Tuesday, June 30, 2009

Firefox taking a long time to start after upgrading to 3.5

Firefox 3.5 (code name Shirekoto) is out, get it while is hot!

After updating it and running it for the first time, I got this little window

It took me a while to even realize that that little guy was somewhere on my screen, I thought Firefox was just taking a long time to start after the update, then I clicked on the taskbar icon and that's when I saw the little window, I tried to re-size it but it didn't allow me to, all I could do was close the little (attempt of) window, then Firefox ran normal

thought I would blog about it in case this happens to more people

symptom: Firefox takes a long time to start after updating to 3.5
solution: click on the taskbar firefox icon and find the little window on your screen, close it, then Firefox will run normal

Wednesday, June 24, 2009

Are we done with XML yet? hello JINI

I have never been a fan of XML, I've hated it from "day one", the format is too inefficient, too bloated; I have had some ideas on things that can be done to reduce the size on xml data sets, such as having some kind of header definition embedded at the top of the file and other ideas, but the point of this post is not to talk those ideas, but rather my own proposal to start getting rid of xml.

One area where I find particularly painful to use XML is on configuration files and my proposal to fix this problem is to formalize a new format, which I call JINI for now. JINI is a subset of JSON similar in simplicity to the good old .ini files and it's used specifically for configuration purposes to replace all those app.config xml configuration files

Benefits
- Simple
- No brackets
- Simple
- Shorter
- Easier to read
- Simpler
- Everything that can be expressed in XML, can be expressed in JINI, but simpler

Too bad "jini" is kinda already taken (2 million results on Google and 714K on Bing say so)

I thought someone would have come up with this idea already, but a quick search gave me nothing, so I thought I would get the ball rolling

what do you think?

Monday, June 22, 2009

How to: Convert DIB to Bitmap

During the weekend a friend asked for some help on code to convert a DIB to a Bitmap in .NET, he had found some code in internet and it almost did everything he needed but the image was getting cut off along the sides, we figured the issue had something to do with getting the pointer to the bitmap, so we digged more and found some more code that then we had to make a couple fixes and changes to get it working. Since I was not able to find a full working solution to the problem I decided to write this post, I saw many people asking the same question in multiple forums so hopefully this will help

First, you’ll need to add System.Drawing.dll to the references of your project, then you’ll need to add the following uses clauses

//Name spaces needed
using System;
using System.Drawing;
using System.Drawing.Imaging;
using System.IO;
using System.Reflection;
using System.Runtime.InteropServices;

the next step is to declare the BITMAPINFOHEADER structure, this can be declared outside your class:

[StructLayout(LayoutKind.Sequential, Pack = 1)]
public struct BITMAPINFOHEADER
{
public uint biSize;
public int biWidth;
public int biHeight;
public ushort biPlanes;
public ushort biBitCount;
public uint biCompression;
public uint biSizeImage;
public int biXPelsPerMeter;
public int biYPelsPerMeter;
public uint biClrUsed;
public uint biClrImportant;

public void Init()
{
biSize = (uint)Marshal.SizeOf(this);
}
}

Then you need to import a function from GdiPlus.dll
//GDI External method needed Place it within your class
[DllImport("GdiPlus.dll", CharSet = CharSet.Unicode, ExactSpelling = true)]
private static extern int GdipCreateBitmapFromGdiDib(IntPtr pBIH,
IntPtr pPix, out IntPtr pBitmap);

Now we can get to the actual code to conver the DIB to Bitmap, note the use of a helper function GetPixelInfo, this was the culprit of our issues, and the one we had a hard time finding then getting to work, the function BitmapFromDIB that you’ll find all over the web doesn’t have this code and so it doesn’t work in many cases, in fact the function that you'll find all over the place has a parameter pPix but it doesn't specify how to get this value. This code needs to be declared inside your class

//THIS METHOD SAVES THE CONTENTS OF THE DIB POINTER INTO A BITMAP OBJECT
private static Bitmap BitmapFromDIB(IntPtr pDIB)
{
//get pointer to bitmap header info
IntPtr pPix = GetPixelInfo(pDIB);

//Call external GDI method
MethodInfo mi = typeof(Bitmap).GetMethod("FromGDIplus", BindingFlags.Static | BindingFlags.NonPublic);
if (mi == null)
return null;

// Initialize memory pointer where Bitmap will be saved
IntPtr pBmp = IntPtr.Zero;

//Call external methosd that saves bitmap into pointer
int status = GdipCreateBitmapFromGdiDib(pDIB, pPix, out pBmp);

//If success return bitmap, if failed return null
if ((status == 0) && (pBmp != IntPtr.Zero))
return (Bitmap)mi.Invoke(null, new object[] { pBmp });
else
return null
;
}

//THIS METHOD GETS THE POINTER TO THE BITMAP HEADER INFO
private static IntPtr GetPixelInfo(IntPtr bmpPtr)
{
BITMAPINFOHEADER bmi = (BITMAPINFOHEADER)Marshal.PtrToStructure(bmpPtr, typeof(BITMAPINFOHEADER));

if (bmi.biSizeImage == 0)
bmi.biSizeImage = (uint)(((((bmi.biWidth * bmi.biBitCount) + 31) & ~31) >> 3) * bmi.biHeight);

int p = (int)bmi.biClrUsed;
if ((p == 0) && (bmi.biBitCount <= 8))
p = 1 << bmi.biBitCount;
p = (p * 4) + (int)bmi.biSize + (int)bmpPtr;
return (IntPtr)p;
}

finally, as a plus, what my friend actually needed was to convert from DIB to TIFF, so he wrote a function to do that, this function reuses the BitmapFromDIB function and allows you to set the image resolution

private void SavehDibToTiff(int hDIB, string fileName, int xRes, int yRes)
{
//Identify the memory pointer to the DIB Handler (hDIB)
IntPtr dibPtr = new IntPtr(hDIB);

//Save the contents of DIB pointer into bitmap object
Bitmap myBitmap = BitmapFromDIB(dibPtr);

//Set resolution if needed
if (xRes >0 && yRes>0)
myBitmap.SetResolution(xRes, yRes);

//Create an instance of the windows TIFF encoder
ImageCodecInfo ici = GetEncoderInfo("image/tiff");

//Define encoder parameters
EncoderParameters eps = new EncoderParameters(1); // only one parameter in this case (compression)

//Create an Encoder Value for TIFF compression Group 4
long ev = (long)EncoderValue.CompressionCCITT4;
eps.Param[0] = new EncoderParameter(System.Drawing.Imaging.Encoder.Compression, ev);

//Save file
myBitmap.Save(fileName, ici, eps);
}
//Helper to get Encoder from Windows for file type.
private static ImageCodecInfo GetEncoderInfo(String mimeType)
{
ImageCodecInfo[] encoders = ImageCodecInfo.GetImageEncoders();
for (int j = 0; j < encoders.Length; ++j)
{
if (encoders[j].MimeType == mimeType)
return encoders[j];
}
return null;
}

and that is it, hoping this will help someone out there

Saturday, June 13, 2009

What are those colors on IE tabs?

colored tabs

I thought this was one of the best features in Internet Explorer 8, but not everyone gets the meaning of the colors on the tabs. The colors represent that the tabs are related, meaning that you followed a link from one page, IE opened another tab and assigned the same color as the previous tab, so that way the tabs that are derived from other tabs are now all grouped and it’s very easy to see this, very useful when you have multiple tabs open for different purposes. I just had to blog this because someone actually asked me what those colors were for, so I figured there’s more people who may have the same question.