如何以http格式读取文件内容？

Question

我使用UNO来读取本地机器中的文件内容。没关系。但是当UNO以http格式读取文件内容时，它会显示错误：

Error (<class '__main__.com.sun.star.lang.IllegalArgumentException'>) during conversion:Unsupported URL <http://crypto.stanford.edu/DRM2002/darknet5.doc>: "type detection failed".

如何以http格式读取文件内容？ 这是http://www.openoffice.org/udk/python/python-bridge.html

中的代码

import getopt, sys
import uno

from unohelper import Base, systemPathToFileUrl, absolutize
from os import getcwd
from os.path import splitext
from com.sun.star.beans import PropertyValue
from com.sun.star.uno import Exception as UnoException
from com.sun.star.io import IOException, XOutputStream

class OutputStream( Base, XOutputStream ):
    def __init__( self ):
        self.closed = 0
    def closeOutput(self):
        self.closed = 1
    def writeBytes( self, seq ):
        print(seq.value)
    def flush( self ):
        pass

def main():
    retVal = 0
    doc = None
    stdout = False

    try:
        opts, args = getopt.getopt(sys.argv[1:], "hc:",
            ["help", "connection-string=" , "html", "pdf", "stdout" ])
        url = "uno:socket,host=0.0.0.0,port=2002;urp;StarOffice.ComponentContext"
        filterName = "Text (Encoded)"
        extension  = "txt"
        for o, a in opts:
            if o in ("-h", "--help"):
                usage()
                sys.exit()
            if o in ("-c", "--connection-string" ):
                url = "uno:" + a + ";urp;StarOffice.ComponentContext"
            if o == "--html":
                filterName = "HTML (StarWriter)"
                extension  = "html"
            if o == "--pdf":
                filterName = "writer_pdf_Export"
                extension  = "pdf"
            if o == "--stdout":
                stdout = True

        if not len( args ):
            usage()
            sys.exit()

        ctxLocal = uno.getComponentContext()
        smgrLocal = ctxLocal.ServiceManager

        resolver = smgrLocal.createInstanceWithContext(
                 "com.sun.star.bridge.UnoUrlResolver", ctxLocal )
        ctx = resolver.resolve( url )
        smgr = ctx.ServiceManager

        desktop = smgr.createInstanceWithContext("com.sun.star.frame.Desktop", ctx )

        cwd = systemPathToFileUrl( getcwd() )
        outProps = (
            PropertyValue( "FilterName" , 0, filterName , 0 ),
        PropertyValue( "Overwrite" , 0, True , 0 ),
            PropertyValue( "OutputStream", 0, OutputStream(), 0)
        )

        inProps = PropertyValue( "Hidden" , 0 , True, 0 ),
        for path in args:
            try:
                fileUrl = absolutize( cwd, systemPathToFileUrl(path) )
                doc = desktop.loadComponentFromURL( fileUrl , "_blank", 0, inProps )

                if not doc:
                    raise UnoException( "Couldn't open stream for unknown reason", None )

                if not stdout:
                    (dest, ext) = splitext(path)
                    dest = dest + "." + extension
                    destUrl = absolutize( cwd, systemPathToFileUrl(dest) )
                    sys.stderr.write(destUrl + "\n")
                    doc.storeToURL(destUrl, outProps)
                else:
                    doc.storeToURL("private:stream",outProps)
            except IOException as e:
                sys.stderr.write( "Error during conversion: " + e.Message + "\n" )
                retVal = 1
            except UnoException as e:
                sys.stderr.write( "Error ("+repr(e.__class__)+") during conversion:" + e.Message + "\n" )
                retVal = 1
            if doc:
                doc.dispose()

    except UnoException as e:
        sys.stderr.write( "Error ("+repr(e.__class__)+") :" + e.Message + "\n" )
        retVal = 1
    except getopt.GetoptError as e:
        sys.stderr.write( str(e) + "\n" )
        usage()
        retVal = 1

    sys.exit(retVal)

def usage():
    sys.stderr.write( "usage: ooextract.py --help | --stdout\n"+
                  "       [-c <connection-string> | --connection-string=<connection-string>\n"+
          "       [--html|--pdf]\n"+
          "       [--stdout]\n"+
                  "       file1 file2 ...\n"+
                  "\n" +
                  "Extracts plain text from documents and prints it to a file (unless --stdout is specified).\n" +
                  "Requires an OpenOffice.org instance to be running. The script and the\n"+
                  "running OpenOffice.org instance must be able to access the file with\n"+
                  "by the same system path. [ To have a listening OpenOffice.org instance, just run:\n"+
          "openoffice \"-accept=socket,host=localhost,port=2002;urp;\" \n"
                  "\n"+
          "--stdout \n" +
          "         Redirect output to stdout. Avoids writing to a file directly\n" + 
                  "-c <connection-string> | --connection-string=<connection-string>\n" +
                  "        The connection-string part of a uno url to where the\n" +
                  "        the script should connect to in order to do the conversion.\n" +
                  "        The strings defaults to socket,host=localhost,port=2002\n"
                  "--html \n"
                  "        Instead of the text filter, the writer html filter is used\n"
                  "--pdf \n"
                  "        Instead of the text filter, the pdf filter is used\n"
                  )

main()

这是我执行程序的方式：

1. openoffice.org --accept="socket,host=0.0.0.0,port=2002;urp;StarOffice.ServiceManager"
2. python ooextract.py --stdout /home/tungdq/hello.doc -> ok
3. python ooextract.py --stdout http://crypto.stanford.edu/DRM2002/darknet5.doc -> error