private void ImportFromWeb(Uri url)
{
//
// Download
//
string content;
var wc = new WebClient();
wc.Encoding = Encoding.UTF8;
try
{
using (CurrentCursorScope.EnterWait())
content = wc.DownloadString(url);
}
catch (WebException e)
{
Program.ShowExceptionDialog(e, "Import Error", this);
return;
}
//
// Make sure it's HTML otherwise get confirmation to proceed.
//
var typeHeader = wc.ResponseHeaders[HttpResponseHeader.ContentType];
var contentType = new ContentType(typeHeader);
if (string.IsNullOrEmpty(typeHeader) || !contentType.IsHtml())
{
var msg = string.Format(
"The downloaded resource is \u201c{0}\u201d rather than HTML, "
+ "which could produce undesired results. Proceed anyway?",
contentType.MediaType);
if (DialogResult.Yes != MessageBox.Show(this, msg, "Non-HTML Content", MessageBoxButtons.YesNo, MessageBoxIcon.Exclamation))
{
return;
}
}
//
// If it's too big (> 512KB), get confirmation to proceed.
//
var lengthHeader = wc.ResponseHeaders[HttpResponseHeader.ContentLength];
long size;
if (long.TryParse(lengthHeader, NumberStyles.None, CultureInfo.InvariantCulture, out size) &&
size > 512 * 1024)
{
var msg = string.Format(
"The downloaded resource is rather large ({0} bytes). Proceed anyway?",
size.ToString("N0"));
if (DialogResult.Yes != MessageBox.Show(this, msg, "Large Content", MessageBoxButtons.YesNo, MessageBoxIcon.Exclamation))
{
return;
}
}
//
// Load 'er up!
//
using (CurrentCursorScope.EnterWait())
{
var document = new HtmlDocument();
document.LoadHtml(content);
Open(document);
}
_lastKnownGoodImportedUrl = url;
}
private void ImportFromWebMenu_Click(object sender, EventArgs args)
{
Uri url = null;
var input = _lastKnownGoodImportedUrl != null
? _lastKnownGoodImportedUrl.ToString()
: string.Empty;
do
{
input = Interaction.InputBox("Enter URL:", "Import From Web", input,
(int)(Location.X + (Size.Height / 10f)),
(int)(Location.Y + (Size.Height / 10f))).Trim();
if (string.IsNullOrEmpty(input))
{
return;
}
//
// If some entered just the DNS name to get the home page
// then we prepend "http://" for the user to prevent typing.
//
// http://www.shauninman.com/archive/2006/05/08/validating_domain_names
//
if (Regex.IsMatch(input, @"^([a-z0-9] ([-a-z0-9]*[a-z0-9])? \.)+
( (a[cdefgilmnoqrstuwxz]|aero|arpa)
|(b[abdefghijmnorstvwyz]|biz)
|(c[acdfghiklmnorsuvxyz]|cat|com|coop)
|d[ejkmoz]
|(e[ceghrstu]|edu)
|f[ijkmor]
|(g[abdefghilmnpqrstuwy]|gov)
|h[kmnrtu]
|(i[delmnoqrst]|info|int)
|(j[emop]|jobs)
|k[eghimnprwyz]
|l[abcikrstuvy]
|(m[acdghklmnopqrstuvwxyz]|mil|mobi|museum)
|(n[acefgilopruz]|name|net)
|(om|org)
|(p[aefghklmnrstwy]|pro)
|qa
|r[eouw]
|s[abcdeghijklmnortvyz]
|(t[cdfghjklmnoprtvwz]|travel)
|u[agkmsyz]
|v[aceginu]
|w[fs]
|y[etu]
|z[amw])", RegexOptions.IgnoreCase | RegexOptions.CultureInvariant | RegexOptions.IgnorePatternWhitespace | RegexOptions.ExplicitCapture))
{
input = "http://" + input;
}
if (!Uri.IsWellFormedUriString(input, UriKind.Absolute))
{
MessageBox.Show(this, "The entered URL does not appear to be correctly formatted.", "Invalid URL", MessageBoxButtons.OK, MessageBoxIcon.Exclamation);
}
else
{
url = new Uri(input, UriKind.Absolute);
}
}while (url == null);
//
// Download
//
string content;
var wc = new WebClient();
bool proxyAuthAttempted = false;
while (true)
{
try
{
using (CurrentCursorScope.EnterWait())
content = wc.DownloadStringUsingResponseEncoding(url);
break;
}
catch (WebException e)
{
if (e.Status == WebExceptionStatus.ProtocolError &&
!proxyAuthAttempted)
{
var httpResponse = e.Response as HttpWebResponse;
if (httpResponse != null &&
HttpStatusCode.ProxyAuthenticationRequired == httpResponse.StatusCode)
{
proxyAuthAttempted = true;
wc.AddHttpWebRequestModifier(hwr => hwr.Proxy.Credentials = CredentialCache.DefaultCredentials);
continue;
}
}
Program.ShowExceptionDialog(e, "Import Error", this);
return;
}
}
//
// Make sure it's HTML otherwise get confirmation to proceed.
//
var typeHeader = wc.ResponseHeaders[HttpResponseHeader.ContentType];
var contentType = new ContentType(typeHeader);
if (string.IsNullOrEmpty(typeHeader) || !contentType.IsHtml())
{
var msg = string.Format(
"The downloaded resource is \u201c{0}\u201d rather than HTML, "
+ "which could produce undesired results. Proceed anyway?",
contentType.MediaType);
if (DialogResult.Yes != MessageBox.Show(this, msg, "Non-HTML Content", MessageBoxButtons.YesNo, MessageBoxIcon.Exclamation))
{
return;
}
}
//
// If it's too big (> 512KB), get confirmation to proceed.
//
var lengthHeader = wc.ResponseHeaders[HttpResponseHeader.ContentLength];
long size;
if (long.TryParse(lengthHeader, NumberStyles.None, CultureInfo.InvariantCulture, out size) &&
size > 512 * 1024)
{
var msg = string.Format(
"The downloaded resource is rather large ({0} bytes). Proceed anyway?",
size.ToString("N0"));
if (DialogResult.Yes != MessageBox.Show(this, msg, "Large Content", MessageBoxButtons.YesNo, MessageBoxIcon.Exclamation))
{
return;
}
}
//
// Load 'er up!
//
using (CurrentCursorScope.EnterWait())
{
var document = new HtmlDocument();
document.LoadHtml2(content);
Open(document, content);
}
_lastKnownGoodImportedUrl = url;
}
