无论页数如何，pyPdf 输出文件的大小都相同

Question

我正在尝试使用 pyPdf 将大型 pdf 中的几页提取到单独的文件中。每当我这样做时，生成的文件大小几乎与源文件相同。我认为这与文件内的书签有关，因为如果页面不包含任何链接，输出文件的大小会非常小。我不知道如何从输出文件中排除书签。

from pyPdf import PdfFileWriter as writer, PdfFileReader as reader
w = writer()
r = reader(open('9.pdf'))

for p in xrange(5):
    w.addPage(r.getPage(p))
with open('out.pdf', 'wb') as stream:
    w.write(stream)

w._objects
# prints:

{'/Kids': [IndirectObject(4, 0), IndirectObject(5, 0), IndirectObject(6, 0), IndirectObject(7, 0), IndirectObject(8, 0)], '/Type': '/Pages', '/Count': 5}
{'/Producer': u'Python PDF Library - http://pybrary.net/pyPdf/'}
{'/Type': '/Catalog', '/Pages': IndirectObject(1, 0)}
{'/Parent': IndirectObject(1, 0), '/Rotate': 0, '/Contents': IndirectObject(4307, 0), '/Resources': {'/ColorSpace': {'/CS1': IndirectObject(4309, 0), '/CS0': IndirectObject(4305, 0)}, '/XObject': {'/Im0': IndirectObject(4312, 0)}, '/ExtGState': {'/GS2': IndirectObject(4324, 0), '/GS1': IndirectObject(4323, 0), '/GS0': IndirectObject(4306, 0)}, '/Font': {'/T1_2': IndirectObject(4308, 0), '/T1_0': IndirectObject(4303, 0), '/T1_1': IndirectObject(4304, 0)}, '/ProcSet': ['/PDF', '/Text', '/ImageB']}, '/CropBox': [0, 0, 612, 792], '/BCLPrivAnnots': {'/BCLC_BCL_Jade': []}, '/MediaBox': [0, 0, 612, 792], '/Annots': IndirectObject(4301, 0), '/Type': '/Page'}
{'/Parent': IndirectObject(1, 0), '/Contents': IndirectObject(2, 0), '/Resources': {'/ColorSpace': {'/CS1': IndirectObject(4309, 0), '/CS0': IndirectObject(4305, 0)}, '/ExtGState': {'/GS2': IndirectObject(3417, 0), '/GS1': IndirectObject(3412, 0), '/GS0': IndirectObject(4306, 0)}, '/Font': {'/T1_2': IndirectObject(3413, 0), '/T1_0': IndirectObject(3415, 0), '/T1_1': IndirectObject(3416, 0)}, '/ProcSet': ['/PDF', '/Text']}, '/Rotate': 0, '/CropBox': [0, 0, 612, 792], '/BCLPrivAnnots': {'/BCLC_BCL_Jade': []}, '/MediaBox': [0, 0, 612, 792], '/Thumb': IndirectObject(3920, 0), '/Type': '/Page'}
{'/Parent': IndirectObject(1, 0), '/Contents': IndirectObject(4, 0), '/Resources': {'/ColorSpace': {'/CS0': IndirectObject(4305, 0)}, '/ExtGState': {'/GS0': IndirectObject(4306, 0)}, '/Font': {'/T1_2': IndirectObject(3425, 0), '/T1_3': IndirectObject(3428, 0), '/T1_0': IndirectObject(3426, 0), '/T1_1': IndirectObject(3427, 0)}, '/ProcSet': ['/PDF', '/Text']}, '/Rotate': 0, '/CropBox': [0, 0, 612, 792], '/BCLPrivAnnots': {'/BCLC_BCL_Jade': []}, '/MediaBox': [0, 0, 612, 792], '/Thumb': IndirectObject(3921, 0), '/Type': '/Page'}
{'/Parent': IndirectObject(1, 0), '/Contents': IndirectObject(6, 0), '/Resources': {}, '/Rotate': 0, '/CropBox': [0, 0, 612, 792], '/BCLPrivAnnots': {'/BCLC_BCL_Jade': []}, '/MediaBox': [0, 0, 612, 792], '/Thumb': IndirectObject(3922, 0), '/Type': '/Page'}
{'/Parent': IndirectObject(1, 0), '/Contents': IndirectObject(9, 0), '/Resources': IndirectObject(8, 0), '/Rotate': 0, '/CropBox': [0, 0, 612, 792], '/BCLPrivAnnots': {'/BCLC_BCL_Jade': []}, '/MediaBox': [0, 0, 612, 792], '/Thumb': IndirectObject(3923, 0), '/Type': '/Page'}

Answer 1

使用 PyPdf 时，输出文件几乎包含所有格式。

具体来说，替换：

with open('out.pdf', 'wb') as stream:
    w.write(stream)

for：

stream = file('out.pdf', 'wb')
w.write(stream)
stream.close()

然后查看最终结果。

也是一个很好的做法。

fin = open('9.pdf')
r = reader(fin)
fin.close()

写 :而不是：

r = reader(open('9.pdf'))