We Keep Coding

c#, check area of pdf

I need to enter a text to existing pdf (in top or bottom of the page) in c#.
I need to make sure that I dont overwrite any visible text or image.
Is there any way I could check an area in pdf if it contains text, image, control etc? I understand it will not be 100% accurate

You're going to need a full PDF consumer at the very least, because the only way to find out where the marks are on the page is to parse (and possibly render) the PDF.
There are complications which you haven't covered (possibly they have not occurred to you); what do you consider to be the area of the PDF file ? The MediaBox ? CropBox, TrimBox, ArtBox, BleedBox ? What if the PDF file contains, for example, a rectangular fill with white which covers the page ? What about a /Separation space called /White ? is that white (it generally gets rendered that way on the output) or not ? And yes, this is a widely used ink in the T-shirt printing industry amongst others.
To me the simplest solution would seem to be to use a tool which will give you the BoundingBox of marks on the page. I know the Ghostscript bbox device can do this, I imagine there are other tools which can do so. But note (for Ghostscript at least); if there are any marks in white (whatever the colour space), these are considered as marking the page and will be counted into the bbox.
The same tool ought to be able to give the size of the various Boxes in the PDF file (you'd need the pdf_info.ps program for Ghostscript to get this, currently). You can then quickly calculate which areas are unmarked.
But 'unmarked' isn't the same things as 'white'. If you want to not count areas which are painted in 'white' then the problem becomes greater. You really need to render the content and then look at each image sample in the output to see if its white or not, recording the maxima and minima of the x and y co-ordinates to determine the 'non-white' area of the page.
This is because there are complications like transfer functions, transparency blending, colour management, and image masking, any or all of which might cause an area which is marked with a non-white colour to be rendered white (a transparency SMask for example) or an area marked with white to be rendered non-white (eg a transfer function).
Your question is unclear because you haven't defined whether any of these issues are important to you, and how you want to treat them.

How to render dynamic data in a custom WPF control, such as a line graph?

I'm working on a LineGraph control which consists many DependencyProperties that affect how the control should display its data. For example, the control contains the following properties to affect its axes:
AxisStroke - Color of the axes.
AxisThickness - Stroke thickness of the axes.
It also contains properties for display numbers & tick marks
VerticalTicks - True/False to indicate whether or not ticks appear along the vertical axis
HorizontalTicks - True/False to indicate whether or not ticks appear along the horizontal axis
VerticalMin - Minimum value on the vertical axis (numeric)
VerticalStep - The distance in between each vertical tick
VerticalMax - Maximum value on the vertical axis (numeric)
HorizontalMin - Minimum value on the horizontal axis (numeric)
HorizontalStep - The distance in between each horizontal tick
HorizontalMax - Maximum value on the horizontal axis (numeric)
And many more properties exist to allow for different line styles on a single graph (LineColor, LineThickness, DataPointShape, and DataPointIcon to name a few).
My goal is to be able to call out my LineGraph in XAML to insert it into a Window. I would like to be able to specify each of these settings inside the XAML as well, and see the new rendered image of the control in the WPF designer.
Now, given there is a lot of geometric shapes to render on the LineGraph, I though using a Canvas would be a good choice to render the data. Unfortunately, when I'm working in XAML, I cannot perform computations for the locations of shapes based on the control's width & height.
And yes, the shapes' locations must be computed because the data points for the graph are dynamic and the tick-related information is dynamic. Not to mention, I would like to display the actual values along each axis of the LineGraph.
So, I thought I might be able to display this control as if I was doing the rendering in C# code. Other windowing frameworks sometimes provide a Render method that can be used for laying out all of the sub-components.
Doing this, however, doesn't seem possible since WPF relies heavily on XAML for the visual appearance of controls. Also, requiring that the WPF designer must display the LineGraph based on the properties and data specified, it doesn't seem like C# code would solve the problem.
I suppose my questions are these:
How can I render data dynamically inside of a WPF control?
Am I able to specify in C# how my control is rendered, allowing the WPF designer to reflect it?
Side Note:
I've done quite a bit of research, but I am only finding information on how to implement more simple types of controls. If you know of any references that contain information on this topic, please feel free to post them in addition to your answers. I will be more than happy to learn how to do this completely.
EDIT:
I've created a graph using Excel to elaborate what the LineGraph control might look like if it has correct data and properties.

I will answer this based on my experience on implementing custom built graphing libraries in WIN32, WinForm, WPF, WinCE, WP8+WinRT, ....and even on a FPGA :)
It's extremely difficult to implement one from scratch. It may seem easy at first but you will run into a lot of "What should I do if this happens?". For example, in your above graph it seems you got a DataPoint # (5,100) it graphs it pretty well. But lets say, I add another DataPoint # (5.000000005, 0). How would you handle that in your code? Would you say that each pixel on the graph represents an exact value on the X-Axis, or does each pixel represent a range of X-Values?
I would recommend that you use an already establish library to do what you want to do unless you need something very specific like lets say you need horizontal cursors on the graph (think Tektronix Oscilloscope) and you need to calculate some values in between the two cursors.. then maybe you need to implement your own custom one or build on top of an open source one.
So, if you are still adamant of creating your own custom control here are answers to your questions.
How can I render data dynamically inside of a WPF control?
You can use a WriteableBitmap and create your own primitive drawing library from that. After you're done rendering, set it as the ImageSource of your control.
Or you can use WriteableBitmapEx which has GDI like drawing functions already implemented for you.
WriteableBitmapEx CodePlex Page, I also think you can just get it from NuGet as well.
You can also use a <Canvas> and add UI elements to that as well.
Am I able to specify in C# how my control is rendered, allowing the WPF designer to reflect it?
This depends on how you create your controls, but yes you can create Properties in your custom control that will appear in the Designer. Allowing you to change it thus updating the display. I would read a lot of tutorials about writing your own custom user control library. They can explain it better than I can in a SO answer. If you implement the properties correctly it should like so.....
Full Size Image: http://i.stack.imgur.com/pmevo.png
After changing the Number of Rows from 15 to 10 and the starting Y offset to -1 (thus moving the graph up and making the rows a lot taller)
Full Size Image: http://i.stack.imgur.com/0RKnA.png

MS Reports rdlc c# win forms multiple columns

I am trying to create a Report with two columns.
Records should add in as below...
1.Bob 6.Sarch
2.Sue 7.Barrie
3.Adam 8.James
4.Dave 9.Steve
5.Robin 10.Euan
11.Fred
12.Heidi
13.Liz
For the first column (1-5 in this example), a page break should wrap the data to the second column (6-10). After the second column is full (11-13) the the data should page wrap.
The data used to generate the report will have a fixed number of columns. The height and width required for the data will not change.
I am using with C#, WinForms, .net4. I have tried several various approaches with no success. Thus far I have only managed
1.Bob 2.Sue
3.Adam...
Please help or point me in the right direction :-)
Many Thanks

TL;DR ---> This isn't possible with SSRS, afaik.
If I understand correctly, you want to list the items in a data set, that when rendered with a hard page break renderer will render with these requirements:
items are added vertically as long as there's space on the page
a new column of items is started when the first column ran out of space, vertically
a page break is inserted when the second column filled up the vertical space, thus starting a new, first column on a new page
Now, there's only these controls in SSRS:
Textbox
Table, Matrix
Chart
Gauge
List
Image
Subreport
Line, Rectangle
Out of these only the Tablix (Table/Matrix) and List are remotely capable of such a thing, but they can't satisfy your requirements. I'm afraid the answer to your question is that this can't be (easily) done.
The hard way that this may still be possible is perhaps by creating a custom report item and/or by creating a custom rendering extension. But that may be overkill for just wanting to create a nice little list.

Overlaying a 'dot' on to an image on click

I am trying to write a simple program that lets me overlay a dot on top of an image when the image is clicked. I can save the X and Y data back to my database but then I will want to be able to call that information back at a later date and overlay the dots again via code unlike the first time when the user had to click the image.
I got as far as capturing the X and Y of the click no problem but I am having trouble finding examples specifically for what I am trying to do. All of the examples online seem to be for saving the image with the added graphic but I do not need to do that as it will be the same image every time.
Once I can do this, I also need to work out a way that I can detect what area of the image has been clicked. The areas I need to mark out vary in shape and size so I need to try and work out a way to 'map' these areas and then cross reference with the co-ordinates of the users click (I assume that I may need to do some clever geometry stuff for that?)
If anyone has any suggestions of what subjects/classes/methods etc. to research for either of my queries, I would very grateful.
Thanks in advance

You can use the System.Drawing namespace to achieve this.
Create a control and override OnPaint and OnPaintBackground. Store your clicks in a List
In OnPaintBackground, draw the image using DrawImageUnscaled using the graphics object which is passed to you as a parameter.
In OnPaint, loop through your points array and call graphics.FillElipse or similar to draw a little dot.
Because this isnt a retained mode graphics system, you need to keep drawing these items so this may not suit a large number of dots. In that case, you can create an in memory bitmap and get a graphics drawing object using graphics.FromImage.

Windows Form Ordering using MDILayout

I have a very specific problem using C# and a Windows MDI Form application. I want to display two (or more) images to the user, a 'left' and a 'right' image. The names of the images are concealed from the user, and then the user selects which image they prefer (this is part of a study involving medical image quality, so the user has to be blinded from possibly relevant capture parameters which might be revealed in the image name). Instead of showing the actual names, substitute names like 'image 0' and 'image 1' (etc) are shown to the user.
Whenever I use the standard MDILayout.TileVertical or TileHorizontal, the images are loaded in reverse order. For example, if I have image 0 and image 1, they are displayed
Image 1 Image 0
Three or more images would be something like
2 1 0
or
3 2
1 0
And so forth. The problem is, my users are confused by this right to leftness, and if I have another dialog box that asks them which image is better (or to rate the displayed images), they always confuse the order of images on the screen with the order of images in the dialog box. That is, if I just order the images 0 1 2 3 etc in a ratings dialog, they assume that image 3 as it's displayed is image 0 in the MDI parent window, image 2 is image 1, etc-- they read left to right, and the images are being displayed right to left. If I reorder the tabs in the ratings dialog box to reflect the order on the screen, that just confuses them further ("Why is image 3 before image 2?") and the results come out in the wrong order, and are generally unusable.
So, how do I force the ordering of displayed windows using MDILayout in C#? Do I have to do it by hand, or is there some switch I can send to the layout manager?
Thanks!

Why are you using an MDI interface? Surely a single window with a TableLayoutPanel or similar providing layout would be more suitable. The only reason you'd want to use a MDI layout is to allow the users to move the windows, which as far as I can tell from your description of the problem isn't desirable anyway?

Another idea would be to put the actual rating mechanism at the bottom of each child window. So the answer is actually attached to the picture on their child windows instead of having the answers in their own area.

Could you avoid this problem by (before displaying the images) you:
Put the image references in a structure (array or similar).
Have a recursive function build a reverse order structure (or reorder the original).
Use the new reversed order structure to build your child windows as before.
It would add one more layer but might solve your problem if no one finds the reverse layout order switch soon enough.

I strongly recommend following Groky's advice and using a single-form interface rather than MDI for this.
If you must use MDI, you need to know that the MDI layout methods use the Z-order of MDI forms to determine where the forms end up. For example, if image 2 is behind image 1, then image 1 will be on the left and image 2 will be on the right. The most logical way to cause this to happen would be to load image 2's form, then image 1's form, then do the MDI layout. You can also use the ActivateMdiChild method to put the forms in a particular order (activating one form puts the other forms behind it).
It's complicated and error-prone, and I strongly recommend having a two-pane interface on a single form instead, but this will work.

Thanks Owen and Groky, but the Single-Form interface is just not going to work. First, I already have the display code in the MDI format, so that rewrite would require a very, very large rewrite of the code. It took me about three weeks to write the basics of the app a while ago; these aren't jpgs I'm showing here, these are DCM images, and each one is a good 30 mb, with a variety of support tools that I haven't seen outside of medical imaging.
Second, some radiologists don't like split screening for image comparison, and others require it. As such, to accommodate both kinds of users, I set this up with tiling, but then the user can maximize images and then switch between them. So, MDI is the right approach for that differing set of tastes; a single interface with a very complicated set of tab controls just sounds like a nightmare compared to an already extant and (for the most part) working system.
However, since I do control the way in which images are displayed, I can force the z-ordering, and then that should work, right? That's the basis of Fred and Owen's answers, if I'm reading them properly. The user enters 'evaluation mode', and then the program loads the images, shows them, and only once the user has entered an evaluation are the images closed. Given that constraint, I can probably enforce a particular z ordering (maybe by looping from length to 0 rather than from 0 to length).