Code Download
This project requires that you download the HtmlAgilityPack separately and place the HtmlAgilityPack.dll into the appropriate Debug and/or Release folder.
Introduction
As far as I could tell, the NOAA weather webservice does not provide a function for obtaining the weather radar image of particular location, nor is this feature readily provided by other weather web services. So here I present a "bare-bones" implementation, using HtmlAgilityPack for some simple XPath queries.
The radar.weather.gov Website
If you inspect a page displaying a radar image on this website, you'll notice that the image is actually a composite of 8 selectable pieces:
The first image is a JPG, and the remaining are GIF's:
Topography
|
Doppler Radar
|
Counties
|
Rivers
|
Roads
|
Cities
| Warnings (none at this moment)
|
Legend
|
So, given you're nearest radar station (pick one on http://radar.weather.gov) we need to obtain these images and combine them together to show the final composite image. There's a slight twist - two of the images come from a different source and are computed by the Javascript on the page, so we have special handlers for these. There are five steps:
Step 1: Acquiring the Page HTML
Given the URL entered in the textbox, we first acquire the HTML for the page:
protected string GetPageHtml(string url)
{
using (WebClient client = new WebClient())
{
return client.DownloadString(url);
}
}
Step 2: Figuring out the Image URL's
This is the most complex part of the process, as we have to programmatically generate two of the image URL's. You'll also see how we use the HtmlAgilityPack to extract the specific node values for the div
sections you saw earlier.
protected List<string> GetPageImageUrls(string html, string url)
{
List<string> ret = new List<string>();
HtmlAgilityPack.HtmlDocument doc = new HtmlAgilityPack.HtmlDocument();
doc.LoadHtml(html);
WebClient webClient = new WebClient();
int n = 0;
while (true)
{
var nodes = doc.DocumentNode.SelectNodes(String.Format("//div[@id='image{0}']/img", n++));
if ((nodes != null) && (nodes.Count == 1))
{
string imageUrl = nodes.Select(node => node.Attributes["src"].Value).Single();
if (imageUrl == "#")
{
string name = nodes.Select(node => node.Attributes["name"].Value).Single();
string rid = url.Between("rid=", "&").ToUpper();
string product = url.Between("product=", "&").ToUpper();
switch (name)
{
case "conditionalimage":
imageUrl = String.Format("RadarImg/{0}/{1}_{0}_0.gif", product, rid);
break;
case "conditionallegend":
imageUrl = String.Format("Legend/{0}/{1}_{0}_Legend_0.gif", product, rid);
break;
}
}
ret.Add(imageUrl);
}
else
{
break;
}
}
webClient.Dispose();
return ret;
}
Step 3: Downloading the Images
Here we simply download the images, saving the Image
and its associated MemoryStream
instance. While I could clone the image at this point and free up the memory stream and the original image, I opted to keep the two together as a package for later cleanup.
protected List<ImageData> DownloadImages(List<string> imageUrls)
{
List<ImageData> ret = new List<ImageData>();
WebClient webClient = new WebClient();
foreach (string url in imageUrls)
{
byte[] data = webClient.DownloadData("http://radar.weather.gov/" + url);
MemoryStream stream = new MemoryStream(data);
Image image = Image.FromStream(stream);
ret.Add(new ImageData() { Image = image, BackingStream = stream });
}
webClient.Dispose();
return ret;
}
Step 4: For fun, we Save the Images
You'll note my assumption that the first image is a JPG and the rest are GIF's. Yes, we could figure this out by inspecting the raw image format, but that's more work than I wanted to put into the code at the moment.
protected void WriteImages(List<ImageData> images)
{
int n=0;
images.ForEach(img => img.Image.Save("img" + n++ + (n==1 ? ".jpg" : ".gif")));
}
Step 5: Combine the Images
Here we combine the images, returning a new image consisting of the composite image data:
protected Image CombineImages(Graphics gr, List<ImageData> images, Size size)
{
Image baseImage = (Image)images[0].Image.Clone();
gr = Graphics.FromImage(baseImage);
for (int i=1; i<images.Count; i++)
{
gr.DrawImage(images[i].Image, new Point(0, 0));
}
gr.Dispose();
return baseImage;
}
Step 6: Cleanup
Here we dispose of the images and their the associated memory streams.
protected void Cleanup(List<ImageData> images)
{
images.ForEach(img =>
{
img.BackingStream.Dispose();
img.Image.Dispose();
});
}
Putting it all Together
When you click on the Go button, an async process is started, and we provide a callback to display the progress on the status bar:
protected async void btnGo_Click(object sender, EventArgs e)
{
btnGo.Enabled = false;
string url = tbUrl.Text;
Graphics gr = pbRadar.CreateGraphics();
Image bitmap = await Task.Run(() => GetRadarImage(gr,
url,
(progress) => this.BeginInvoke(() => tsLabel.Text = progress)));
pbRadar.Image = bitmap;
tsLabel.Text = "Done";
btnGo.Enabled = true;
}
And the implementation of GetRadarImage chains together the processes we described above:
protected Image GetRadarImage(Graphics gr, string url, Action<string> progressCallback)
{
progressCallback("Acquiring page...");
string html = GetPageHtml(url);
progressCallback("Scraping page...");
List<string> imageUrls = GetPageImageUrls(html, url);
progressCallback("Downloading images...");
List<ImageData> images = DownloadImages(imageUrls);
progressCallback("Writing images...");
WriteImages(images);
progressCallback("Combining images...");
Image bitmap = CombineImages(gr, images, pbRadar.Size);
progressCallback("Cleanup...");
Cleanup(images);
return bitmap;
}
Extension Methods
To make my life a bit easier, I borrowed some extension methods that I use in other projects:
public static class Extensions
{
public static string Between(this String src, string s1, string s2)
{
return src.RightOf(s1).LeftOf(s2);
}
public static string RightOf(this String src, string s)
{
string ret = String.Empty;
int idx = src.IndexOf(s);
if (idx != -1)
{
ret = src.Substring(idx + s.Length);
}
return ret;
}
public static string LeftOf(this String src, string s)
{
string ret = src;
int idx = src.IndexOf(s);
if (idx != -1)
{
ret = src.Substring(0, idx);
}
return ret;
}
public static void BeginInvoke(this Control control, Action action)
{
if (control.InvokeRequired)
{
control.BeginInvoke((Delegate)action);
}
else
{
action();
}
}
}
Conclusion
And there you have it, a bare-bones approach to acquiring a radar image in your own application!