Fix #71, #72

Author: hhrutter

README.md

@@ -69,12 +69,18 @@ Even if you want to dive right into pdfcpu backend integration it is highly reco
 ## Status
 
 [Version: 0.1.22](https://github.com/hhrutter/pdfcpu/releases/tag/v0.1.22)
+
 * Support for `go mod`
 * Command completion
 * `pages insert` inserts empty pages before all selected pages
 * `pages remove` removes all selected pages
 * Bug fixes #64, #65, #68, #69
 
+## Reminder
+
+Always make sure your work is based on the latest commit!<br>
+pdfcpu is still *Alpha* - bugfixes are committed on the fly and will be mentioned on the next release notes.<br>
+
 ## Demo Screencast
 
 (using older version with a smaller command set)
@@ -112,7 +118,7 @@ pdfcpu ve
 * Feature requests - always welcome!
 * Bug fixes - always welcome!
 * PRs - also welcome, although I can't promise a merge-in right now.
-* pdfcpu is stable but still alpha and occasionally undergoing heavy changes.
+* pdfcpu is stable but still *Alpha* and occasionally undergoing heavy changes.
 
 ### How
 

pkg/api/api.go

@@ -207,7 +207,6 @@ func writeSinglePagePDF(ctx *pdf.Context, pageNr int, dirOut string) error {
 	ctx.ResetWriteContext()
 
 	w := ctx.Write
-	w.Command = "ExtractPages"
 	w.SelectedPages[pageNr] = true
 	w.DirName = dirOut + "/"
 	w.FileName = singlePageFileName(ctx, pageNr)
@@ -330,7 +329,6 @@ func pageRangeFileName(ctx *pdf.Context, from, thru int) string {
 func writeSpan(ctx *pdf.Context, from, thru int, dirOut string) error {
 	ctx.ResetWriteContext()
 	w := ctx.Write
-	w.Command = "Split"
 	w.SelectedPages = selectedPageRange(from, thru)
 	w.DirName = dirOut + "/"
 	w.FileName = pageRangeFileName(ctx, from, thru)
@@ -457,8 +455,6 @@ func Merge(cmd *Command) ([]string, error) {
 		return nil, err
 	}
 
-	ctxDest.Write.Command = "Merge"
-
 	dirName, fileName := filepath.Split(fileOut)
 	ctxDest.Write.DirName = dirName
 	ctxDest.Write.FileName = fileName
@@ -995,7 +991,6 @@ func Trim(cmd *Command) ([]string, error) {
 		return nil, err
 	}
 
-	ctx.Write.Command = "Trim"
 	ctx.Write.SelectedPages = pages
 
 	dirName, fileName := filepath.Split(fileOut)
@@ -1363,7 +1358,6 @@ func ImportImages(cmd *Command) ([]string, error) {
 		return nil, err
 	}
 
-	ctx.Write.Command = "Import"
 	dirName, fileName := filepath.Split(fileOut)
 	ctx.Write.DirName = dirName
 	ctx.Write.FileName = fileName
@@ -1577,7 +1571,6 @@ func NUp(cmd *Command) ([]string, error) {
 		return nil, err
 	}
 
-	ctx.Write.Command = "N-Up"
 	dirName, fileName := filepath.Split(fileOut)
 	ctx.Write.DirName = dirName
 	ctx.Write.FileName = fileName

pkg/api/process_test.go

@@ -66,6 +66,8 @@ func ExampleReadContext() {
 	// This allows to run pdf as a backend to an http server for on the fly pdf processing.
 
 	config := pdf.NewDefaultConfiguration()
+	config.Cmd = pdf.OPTIMIZE
+
 	fileIn := filepath.Join(inDir, "CenterOfWhy.pdf")
 	fileOut := filepath.Join(outDir, "test.pdf")
 
@@ -125,15 +127,17 @@ func ExampleReadContext() {
 
 }
 
-func TestReadSeekerAndWriter(t *testing.T) {
+func TestOptimizeUsingReadSeekerAndWriter(t *testing.T) {
 
 	config := pdf.NewDefaultConfiguration()
+	config.Cmd = pdf.OPTIMIZE
+
 	fileIn := filepath.Join(inDir, "CenterOfWhy.pdf")
 	fileOut := filepath.Join(outDir, "test.pdf")
 
 	f, err := os.Open(fileIn)
 	if err != nil {
-		t.Fatalf("TestReadSeekerAndWriter Open:  %v\n", err)
+		t.Fatalf("TestOptimizeUsingReadSeekerAndWriter Open:  %v\n", err)
 	}
 
 	defer func() {
@@ -142,27 +146,27 @@ func TestReadSeekerAndWriter(t *testing.T) {
 
 	fileInfo, err := f.Stat()
 	if err != nil {
-		t.Fatalf("TestReadSeekerAndWriter Stat:  %v\n", err)
+		t.Fatalf("TestOptimizeUsingReadSeekerAndWriter Stat:  %v\n", err)
 	}
 
 	ctx, err := ReadContext(f, fileIn, fileInfo.Size(), config)
 	if err != nil {
-		t.Fatalf("TestReadSeekerAndWriter Read:  %v\n", err)
+		t.Fatalf("TestOptimizeUsingReadSeekerAndWriter Read:  %v\n", err)
 	}
 
 	err = ValidateContext(ctx)
 	if err != nil {
-		t.Fatalf("TestReadSeekerAndWriter Validate:  %v\n", err)
+		t.Fatalf("TestOptimizeUsingReadSeekerAndWriter Validate:  %v\n", err)
 	}
 
 	err = OptimizeContext(ctx)
 	if err != nil {
-		t.Fatalf("TestReadSeekerAndWriter Optimize:  %v\n", err)
+		t.Fatalf("TestOptimizeUsingReadSeekerAndWriter Optimize:  %v\n", err)
 	}
 
 	w, err := os.Create(fileOut)
 	if err != nil {
-		t.Fatalf("TestReadSeekerAndWriter Create:  %v\n", err)
+		t.Fatalf("TestOptimizeUsingReadSeekerAndWriter Create:  %v\n", err)
 
 	}
 
@@ -183,21 +187,23 @@ func TestReadSeekerAndWriter(t *testing.T) {
 
 	err = WriteContext(ctx, w)
 	if err != nil {
-		t.Fatalf("TestReadSeekerAndWriter Write:  %v\n", err)
+		t.Fatalf("TestOptimizeUsingReadSeekerAndWriter Write:  %v\n", err)
 	}
 
 }
 
-func TestTrimUsingReadSeekerCloser(t *testing.T) {
+func TestTrimUsingReadSeekerAndWriter(t *testing.T) {
 
 	config := pdf.NewDefaultConfiguration()
+	config.Cmd = pdf.TRIM
+
 	fileIn := filepath.Join(inDir, "pike-stanford.pdf")
 	fileOut := filepath.Join(outDir, "test.pdf")
 	pageSelection := []string{"-2"}
 
 	f, err := os.Open(fileIn)
 	if err != nil {
-		t.Fatalf("TestTrimUsingReadSeekerCloser Open:  %v\n", err)
+		t.Fatalf("TestTrimUsingReadSeekerAndWriter Open:  %v\n", err)
 	}
 
 	defer func() {
@@ -206,22 +212,22 @@ func TestTrimUsingReadSeekerCloser(t *testing.T) {
 
 	ctx, err := ReadContext(f, "", 0, config)
 	if err != nil {
-		t.Fatalf("TestTrimUsingReadSeekerCloser Read:  %v\n", err)
+		t.Fatalf("TestTrimUsingReadSeekerAndWriter Read:  %v\n", err)
 	}
 
 	err = ValidateContext(ctx)
 	if err != nil {
-		t.Fatalf("TestTrimUsingReadSeekerCloser Validate:  %v\n", err)
+		t.Fatalf("TestTrimUsingReadSeekerAndWriter Validate:  %v\n", err)
 	}
 
 	err = OptimizeContext(ctx)
 	if err != nil {
-		t.Fatalf("TestTrimUsingReadSeekerCloser Optimize:  %v\n", err)
+		t.Fatalf("TestTrimUsingReadSeekerAndWriter Optimize:  %v\n", err)
 	}
 
 	w, err := os.Create(fileOut)
 	if err != nil {
-		t.Fatalf("TestTrimUsingReadSeekerCloser Create:  %v\n", err)
+		t.Fatalf("TestTrimUsingReadSeekerAndWriter Create:  %v\n", err)
 
 	}
 
@@ -240,22 +246,20 @@ func TestTrimUsingReadSeekerCloser(t *testing.T) {
 
 	}()
 
-	ctx.Write.Command = "Trim"
-
 	pages, err := pagesForPageSelection(ctx.PageCount, pageSelection)
 	if err != nil {
-		t.Fatalf("TestTrimUsingReadSeekerCloser pageSelection:  %v\n", err)
+		t.Fatalf("TestTrimUsingReadSeekerAndWriter pageSelection:  %v\n", err)
 	}
 	ctx.Write.SelectedPages = pages
 
 	err = WriteContext(ctx, w)
 	if err != nil {
-		t.Fatalf("TestTrimUsingReadSeekerCloser Write:  %v\n", err)
+		t.Fatalf("TestTrimUsingReadSeekerAndWriter Write:  %v\n", err)
 	}
 
 }
 
-func TestMergeUsingReadSeekerCloser(t *testing.T) {
+func TestMergeUsingReadSeekerAndWriter(t *testing.T) {
 
 	rr := []pdf.ReadSeekerCloser{}
 
@@ -265,7 +269,7 @@ func TestMergeUsingReadSeekerCloser(t *testing.T) {
 
 		f, err := os.Open(fileIn)
 		if err != nil {
-			t.Fatalf("TestMergeUsingReadSeekerCloser Open:  %v\n", err)
+			t.Fatalf("TestMergeUsingReadSeekerAndWriter Open:  %v\n", err)
 		}
 
 		rr = append(rr, f)
@@ -278,17 +282,18 @@ func TestMergeUsingReadSeekerCloser(t *testing.T) {
 	}()
 
 	config := pdf.NewDefaultConfiguration()
+	config.Cmd = pdf.MERGE
 
 	ctx, err := MergeContexts(rr, config)
 	if err != nil {
-		t.Fatalf("TestMergeUsingReadSeekerCloser Open:  %v\n", err)
+		t.Fatalf("TestMergeUsingReadSeekerAndWriter Open:  %v\n", err)
 	}
 
 	fileOut := filepath.Join(outDir, "test.pdf")
 
 	w, err := os.Create(fileOut)
 	if err != nil {
-		t.Fatalf("TestMergeUsingReadSeekerCloser create output file: %v\n", err)
+		t.Fatalf("TestMergeUsingReadSeekerAndWriter create output file: %v\n", err)
 	}
 
 	defer func() {
@@ -308,7 +313,7 @@ func TestMergeUsingReadSeekerCloser(t *testing.T) {
 
 	err = WriteContext(ctx, w)
 	if err != nil {
-		t.Fatalf("TestMergeUsingReadSeekerCloser Write output: %v\n", err)
+		t.Fatalf("TestMergeUsingReadSeekerAndWriter write output: %v\n", err)
 	}
 
 }

pkg/pdfcpu/context.go

@@ -548,7 +548,6 @@ type WriteContext struct {
 	DirName             string
 	FileName            string
 	FileSize            int64
-	Command             string        // The processing command in effect.
 	SelectedPages       IntSet        // For split, trim and extract.
 	BinaryTotalSize     int64         // total stream data, counts 100% all stream data written.
 	BinaryImageSize     int64         // total image stream data written = Read.BinaryImageSize.

pkg/pdfcpu/dict.go

@@ -40,8 +40,10 @@ func (d Dict) Len() int {
 
 // Insert adds a new entry to this PDFDict.
 func (d Dict) Insert(key string, value Object) (ok bool) {
-	if _, found := d.Find(key); found {
-		return false
+	if o, found := d.Find(key); found {
+		if o.PDFString() != value.PDFString() {
+			return false
+		}
 	}
 	d[key] = value
 	return true

pkg/pdfcpu/read.go

@@ -777,7 +777,14 @@ func scanLine(s *bufio.Scanner) (string, error) {
 			break
 		}
 	}
-	return s.Text(), nil
+	// Remove comment.
+	s1 := s.Text()
+	i := strings.Index(s1, "%")
+	if i >= 0 {
+		s1 = s1[:i]
+	}
+
+	return s1, nil
 }
 
 func scanTrailer(s *bufio.Scanner, line string) (string, error) {

pkg/pdfcpu/validate/structTree.go

@@ -88,10 +88,14 @@ func validateObjectReferenceDict(xRefTable *pdf.XRefTable, d pdf.Dict) error {
 
 	// Obj: required, indirect reference
 	ir := d.IndirectRefEntry("Obj")
-	if ir == nil {
+	if xRefTable.ValidationMode == pdf.ValidationStrict && ir == nil {
 		return errors.New("validateObjectReferenceDict: missing required entry \"Obj\"")
 	}
 
+	if ir == nil {
+		return nil
+	}
+
 	obj, err := xRefTable.Dereference(*ir)
 	if err != nil {
 		return err

pkg/pdfcpu/writePages.go

@@ -182,7 +182,7 @@ func writeKids(ctx *Context, a Array, pageNr *int) (Array, int, error) {
 				if ctx.Cmd == REMOVEPAGES {
 					writePage = !writePage
 				}
-				if ctx.Write.SelectedPages[*pageNr] {
+				if writePage {
 					log.Write.Printf("writeKids: writing page:%d\n", *pageNr)
 					err = writePageDict(ctx, ir, d, *pageNr)
 					kids = append(kids, o)