IOError:在Linux上的ReportLab中使用matplotlib PNG"解码器zip不可用",适用于Windows

Question

我正在使用ReportLab打印matplotlib生成的图表.

我能够在我的Windows开发机器上做到这一点没有问题.但是,当我部署到Ubuntu服务器时,渲染失败并显示错误.我假设我错过了一个Python模块,但我不知道哪个模块.我相信我的开发机器和服务器上的Python,matplotlib,ReportLab和PIL的版本是相同的.

将matplotlib图(称为图表)转换为PNG并返回的代码:

img_stream = StringIO.StringIO()
chart.savefig(img_stream, format = 'png')
img_stream.seek(0)

return img_stream

使用图片的代码:

    res_img = charts.CreateProjectionChart(doc.fund) #calls above code
    if res_img:
        img = ImageReader(res_img)
        canvas.drawImage(img, FromLeft(first_col), FromTop(3.5, 2), width - (.1 * inch), 1.75 * inch, preserveAspectRatio=True, anchor='c')

在Windows上运行时,这是有效的.在Linux上运行时会产生以下错误:

  File "/home/web-server/reports.py", line 913, in FirstPageSetup
    canvas.drawImage(img, FromLeft(first_col), FromTop(3.5, 2), width - (.1 * inch), 1.75 * inch, preserveAspectRatio=True, anchor='c')
  File "/usr/local/lib/python2.7/dist-packages/reportlab-2.5-py2.7-linux-x86_64.egg/reportlab/pdfgen/canvas.py", line 840, in drawImage
    rawdata = image.getRGBData()
  File "/usr/local/lib/python2.7/dist-packages/reportlab-2.5-py2.7-linux-x86_64.egg/reportlab/lib/utils.py", line 658, in getRGBData
    annotateException('\nidentity=%s'%self.identity())
  File "/usr/local/lib/python2.7/dist-packages/reportlab-2.5-py2.7-linux-x86_64.egg/reportlab/lib/utils.py", line 648, in getRGBData
    if Image.VERSION.startswith('1.1.7'): im.load()
  File "/usr/local/lib/python2.7/dist-packages/PIL-1.1.7-py2.7-linux-x86_64.egg/ImageFile.py", line 189, in load
    d = Image._getdecoder(self.mode, d, a, self.decoderconfig)
  File "/usr/local/lib/python2.7/dist-packages/PIL-1.1.7-py2.7-linux-x86_64.egg/Image.py", line 385, in _getdecoder
    raise IOError("decoder %s not available" % decoder_name)
IOError: decoder zip not available
identity=[ImageReader@0x30336d0]
   handle_pageBegin args=()

Answer 1

显然PIL setup.py不知道如何找到libz.so. PIL预计libz.so将位于/usr/lib不/usr/lib/i386-linux-gnu/libz.so.

解决问题

1)使用查找系统libz.so的位置find . -name libz.so.

2)使用创建从libz.so到/ usr/lib的软链接sudo ln -s /usr/lib/i386-linux-gnu/libz.so /usr/lib.

3)正如@Larry建议你在sym链接sol之后重新安装PIL.

要解决64位系统的这个问题,请看这里http://www.foxhop.net/ubuntu-python-easy_install-pil-does-not-install-zlib-support