feat(sync): add CFI converter for CREngine XPointer to standard epubcfi
Implements ConvertCREToStandard which converts CREngine XPointers (e.g. /body/DocFragment[6]/body/div/p[47]/text().2399) to standard epubcfi format (e.g. epubcfi(/6/12!/4/2[id]/4/1:7)). Key components: - indexChildNodes: faithful port of foliate-js's epubcfi.js algorithm for computing CFI-compatible child node indices including virtual positions, null positions between adjacent elements, and text chunks - preprocessXHTML: converts XHTML self-closing tags (e.g. <a id="x"/>) to open/close pairs so Go's HTML parser produces the same DOM as the browser's XHTML parser - buildCFI: walks up from a text node to body, computing CFI indices at each level using indexChildNodes - findTextInNode: regex-based whitespace-flexible text search for context_text fallback positioning - convertByPercentageOffset: estimates position via book-wide character counts when no context_text is available - ConvertCREToStandard: orchestrates text search → percentage fallback Supports CREngine XPointer format, CREngine fragment ID format (#_doc_fragment_N_anchor), and includes round-trip test coverage for 1984 and Crime and Punishment EPUBs.
This commit is contained in:
File diff suppressed because it is too large
Load Diff
@@ -0,0 +1,353 @@
|
|||||||
|
package sync
|
||||||
|
|
||||||
|
import (
|
||||||
|
"testing"
|
||||||
|
)
|
||||||
|
|
||||||
|
func TestParseCREXPointer(t *testing.T) {
|
||||||
|
tests := []struct {
|
||||||
|
input string
|
||||||
|
wantFrag int
|
||||||
|
wantPath int
|
||||||
|
wantChar int
|
||||||
|
}{
|
||||||
|
{"/body/DocFragment[0]/body/div[4]/p[38]/text().541", 0, 2, 541},
|
||||||
|
{"/body/DocFragment[2]/body/div/p[5]/text()[2].16", 2, 2, 16},
|
||||||
|
{"/body/DocFragment[1]/body", 1, 0, 0},
|
||||||
|
{"/body/DocFragment[5]/body/div[3]/p[28]", 5, 2, 0},
|
||||||
|
}
|
||||||
|
|
||||||
|
for _, tt := range tests {
|
||||||
|
xp, err := ParseCREXPointer(tt.input)
|
||||||
|
if err != nil {
|
||||||
|
t.Errorf("ParseCREXPointer(%q) error: %v", tt.input, err)
|
||||||
|
continue
|
||||||
|
}
|
||||||
|
if xp.FragmentIndex != tt.wantFrag {
|
||||||
|
t.Errorf("FragmentIndex = %d, want %d", xp.FragmentIndex, tt.wantFrag)
|
||||||
|
}
|
||||||
|
if len(xp.ElementPath) != tt.wantPath {
|
||||||
|
t.Errorf("len(ElementPath) = %d, want %d (got %+v)", len(xp.ElementPath), tt.wantPath, xp.ElementPath)
|
||||||
|
}
|
||||||
|
if xp.CharOffset != tt.wantChar {
|
||||||
|
t.Errorf("CharOffset = %d, want %d", xp.CharOffset, tt.wantChar)
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestParseCREXPointerInvalid(t *testing.T) {
|
||||||
|
_, err := ParseCREXPointer("epubcfi(/6/4!/4/2/1:0)")
|
||||||
|
if err == nil {
|
||||||
|
t.Error("expected error for standard epubcfi")
|
||||||
|
}
|
||||||
|
_, err = ParseCREXPointer("")
|
||||||
|
if err == nil {
|
||||||
|
t.Error("expected error for empty string")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestIsCREXPointer(t *testing.T) {
|
||||||
|
if !IsCREXPointer("/body/DocFragment[0]/body/div/p") {
|
||||||
|
t.Error("should recognize CRE XPointer")
|
||||||
|
}
|
||||||
|
if !IsCREXPointer("#_doc_fragment_5_ link2HCH0002") {
|
||||||
|
t.Error("should recognize CRE fragment ID")
|
||||||
|
}
|
||||||
|
if IsCREXPointer("epubcfi(/6/4!/4/2/1:0)") {
|
||||||
|
t.Error("should not recognize standard epubcfi as CRE")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestIsCREFragmentID(t *testing.T) {
|
||||||
|
if !IsCREFragmentID("#_doc_fragment_5_ link2HCH0002") {
|
||||||
|
t.Error("should recognize fragment ID")
|
||||||
|
}
|
||||||
|
if IsCREFragmentID("/body/DocFragment[2]/body") {
|
||||||
|
t.Error("should not recognize XPointer as fragment ID")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestParseCREFragmentID(t *testing.T) {
|
||||||
|
tests := []struct {
|
||||||
|
input string
|
||||||
|
wantSpine int
|
||||||
|
wantAnchor string
|
||||||
|
}{
|
||||||
|
{"#_doc_fragment_5_ link2HCH0002", 5, "link2HCH0002"},
|
||||||
|
{"#_doc_fragment_0_", 0, ""},
|
||||||
|
{"#_doc_fragment_12_someid123", 12, "someid123"},
|
||||||
|
}
|
||||||
|
|
||||||
|
for _, tt := range tests {
|
||||||
|
frag, err := ParseCREFragmentID(tt.input)
|
||||||
|
if err != nil {
|
||||||
|
t.Errorf("ParseCREFragmentID(%q) error: %v", tt.input, err)
|
||||||
|
continue
|
||||||
|
}
|
||||||
|
if frag.SpineIndex != tt.wantSpine {
|
||||||
|
t.Errorf("SpineIndex = %d, want %d", frag.SpineIndex, tt.wantSpine)
|
||||||
|
}
|
||||||
|
if frag.Anchor != tt.wantAnchor {
|
||||||
|
t.Errorf("Anchor = %q, want %q", frag.Anchor, tt.wantAnchor)
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestParseCREFragmentIDInvalid(t *testing.T) {
|
||||||
|
_, err := ParseCREFragmentID("/body/DocFragment[2]/body")
|
||||||
|
if err == nil {
|
||||||
|
t.Error("expected error for XPointer input")
|
||||||
|
}
|
||||||
|
_, err = ParseCREFragmentID("#_doc_fragment_")
|
||||||
|
if err == nil {
|
||||||
|
t.Error("expected error for missing index")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestIsStandardEPUBCFI(t *testing.T) {
|
||||||
|
if !IsStandardEPUBCFI("epubcfi(/6/4!/4/2/1:0)") {
|
||||||
|
t.Error("should recognize standard epubcfi")
|
||||||
|
}
|
||||||
|
if IsStandardEPUBCFI("/body/DocFragment[0]/body") {
|
||||||
|
t.Error("should not recognize CRE as standard")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestConvert1984(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/uploads/Ebooks/George Orwell/1984 (126)/1984 - George Orwell.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
xp := "/body/DocFragment[2]/body/div/p[5]/text().500"
|
||||||
|
result, err := c.ConvertCREToStandard(xp, 0.01, "")
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("ConvertCREToStandard error: %v", err)
|
||||||
|
}
|
||||||
|
t.Logf("Input: %s", xp)
|
||||||
|
t.Logf("EPUBCFI: %s", result.EPUBCFI)
|
||||||
|
t.Logf("Href: %s", result.Href)
|
||||||
|
t.Logf("Precision: %s", result.Precision)
|
||||||
|
t.Logf("Percentage: %.4f", result.Percentage)
|
||||||
|
|
||||||
|
if result.Precision == "percentage" {
|
||||||
|
t.Error("expected better than percentage precision")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestConvertCrimeAndPunishmentFragmentID(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/uploads/Ebooks/Fyodor Dostoyevsky/Crime and Punishment (103)/Crime and Punishment - Fyodor Dostoyevsky.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
xp := "#_doc_fragment_5_ link2HCH0002"
|
||||||
|
result, err := c.ConvertCREToStandard(xp, 0.0303, "")
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("ConvertCREToStandard error: %v", err)
|
||||||
|
}
|
||||||
|
t.Logf("Input: %s", xp)
|
||||||
|
t.Logf("EPUBCFI: %s", result.EPUBCFI)
|
||||||
|
t.Logf("Href: %s", result.Href)
|
||||||
|
t.Logf("Precision: %s", result.Precision)
|
||||||
|
t.Logf("Percentage: %.4f", result.Percentage)
|
||||||
|
|
||||||
|
if result.Precision == "percentage" {
|
||||||
|
t.Error("expected better than percentage precision")
|
||||||
|
}
|
||||||
|
if result.Href == "" {
|
||||||
|
t.Error("expected non-empty href")
|
||||||
|
}
|
||||||
|
if result.Precision != "element" {
|
||||||
|
t.Errorf("expected element precision, got %s", result.Precision)
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestConvertHessBook(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/William L Hess/Crushing the Great Serpent_ Did God (42)/Crushing the Great Serpent_ Did - William L Hess.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
xp := "/body/DocFragment[5]/body/div[3]/p[28]/text().500"
|
||||||
|
result, err := c.ConvertCREToStandard(xp, 0.08, "")
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("ConvertCREToStandard error: %v", err)
|
||||||
|
}
|
||||||
|
t.Logf("Input: %s", xp)
|
||||||
|
t.Logf("EPUBCFI: %s", result.EPUBCFI)
|
||||||
|
t.Logf("Href: %s", result.Href)
|
||||||
|
t.Logf("Precision: %s", result.Precision)
|
||||||
|
t.Logf("Percentage: %.4f", result.Percentage)
|
||||||
|
|
||||||
|
if result.Precision == "percentage" {
|
||||||
|
t.Error("expected better than percentage precision")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestConvertCPByTextSearch(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/uploads/Ebooks/Fyodor Dostoyevsky/Crime and Punishment (103)/Crime and Punishment - Fyodor Dostoyevsky.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
xp := "/body/DocFragment[6]/body/div/p[47]/text().1"
|
||||||
|
contextText := "Raskolnikov was not used to crowds, and, as we said before, he avoided society of every sort, more especially of l"
|
||||||
|
result, err := c.ConvertCREToStandard(xp, 0.0579, contextText)
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("ConvertCREToStandard error: %v", err)
|
||||||
|
}
|
||||||
|
t.Logf("Input: %s", xp)
|
||||||
|
t.Logf("ContextText: %s", contextText)
|
||||||
|
t.Logf("EPUBCFI: %s", result.EPUBCFI)
|
||||||
|
t.Logf("Href: %s", result.Href)
|
||||||
|
t.Logf("Precision: %s", result.Precision)
|
||||||
|
t.Logf("Percentage: %.4f", result.Percentage)
|
||||||
|
|
||||||
|
if result.Precision != "exact" {
|
||||||
|
t.Errorf("expected exact precision, got %s", result.Precision)
|
||||||
|
}
|
||||||
|
if result.EPUBCFI == "" {
|
||||||
|
t.Error("expected non-empty EPUBCFI")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestParseEPUBCFI(t *testing.T) {
|
||||||
|
tests := []struct {
|
||||||
|
input string
|
||||||
|
wantSpine int
|
||||||
|
wantSteps int
|
||||||
|
}{
|
||||||
|
{"epubcfi(/6/12!/4/2/90/1:7)", 5, 4},
|
||||||
|
{"epubcfi(/6/4!/4/2/1:0)", 1, 3},
|
||||||
|
{"epubcfi(/6/2!/4)", 0, 1},
|
||||||
|
}
|
||||||
|
|
||||||
|
for _, tt := range tests {
|
||||||
|
spineIndex, steps, err := parseEPUBCFI(tt.input)
|
||||||
|
if err != nil {
|
||||||
|
t.Errorf("parseEPUBCFI(%q) error: %v", tt.input, err)
|
||||||
|
continue
|
||||||
|
}
|
||||||
|
if spineIndex != tt.wantSpine {
|
||||||
|
t.Errorf("spineIndex = %d, want %d", spineIndex, tt.wantSpine)
|
||||||
|
}
|
||||||
|
if len(steps) != tt.wantSteps {
|
||||||
|
t.Errorf("len(steps) = %d, want %d", len(steps), tt.wantSteps)
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestParseEPUBCFIRange(t *testing.T) {
|
||||||
|
spineIndex, steps, err := parseEPUBCFI("epubcfi(/6/40!/4,/24/20,/40/5:61)")
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("parseEPUBCFI range error: %v", err)
|
||||||
|
}
|
||||||
|
if spineIndex != 19 {
|
||||||
|
t.Errorf("spineIndex = %d, want 19", spineIndex)
|
||||||
|
}
|
||||||
|
t.Logf("Range CFI steps: %d", len(steps))
|
||||||
|
for i, s := range steps {
|
||||||
|
t.Logf(" step %d: index=%d offset=%d hasOffset=%v", i, s.Index, s.Offset, s.HasOffset)
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestParseEPUBCFIInvalid(t *testing.T) {
|
||||||
|
_, _, err := parseEPUBCFI("not-a-cfi")
|
||||||
|
if err == nil {
|
||||||
|
t.Error("expected error for invalid CFI")
|
||||||
|
}
|
||||||
|
_, _, err = parseEPUBCFI("epubcfi(/6/12)")
|
||||||
|
if err == nil {
|
||||||
|
t.Error("expected error for CFI without indirection")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestRoundTrip1984(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/uploads/Ebooks/George Orwell/1984 (126)/1984 - George Orwell.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
originalXP := "/body/DocFragment[2]/body/div/p[5]/text().500"
|
||||||
|
forward, err := c.ConvertCREToStandard(originalXP, 0.01, "")
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("forward conversion error: %v", err)
|
||||||
|
}
|
||||||
|
if forward.EPUBCFI == "" {
|
||||||
|
t.Fatal("forward conversion produced empty epubcfi")
|
||||||
|
}
|
||||||
|
t.Logf("Forward: %s → %s", originalXP, forward.EPUBCFI)
|
||||||
|
|
||||||
|
reverse, err := c.ConvertStandardToCRE(forward.EPUBCFI, forward.Percentage, "")
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("reverse conversion error: %v", err)
|
||||||
|
}
|
||||||
|
if reverse.XPointer == "" {
|
||||||
|
t.Fatal("reverse conversion produced empty XPointer")
|
||||||
|
}
|
||||||
|
t.Logf("Reverse: %s → %s", forward.EPUBCFI, reverse.XPointer)
|
||||||
|
t.Logf("Reverse precision: %s", reverse.Precision)
|
||||||
|
|
||||||
|
if reverse.Precision != "exact" {
|
||||||
|
t.Errorf("expected exact precision, got %s", reverse.Precision)
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestRoundTripCP(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/uploads/Ebooks/Fyodor Dostoyevsky/Crime and Punishment (103)/Crime and Punishment - Fyodor Dostoyevsky.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
originalXP := "/body/DocFragment[6]/body/div/p[47]/text().2399"
|
||||||
|
contextText := "Raskolnikov was not used to crowds, and, as we said before, he avoided society of every sort, more especially of l"
|
||||||
|
forward, err := c.ConvertCREToStandard(originalXP, 0.0579, contextText)
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("forward conversion error: %v", err)
|
||||||
|
}
|
||||||
|
if forward.EPUBCFI == "" {
|
||||||
|
t.Fatal("forward conversion produced empty epubcfi")
|
||||||
|
}
|
||||||
|
t.Logf("Forward: %s → %s", originalXP, forward.EPUBCFI)
|
||||||
|
|
||||||
|
reverse, err := c.ConvertStandardToCRE(forward.EPUBCFI, forward.Percentage, contextText)
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("reverse conversion error: %v", err)
|
||||||
|
}
|
||||||
|
if reverse.XPointer == "" {
|
||||||
|
t.Fatal("reverse conversion produced empty XPointer")
|
||||||
|
}
|
||||||
|
t.Logf("Reverse: %s → %s", forward.EPUBCFI, reverse.XPointer)
|
||||||
|
t.Logf("Reverse precision: %s", reverse.Precision)
|
||||||
|
|
||||||
|
if reverse.Precision != "exact" {
|
||||||
|
t.Errorf("expected exact precision, got %s", reverse.Precision)
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestReverseTextSearchFallback(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/uploads/Ebooks/Fyodor Dostoyevsky/Crime and Punishment (103)/Crime and Punishment - Fyodor Dostoyevsky.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
contextText := "Raskolnikov was not used to crowds, and, as we said before, he avoided society of every sort, more especially of l"
|
||||||
|
reverse, err := c.ConvertStandardToCRE("epubcfi(/6/12!/4/99999/1:0)", 0.0579, contextText)
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("reverse conversion error: %v", err)
|
||||||
|
}
|
||||||
|
t.Logf("Text search fallback XPointer: %s", reverse.XPointer)
|
||||||
|
t.Logf("Precision: %s", reverse.Precision)
|
||||||
|
|
||||||
|
if reverse.Precision != "exact" {
|
||||||
|
t.Errorf("expected exact precision from text search, got %s", reverse.Precision)
|
||||||
|
}
|
||||||
|
if reverse.XPointer == "" {
|
||||||
|
t.Error("expected non-empty XPointer from text search")
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
func TestReversePercentageFallback(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/uploads/Ebooks/George Orwell/1984 (126)/1984 - George Orwell.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
reverse, err := c.ConvertStandardToCRE("epubcfi(/6/12!/4/99999/1:0)", 0.5, "")
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("reverse conversion error: %v", err)
|
||||||
|
}
|
||||||
|
t.Logf("Percentage fallback precision: %s", reverse.Precision)
|
||||||
|
|
||||||
|
if reverse.Precision != "percentage" {
|
||||||
|
t.Errorf("expected percentage precision, got %s with XPointer %s", reverse.Precision, reverse.XPointer)
|
||||||
|
}
|
||||||
|
if reverse.XPointer != "" {
|
||||||
|
t.Error("expected empty XPointer for percentage fallback")
|
||||||
|
}
|
||||||
|
}
|
||||||
@@ -0,0 +1,37 @@
|
|||||||
|
package sync
|
||||||
|
|
||||||
|
import "testing"
|
||||||
|
|
||||||
|
func TestDebugHess(t *testing.T) {
|
||||||
|
epubPath := "/home/nymusicman/Code/bookhoard/William L Hess/Crushing the Great Serpent_ Did God (42)/Crushing the Great Serpent_ Did - William L Hess.epub"
|
||||||
|
c := NewCFIConverter(epubPath)
|
||||||
|
|
||||||
|
xp, err := ParseCREXPointer("/body/DocFragment[5]/body/div[3]/p[28]/text().500")
|
||||||
|
if err != nil {
|
||||||
|
t.Fatal(err)
|
||||||
|
}
|
||||||
|
t.Logf("FragmentIndex: %d, CharOffset: %d", xp.FragmentIndex, xp.CharOffset)
|
||||||
|
|
||||||
|
doc, href, err := c.getContentDoc(xp.FragmentIndex)
|
||||||
|
if err != nil {
|
||||||
|
t.Fatalf("getContentDoc error: %v", err)
|
||||||
|
}
|
||||||
|
t.Logf("href: %s", href)
|
||||||
|
|
||||||
|
body := findBody(doc)
|
||||||
|
if body == nil {
|
||||||
|
t.Fatal("no body")
|
||||||
|
}
|
||||||
|
|
||||||
|
totalChars := countTextChars(body)
|
||||||
|
t.Logf("total chars in doc: %d", totalChars)
|
||||||
|
|
||||||
|
target, foundOffset := findNodeAtCharOffset(body, 500)
|
||||||
|
if target == nil {
|
||||||
|
t.Fatal("target is nil")
|
||||||
|
}
|
||||||
|
t.Logf("foundOffset: %d target: <%s>", foundOffset, target.Data)
|
||||||
|
|
||||||
|
cfi, err := buildCFI(5, target, 0)
|
||||||
|
t.Logf("cfi: %q err: %v", cfi, err)
|
||||||
|
}
|
||||||
Reference in New Issue
Block a user