil + } +} + +// Get fetches url, retrying transient failures. +func (c *Client) Get(ctx context.Context, url string) ([]byte, error) { + var lastErr error + for attempt := 0; attempt <= c.retries; attempt++ { + if attempt > 0 { + backoff := time.Duration(1< 0 { + c.logf("warning: %d chapter(s) appear only in the chapter dropdown and were not "+ + "in the landing page list; verify the export is complete", len(extra)) + } + if len(final) != len(novel.Chapters) { + c.logf("chapter list reconciled to %d chapters using the in-chapter dropdown", len(final)) + } + novel.Chapters = final + return novel, nil +} + +// Chapters fetches every chapter concurrently and returns them in reading +// order. Any chapter that cannot be fetched or parsed fails the whole run +// rather than yielding a book with a hole in it. +func (c *Crawler) Chapters(ctx context.Context, novel *Novel) ([]*Chapter, error) { + chapters := make([]*Chapter, len(novel.Chapters)) + + workers := c.Workers + if workers < 1 { + workers = 1 + } + + group, groupCtx := errgroup.WithContext(ctx) + group.SetLimit(workers) + + var mu sync.Mutex + done := 0 + + for i, ref := range novel.Chapters { + i, ref := i, ref + group.Go(func() error { + page, err := c.page(groupCtx, ref.URL) + if err != nil { + return err + } + chapter, err := ParseChapter(page, ref) + if err != nil { + return err + } + chapters[i] = chapter + + mu.Lock() + done++ + c.logf("fetched %d/%d: %s (%d words)", done, len(novel.Chapters), + chapter.Heading(), chapter.WordCount()) + mu.Unlock() + return nil + }) + } + + if err := group.Wait(); err != nil { + return nil, err + } + return chapters, nil +} + +// page returns a page from the cache when available, otherwise fetches and +// caches it. +func (c *Crawler) page(ctx context.Context, pageURL string) ([]byte, error) { + path := c.cachePath(pageURL) + if path != "" { + if body, err := os.ReadFile(path); err == nil && len(body) > 0 { + return body, nil + } + } + + body, err := c.Client.Get(ctx, pageURL) + if err != nil { + return nil, err + } + + if path != "" { + if err := os.MkdirAll(filepath.Dir(path), 0o755); err == nil { + // A failed cache write must not fail the crawl. + _ = os.WriteFile(path, body, 0o644) + } + } + return body, nil +} + +// unsafeFileChars matches everything not allowed in a cache file name. +var unsafeFileChars = regexp.MustCompile(`[^A-Za-z0-9._-]+`) + +// cachePath maps a page URL to a cache file, or "" when caching is disabled. +func (c *Crawler) cachePath(pageURL string) string { + if c.CacheDir == "" { + return "" + } + u, err := url.Parse(pageURL) + if err != nil { + return "" + } + name := unsafeFileChars.ReplaceAllString(strings.Trim(u.Path, "/"), "_") + if name == "" { + return "" + } + if !strings.HasSuffix(name, ".html") { + name += ".html" + } + return filepath.Join(c.CacheDir, name) +} + +// TotalWords sums the word count across chapters. +func TotalWords(chapters []*Chapter) int { + n := 0 + for _, ch := range chapters { + n += ch.WordCount() + } + return n +} + +// Describe renders a one-line summary of a crawl result. +func Describe(novel *Novel, chapters []*Chapter) string { + return fmt.Sprintf("%s — %d chapters, %d words", novel.Title, len(chapters), TotalWords(chapters)) +} diff --git a/internal/monkeyd/css_words.go b/internal/monkeyd/css_words.go new file mode 100644 index 0000000..6317b25 --- /dev/null +++ b/internal/monkeyd/css_words.go @@ -0,0 +1,50 @@ +package monkeyd + +import ( + "regexp" + "strconv" + "strings" +) + +// The site hides part of every chapter behind CSS rather than putting it in the +// markup. Chapter HTML carries empty elements such as +// +// Nghe trưởng tử +// +// and the page stylesheet supplies the missing word: +// +// .t-3e625e...:before { content: "vị"; } +// +// Reading DOM text alone therefore drops hundreds of words per chapter without +// any visible error. wordRule finds those rules so the words can be put back. +var wordRule = regexp.MustCompile(`\.([A-Za-z0-9_-]+)\s*::?before\s*\{[^}]*?content\s*:\s*"((?:[^"\\]|\\.)*)"`) + +// cssEscape matches a CSS character escape: a hex code point, optionally +// followed by one whitespace terminator, or an escaped literal character. +var cssEscape = regexp.MustCompile(`\\([0-9A-Fa-f]{1,6})\s?|\\(.)`) + +// ParseWordClasses maps CSS class name to the word its :before rule injects. +func ParseWordClasses(page []byte) map[string]string { + words := make(map[string]string) + for _, m := range wordRule.FindAllSubmatch(page, -1) { + words[string(m[1])] = decodeCSSString(string(m[2])) + } + return words +} + +// decodeCSSString resolves the escape sequences allowed inside a CSS string. +func decodeCSSString(s string) string { + if !strings.Contains(s, `\`) { + return s + } + return cssEscape.ReplaceAllStringFunc(s, func(esc string) string { + m := cssEscape.FindStringSubmatch(esc) + if m[1] != "" { + if cp, err := strconv.ParseInt(m[1], 16, 32); err == nil && cp > 0 { + return string(rune(cp)) + } + return "" + } + return m[2] + }) +} diff --git a/internal/monkeyd/html_nodes.go b/internal/monkeyd/html_nodes.go new file mode 100644 index 0000000..7d5d16d --- /dev/null +++ b/internal/monkeyd/html_nodes.go @@ -0,0 +1,95 @@ +package monkeyd + +import ( + "strings" + + "golang.org/x/net/html" +) + +// attr returns the value of the named attribute, or "" when absent. +func attr(n *html.Node, name string) string { + for _, a := range n.Attr { + if a.Key == name { + return a.Val + } + } + return "" +} + +// hasClass reports whether the node carries the given class token. +func hasClass(n *html.Node, class string) bool { + for _, tok := range strings.Fields(attr(n, "class")) { + if tok == class { + return true + } + } + return false +} + +// findNode returns the first node in document order satisfying match. +func findNode(root *html.Node, match func(*html.Node) bool) *html.Node { + if match(root) { + return root + } + for c := root.FirstChild; c != nil; c = c.NextSibling { + if found := findNode(c, match); found != nil { + return found + } + } + return nil +} + +// findAllNodes returns every node satisfying match, in document order. +func findAllNodes(root *html.Node, match func(*html.Node) bool) []*html.Node { + var out []*html.Node + var walk func(*html.Node) + walk = func(n *html.Node) { + if match(n) { + out = append(out, n) + } + for c := n.FirstChild; c != nil; c = c.NextSibling { + walk(c) + } + } + walk(root) + return out +} + +// elementByID finds an element by its id attribute. +func elementByID(root *html.Node, id string) *html.Node { + return findNode(root, func(n *html.Node) bool { + return n.Type == html.ElementNode && attr(n, "id") == id + }) +} + +// elementByTag finds the first element with the given tag name. +func elementByTag(root *html.Node, tag string) *html.Node { + return findNode(root, func(n *html.Node) bool { + return n.Type == html.ElementNode && n.Data == tag + }) +} + +// nodeText collects the descendant text of a node with whitespace collapsed. +func nodeText(n *html.Node) string { + if n == nil { + return "" + } + var b strings.Builder + var walk func(*html.Node) + walk = func(n *html.Node) { + if n.Type == html.TextNode { + b.WriteString(n.Data) + } + for c := n.FirstChild; c != nil; c = c.NextSibling { + walk(c) + } + } + walk(n) + return collapseSpaces(b.String()) +} + +// collapseSpaces trims the string and reduces every whitespace run, including +// the non-breaking spaces the site emits as  , to a single space. +func collapseSpaces(s string) string { + return strings.Join(strings.Fields(s), " ") +} diff --git a/internal/monkeyd/novel.go b/internal/monkeyd/novel.go new file mode 100644 index 0000000..35a4b9e --- /dev/null +++ b/internal/monkeyd/novel.go @@ -0,0 +1,189 @@ +package monkeyd + +import ( + "bytes" + "fmt" + "net/url" + "strings" + + "golang.org/x/net/html" +) + +// ChapterRef points at a single chapter listed on a novel page. +type ChapterRef struct { + Label string // as shown on the site, e.g. "14" or "Chương 12" + URL string +} + +// Novel is a novel's landing page: its title and its chapters in reading order. +type Novel struct { + Title string + Slug string + URL string + Chapters []ChapterRef +} + +// ParseNovelPage reads the title and chapter list from a novel landing page. +// +// Chapter URLs are always taken from the anchors on the page. Slugs are not +// uniform across novels ("/14.html" on one, "/chuong-12.html" on another) and +// numbering has gaps, so generating URLs from a chapter count would fetch 404s +// and miss real chapters. +func ParseNovelPage(page []byte, pageURL string) (*Novel, error) { + doc, err := html.Parse(bytes.NewReader(page)) + if err != nil { + return nil, fmt.Errorf("parse novel page: %w", err) + } + base, err := url.Parse(pageURL) + if err != nil { + return nil, fmt.Errorf("parse novel url: %w", err) + } + + novel := &Novel{ + Title: novelTitle(doc), + Slug: slugFromNovelURL(base), + URL: pageURL, + Chapters: chapterRefsFromList(doc, base), + } + if novel.Title == "" { + novel.Title = novel.Slug + } + if len(novel.Chapters) == 0 { + return nil, fmt.Errorf("no chapters found on %s (page layout may have changed)", pageURL) + } + return novel, nil +} + +// novelTitle prefers the

heading and falls back to the document title. +func novelTitle(doc *html.Node) string { + if h1 := elementByTag(doc, "h1"); h1 != nil { + if t := nodeText(h1); t != "" { + return t + } + } + return nodeText(elementByTag(doc, "title")) +} + +// slugFromNovelURL turns https://host/tro-lai-nam-thang-cu.html into +// "tro-lai-nam-thang-cu". +func slugFromNovelURL(u *url.URL) string { + seg := strings.Trim(u.Path, "/") + if i := strings.LastIndex(seg, "/"); i >= 0 { + seg = seg[i+1:] + } + return strings.TrimSuffix(seg, ".html") +} + +// chapterRefsFromList reads the "list-chapters" block on the landing page. +// The site lists newest first, so the result is reversed into reading order. +func chapterRefsFromList(doc *html.Node, base *url.URL) []ChapterRef { + list := findNode(doc, func(n *html.Node) bool { + return n.Type == html.ElementNode && hasClass(n, "list-chapters") + }) + if list == nil { + return nil + } + + titles := findAllNodes(list, func(n *html.Node) bool { + return n.Type == html.ElementNode && hasClass(n, "episode-title") + }) + + var refs []ChapterRef + for _, title := range titles { + link := elementByTag(title, "a") + if link == nil { + continue + } + href := strings.TrimSpace(attr(link, "href")) + if href == "" { + continue + } + abs, err := base.Parse(href) + if err != nil { + continue + } + refs = append(refs, ChapterRef{Label: nodeText(link), URL: abs.String()}) + } + return reverseRefs(refs) +} + +// ChapterRefsFromSelect reads the chapter dropdown embedded in every chapter +// page, whose options hold "novel-slug,chapter-slug" pairs. This is a second, +// independent view of the chapter list used to cross-check the landing page. +func ChapterRefsFromSelect(page []byte, base *url.URL) ([]ChapterRef, error) { + doc, err := html.Parse(bytes.NewReader(page)) + if err != nil { + return nil, fmt.Errorf("parse chapter page: %w", err) + } + sel := elementByID(doc, "selected_chapter") + if sel == nil { + return nil, nil + } + + var refs []ChapterRef + for _, opt := range findAllNodes(sel, func(n *html.Node) bool { + return n.Type == html.ElementNode && n.Data == "option" + }) { + novelSlug, chapterSlug, ok := strings.Cut(attr(opt, "value"), ",") + if !ok || novelSlug == "" || chapterSlug == "" { + continue + } + abs, err := base.Parse("/" + novelSlug + "/" + chapterSlug + ".html") + if err != nil { + continue + } + refs = append(refs, ChapterRef{Label: nodeText(opt), URL: abs.String()}) + } + return reverseRefs(refs), nil +} + +// ReconcileChapterRefs picks the chapter list to crawl from the landing page +// list and the in-chapter dropdown. +// +// Both views come from the same site ordering, so when the dropdown is a +// superset it is preferred: that keeps the run correct even if the landing page +// ever truncates or paginates its list. Anything the dropdown alone knows about +// while disagreeing on order is returned as extra so the caller can warn rather +// than silently export a short book. +func ReconcileChapterRefs(fromList, fromSelect []ChapterRef) (final, extra []ChapterRef) { + if len(fromSelect) == 0 { + return fromList, nil + } + + inSelect := refURLSet(fromSelect) + if listIsSubsetOf(fromList, inSelect) { + return fromSelect, nil + } + + inList := refURLSet(fromList) + for _, ref := range fromSelect { + if !inList[ref.URL] { + extra = append(extra, ref) + } + } + return fromList, extra +} + +func refURLSet(refs []ChapterRef) map[string]bool { + set := make(map[string]bool, len(refs)) + for _, r := range refs { + set[r.URL] = true + } + return set +} + +func listIsSubsetOf(refs []ChapterRef, set map[string]bool) bool { + for _, r := range refs { + if !set[r.URL] { + return false + } + } + return true +} + +func reverseRefs(refs []ChapterRef) []ChapterRef { + for i, j := 0, len(refs)-1; i < j; i, j = i+1, j-1 { + refs[i], refs[j] = refs[j], refs[i] + } + return refs +} diff --git a/internal/monkeyd/novel_test.go b/internal/monkeyd/novel_test.go new file mode 100644 index 0000000..112d344 --- /dev/null +++ b/internal/monkeyd/novel_test.go @@ -0,0 +1,119 @@ +package monkeyd + +import ( + "net/url" + "testing" +) + +// novelFixture reproduces the two traits that break naive crawlers: the list is +// newest-first, and chapter numbering has a gap (no chapter 4). +const novelFixture = `TEN TRUYEN +

TEN TRUYEN

+
+ + + + +
` + +func TestParseNovelPageOrdersChaptersForReading(t *testing.T) { + novel, err := ParseNovelPage([]byte(novelFixture), "https://monkeydd.com/n.html") + if err != nil { + t.Fatalf("ParseNovelPage: %v", err) + } + + if novel.Title != "TEN TRUYEN" { + t.Errorf("Title = %q", novel.Title) + } + if novel.Slug != "n" { + t.Errorf("Slug = %q, want %q", novel.Slug, "n") + } + + wantLabels := []string{"1", "2", "3", "5"} + if len(novel.Chapters) != len(wantLabels) { + t.Fatalf("got %d chapters, want %d", len(novel.Chapters), len(wantLabels)) + } + for i, want := range wantLabels { + if novel.Chapters[i].Label != want { + t.Errorf("chapter %d label = %q, want %q", i, novel.Chapters[i].Label, want) + } + } + // Relative hrefs must resolve against the novel URL. + if got, want := novel.Chapters[1].URL, "https://monkeydd.com/n/2.html"; got != want { + t.Errorf("chapter 2 URL = %q, want %q", got, want) + } +} + +func TestParseNovelPageNoChapters(t *testing.T) { + if _, err := ParseNovelPage([]byte(`x`), + "https://monkeydd.com/n.html"); err == nil { + t.Fatal("want an error when no chapters are listed") + } +} + +const selectFixture = ` +` + +func TestChapterRefsFromSelect(t *testing.T) { + base, err := url.Parse("https://monkeydd.com/n.html") + if err != nil { + t.Fatal(err) + } + refs, err := ChapterRefsFromSelect([]byte(selectFixture), base) + if err != nil { + t.Fatalf("ChapterRefsFromSelect: %v", err) + } + + if len(refs) != 3 { + t.Fatalf("got %d refs %+v, want 3 (malformed option skipped)", len(refs), refs) + } + if got, want := refs[0].URL, "https://monkeydd.com/n/1.html"; got != want { + t.Errorf("first ref URL = %q, want %q", got, want) + } + if got, want := refs[1].URL, "https://monkeydd.com/n/chuong-3.html"; got != want { + t.Errorf("second ref URL = %q, want %q", got, want) + } +} + +func TestReconcileChapterRefs(t *testing.T) { + ref := func(u string) ChapterRef { return ChapterRef{Label: u, URL: u} } + + t.Run("dropdown superset wins", func(t *testing.T) { + list := []ChapterRef{ref("a"), ref("b")} + sel := []ChapterRef{ref("a"), ref("b"), ref("c")} + + final, extra := ReconcileChapterRefs(list, sel) + if len(final) != 3 { + t.Errorf("got %d chapters, want the 3 from the dropdown", len(final)) + } + if len(extra) != 0 { + t.Errorf("got %d extra, want 0", len(extra)) + } + }) + + t.Run("disagreement is reported not silently merged", func(t *testing.T) { + list := []ChapterRef{ref("a"), ref("z")} + sel := []ChapterRef{ref("a"), ref("b")} + + final, extra := ReconcileChapterRefs(list, sel) + if len(final) != 2 || final[1].URL != "z" { + t.Errorf("final = %+v, want the landing page list", final) + } + if len(extra) != 1 || extra[0].URL != "b" { + t.Errorf("extra = %+v, want [b] so the caller can warn", extra) + } + }) + + t.Run("empty dropdown falls back to the list", func(t *testing.T) { + list := []ChapterRef{ref("a")} + final, extra := ReconcileChapterRefs(list, nil) + if len(final) != 1 || len(extra) != 0 { + t.Errorf("final = %+v, extra = %+v", final, extra) + } + }) +} diff --git a/internal/pdfout/font.go b/internal/pdfout/font.go new file mode 100644 index 0000000..cd4253b --- /dev/null +++ b/internal/pdfout/font.go @@ -0,0 +1,54 @@ +package pdfout + +import ( + "fmt" + "os" + "path/filepath" + "runtime" +) + +// Vietnamese text needs the Latin Extended Additional block (ư, ạ, ế, ộ …). +// Every font listed here ships with its platform and covers it; fonts with only +// basic Latin would silently drop the diacritics. +func fontCandidates() []string { + switch runtime.GOOS { + case "windows": + dir := filepath.Join(os.Getenv("SystemRoot"), "Fonts") + if os.Getenv("SystemRoot") == "" { + dir = `C:\Windows\Fonts` + } + return []string{ + filepath.Join(dir, "segoeui.ttf"), + filepath.Join(dir, "arial.ttf"), + filepath.Join(dir, "calibri.ttf"), + filepath.Join(dir, "tahoma.ttf"), + filepath.Join(dir, "times.ttf"), + } + case "darwin": + return []string{ + "/System/Library/Fonts/Supplemental/Arial.ttf", + "/Library/Fonts/Arial.ttf", + "/System/Library/Fonts/Supplemental/Times New Roman.ttf", + } + default: + return []string{ + "/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf", + "/usr/share/fonts/truetype/noto/NotoSans-Regular.ttf", + "/usr/share/fonts/TTF/DejaVuSans.ttf", + "/usr/share/fonts/dejavu/DejaVuSans.ttf", + "/usr/share/fonts/liberation/LiberationSans-Regular.ttf", + } + } +} + +// FindFont returns the first available system font suitable for Vietnamese. +func FindFont() (string, error) { + candidates := fontCandidates() + for _, path := range candidates { + if info, err := os.Stat(path); err == nil && !info.IsDir() { + return path, nil + } + } + return "", fmt.Errorf("no Vietnamese-capable system font found (looked for %v); "+ + "pass -font with a path to a .ttf file", candidates) +} diff --git a/internal/pdfout/pdf.go b/internal/pdfout/pdf.go new file mode 100644 index 0000000..d4a2356 --- /dev/null +++ b/internal/pdfout/pdf.go @@ -0,0 +1,144 @@ +package pdfout + +import ( + "fmt" + "sort" + "strings" + + "github.com/go-pdf/fpdf" +) + +// pointsToMM converts typographic points to millimetres. +const pointsToMM = 25.4 / 72.0 + +// bodyFont is the internal family name registered with the PDF. +const bodyFont = "body" + +// PageSize is a page in millimetres. +type PageSize struct { + Name string + W, H float64 +} + +// Presets are the selectable page geometries. +// +// Phone reading depends far more on page shape than on font size. A phone +// viewer scales a whole page to fit the screen, so a large font on an A4 page +// still ends up tiny: the page is ~3x wider than the screen and gets shrunk to +// match. A page cut to the phone's own aspect ratio fills the screen at 100%, +// which is why "phone" is a small 9:16 page rather than A4 with big type. +var Presets = map[string]PageSize{ + "phone": {"phone", 90, 160}, + "a5": {"a5", 148, 210}, + "a4": {"a4", 210, 297}, +} + +// PresetNames lists preset keys in a stable order for help text. +func PresetNames() []string { + names := make([]string, 0, len(Presets)) + for name := range Presets { + names = append(names, name) + } + sort.Strings(names) + return names +} + +// Chapter is a chapter ready to render. +type Chapter struct { + Heading string + Paragraphs []string +} + +// Options controls the exported PDF. +type Options struct { + Page PageSize + Margin float64 // mm + FontFile string + FontSize float64 // pt + LineSpacing float64 // multiple of font size + Title string + SourceURL string +} + +// footerReserve is the vertical space kept clear for the page number. +const footerReserve = 6.0 + +// Write renders the chapters to a PDF at path. +func Write(path string, opts Options, chapters []Chapter) error { + pdf := fpdf.NewCustom(&fpdf.InitType{ + UnitStr: "mm", + Size: fpdf.SizeType{Wd: opts.Page.W, Ht: opts.Page.H}, + }) + + pdf.SetMargins(opts.Margin, opts.Margin, opts.Margin) + pdf.SetAutoPageBreak(true, opts.Margin+footerReserve) + + // AddUTF8Font embeds a subset of the TrueType file, which is what makes the + // Vietnamese diacritics render instead of falling back to "?". + pdf.AddUTF8Font(bodyFont, "", opts.FontFile) + pdf.SetFont(bodyFont, "", opts.FontSize) + pdf.SetTitle(opts.Title, true) + + lineHeight := opts.FontSize * opts.LineSpacing * pointsToMM + paragraphGap := lineHeight * 0.45 + headingSize := opts.FontSize * 1.35 + + addFooter(pdf, opts) + writeTitlePage(pdf, opts, len(chapters)) + + for _, ch := range chapters { + pdf.AddPage() + + pdf.SetFontSize(headingSize) + pdf.MultiCell(0, headingSize*1.3*pointsToMM, ch.Heading, "", "L", false) + pdf.Ln(paragraphGap * 1.6) + + pdf.SetFontSize(opts.FontSize) + for _, p := range ch.Paragraphs { + // "J" justifies, which keeps the short measure of a phone page tidy. + pdf.MultiCell(0, lineHeight, p, "", "J", false) + pdf.Ln(paragraphGap) + } + } + + if err := pdf.OutputFileAndClose(path); err != nil { + return fmt.Errorf("write pdf %s: %w", path, err) + } + return nil +} + +// addFooter prints a centred page number, restoring the body font size so the +// footer callback cannot leak its own size into the following content. +func addFooter(pdf *fpdf.Fpdf, opts Options) { + pdf.SetFooterFunc(func() { + if pdf.PageNo() <= 1 { + return + } + pdf.SetY(-(opts.Margin + footerReserve*0.6)) + pdf.SetFontSize(opts.FontSize * 0.75) + pdf.SetTextColor(120, 120, 120) + pdf.CellFormat(0, 4, fmt.Sprintf("%d", pdf.PageNo()-1), "", 0, "C", false, 0, "") + pdf.SetTextColor(0, 0, 0) + pdf.SetFontSize(opts.FontSize) + }) +} + +func writeTitlePage(pdf *fpdf.Fpdf, opts Options, chapterCount int) { + pdf.AddPage() + pdf.SetY(opts.Page.H * 0.30) + + titleSize := opts.FontSize * 1.9 + pdf.SetFontSize(titleSize) + pdf.MultiCell(0, titleSize*1.35*pointsToMM, strings.ToUpper(opts.Title), "", "C", false) + + pdf.Ln(opts.FontSize * pointsToMM * 2) + pdf.SetFontSize(opts.FontSize * 0.85) + pdf.SetTextColor(90, 90, 90) + pdf.MultiCell(0, opts.FontSize*1.4*pointsToMM, + fmt.Sprintf("%d chương", chapterCount), "", "C", false) + if opts.SourceURL != "" { + pdf.MultiCell(0, opts.FontSize*1.4*pointsToMM, opts.SourceURL, "", "C", false) + } + pdf.SetTextColor(0, 0, 0) + pdf.SetFontSize(opts.FontSize) +} diff --git a/internal/pdfout/pdf_test.go b/internal/pdfout/pdf_test.go new file mode 100644 index 0000000..956ce53 --- /dev/null +++ b/internal/pdfout/pdf_test.go @@ -0,0 +1,96 @@ +package pdfout + +import ( + "os" + "path/filepath" + "testing" +) + +func testOptions(t *testing.T) Options { + t.Helper() + font, err := FindFont() + if err != nil { + t.Skipf("no system font available: %v", err) + } + return Options{ + Page: Presets["phone"], + Margin: 6, + FontFile: font, + FontSize: 12, + LineSpacing: 1.55, + Title: "TRỞ LẠI NĂM THÁNG CŨ", + SourceURL: "https://example.test/n.html", + } +} + +func TestWriteProducesReadablePDF(t *testing.T) { + path := filepath.Join(t.TempDir(), "out.pdf") + + chapters := []Chapter{ + {Heading: "Chương 1", Paragraphs: []string{ + "Nghe vị trưởng tử nói chuyện với nàng.", + "Một đoạn văn khác để kiểm tra ngắt dòng tự động trên trang nhỏ.", + }}, + {Heading: "Chương 2", Paragraphs: []string{"Đoạn cuối."}}, + } + + if err := Write(path, testOptions(t), chapters); err != nil { + t.Fatalf("Write: %v", err) + } + + info, err := os.Stat(path) + if err != nil { + t.Fatalf("stat output: %v", err) + } + if info.Size() == 0 { + t.Fatal("wrote an empty PDF") + } + + header := make([]byte, 5) + f, err := os.Open(path) + if err != nil { + t.Fatal(err) + } + defer f.Close() + if _, err := f.Read(header); err != nil { + t.Fatal(err) + } + if string(header) != "%PDF-" { + t.Errorf("output does not start with a PDF header, got %q", header) + } +} + +// A novel with many chapters must not overflow a page; auto page break plus the +// footer reserve handles that, so a long chapter should span several pages. +func TestWriteHandlesLongChapters(t *testing.T) { + path := filepath.Join(t.TempDir(), "long.pdf") + + paragraphs := make([]string, 200) + for i := range paragraphs { + paragraphs[i] = "Một đoạn văn dài để buộc trình kết xuất phải sang trang mới nhiều lần." + } + + if err := Write(path, testOptions(t), []Chapter{{Heading: "Chương 1", Paragraphs: paragraphs}}); err != nil { + t.Fatalf("Write: %v", err) + } + info, err := os.Stat(path) + if err != nil { + t.Fatal(err) + } + if info.Size() < 2000 { + t.Errorf("output suspiciously small (%d bytes) for 200 paragraphs", info.Size()) + } +} + +func TestPresetNamesIsStable(t *testing.T) { + got := PresetNames() + want := []string{"a4", "a5", "phone"} + if len(got) != len(want) { + t.Fatalf("PresetNames() = %v, want %v", got, want) + } + for i := range want { + if got[i] != want[i] { + t.Errorf("PresetNames()[%d] = %q, want %q", i, got[i], want[i]) + } + } +}