Hey, I've got this nice little piece of code, much like all the other versions of this method of upload using WSS WebServices. I've got one major problem though - once I have uploaded a file into my doc list, and updated the list item to write a comment/description, the file is stuck there. What I mean is that this method will not overwrite the file once I've uploaded it. Nobody else out there seems to have posted this issue yet, so .. anyone?
I have another version of the method which uses a byte[] instead of a Stream .. same issue though.
Note: I have switched off the 'require documents to be checked out before they can be edited' option for the library. No luck tho .. The doc library does have versioning turned on though, with a major version being created for each update.
private void UploadStream(string fullPath, Stream uploadStream)
{
WebRequest request = WebRequest.Create(fullPath);
request.Credentials = CredentialCache.DefaultCredentials; // User must have 'Contributor' access to the document library
request.Method = "PUT";
request.Headers.Add("Overwrite", "t");
byte[] buffer = new byte[4096];
using (Stream stream = request.GetRequestStream())
{
for (int i = uploadStream.Read(buffer, 0, buffer.Length); i > 0; i = uploadStream.Read(buffer, 0, buffer.Length))
{
stream.Write(buffer, 0, i);
}
}
WebResponse response = request.GetResponse(); // Upload the file
response.Close();
}
Original credits to: http://geek.hubkey.com/2007/10/upload-file-to-sharepoint-document.html
EDIT -- major finding .. when I call it from my nUnit test project it works fine. It seems it only fails when I call it from my WCF application (nUnit running under logged on user account, WCF app has app pool running under that same user -- my account, which also has valid permissions in SharePoint).
Nuts. "Now where to start?!", I muses to myself.
SOLVED -- I found a little bug - the file was being created in the right place, but the update path was wrong.. I ended up finding a folder full of files with many, many new versions.. doh!
Why not use the out-of-the-box SharePoint webservice, Lists.asmx? You'll find it in
http://SITEURL/___vti_bin/Lists.asmx
Edit, I checked out the link and it seems you are calling the out of the box web service. This has got be versioning related then. Can you check out the different versions that exist in the doc lib of the specific file? see if it perhaps gets added as a minor version through the service?
Have you tried using a capital T? SharePoint's webdav header processing is not very likely to be case-sensitive, but the protocol does specify a capital T. Oh, and what is the response? A 412 error code or something altogether different?
Related
I have a Sitecore 6.4 setup where an editor can click a button to generate a Word doc. I was adding the file to the media library and that was working fine. The editor would click the button in the content editor, the file was generated, media item was generated, then the content editor would show the new item in the media library and the editor could click the "download" button on the ribbon or the item to download it. However, my media library was getting unnecessarily filled up so I am trying to bypass the media library.
Instead of making the file in an arbitrary location as before, I am putting it in the temp directory like this:
wordOutputPath = Sitecore.IO.FileUtil.GetWorkFilename(Sitecore.Configuration.Settings.TempFolderPath, printItem.Name, ".docx");
File.Copy(wordTemplatePath, wordOutputPath);
WordprocessingDocument doc = WordprocessingDocument.Open(wordOutputPath, true);
After I "fill in" the file with content, I do this:
Sitecore.Context.ClientPage.ClientResponse.Download(wordFilePath);
Now if I am logged in as a Sitecore admin I get the browser's download dialog and can download the file. However, if I am logged in as a non-admin user, I get a little clicking and whirring, so to speak, and the file is generated, but the save file dialog never comes up in the browser. I can go in through the file system and see & open the Word doc, and it looks fine.
I found something in the Sitecore release notes for 6.6:
Released Sitecore CMS and DMS 6.6.0 rev. 130111 (6.6.0 Update-3)
[...]
Miscellaneous
Only Administrators were allowed to download files. (316774, 348557)
This was a problem in several areas of the system, for example the Package Generator and the Export Language Wizard in the CMS. It also affected the Export Users Wizard in the ECM module.
So I tried using SecurityDisabler (no longer have the code handy) and UserSwitcher like this:
using (new Sitecore.Security.Accounts.UserSwitcher(Sitecore.Security.Accounts.User.FromName("sitecore\admin", false)))
{
Sitecore.Context.ClientPage.ClientResponse.Download(wordFilePath);
}
The IUSR and IIS_IUSRS accounts both have read, list & read/execute permissions on the temp folder and the individual files show read & read/execute permissions for those two accounts as well.
What can I do to allow non-admin users to download these files? Does the bug that was fixed in 6.6 have anything to do with it?
Thank you.
Calling Sitecore.Context.ClientPage.ClientResponse.Download() method will result in a Download command sent back to the Sitecore Client in the Web browser, which in turn will make a call back to the server to execute it. This will be why using a security switcher (or a security disabler for that matter) before calling the Sitecore.Context.ClientPage.ClientResponse.Download() method has no effect, as the admin only restriction is being applied to the download else where.
If you search through sitecore\shell directory in your webroot, you will find that there is a file called download.aspx. This should be the actual page the download request is sent to. This web form page inherits the Sitecore.Shell.DownloadPage class, so if you take a look at that class's implementation, you will find that the OnLoad method is implemented as such:
protected override void OnLoad(EventArgs e)
{
Assert.ArgumentNotNull(e, "e");
base.OnLoad(e);
string fileHandle = StringUtil.GetString(new string[] { base.Request.QueryString["file"] });
bool flag = false;
string filename = FileHandle.GetFilename(fileHandle);
if (!string.IsNullOrEmpty(filename))
{
fileHandle = filename;
flag = true;
}
if (!string.IsNullOrEmpty(fileHandle))
{
if (MediaManager.IsMediaUrl(fileHandle))
{
this.DownloadMediaFile(fileHandle);
}
else if (fileHandle.IndexOf("id=", StringComparison.OrdinalIgnoreCase) >= 0)
{
this.DownloadMediaById(fileHandle);
}
else if (flag || Context.IsAdministrator)
{
this.DownloadFile(fileHandle);
}
}
}
As you can see, the last if block has an IsAdministrator check which would be where your non-admin users are getting blocked.
Although I haven't 100% verified that this is where the problem lies, the download restriction issue can be resolved, by creating a new class that inherits Sitecore.Shell.DownloadPage and overrides the OnLoad implementation with some extra security checks as required. I recommend that you don't remove the IsAdministrator check altogether as it is there for a good reason. Update the Inherits attribute in the download.aspx to make it use it instead.
I am new to the wsdl\soapmessage query\reply world( if i can put it in this way), and I am facing some difficulties using the following wsdl( which I really really hope, one will be so kind to look at at least one of the services described there)
http://almdemo.polarion.com/polarion/ws/services/TrackerWebService?wsdl
which was provided to me to develop a matlab webinterface. Right now my matlab code looks like this:
targetNamespace = 'http://ws.polarion.com/TrackerWebService';
method = 'queryWorkItems';
values= {'Query','Sort'}
names = {'query', 'sort'}
types ={'xsd:string','xsd:string'}
message = createSoapMessage( targetNamespace, method, values, names, types)
response = callSoapService('http://almdemo.polarion.com/polarion/ws/services',...
% Service's endpoint
'http://almdemo.polarion.com/polarion/#/workitems',...
% Server method to run
message)
% SOAP message created using createSoapMessage
author = parseSoapResponse(response)
Herewith to save you time I will just enonce my two problems:
Is the code correct?
Could someone tell me if such a wsdl link is just a definition of webservices or it is also a service's endpoint?
Normally to execute manually\per clicks this services on the weppage
http://almdemo.polarion.com/polarion, you have to login!
So how do I send a message in matlab which first log me in? Or must such a service be introduced into that wsdl for me to do it?? Could you be kind enough to write how it must be defined, because I don't really
write wsdl files, but I shall learn!**
I will be really thankful for your help and I wish you a happy week(-end) guys(& girls)!!!
Regards
Chrysmac
ps: I tried to use Soapui and gave that webpage as endpoint, but the toool crashes each time I enter my credentials! Maybe because of the dataload!??
I need to create a simple web service (being the "server"). The goal is to provide some data I do read in an Qt / C++ application as JSON data. Basically a JavaScript application in the browser shall read its data from the Qt app. It is usually a single user scenario, so the user runs a Google Maps application in her browser, while additional data come from the Qt application.
So far I have found these libs:
Qxt: http://libqxt.bitbucket.org/doc/0.6/index.html but being a newbie on C++/Qt I miss some examples. Added: I have found one example here
gSoap: http://www.cs.fsu.edu/~engelen/soap.html has more examples and documentation and also seems to support JSON
KD SOAP: http://www.kdab.com/kdab-products/kd-soap/ with no example as far as I can tell, docu is here
Qt features itself, but it is more about acting as a client: http://qt-project.org/videos/watch/qt-networking-web-services
Checking SO gives me basically links to the above libs
webservice with Qt with an example I do not really get.
How to Create a webservice by Qt
So basically I do have the following questions:
Which lib would you use? I want to keep it as simple as possible and would need an example.
Is there another (easy!) way to provide the JSON data to the JavaScript Web page besides the WebService?
-- Edit, remarks: ---
Needs to be application intrinsic. No web server can be installed, no extra run time can be used. The user just runs the app. Maybe the Qt WebKit could be an approach....
-- Edit 2 --
Currently checking the tiny web servers as of SO " Qt HTTP Server? "
As of my tests, currently I am using QtWebApp: http://stefanfrings.de/qtwebapp/index-en.html This is one of the answers of Edit 2 ( Qt HTTP Server? )
Stefan's small WebServer has some well documented code, is written in "Qt C++" and easy to use, especially if you have worked with servlets already. Since it can be easily integrated in my Qt project, I'll end up with an internal WebServer.
Some demo code from my JSON tests, showing that generating the JSON content is basically creating a QString.
void WebServiceController::service(HttpRequest& request, HttpResponse& response) {
// set some headers
response.setHeader("Content-Type", "application/json; charset=ISO-8859-1");
response.setCookie(HttpCookie("wsTest","CreateDummyPerson",600));
QString dp = WebServiceController::getDummyPerson();
QByteArray ba = dp.toLocal8Bit();
const char *baChar = ba.data();
response.write(ba);
}
If someone has easy examples with other libs to share, please let me know.
QByteArray ba = dp.toLocal8Bit();
const char *baChar = ba.data();
You don't need to convert the QByteArray to char array. Response.write() can also be called with a QByteArray.
By the way: qPrintable(dp) is a shortcut to convert from QString to char array.
I'm trying to use the libcurl in a C/C++ application to post files to DropBox.
I would like to use the "/files (POST)" API as documented here...
https://www.dropbox.com/developers/reference/api#files-POST
I am having problems with properly authenticating (OAuth) this call. It is unclear to me how to properly create the authentication signature.
From some a sample I saw, it looked like they were reading in the whole file to create the HMAC-SHA1 encoding on. This seems problematic on large files.
Does anyone have experience or insight using this API or something similar?
I have just use the libouth and libcurl to get information from sina weibo. here is my example for you refer. you can also refer the liboauth test programmer in the tests dir, oauthtest.c
if (use_post)
{
req_url = oauth_sign_url2(test_call_uri, &postarg, OA_HMAC, NULL, c_key, c_secret, t_key, t_secret);
reply = oauth_http_post(req_url,postarg);
}
I suggest using BOOST ASIO . Makes uploading and downloading a breeze.
The below web services code has worked properly for me for over a year. We have updated our SharePoint servers, and now the below code throws an exception (at the bottom line of code) "Object reference not set to an instance of an object"
UserProfileWS.UserProfileService userProfileService = new UserProfileWS.UserProfileService();
userProfileService.Credentials = System.Net.CredentialCache.DefaultNetworkCredentials;
string serviceloc = "/_vti_bin/UserProfileService.asmx";
userProfileService.Url = _webUrl + serviceloc;
UserProfileWS.PropertyData[] info = userProfileService.GetUserProfileByName(null);
EDIT: The service is still there. I browse http:///_vti_bin/UserProfileService.asmx, and the information for the service is still there, including the full description of the GetUserProfileByName call.
EDIT2: This does appear to be due to a change in SharePoint. I loaded a previous version of my software (known to be working), and it exhibits the same erroneous behavior.
try
UserProfileWS.PropertyData[] info = userProfileService.GetUserProfileByName(userName);
as specified http://msdn.microsoft.com/en-us/library/microsoft.office.server.userprofiles.userprofileservice.getuserprofilebyname(v=office.12).aspx
When was the farm updated? Was the WSS updates installed before the MOSS updates? If you believe it to be a problem as a result of infrastructure updates, build a test farm and try the code against pre-updates (go back as far as a year ago to start off).