quietened logging

pdf/crossrefs.go

@@ -205,6 +205,7 @@ func (this *PdfParser) lookupByNumberWrapper(objNumber int, attemptRepairs bool)
 	return obj, inObjStream, nil
 }
 
+// getObjectNumber returns object number, generation number, error for `obj`
 func getObjectNumber(obj PdfObject) (int64, int64, error) {
 	if io, isIndirect := obj.(*PdfIndirectObject); isIndirect {
 		return io.ObjectNumber, io.GenerationNumber, nil

pdf/page.go

@@ -233,7 +233,7 @@ func NewPdfPage() *PdfPage {
 func (reader *PdfReader) newPdfPageFromDict(p *PdfObjectDictionary) (*PdfPage, error) {
 	page := NewPdfPage()
 
-	fmt.Printf("newPdfPageFromDict %+v\n", p)
+	// fmt.Printf("newPdfPageFromDict %+v\n", p)
 
 	d := *p
 
@@ -614,7 +614,7 @@ func (this *PdfPage) AddWatermarkImage(ximg *XObjectImage, opt WatermarkImageOpt
 	pWidth := bbox.Urx - bbox.Llx
 	pHeight := bbox.Ury - bbox.Lly
 
-	fmt.Printf("AddWatermarkImage: bbox=%+v opt=%+v\n", bbox, opt)
+	// fmt.Printf("AddWatermarkImage: bbox=%+v opt=%+v\n", bbox, opt)
 	var wWidth float64
 	var wHeight float64
 	var xOffset float64
@@ -661,7 +661,7 @@ func (this *PdfPage) AddWatermarkImage(ximg *XObjectImage, opt WatermarkImageOpt
 		"%.0f 0 0 %.0f %.4f %.4f cm\n"+
 		"/%s Do\n"+
 		"Q", wWidth, wHeight, xOffset, yOffset, imgName)
-	fmt.Printf("AddWatermarkImage: contentStr=%q\n", contentStr)
+	// fmt.Printf("AddWatermarkImage: contentStr=%q\n", contentStr)
 	this.AddContentStreamByString(contentStr)
 
 	return nil

pdf/pdfobject.go

@@ -100,11 +100,19 @@ func (this *PdfObjectInteger) DefaultWriteString() string {
 }
 
 func (this *PdfObjectFloat) String() string {
-	return fmt.Sprintf("%f", *this)
+	return fmt.Sprintf("%f", pinNumber(*this))
 }
 
 func (this *PdfObjectFloat) DefaultWriteString() string {
-	return fmt.Sprintf("%f", *this)
+	return fmt.Sprintf("%f", pinNumber(*this))
+}
+
+func pinNumber(x float64) float64 {
+	// n := math.Floor(x + 0.5)
+	// if math.Abs(x) > 1.0 && math.Abs(x-n) < 1e-5 {
+	// 	x = n
+	// }
+	return x
 }
 
 func (this *PdfObjectString) String() string {
@@ -220,7 +228,7 @@ func (this *PdfObjectArray) DefaultWriteString() string {
 }
 
 // byName sorts slices of PdfObjectName. It is needed because sort.Strings(keys) gives a typecheck
-// error which I find strange because a PdfObjectName is a string.
+// error, which I find strange because a PdfObjectName is a string.
 type byName []PdfObjectName
 
 func (x byName) Len() int { return len(x) }

pdf/pdfparser.go

@@ -21,16 +21,18 @@ import (
 )
 
 // Regular Expressions for parsing and identifying object signatures.
-var rePdfVersion = regexp.MustCompile(`%PDF-(\d\.\d)`)
+var (
-var reEOF = regexp.MustCompile("%%EOF")
+	rePdfVersion     = regexp.MustCompile(`%PDF-(\d\.\d)`)
-var reXrefTable = regexp.MustCompile(`\s*xref\s*`)
+	reEOF            = regexp.MustCompile("%%EOF")
-var reStartXref = regexp.MustCompile(`startx?ref\s*(\d+)`)
+	reXrefTable      = regexp.MustCompile(`\s*xref\s*`)
-var reNumeric = regexp.MustCompile(`^[\+-.]*([0-9.]+)`)
+	reStartXref      = regexp.MustCompile(`startx?ref\s*(\d+)`)
-var reExponential = regexp.MustCompile(`^[\+-.]*([0-9.]+)e[\+-.]*([0-9.]+)`)
+	reNumeric        = regexp.MustCompile(`^[\+-.]*([0-9.]+)`)
-var reReference = regexp.MustCompile(`^\s*(\d+)\s+(\d+)\s+R`)
+	reExponential    = regexp.MustCompile(`^[\+-.]*([0-9.]+)e[\+-.]*([0-9.]+)`)
-var reIndirectObject = regexp.MustCompile(`(\d+)\s+(\d+)\s+obj`)
+	reReference      = regexp.MustCompile(`^\s*(\d+)\s+(\d+)\s+R`)
-var reXrefSubsection = regexp.MustCompile(`(\d+)\s+(\d+)\s*$`)
+	reIndirectObject = regexp.MustCompile(`(\d+)\s+(\d+)\s+obj`)
-var reXrefEntry = regexp.MustCompile(`(\d+)\s+(\d+)\s+([nf])\s*$`)
+	reXrefSubsection = regexp.MustCompile(`(\d+)\s+(\d+)\s*$`)
+	reXrefEntry      = regexp.MustCompile(`(\d+)\s+(\d+)\s+([nf])\s*$`)
+)
 
 type PdfParser struct {
 	rs               io.ReadSeeker
@@ -44,30 +46,18 @@ type PdfParser struct {
 	repairsAttempted bool // Avoid multiple attempts for repair.
 }
 
-func isWhiteSpace(ch byte) bool {
+func isWhiteSpace(c byte) bool {
 	// Table 1 white-space characters (7.2.2 Character Set)
 	// spaceCharacters := string([]byte{0x00, 0x09, 0x0A, 0x0C, 0x0D, 0x20})
-	if (ch == 0x00) || (ch == 0x09) || (ch == 0x0A) || (ch == 0x0C) || (ch == 0x0D) || (ch == 0x20) {
+	return (c == 0x00) || (c == 0x09) || (c == 0x0A) || (c == 0x0C) || (c == 0x0D) || (c == 0x20)
-		return true
-	} else {
-		return false
-	}
 }
 
 func isDecimalDigit(c byte) bool {
-	if c >= '0' && c <= '9' {
+	return '0' <= c && c <= '9'
-		return true
-	} else {
-		return false
-	}
 }
 
 func isOctalDigit(c byte) bool {
-	if c >= '0' && c <= '7' {
+	return '0' <= c && c <= '7'
-		return true
-	} else {
-		return false
-	}
 }
 
 // Skip over any spaces.
@@ -836,6 +826,8 @@ func (this *PdfParser) parseXrefStream(xstm *PdfObjectInteger) (*PdfObjectDictio
 		common.Log.Debug("ERROR: Unable to decode stream")
 		return nil, err
 	}
+	fmt.Printf("@@ ds=%d %T %+v\n", len(ds), ds, ds)
+	// panic(fmt.Errorf("WWWWWWWW"))
 
 	s0 := int(b[0])
 	s1 := int(b[0] + b[1])
@@ -1040,7 +1032,7 @@ func (this *PdfParser) loadXrefs() (*PdfObjectDictionary, error) {
 	if err != nil {
 		return nil, err
 	}
-	common.Log.Debug("fsize: %d", fSize)
+	common.Log.Debug("loadXrefs: fsize: %d", fSize)
 	if fSize <= offset {
 		offset = fSize
 	}