How do you save images loaded in a Microsoft Web Browser Control?
-
When you get the image URL by enumerating
IHTMLDocument2.images
, make sure you are getting the absolute URL including the host information and scheme. If the server's giving you "Bad Request" then use aHttpWebRequest
which will give you a little more feedback. You could invoke the Save As functionality by usingIWebBrowser.ExecWB
with parameters that are documented in the Platform SDK under "Web Development". Visit http://msdn.microsoft.com/library[^] for details. Basically, execute something like this:axWebBrowser1.ExecWB(
71, // IDM_SAVEAS
0, // OLECMDEXECOPT_DODEFAULT,
path, // The path to save the file
null);The first two parameters - enums - should also be defined in Interop.SHDocVw.dll or Microsoft.mshtml.dll; I really don't remember off-hand. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles]
Ok, if I use HttpWebRequest to reload the url of the image,when I execute GetRequestStream, it throws an exception that says "cannot send a content-body with this verb type". For the webbrowser.ExecWB, I looked over the documentation and it saves that the first parameter is like clicking on the File menu item, and then selecting Save As... What I would need using this implementation would be to right click on an image in the browser and execute Save As... Any ideas? Thanks
-
Ok, if I use HttpWebRequest to reload the url of the image,when I execute GetRequestStream, it throws an exception that says "cannot send a content-body with this verb type". For the webbrowser.ExecWB, I looked over the documentation and it saves that the first parameter is like clicking on the File menu item, and then selecting Save As... What I would need using this implementation would be to right click on an image in the browser and execute Save As... Any ideas? Thanks
There's no support for that via
IOleCommandTarget
(see the docs). You would be able to, however, copy the filename from the save location + "filename_files" directory. Most often, the image retains the filename (only doesn't when a collision occurs). To simulate a user click would require getting the client X and Y coordinates, translating those to screen coordinates, then simulating a right mouse click. You would then have to guess where "Save Image As..." is located (tends to change positions with versions) and simulate a click. This is all very error-prone. Why are you usingGetRequestStream
, though? You do something like this:HttpWebRequest request = (HttpWebRequest)WebRequest.Create(
"https://somesecureserver/path/image.gif");
HttpWebResponse response = (HttpWebResponse)request.GetResponse();
// Now write out the response to a fileUnless you plan on POST'ing information to the site, you don't need the request stream. You should look at the examples in the .NET Framework SDK documentation. For
HttpWebRequest
andHttpWebResponse
there's quite a unique examples. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles] -
There's no support for that via
IOleCommandTarget
(see the docs). You would be able to, however, copy the filename from the save location + "filename_files" directory. Most often, the image retains the filename (only doesn't when a collision occurs). To simulate a user click would require getting the client X and Y coordinates, translating those to screen coordinates, then simulating a right mouse click. You would then have to guess where "Save Image As..." is located (tends to change positions with versions) and simulate a click. This is all very error-prone. Why are you usingGetRequestStream
, though? You do something like this:HttpWebRequest request = (HttpWebRequest)WebRequest.Create(
"https://somesecureserver/path/image.gif");
HttpWebResponse response = (HttpWebResponse)request.GetResponse();
// Now write out the response to a fileUnless you plan on POST'ing information to the site, you don't need the request stream. You should look at the examples in the .NET Framework SDK documentation. For
HttpWebRequest
andHttpWebResponse
there's quite a unique examples. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles]Ok, after parsing the html code in WebBrowser for the url of the image. I execute
HttpWebRequest request = (HttpWebRequest)WebRequest.Create(url);
HttpWebResponse response = (HttpWebResponse)request.GetResponse();Right in the call to GetResponse I get the Bad Request Exception that I got using the WebClient class. Is there something else I must assign to my request so that it knows this request is being made from the same app as the WebBrowser? Thanks
-
Ok, after parsing the html code in WebBrowser for the url of the image. I execute
HttpWebRequest request = (HttpWebRequest)WebRequest.Create(url);
HttpWebResponse response = (HttpWebResponse)request.GetResponse();Right in the call to GetResponse I get the Bad Request Exception that I got using the WebClient class. Is there something else I must assign to my request so that it knows this request is being made from the same app as the WebBrowser? Thanks
The WebBrowser is an in-process component of your application - it is in the same process. What you're asking for is very ambiguous. You can set the
UserAgent
to the same as that for the WebBrowser control, but that doesn't necessarily mean it'll work. You need to be more specific. Is an actual exception being thrown? If so, what is it? If not, is the server returning a page? Have you debugged your code and made sure that your URL is defined correctly while stepping through your code. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles] -
The WebBrowser is an in-process component of your application - it is in the same process. What you're asking for is very ambiguous. You can set the
UserAgent
to the same as that for the WebBrowser control, but that doesn't necessarily mean it'll work. You need to be more specific. Is an actual exception being thrown? If so, what is it? If not, is the server returning a page? Have you debugged your code and made sure that your URL is defined correctly while stepping through your code. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles]I've checked my code, and the url for the image is correct. When I execute GetResponse for the request, a System.Net.WebException is thrown with the message {"The remote server returned an error: (400) Bad Request." } Do you know how I would get the UserAgent of the WebBrowser control? Thank you,
-
I've checked my code, and the url for the image is correct. When I execute GetResponse for the request, a System.Net.WebException is thrown with the message {"The remote server returned an error: (400) Bad Request." } Do you know how I would get the UserAgent of the WebBrowser control? Thank you,
You don't - nor can you (at least not in its entirety) - set the WebBrowser's user-agent. You set it on the
HttpWebRequest
. See theHttpWebBrowser.UserAgent
property in the .NET Framework SDK documentation. You could programmatically set this to be the same as the WebBrowser control's with something like this:IHTMLWindow2 window = (IHTMLWindow2)axWebBrowser;
if (window != null)
{
httpRequest.UserAgent = window.navigator.userAgent;
}That still may not solve the problem. You need to figure out why the server is returning HTTP error code 400. Setting the user-agent may help, but there's many other things that could be wrong. If you have access to the server, check the lots. If it's ASP.NET, open trace.axd in your browser off the web application's root (like http://localhost/myapp/trace.axd). Tracing has to be turned on for that to work, though, and has to allow connections from your host. All that information is in the .NET Framework SDK under the ASP.NET configuration documentation. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles]
-
You don't - nor can you (at least not in its entirety) - set the WebBrowser's user-agent. You set it on the
HttpWebRequest
. See theHttpWebBrowser.UserAgent
property in the .NET Framework SDK documentation. You could programmatically set this to be the same as the WebBrowser control's with something like this:IHTMLWindow2 window = (IHTMLWindow2)axWebBrowser;
if (window != null)
{
httpRequest.UserAgent = window.navigator.userAgent;
}That still may not solve the problem. You need to figure out why the server is returning HTTP error code 400. Setting the user-agent may help, but there's many other things that could be wrong. If you have access to the server, check the lots. If it's ASP.NET, open trace.axd in your browser off the web application's root (like http://localhost/myapp/trace.axd). Tracing has to be turned on for that to work, though, and has to allow connections from your host. All that information is in the .NET Framework SDK under the ASP.NET configuration documentation. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles]
-
Hmm, when I try to do that cast,(IHTMLWindow2)axWebBrowser;, it throws an invalid cast exception. Is there a different way to cast it? Thanks
Then the
WebBrowser
control doesn't implement theIHTMLWindow2
interface (from a COM perspective - not from .NET*). My mistake. What you'll have to do then is castWebBrowser.Document
to anIHTMLDocument2
(defined in Microsoft.mshtml.dll) then useIHTMLDocument2.parentWindow
and cast that to anIHTMLWindow2
. That will get you the reference you need (don't forget to check fornull
!) tonavigator.userAgent
, just like you would in script in DHTML. * When theComImportAttribute
is present on a type, the CLR does aQueryInterface
for an interface on an object instead of a cast as performed in IL instructions. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles] -
Then the
WebBrowser
control doesn't implement theIHTMLWindow2
interface (from a COM perspective - not from .NET*). My mistake. What you'll have to do then is castWebBrowser.Document
to anIHTMLDocument2
(defined in Microsoft.mshtml.dll) then useIHTMLDocument2.parentWindow
and cast that to anIHTMLWindow2
. That will get you the reference you need (don't forget to check fornull
!) tonavigator.userAgent
, just like you would in script in DHTML. * When theComImportAttribute
is present on a type, the CLR does aQueryInterface
for an interface on an object instead of a cast as performed in IL instructions. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles]Interesting, I've definitely learned a lot. I was able to make the cast and pass in the user agent. However, it still sends the 400 bad request exception. What I found though, which is strange is that I can navigate to the image using the webbrowser control using the same or even a different Web Browser Control. Once the browser has navigated I can call this getimage() function, which grabs the picture indirectly from the graphics card buffer. The downside to this is that the browser has to be in focus of course. I still don't get why it would send a bad request to the HttpWebRequest and not to a browser object, but I'm still glad that I can at least grab the picture indirectly. Thanks,
private void getimage() { Control c = this.axWebBrowser1 as Control; Bitmap bmp = CaptureControl(c); bmp.Save("codeword.jpeg",ImageFormat.Jpeg); } [System.Runtime.InteropServices.DllImport("gdi32.dll")] private static extern bool BitBlt(IntPtr hdcDest, int nXDest, int nYDest, int nWidth, int nHeight, IntPtr hdcSrc, int nXSrc, int nYSrc, System.Int32 dwRop); public static Bitmap CaptureControl(System.Windows.Forms.Control control) { Graphics g1 = control.CreateGraphics(); Bitmap bitmap = new Bitmap(control.ClientRectangle.Width, control.ClientRectangle.Height, g1); Graphics g2 = Graphics.FromImage(bitmap); IntPtr dc1 = g1.GetHdc(); try { IntPtr dc2 = g2.GetHdc(); try { BitBlt(dc2, 0, 0, control.ClientRectangle.Width, control.ClientRectangle.Height, dc1, 0, 0, 13369376); } finally { g2.ReleaseHdc(dc2); } } finally { g1.ReleaseHdc(dc1); } return bitmap; }
-
Interesting, I've definitely learned a lot. I was able to make the cast and pass in the user agent. However, it still sends the 400 bad request exception. What I found though, which is strange is that I can navigate to the image using the webbrowser control using the same or even a different Web Browser Control. Once the browser has navigated I can call this getimage() function, which grabs the picture indirectly from the graphics card buffer. The downside to this is that the browser has to be in focus of course. I still don't get why it would send a bad request to the HttpWebRequest and not to a browser object, but I'm still glad that I can at least grab the picture indirectly. Thanks,
private void getimage() { Control c = this.axWebBrowser1 as Control; Bitmap bmp = CaptureControl(c); bmp.Save("codeword.jpeg",ImageFormat.Jpeg); } [System.Runtime.InteropServices.DllImport("gdi32.dll")] private static extern bool BitBlt(IntPtr hdcDest, int nXDest, int nYDest, int nWidth, int nHeight, IntPtr hdcSrc, int nXSrc, int nYSrc, System.Int32 dwRop); public static Bitmap CaptureControl(System.Windows.Forms.Control control) { Graphics g1 = control.CreateGraphics(); Bitmap bitmap = new Bitmap(control.ClientRectangle.Width, control.ClientRectangle.Height, g1); Graphics g2 = Graphics.FromImage(bitmap); IntPtr dc1 = g1.GetHdc(); try { IntPtr dc2 = g2.GetHdc(); try { BitBlt(dc2, 0, 0, control.ClientRectangle.Width, control.ClientRectangle.Height, dc1, 0, 0, 13369376); } finally { g2.ReleaseHdc(dc2); } } finally { g1.ReleaseHdc(dc1); } return bitmap; }
You have a major error in your code - though it's a good workaround to solving your problem: anything that implements
IDisposable
- including (and especially!)Graphics
andBitmap
- should be disposed when you're finished. A good way in C# is like so:using (Graphics g1 = control.CreateGraphics())
{
// ...
}The
using
block makes sure thatDispose
is called, even in case of error. It amounts to this:Graphics g1 = control.CreateGraphics();
try
{
// ...
}
finally
{
if (g1 != null) g1.Dispose();
}Actually, the object is always cast to
IDisposable
in thefinally
block so that explicit interface implementations are handled correctly, but I didn't want to confuse you. Other than that, this should work. If you don't dispose yourGraphics
andBitmaps
above, you'll be leaking resources (memory). Note that you don't have to dispose of controls, though. AllControl
derivatives encapsulates window handles (HWND
s) that are automatically destroyed. Disposing them is not necessary. In order to see why you're getting HTTP 400, you need to take a look at the HTTP request and response. A simple packet sniffer will help. Make sure that you're also requesting https://... A poorly implemented HTTP daemon expecting an SSL socket connection may return 400. There's many other reasons why this might be failing, however. If you're POSTing data from a web browser, for example, you need to do it again (but can be dangerous! you don't want to order a $4,500 segway twice now, do you?!). Like I said, the source of your error judging by your description is impossible to determine. You just have to get down and dirty and debug your code, including the HTTP request and responses sent from both your web browser (i.e., the WebBrowser control that you're embedding) as well as yourHttpWebRequest
andHttpWebResponse
. This posting is provided "AS IS" with no warranties, and confers no rights. Software Design Engineer Developer Division Sustained Engineering Microsoft [My Articles]