MCPcopy
hub / github.com/tdewolff/canvas / NewPDFReader

Function NewPDFReader

cmd/pdftext/reader.go:32–74  ·  view source on GitHub ↗
(reader io.Reader, password string)

Source from the content-addressed store, hash-verified

30}
31
32func NewPDFReader(reader io.Reader, password string) (*pdfReader, error) {
33 data, err := io.ReadAll(reader)
34 if err != nil {
35 return nil, err
36 } else if len(data) < 8 || !bytes.Equal(data[:7], []byte("%PDF-1.")) || data[7] < '0' || '7' < data[7] {
37 return nil, fmt.Errorf("invalid PDF file: bad version")
38 }
39
40 r := &pdfReader{
41 data: data,
42 objects: map[pdfRef]pdfObject{},
43 cache: map[pdfRef]any{},
44 }
45
46 // get startxref
47 var line []byte
48 lrr := newLineReaderReverse(r.data, len(r.data))
49 if line = lrr.Next(); !bytes.Equal(line, []byte("%%EOF")) {
50 return nil, fmt.Errorf("invalid PDF file")
51 }
52 if r.data[lrr.Pos()] == '\r' && r.data[lrr.Pos()+1] == '\n' {
53 r.eol = r.data[lrr.Pos() : lrr.Pos()+2]
54 } else {
55 r.eol = r.data[lrr.Pos() : lrr.Pos()+1]
56 }
57 num, _ := strconv.ParseUint(lrr.Next())
58 if num == 0 {
59 return nil, fmt.Errorf("invalid PDF file")
60 } else if line = lrr.Next(); !bytes.Equal(bytes.TrimSpace(line), []byte("startxref")) {
61 return nil, fmt.Errorf("invalid PDF file")
62 }
63 startxref := int(num)
64 //endtrailer := lrr.Pos()
65
66 if r.trailer, err = r.readTrailer(startxref); err != nil {
67 return nil, err
68 } else if err := r.readEncrypt([]byte(password)); err != nil {
69 return nil, err
70 } else if err := r.readKids(); err != nil {
71 return nil, err
72 }
73 return r, nil
74}
75
76func (r *pdfReader) readCrossReferenceTable() (pdfDict, error) {
77 var starttrailer int

Callers 2

RunMethod · 0.85
RunMethod · 0.85

Calls 7

readTrailerMethod · 0.95
readEncryptMethod · 0.95
readKidsMethod · 0.95
newLineReaderReverseFunction · 0.85
EqualMethod · 0.80
NextMethod · 0.45
PosMethod · 0.45

Tested by

no test coverage detected