Parserfixes std (#11)

* clean svg and cdata
* refactored ebook preparation, separated from calling the pager
* added better unit tests
* add free ebooks for testing

README.md

@@ -1,3 +1,8 @@
+[![Actions](https://github.com/tlinden/epuppy/actions/workflows/ci.yaml/badge.svg)](https://github.com/tlinden/epuppy/actions)
+[![License](https://img.shields.io/badge/license-GPL-blue.svg)](https://github.com/tlinden/epuppy/blob/master/LICENSE)
+[![Go Report Card](https://goreportcard.com/badge/github.com/tlinden/epuppy)](https://goreportcard.com/report/github.com/tlinden/epuppy)
+
+
 # epuppy - terminal epub reader
 
 This is a little TUI epub ebook reader. This is a work in progress and

cmd/config.go

@@ -32,7 +32,7 @@ import (
 )
 
 const (
-	Version string = `v0.0.4`
+	Version string = `v0.0.7`
 	Usage   string = `This is epuppy, a terminal ui ebook viewer.
 
 Usage: epuppy [options] <epub file>
@@ -42,6 +42,7 @@ Options:
 -s --store-progress      remember reading position
 -n --line-numbers        add line numbers
 -c --config <file>       use config <file>
+-i --cover-image         display cover image
 -t --txt                 dump readable content to STDOUT
 -x --xml                 dump source xml to STDOUT
 -N --no-color            disable colors (or use $NO_COLOR env var)
@@ -63,6 +64,7 @@ type Config struct {
 	ColorDark       ColorSetting `koanf:"colordark"`      // comes from config file only
 	ColorLight      ColorSetting `koanf:"colorlight"`     // comes from config file only
 	ShowHelp        bool         `koanf:"help"`
+	ShowCover       bool         `koanf:"cover-image"` // -i
 	Colors          Colors       // generated from user config file or internal defaults, respects dark mode
 	Document        string
 	InitialProgress int // lines
@@ -89,6 +91,7 @@ func InitConfig(output io.Writer) (*Config, error) {
 	flagset.BoolP("txt", "t", false, "dump readable content to STDOUT")
 	flagset.BoolP("xml", "x", false, "dump xml to STDOUT")
 	flagset.BoolP("no-color", "N", false, "disable colors")
+	flagset.BoolP("cover-image", "i", false, "show cover image")
 	flagset.BoolP("help", "h", false, "show help")
 	flagset.StringP("config", "c", "", "read config from file")
 

cmd/pager.go

@@ -21,10 +21,13 @@ package cmd
 // https://github.com/charmbracelet/bubbletea/tree/main/examples/pager
 
 import (
+	"bytes"
 	"fmt"
+	"image"
 	"os"
 	"strings"
 
+	"github.com/blacktop/go-termimg"
 	"github.com/charmbracelet/bubbles/help"
 	"github.com/charmbracelet/bubbles/key"
 	"github.com/charmbracelet/bubbles/viewport"
@@ -132,6 +135,7 @@ type Doc struct {
 	hideUi       bool
 	meta         *Meta
 	config       *Config
+	Cover        *termimg.ImageWidget
 
 	keys keyMap
 	help help.Model
@@ -220,7 +224,12 @@ func (m *Doc) Rewrap() {
 
 	// wrapping has changed, update viewport and line count
 	if content != "" {
-		m.viewport.SetContent(content)
+		if m.Cover != nil {
+			cover, _ := m.Cover.Render()
+			m.viewport.SetContent(cover + "\n\n" + content)
+		} else {
+			m.viewport.SetContent(content)
+		}
 		m.meta.lines = len(strings.Split(content, "\n"))
 	}
 
@@ -270,7 +279,15 @@ func max(a, b int) int {
 	return b
 }
 
-func Pager(conf *Config, title, message string) (int, error) {
+type Ebook struct {
+	Config    *Config
+	Title     string
+	Body      string
+	Cover     []byte
+	MediaType string
+}
+
+func Pager(book *Ebook) (int, error) {
 	width := 80
 	scrollto := 0
 
@@ -281,32 +298,44 @@ func Pager(conf *Config, title, message string) (int, error) {
 		}
 	}
 
-	if conf.StoreProgress {
+	if book.Config.StoreProgress {
-		scrollto = conf.InitialProgress
+		scrollto = book.Config.InitialProgress
 	}
 
-	if conf.LineNumbers {
+	if book.Config.LineNumbers {
 		catn := ""
-		for idx, line := range strings.Split(message, "\n") {
+		for idx, line := range strings.Split(book.Body, "\n") {
 			catn += fmt.Sprintf("%4d: %s\n", idx, line)
 		}
-		message = catn
+		book.Body = catn
 	}
 
 	meta := Meta{
 		initialprogress: scrollto,
-		lines:           len(strings.Split(message, "\n")),
+		lines:           len(strings.Split(book.Body, "\n")),
+	}
+
+	doc := Doc{
+		content:      book.Body,
+		title:        book.Title,
+		initialwidth: width,
+		meta:         &meta,
+		config:       book.Config,
+		keys:         keys,
+	}
+
+	if book.MediaType != "" && book.Config.ShowCover {
+		img, _, err := image.Decode(bytes.NewReader(book.Cover))
+		if err != nil {
+			return 0, fmt.Errorf("failed to load cover image: %w", err)
+		}
+
+		doc.Cover = termimg.NewImageWidgetFromImage(img)
+		doc.Cover.SetSize(width, width).SetProtocol(termimg.Auto)
 	}
 
 	p := tea.NewProgram(
-		Doc{
+		doc,
-			content:      message,
-			title:        title,
-			initialwidth: width,
-			meta:         &meta,
-			config:       conf,
-			keys:         keys,
-		},
 		tea.WithAltScreen(),       // use the full size of the terminal in its "alternate screen buffer"
 		tea.WithMouseCellMotion(), // turn on mouse support so we can track the mouse wheel
 	)
@@ -315,7 +344,7 @@ func Pager(conf *Config, title, message string) (int, error) {
 		return 0, fmt.Errorf("could not run pager: %w", err)
 	}
 
-	if conf.Debug {
+	if book.Config.Debug {
 		fmt.Printf("scrollto: %d, last: %d, diff: %d\n",
 			scrollto, meta.currentline, scrollto-meta.currentline)
 	}

cmd/prepare.go

@@ -22,35 +22,44 @@ import (
 	"path/filepath"
 	"strings"
 
+	"github.com/alecthomas/repr"
 	"github.com/tlinden/epuppy/pkg/epub"
 )
 
-func View(conf *Config) (int, error) {
+func Prepare(conf *Config) (*Ebook, error) {
 	switch filepath.Ext(conf.Document) {
 	case ".epub":
-		return ViewEpub(conf)
+		return PrepareEpub(conf)
 	default:
-		return ViewText(conf)
+		return PrepareText(conf)
 	}
 }
 
-func ViewText(conf *Config) (int, error) {
+func PrepareText(conf *Config) (*Ebook, error) {
 	data, err := os.ReadFile(conf.Document)
 	if err != nil {
-		return 0, err
+		return nil, err
 	}
 
-	if conf.Dump {
+	return &Ebook{
-		return fmt.Println(string(data))
+		Config: conf,
-	}
+		Title:  conf.Document,
-
+		Body:   string(data),
-	return Pager(conf, conf.Document, string(data))
+	}, nil
 }
 
-func ViewEpub(conf *Config) (int, error) {
+func PrepareEpub(conf *Config) (*Ebook, error) {
 	book, err := epub.Open(conf.Document, conf.XML)
 	if err != nil {
-		return 0, err
+		return nil, err
+	}
+
+	if conf.Debug {
+		repr.Println("book.Files()")
+		repr.Println(book.Files())
+		repr.Println(book.Ncx)
+		repr.Println(book.Sections)
+		repr.Println(book.Opf.Manifest)
 	}
 
 	buf := strings.Builder{}
@@ -70,15 +79,15 @@ func ViewEpub(conf *Config) (int, error) {
 
 	fetchByContent(conf, &buf, book)
 
-	if conf.Dump {
+	return &Ebook{
-		return fmt.Println(buf.String())
+		Config:    conf,
-	}
+		Title:     head.String(),
-
+		Body:      buf.String(),
-	return Pager(conf, head.String(), buf.String())
+		Cover:     book.CoverImage,
+		MediaType: book.CoverMediaType,
+	}, nil
 }
 
-// FIXME:  since the  switch to  book.Files() in  epub.Open() this
-// returns invalid chapter numbering
 func fetchByContent(conf *Config, buf *strings.Builder, book *epub.Book) bool {
 	chapter := 1
 	var gotbody bool

cmd/root.go

@@ -59,7 +59,21 @@ func Execute(output io.Writer) int {
 		}
 	}
 
-	progress, err := View(conf)
+	ebook, err := Prepare(conf)
+	if err != nil {
+		return Die(err)
+	}
+
+	if conf.Dump {
+		fmt.Println(ebook.Body)
+		return 0
+	}
+
+	if conf.Debug || conf.XML {
+		return 0
+	}
+
+	progress, err := Pager(ebook)
 	if err != nil {
 		return Die(err)
 	}

cmd/view_test.go

@@ -0,0 +1,118 @@
+package cmd
+
+import (
+	"fmt"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestPrepare(t *testing.T) {
+	var tests = []struct {
+		file string
+		body string
+	}{
+		{
+			"t/epub/basic-v3plus2.epub",
+			`shirt court, an whinny retched a cordage offer groin-murder, picked inner windrow,`,
+		},
+		{
+			"t/epub/childrens-literature.epub",
+			`The child's natural literature. The world has lost certain secrets as the price of an advancing civilization. It is a commonplace of observation that no one can duplicate the success of Mother Goose, whether she be thought of as the maker of jingles or the teller of tales. The conditions of modern life preclude the generally naïve attitude that produced the folk rhymes, ballads, tales, proverbs, fables, and myths. The folk saw things simply and directly. The complex, analytic, questioning mind is not yet, either in or out of stories. The motives from which people act are to them plain and not mixed. Characters are good or bad. They feel no need of elaborately explaining their joys and sorrows. Such experiences come with the day's work. "To-morrow to fresh woods, and pastures new." The zest of life with them is emphatic. Their humor is fresh, unbounded, sincere; there is no trace of cynicism. In folk literature we do not feel the presence of a "writer" who is mightily concerned about maintaining his reputation for wisdom, originality, or style. Hence the freedom from any note of straining after effect, of artificiality. In the midst of a life limited to fundamental needs, their literature deals with fundamentals. On the whole, it was a literature for entertainment. A more learned upper class may have concerned itself then about "problems" and "purposes," as the whole world does now, but the literature of the folk had no such interests.`,
+		},
+		{
+			"t/epub/cole-voyage-of-life.epub",
+			`Thomas Cole is regarded as the founder of the Hudson River School, an American art movement that flourished in the mid-19th century and was concerned with the realistic and detailed portrayal of nature but with a strong influence from Romanticism. This group of American landscape painters worked between about 1825 and 1870 and shared a sense of national pride as well as an interest in celebrating the unique natural beauty found in the United States. The wild, untamed nature found in America was viewed as its special character; Europe had ancient ruins, but America had the uncharted wilderness. As Cole's friend William Cullen Bryant sermonized in verse, so Cole sermonized in paint. Both men saw nature as God's work and as a refuge from the ugly materialism of cities. Cole clearly intended the Voyage of Life to be a didactic, moralizing series of paintings using the landscape as an allegory for religious faith.`,
+		},
+		{
+			"t/epub/epub30-spec.epub",
+			`IDPF Members
+
+Invited Experts/Observers
+
+Version 2.0.1 of this specification was prepared by the International Digital Publishing Forum’s EPUB Maintenance Working Group under the leadership of:
+
+Active members of the working group at the time of publication of revision 2.0.1 were:
+
+Version 1.0 of this specification was prepared by the International Digital Publishing Forum’s Unified OEBPS Container Format Working Group under the leadership of:
+
+Active members of the working group at the time of publication of revision 1.0 were:`,
+		},
+		{
+			"t/epub/epub_sample_file_50KB.epub",
+			`magna aliqua. Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna`,
+		},
+		{
+			"t/epub/Fundamental-Accessibility-Tests-Basic-Functionality-v2.0.0.epub",
+			`Open the Fundamental Accessibility Test book in the reading system.
+
+If the test book is not available in the bookshelf, then open any other book that is available. 
+
+If the reading system also supports side loading, then please provide notes about the accessibility of the side loading feature.
+
+Indicate Pass or Fail.`,
+		},
+		{
+			"t/epub/georgia-cfi.epub",
+			`The Great Valley Region consists of folded sedimentary rocks, extensive erosion having removed the soft layers to form valleys, leaving the hard layers as ridges, both layers running in a N.E.-S.W. direction. In the extreme north-west corner of the state is a small part of the Cumberland Plateau, represented by Lookout and Sand Mts.
+
+On the Blue Ridge escarpment near the N.E. corner of the state is a water-parting separating the waters which find their way respectively N.W. to the Tennessee river, S.W. to the Gulf of Mexico and S.E. to the Atlantic Ocean; indeed, according to B.M. and M.R. Hall (Water Resources of Georgia, p. 2), "there are three springs in north-east Georgia within a stone's throw of each other that send out their waters to Savannah, Ga., to Apalachicola, Fla., and to New Orleans, La." The water-parting between the waters flowing into the`,
+		},
+		{
+			"t/epub/israelsailing.epub",
+			` במשלוח דואר, מה שלפעמים היה נחמד כי גם חשבונות לתשלום לא היו מגיעים. 'טוב, מי`,
+		},
+		{
+			"t/epub/jlreq-in-japanese.epub",
+			` 2.5.1 基本版面からはみ出す例 2.5.2 基本版面で設定した行位置の適用 2.5.3 `,
+		},
+		{
+			"t/epub/minimal-v2.epub",
+			`This is a paragraph.`,
+		},
+		{
+			"t/epub/minimal-v3.epub",
+			`This is a paragraph.`,
+		},
+		{
+			"t/epub/minimal-v3plus2.epub",
+			`This is a paragraph.`,
+		},
+		{
+			"t/epub/moby-dick.epub",
+			`Call me Ishmael. Some years ago—never mind how long precisely—having little or no money in my purse, and nothing particular to interest me on shore, I thought I would sail about a little and see the watery part of the world. It is a way I have of driving off the spleen and regulating the circulation. Whenever I find myself growing grim about the mouth; whenever it is a damp, drizzly November in my soul; whenever I find myself involuntarily pausing before coffin warehouses, and bringing up the rear of every funeral I meet; and especially whenever my hypos get such an upper hand of me, that it requires a strong moral principle to prevent me from deliberately stepping into the street, and methodically knocking people’s hats off—then, I account it high time to get to sea as soon as I can. This is my substitute for pistol and ball. With a philosophical flourish Cato throws himself upon his sword; I quietly take to the ship. There is nothing surprising in this. If they but knew it, almost all men in their degree, some time or other, cherish very nearly the same feelings towards the ocean with me.`,
+		},
+		{
+			"t/epub/sous-le-vent.epub",
+			`SOUS LE VENT`,
+		},
+		{
+			"t/epub/wasteland-otf.epub",
+			`Line 20. Cf. Ezekiel 2:1.
+
+23. Cf. Ecclesiastes 12:5.
+
+31. V. Tristan und Isolde, i, verses 5-8.
+
+42. Id. iii, verse 24.`,
+		},
+		{
+			"pkg/epub/test.epub",
+			`This EPUB file contains 10 hard coded page breaks. Note that these page breaks are different from the reflowed page numbers. If the total number of pages of this book in the reading app is not exactly 10, then you are looking at the reflowed pages. The app may be having a feature to switch between print page and reflowed pages for navigation. Or the print page list may be appearing in the TOC. If print page navigation feature does not exist or does not work then this test should be marked 'Fail'.`,
+		},
+	}
+
+	for _, tt := range tests {
+		testname := fmt.Sprintf("prepare/%s", tt.file)
+		t.Run(testname, func(t *testing.T) {
+			conf := Config{
+				Document: "../" + tt.file,
+			}
+
+			ebook, err := Prepare(&conf)
+
+			assert.NoError(t, err)
+			assert.Contains(t, ebook.Body, tt.body, "expected text not found")
+		})
+	}
+}

go.mod

@@ -41,14 +41,25 @@ require (
 
 require (
 	github.com/antchfx/xpath v1.3.5 // indirect
+	github.com/blacktop/go-termimg v0.1.20 // indirect
+	github.com/charmbracelet/x/mosaic v0.0.0-20250702191427-5bdfc8f2e4ff // indirect
+	github.com/davecgh/go-spew v1.1.1 // indirect
 	github.com/fsnotify/fsnotify v1.9.0 // indirect
 	github.com/go-viper/mapstructure/v2 v2.4.0 // indirect
 	github.com/golang/groupcache v0.0.0-20210331224755-41bb18bfe9da // indirect
 	github.com/knadh/koanf/maps v0.1.2 // indirect
+	github.com/makeworld-the-better-one/dither/v2 v2.4.0 // indirect
+	github.com/mattn/go-sixel v0.0.5 // indirect
 	github.com/mitchellh/copystructure v1.2.0 // indirect
 	github.com/mitchellh/reflectwalk v1.0.2 // indirect
+	github.com/nfnt/resize v0.0.0-20180221191011-83c6a9932646 // indirect
 	github.com/pelletier/go-toml v1.9.5 // indirect
+	github.com/pmezard/go-difflib v1.0.0 // indirect
 	github.com/rogpeppe/go-internal v1.14.1 // indirect
+	github.com/soniakeys/quant v1.0.0 // indirect
+	github.com/stretchr/testify v1.11.1 // indirect
+	golang.org/x/image v0.25.0 // indirect
 	golang.org/x/net v0.44.0 // indirect
 	golang.org/x/tools v0.37.0 // indirect
+	gopkg.in/yaml.v3 v3.0.1 // indirect
 )

go.sum

@@ -8,6 +8,8 @@ github.com/aymanbagabas/go-osc52/v2 v2.0.1 h1:HwpRHbFMcZLEVr42D4p7XBqjyuxQH5SMiE
 github.com/aymanbagabas/go-osc52/v2 v2.0.1/go.mod h1:uYgXzlJ7ZpABp8OJ+exZzJJhRNQ2ASbcXHWsFqH8hp8=
 github.com/aymanbagabas/go-udiff v0.2.0 h1:TK0fH4MteXUDspT88n8CKzvK0X9O2xu9yQjWpi6yML8=
 github.com/aymanbagabas/go-udiff v0.2.0/go.mod h1:RE4Ex0qsGkTAJoQdQQCA0uG+nAzJO/pI/QwceO5fgrA=
+github.com/blacktop/go-termimg v0.1.20 h1:+EAUc3c9hwE/fUYaqRV1BSLvAlOuLySgLTEBzxGbYK4=
+github.com/blacktop/go-termimg v0.1.20/go.mod h1:nwxrOjfFcBjtS358oIGBLfscSLnCpNdRlMVRxsnZwMU=
 github.com/charmbracelet/bubbles v0.21.0 h1:9TdC97SdRVg/1aaXNVWfFH3nnLAwOXr8Fn6u6mfQdFs=
 github.com/charmbracelet/bubbles v0.21.0/go.mod h1:HF+v6QUR4HkEpz62dx7ym2xc71/KBHg+zKwJtMw+qtg=
 github.com/charmbracelet/bubbletea v1.3.10 h1:otUDHWMMzQSB0Pkc87rm691KZ3SWa4KUlvF9nRvCICw=
@@ -22,8 +24,11 @@ github.com/charmbracelet/x/cellbuf v0.0.13-0.20250311204145-2c3ea96c31dd h1:vy0G
 github.com/charmbracelet/x/cellbuf v0.0.13-0.20250311204145-2c3ea96c31dd/go.mod h1:xe0nKWGd3eJgtqZRaN9RjMtK7xUYchjzPr7q6kcvCCs=
 github.com/charmbracelet/x/exp/golden v0.0.0-20241011142426-46044092ad91 h1:payRxjMjKgx2PaCWLZ4p3ro9y97+TVLZNaRZgJwSVDQ=
 github.com/charmbracelet/x/exp/golden v0.0.0-20241011142426-46044092ad91/go.mod h1:wDlXFlCrmJ8J+swcL/MnGUuYnqgQdW9rhSD61oNMb6U=
+github.com/charmbracelet/x/mosaic v0.0.0-20250702191427-5bdfc8f2e4ff h1:OVBKPzoa0k5ZVMoor27BReRZxER1IEDtLHXkRjaHElg=
+github.com/charmbracelet/x/mosaic v0.0.0-20250702191427-5bdfc8f2e4ff/go.mod h1:5qLP4S++M5quSc/xbvWWW8vKkgKwOqOT/IVhAas26XI=
 github.com/charmbracelet/x/term v0.2.1 h1:AQeHeLZ1OqSXhrAWpYUtZyX1T3zVxfpZuEQMIQaGIAQ=
 github.com/charmbracelet/x/term v0.2.1/go.mod h1:oQ4enTYFV7QN4m0i9mzHrViD7TQKvNEEkHUMCmsxdUg=
+github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/erikgeiser/coninput v0.0.0-20211004153227-1c3628e74d0f h1:Y/CXytFA4m6baUTXGLOoWe4PQhGxaX0KpnayAqC48p4=
@@ -47,6 +52,8 @@ github.com/knadh/koanf/v2 v2.3.0 h1:Qg076dDRFHvqnKG97ZEsi9TAg2/nFTa9hCdcSa1lvlM=
 github.com/knadh/koanf/v2 v2.3.0/go.mod h1:gRb40VRAbd4iJMYYD5IxZ6hfuopFcXBpc9bbQpZwo28=
 github.com/lucasb-eyer/go-colorful v1.2.0 h1:1nnpGOrhyZZuNyfu1QjKiUICQ74+3FNCN69Aj6K7nkY=
 github.com/lucasb-eyer/go-colorful v1.2.0/go.mod h1:R4dSotOR9KMtayYi1e77YzuveK+i7ruzyGqttikkLy0=
+github.com/makeworld-the-better-one/dither/v2 v2.4.0 h1:Az/dYXiTcwcRSe59Hzw4RI1rSnAZns+1msaCXetrMFE=
+github.com/makeworld-the-better-one/dither/v2 v2.4.0/go.mod h1:VBtN8DXO7SNtyGmLiGA7IsFeKrBkQPze1/iAeM95arc=
 github.com/mattn/go-isatty v0.0.20 h1:xfD0iDuEKnDkl03q4limB+vH+GxLEtL/jb4xVJSWWEY=
 github.com/mattn/go-isatty v0.0.20/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
 github.com/mattn/go-localereader v0.0.1 h1:ygSAOl7ZXTx4RdPYinUpg6W99U8jWvWi9Ye2JC/oIi4=
@@ -54,6 +61,8 @@ github.com/mattn/go-localereader v0.0.1/go.mod h1:8fBrzywKY7BI3czFoHkuzRoWE9C+Ei
 github.com/mattn/go-runewidth v0.0.12/go.mod h1:RAqKPSqVFrSLVXbA8x7dzmKdmGzieGRCM46jaSJTDAk=
 github.com/mattn/go-runewidth v0.0.16 h1:E5ScNMtiwvlvB5paMFdw9p4kSQzbXFikJ5SQO6TULQc=
 github.com/mattn/go-runewidth v0.0.16/go.mod h1:Jdepj2loyihRzMpdS35Xk/zdY8IAYHsh153qUoGf23w=
+github.com/mattn/go-sixel v0.0.5 h1:55w2FR5ncuhKhXrM5ly1eiqMQfZsnAHIpYNGZX03Cv8=
+github.com/mattn/go-sixel v0.0.5/go.mod h1:h2Sss+DiUEHy0pUqcIB6PFXo5Cy8sTQEFr3a9/5ZLNw=
 github.com/mitchellh/copystructure v1.2.0 h1:vpKXTN4ewci03Vljg/q9QvCGUDttBOGBIa15WveJJGw=
 github.com/mitchellh/copystructure v1.2.0/go.mod h1:qLl+cE2AmVv+CoeAwDPye/v+N2HKCj9FbZEVFJRxO9s=
 github.com/mitchellh/reflectwalk v1.0.2 h1:G2LzWKi524PWgd3mLHV8Y5k7s6XUvT0Gef6zxSIeXaQ=
@@ -66,6 +75,8 @@ github.com/muesli/reflow v0.3.0 h1:IFsN6K9NfGtjeggFP+68I4chLZV2yIKsXJFNZ+eWh6s=
 github.com/muesli/reflow v0.3.0/go.mod h1:pbwTDkVPibjO2kyvBQRBxTWEEGDGq0FlB1BIKtnHY/8=
 github.com/muesli/termenv v0.16.0 h1:S5AlUN9dENB57rsbnkPyfdGuWIlkmzJjbFf0Tf5FWUc=
 github.com/muesli/termenv v0.16.0/go.mod h1:ZRfOIKPFDYQoDFF4Olj7/QJbW60Ol/kL1pU3VfY/Cnk=
+github.com/nfnt/resize v0.0.0-20180221191011-83c6a9932646 h1:zYyBkD/k9seD2A7fsi6Oo2LfFZAehjjQMERAvZLEDnQ=
+github.com/nfnt/resize v0.0.0-20180221191011-83c6a9932646/go.mod h1:jpp1/29i3P1S/RLdc7JQKbRpFeM1dOBd8T9ki5s+AY8=
 github.com/pelletier/go-toml v1.9.5 h1:4yBQzkHv+7BHq2PQUZF3Mx0IYxG7LsP222s7Agd3ve8=
 github.com/pelletier/go-toml v1.9.5/go.mod h1:u1nR/EPcESfeI/szUZKdtJ0xRNbUoANCkoOuaOx1Y+c=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
@@ -76,10 +87,16 @@ github.com/rivo/uniseg v0.4.7 h1:WUdvkW8uEhrYfLC4ZzdpI2ztxP1I582+49Oc5Mq64VQ=
 github.com/rivo/uniseg v0.4.7/go.mod h1:FN3SvrM+Zdj16jyLfmOkMNblXMcoc8DfTHruCPUcx88=
 github.com/rogpeppe/go-internal v1.14.1 h1:UQB4HGPB6osV0SQTLymcB4TgvyWu6ZyliaW0tI/otEQ=
 github.com/rogpeppe/go-internal v1.14.1/go.mod h1:MaRKkUm5W0goXpeCfT7UZI6fk/L7L7so1lCWt35ZSgc=
+github.com/soniakeys/quant v1.0.0 h1:N1um9ktjbkZVcywBVAAYpZYSHxEfJGzshHCxx/DaI0Y=
+github.com/soniakeys/quant v1.0.0/go.mod h1:HI1k023QuVbD4H8i9YdfZP2munIHU4QpjsImz6Y6zds=
 github.com/spf13/pflag v1.0.10 h1:4EBh2KAYBwaONj6b2Ye1GiHfwjqyROoF4RwYO+vPwFk=
 github.com/spf13/pflag v1.0.10/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
+github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/testify v1.6.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
 github.com/stretchr/testify v1.8.1 h1:w7B6lhMri9wdJUVmEZPGGhZzrYTPvgJArz7wNPgYKsk=
 github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o6fzry7u4=
+github.com/stretchr/testify v1.11.1 h1:7s2iGBzp5EwR7/aIZr8ao5+dra3wiQyKjjFuvgVKu7U=
+github.com/stretchr/testify v1.11.1/go.mod h1:wZwfW3scLgRK+23gO65QZefKpKQRnfz6sD981Nm4B6U=
 github.com/xo/terminfo v0.0.0-20220910002029-abceb7e1c41e h1:JVG44RsyaB9T2KIHavMF/ppJZNG9ZpyihvCd0w101no=
 github.com/xo/terminfo v0.0.0-20220910002029-abceb7e1c41e/go.mod h1:RbqR21r5mrJuqunuUZ/Dhy/avygyECGrLceyNeo4LiM=
 github.com/yuin/goldmark v1.4.13/go.mod h1:6yULJ656Px+3vBD8DxQVa3kxgyrAnzto9xy5taEt/CY=
@@ -91,6 +108,8 @@ golang.org/x/crypto v0.23.0/go.mod h1:CKFgDieR+mRhux2Lsu27y0fO304Db0wZe70UKqHu0v
 golang.org/x/crypto v0.31.0/go.mod h1:kDsLvtWBEx7MV9tJOj9bnXsPbxwJQ6csT/x4KIN4Ssk=
 golang.org/x/exp v0.0.0-20220909182711-5c715a9e8561 h1:MDc5xs78ZrZr3HMQugiXOAkSZtfTpbJLDr/lwfgO53E=
 golang.org/x/exp v0.0.0-20220909182711-5c715a9e8561/go.mod h1:cyybsKvd6eL0RnXn6p/Grxp8F5bW7iYuBgsNCOHpMYE=
+golang.org/x/image v0.25.0 h1:Y6uW6rH1y5y/LK1J8BPWZtr6yZ7hrsy6hFrXjgsc2fQ=
+golang.org/x/image v0.25.0/go.mod h1:tCAmOEGthTtkalusGp1g3xa2gke8J6c2N565dTyl9Rs=
 golang.org/x/mod v0.6.0-dev.0.20220419223038-86c51ed26bb4/go.mod h1:jJ57K6gSWd91VN4djpZkiMVwK6gcyfeH4XE8wZrZaV4=
 golang.org/x/mod v0.8.0/go.mod h1:iBbtSCu2XBx23ZKBPSOrRkjjQPZFPuis4dIYUhu/chs=
 golang.org/x/mod v0.12.0/go.mod h1:iBbtSCu2XBx23ZKBPSOrRkjjQPZFPuis4dIYUhu/chs=
@@ -161,5 +180,7 @@ golang.org/x/tools v0.21.1-0.20240508182429-e35e4ccd0d2d/go.mod h1:aiJjzUbINMkxb
 golang.org/x/tools v0.37.0 h1:DVSRzp7FwePZW356yEAChSdNcQo6Nsp+fex1SUW09lE=
 golang.org/x/tools v0.37.0/go.mod h1:MBN5QPQtLMHVdvsbtarmTNukZDdgwdwlO5qGacAzF0w=
 golang.org/x/xerrors v0.0.0-20190717185122-a985d3407aa7/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
+gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
 gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
 gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=

pkg/epub/book.go

@@ -9,14 +9,24 @@ import (
 	"path"
 )
 
+// a section in the book
+type Section struct {
+	File, Title, MediaType string
+}
+
 // Book epub book
 type Book struct {
-	Ncx       Ncx       `json:"ncx"`
+	Ncx            Ncx       `json:"ncx"`
-	Opf       Opf       `json:"opf"`
+	Opf            Opf       `json:"opf"`
-	Container Container `json:"-"`
+	Container      Container `json:"-"`
-	Mimetype  string    `json:"-"`
+	Mimetype       string    `json:"-"`
-	Content   []Content
+	Content        []Content
-	fd        *zip.ReadCloser
+	fd             *zip.ReadCloser
+	CoverImage     []byte
+	CoverFile      string
+	CoverMediaType string
+	Sections       []Section
+	dumpxml        bool
 }
 
 // Open open resource file

pkg/epub/container.go

@@ -1,11 +1,11 @@
 package epub
 
-//Container META-INF/container.xml file
+// Container META-INF/container.xml file
 type Container struct {
 	Rootfile Rootfile `xml:"rootfiles>rootfile" json:"rootfile"`
 }
 
-//Rootfile root file
+// Rootfile root file
 type Rootfile struct {
 	Path string `xml:"full-path,attr" json:"path"`
 	Type string `xml:"media-type,attr" json:"type"`

pkg/epub/content.go

@@ -8,9 +8,11 @@ import (
 )
 
 var (
-	cleanentitles = regexp.MustCompile(`&.+;`)
+	cleanentitles = regexp.MustCompile(`&[a-z]+;`)
 	empty         = regexp.MustCompile(`(?s)^[\s ]*$`)
-	newlines      = regexp.MustCompile(`[\r\n]+`)
+	newlines      = regexp.MustCompile(`[\r\n\s]+`)
+	cleansvg      = regexp.MustCompile(`(<svg.+</svg>|<!\[CDATA\[.+\]\]>)`)
+	cleanmarkup   = regexp.MustCompile(`<[^<>]+>`)
 )
 
 // Content nav-point content
@@ -22,30 +24,46 @@ type Content struct {
 	XML   []byte
 }
 
+// parse XML, look for title and <p>.*</p> stuff
 func (c *Content) String(content []byte) error {
-	// parse XML, look for title and <p>.*</p> stuff
 	doc, err := xmlquery.Parse(
 		strings.NewReader(
-			cleanentitles.ReplaceAllString(string(content), " ")))
+			cleansvg.ReplaceAllString(
+				cleanentitles.ReplaceAllString(string(content), " "), "")))
 	if err != nil {
-		panic(err)
+		return err
 	}
 
-	// extract the title
+	if c.Title == "" {
-	for _, item := range xmlquery.Find(doc, "//title") {
+		// extract the title
-		c.Title = strings.TrimSpace(item.InnerText())
+		for _, item := range xmlquery.Find(doc, "//title") {
+			c.Title = strings.TrimSpace(item.InnerText())
+		}
 	}
 
 	// extract all  paragraphs, ignore any formatting  and re-fill the
-	// paragraph,  that is, we  replaces all newlines inside  with one
+	// paragraph,  that is, we  replace all newlines inside  with one
 	// space.
 	txt := strings.Builder{}
+	var have_p bool
 	for _, item := range xmlquery.Find(doc, "//p") {
 		if !empty.MatchString(item.InnerText()) {
+			have_p = true
 			txt.WriteString(newlines.ReplaceAllString(item.InnerText(), " ") + "\n\n")
 		}
 	}
 
+	if !have_p {
+		// try  <div></div>, which some  ebooks use, so get  all divs,
+		// remove markup and paragraphify the parts
+		for _, item := range xmlquery.Find(doc, "//div") {
+			if !empty.MatchString(item.InnerText()) {
+				cleaned := cleanmarkup.ReplaceAllString(item.InnerText(), "")
+				txt.WriteString(newlines.ReplaceAllString(cleaned, " ") + "\n\n")
+			}
+		}
+	}
+
 	c.Body = strings.TrimSpace(txt.String())
 	c.XML = content
 

pkg/epub/open.go

@@ -4,75 +4,202 @@ import (
 	"archive/zip"
 	"fmt"
 	"log"
-	"os"
+	"path/filepath"
+	"regexp"
 	"strings"
 )
 
-// Open open a epub file
+var (
+	// to find content
+	types = regexp.MustCompile(`application/(xml|html|xhtml|htm)`)
+
+	// cleanup regexes
+	deanchor = regexp.MustCompile(`#.*$`)
+	cleanext = regexp.MustCompile(`^\.`)
+)
+
+// Open open a epub file and return the filled Book structure
 func Open(fn string, dumpxml bool) (*Book, error) {
+	bk, err := openFile(fn, dumpxml)
+	if err != nil {
+		return bk, err
+	}
+
+	defer func() {
+		if err := bk.fd.Close(); err != nil {
+			log.Fatal(err)
+		}
+	}()
+
+	if err := bk.getManifest(); err != nil {
+		return bk, err
+	}
+
+	if err := bk.getSections(); err != nil {
+		return bk, err
+	}
+
+	if err := bk.readSectionContent(); err != nil {
+		return bk, err
+	}
+
+	return bk, nil
+}
+
+// load the epub zip file
+func openFile(fn string, dumpxml bool) (*Book, error) {
 	fd, err := zip.OpenReader(fn)
 	if err != nil {
 		return nil, err
 	}
 
-	defer func() {
+	bk := &Book{fd: fd, dumpxml: dumpxml}
-		if err := fd.Close(); err != nil {
-			log.Fatal(err)
-		}
-	}()
 
-	bk := Book{fd: fd}
+	return bk, nil
+}
+
+// load the manifest
+func (bk *Book) getManifest() error {
 	mt, err := bk.readBytes("mimetype")
 	if err != nil {
-		return &bk, err
+		return err
 	}
 
 	bk.Mimetype = string(mt)
 
+	// contains the root path
 	err = bk.readXML("META-INF/container.xml", &bk.Container)
 	if err != nil {
-		return &bk, err
+		return err
 	}
 
+	// contains the OPF data
 	err = bk.readXML(bk.Container.Rootfile.Path, &bk.Opf)
 	if err != nil {
-		return &bk, err
+		return err
 	}
 
+	// look for TOC (might be incomplete, see below!)
 	for _, mf := range bk.Opf.Manifest {
 		if mf.ID == bk.Opf.Spine.Toc {
 			err = bk.readXML(bk.filename(mf.Href), &bk.Ncx)
 			if err != nil {
-				return &bk, err
+				return err
 			}
+		}
 
-			break
+		if mf.ID == "cover-image" {
+			bk.CoverFile = mf.Href
+			bk.CoverMediaType = mf.MediaType
 		}
 	}
 
-	for _, file := range bk.Files() {
+	return nil
-		content, err := bk.readBytes(file)
+}
-		if err != nil {
+
-			return &bk, err
+// extract the readable sections of the epub
+func (bk *Book) getSections() error {
+	// to store our final content sections
+	sections := []Section{}
+
+	// count the content items in the raw manifest
+	var manifestcount int
+	for _, item := range bk.Opf.Manifest {
+		if types.MatchString(item.MediaType) {
+			manifestcount++
+		}
+	}
+
+	// we have ncx points from the TOC, try those
+	if len(bk.Ncx.Points) > 0 {
+		for _, block := range bk.Ncx.Points {
+			sect := Section{
+				File:  "OEBPS/" + block.Content.Src,
+				Title: block.Text,
+			}
+
+			srcfile := deanchor.ReplaceAllString(block.Content.Src, "")
+
+			for _, file := range bk.Files() {
+				if strings.Contains(file, srcfile) {
+					sect.File = file
+					sect.MediaType = "application/" + cleanext.ReplaceAllString(filepath.Ext(file), "")
+					break
+				}
+			}
+
+			sections = append(sections, sect)
 		}
 
-		ct := Content{Src: file}
+		if len(sections) < manifestcount {
-		if strings.Contains(string(content), "<?xml") {
+			// TOC  was incomplete, restart  from scratch but  use the
-			if err := ct.String(content); err != nil {
+			// OPF Manifest directly
-				return &bk, err
+
+			sections = []Section{}
+
+			for _, item := range bk.Opf.Manifest {
+				if types.MatchString(item.MediaType) {
+					sect := Section{
+						File:      "OEBPS/" + item.Href,
+						MediaType: item.MediaType,
+					}
+
+					srcfile := deanchor.ReplaceAllString(item.Href, "")
+
+					for _, file := range bk.Files() {
+						if strings.Contains(file, srcfile) {
+							sect.File = file
+							break
+						}
+					}
+
+					sections = append(sections, sect)
+				}
 			}
 		}
+	} else {
+		// no TOC, just pull in the files directly
+		for _, file := range bk.Files() {
+			sections = append(sections,
+				Section{
+					File:      file,
+					MediaType: "application/" + cleanext.ReplaceAllString(filepath.Ext(file), ""),
+				})
+		}
+	}
+
+	// final sections to keep
+	bk.Sections = sections
+
+	return nil
+}
+
+func (bk *Book) readSectionContent() error {
+	// now read in the actual xml contents
+	for _, section := range bk.Sections {
+		content, err := bk.readBytes(section.File)
+		if err != nil {
+			return err
+		}
+
+		if strings.Contains(section.File, bk.CoverFile) {
+			bk.CoverImage = content
+		}
+
+		ct := Content{Src: section.File, Title: section.Title}
+
+		if types.MatchString(section.MediaType) {
+			if err := ct.String(content); err != nil {
+				return err
+			}
+		}
+
+		if bk.dumpxml {
+			fmt.Println(string(ct.XML))
+		}
 
 		bk.Content = append(bk.Content, ct)
-
-		if dumpxml {
-			fmt.Println(string(ct.XML))
-		}
 	}
 
-	if dumpxml {
+	return nil
-		os.Exit(0)
-	}
-
-	return &bk, nil
 }

pkg/epub/opf.go

@@ -1,13 +1,13 @@
 package epub
 
-//Opf content.opf
+// Opf content.opf
 type Opf struct {
 	Metadata Metadata   `xml:"metadata" json:"metadata"`
 	Manifest []Manifest `xml:"manifest>item" json:"manifest"`
 	Spine    Spine      `xml:"spine" json:"spine"`
 }
 
-//Metadata metadata
+// Metadata metadata
 type Metadata struct {
 	Title       []string     `xml:"title" json:"title"`
 	Language    []string     `xml:"language" json:"language"`
@@ -53,7 +53,7 @@ type Metafield struct {
 	Content string `xml:"content,attr" json:"content"`
 }
 
-//Manifest manifest
+// Manifest manifest
 type Manifest struct {
 	ID           string `xml:"id,attr" json:"id"`
 	Href         string `xml:"href,attr" json:"href"`