Hi Chris, Thanks! (and thanks also for the rest of the comments/help). Basically I ended up writing my own stream reader. I would detect the change to handle the binary data, and character data. It was rather interesting, but, at least I now have a stream reader that can change encodings. ;-)
Thanks again all. Cheers! Dave ----- Original Message ----- From: "Chris Day" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Monday, March 08, 2004 11:48 PM Subject: Re: [ADVANCED-DOTNET] parsing a java webservice One thing you may want to try is that once you have gotten to the start of the document pass the stream into a BinaryReader. I've found that StreamReaders really don't like binary data in the stream! StreamReader sr = new StreamReader(someXmlStream); // read through to word doc BinaryReader br = new BinaryReader(sr.BaseStream); byte[] buf = br.ReadBytes(length); HTH Chris > -----Original Message----- > From: dave wanta [mailto:[EMAIL PROTECTED] > Sent: Saturday, 6 March 2004 1:50 PM > To: [EMAIL PROTECTED] > Subject: Re: [ADVANCED-DOTNET] parsing a java webservice > > See, that's the problem. > > The doc IS NOT base64 encoded. It is raw binary. That's > right. They just threw in binary data in the middle of this document. > > That's why I have to parse the response by hand. Its not even > valid XML. > Unfortunately, the service is provided by the gov't, so the > way it is provided is totally out of my hands. (Don't ask me > how/why it is built this way). Its like they threw in tags > where ever they thought they wanted too. > Then they said "oh btw, here is a binary doc, and its name is > blah.doc. It ends at the start of the next tag". > > So, I parse the string, until I get to the start of the doc, > then grab the sub string. But when I convert it back to > binary ( Encoding.UTF8.GetBytes( docSubstring ) ), the data > gets corrupt. > > hth, > Dave > PS: I guess the transport is the java webservice, grrr.., and > the real problem I'm having is going byte[] --> string --> > byte[] without corruption. > I'm wondering if I need to build my own reader, so I can > switch encodings on the fly. > > > > > ----- Original Message ----- > From: "Streno, Robert" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Friday, March 05, 2004 4:34 PM > Subject: Re: [ADVANCED-DOTNET] parsing a java webservice > > > > Why are you parsing the raw SOAP? Why not make a call > directly to the > > web service using a proxy constructed from the WSDL for the > web service? > > > > Once you do that, you should be able to reference the parameter as a > string. > > Assuming that the doc has been base64 encoded, it should be a > > relatively simple task to write the individual bytes out to > a disk file. > > > > -Rob > > > > -----Original Message----- > > From: Moderated discussion of advanced .NET topics. > > [mailto:[EMAIL PROTECTED] On Behalf Of dave wanta > > Sent: Friday, March 05, 2004 11:44 AM > > To: [EMAIL PROTECTED] > > Subject: [ADVANCED-DOTNET] parsing a java webservice > > > > > > Hi All, > > Hopefully I can explain this one with out mucking this > up.;-) I've got > > to consume a Java webservice (which I know nothing about). > > > > A co-worker gave me the code that mades the Soap request, using a > WebRequest > > (only way we could get .NET to talk to the service successfully). > > Which works great. > > > > So now I get the ResponseStream, which contains the Soap > XML (of which > > is embedded a binary word document). I need to parse the response > > stream to find where the word document starts and ends, and > extract it. > > > > So I'm reading the stream with a StreamReader, using > Encoding.UTF8. I > locate > > where the word doc starts and stops in the string, and then covert > > that substring back to a byte[], and save it. but the word doc gets > > corrupt in the process. Are there any recommendations on how to > > convert Stream --> String --> SubString --> Byte[] without > > loosing/corrupting any of the > data? > > > > If I go Stream --> byte[], and manually locate the word doc ( I > physically > > counted the offset where the doc starts and ends), and then > save that > > sub array, everything works. Obviously, I can't do this in > production, > > but I verified the word doc is not corrupt. > > > > Any ideas? > > > > Thanks, > > Dave > > > > =================================== > > This list is hosted by DevelopMentor(r) > http://www.develop.com Some > > .NET courses you may be interested in: > > > > NEW! Guerrilla ASP.NET, 17 May 2004, in Los Angeles > > http://www.develop.com/courses/gaspdotnetls > > > > View archives and manage your subscription(s) at > http://discuss.develop.com > > > > =================================== > > This list is hosted by DevelopMentor(r) http://www.develop.com Some > > .NET courses you may be interested in: > > > > NEW! Guerrilla ASP.NET, 17 May 2004, in Los Angeles > > http://www.develop.com/courses/gaspdotnetls > > > > View archives and manage your subscription(s) at > http://discuss.develop.com > > > > =================================== > This list is hosted by DevelopMentor(r) http://www.develop.com > Some .NET courses you may be interested in: > > NEW! Guerrilla ASP.NET, 17 May 2004, in Los Angeles > http://www.develop.com/courses/gaspdotnetls > > View archives and manage your subscription(s) at > http://discuss.develop.com > > =================================== This list is hosted by DevelopMentorŪ http://www.develop.com Some .NET courses you may be interested in: NEW! Guerrilla ASP.NET, 17 May 2004, in Los Angeles http://www.develop.com/courses/gaspdotnetls View archives and manage your subscription(s) at http://discuss.develop.com =================================== This list is hosted by DevelopMentorŪ http://www.develop.com Some .NET courses you may be interested in: NEW! Guerrilla ASP.NET, 17 May 2004, in Los Angeles http://www.develop.com/courses/gaspdotnetls View archives and manage your subscription(s) at http://discuss.develop.com
