Add GetObjectNums

pdf/core/crossrefs.go

@@ -42,6 +42,7 @@ type XrefObject struct {
 
 // XrefTable is a map between object number and corresponding XrefObject.
 // TODO (v3): Unexport.
+// TODO: Consider changing to a slice, so can maintain the object order without sorting when analyzing.
 type XrefTable map[int]XrefObject
 
 // ObjectStream represents an object stream's information which can contain multiple indirect objects.

pdf/core/utils.go

@@ -34,6 +34,20 @@ func (parser *PdfParser) Inspect() (map[string]int, error) {
 	return parser.inspect()
 }
 
+// GetObjectNums returns a sorted list of object numbers of the PDF objects in the file.
+func (parser *PdfParser) GetObjectNums() []int {
+	objNums := []int{}
+	for _, x := range parser.xrefs {
+		objNums = append(objNums, x.objectNumber)
+	}
+
+	// Sort the object numbers to give consistent ordering of PDF objects in output.
+	// Needed since parser.xrefs is a map.
+	sort.Ints(objNums)
+
+	return objNums
+}
+
 func getUniDocVersion() string {
 	return common.Version
 }

pdf/model/reader.go

@@ -756,6 +756,15 @@ func (this *PdfReader) Inspect() (map[string]int, error) {
 	return this.parser.Inspect()
 }
 
+// GetObjectNums returns the object numbers of the PDF objects in the file
+// Numbered objects are either indirect objects or stream objects.
+// e.g. objNums := pdfReader.GetObjectNums()
+// The underlying objects can then be accessed with
+// pdfReader.GetIndirectObjectByNumber(objNums[0]) for the first available object.
+func (r *PdfReader) GetObjectNums() []int {
+	return r.parser.GetObjectNums()
+}
+
 // Get specific object number.
 func (this *PdfReader) GetIndirectObjectByNumber(number int) (PdfObject, error) {
 	obj, err := this.parser.LookupByNumber(number)