I have a dynamically generated PDF that can be in one of two modes. If in a draft mode, the only real difference is a watermark that states DRAFT on each page.
I am using a page event, so on the startpage I print the image, so that it is under everything. When I did this on the endpage event then it would be on top of everything, which was bad.
I would like to have alternating colors on the tables so that it is easier to see, but if I set the gray background to be 0.9 then it is opaque and the watermark is covered. If I set the background color to be null and the gray color is 1.0 then it is transparent.
So, my question is, on the non-white alternating rows can I get the watermark to be seen.
Ideally it would be nice to be able to place the image on top of the layer with the borders and background colors, so it would be under the text, but, it appears that isn't an option, as, when I use the stamp with the under content layer it would be under all the cells, since they all had colors set, so nothing was transparent.
UPDATE:
I haven't tried this yet, but it appears that if I save the PDF then using the PdfStamper I can save the graphics state, set the opacity for the image, then restore the graphics state, and it may work.
http://www.vbforums.com/showthread.php?t=518686
But, I don't see any way to do what I would like without saving it first, as I am actually generating and streaming directly to the browser, and would prefer not to save, if possible.
By using the link in my final update I was able to get the watermark to work.
So, I save the file, then I use a MemoryStream to read in the file and I save the graphics state, set the opacity, set the watermark and then restore the graphics state.
I then stream this back to the browser and delete the file.
I haven't found any way to do it that works better than this. What is nice is that an image is not needed for the watermark, so I set the font size to 96pts, and that seemed to work fine, in landscape and portrait mode.
Related
i can't remove the white background when saving as png or jpeg.
tried everything what suggested me on the web, but still get the diminsions i set up as whitebackground.
How can i save image without the background?
see image below for settings.
enter image description here
if you want to remove all the white from the picture, better go to color range and sample white. layer via cut and delete the layer. and save it in png as #cybernatic.nomad said, supports transparency.
Make sure you have created your image on a new layer and delete (or hide) the first white layer. I just created a simple text on a new layer with a transparent background and then loaded the image back into Photoshop to confirm that it works properly. As per previous comment I use .png format to preserve transparency.
I have a radio button that should display an image (style BS_AUTORADIOBUTTON|BS_PUSHLIKE|BS_BITMAP).
I create a bitmap via CreateDIBSection (using a BITMAPINFO with BI_RGB) and obtain a pointer to the raw pixels via the ppvBits, so that I can manipulate them freely.
I use BM_SETIMAGE to set the button's image to the bitmap I created.
So far, I can set the RGB and alpha by manipulating the pixels by hand. I tested that even semi-transparent (non-premultiplied) alpha values look good.
As far as I can tell, everything works, except if all pixels in the image are transparent. In that case, the button apparently ignores the alpha value, simply displaying a rectangle with each pixel having the respective color with full opacity.
I found a hint that Windows - at least in some cases - actually seems to interpret images whose pixels' alpha values are all 0 as completely opaque images:
When the window manager sees a 32bpp bitmap, it looks at the alpha
channel. If it's all zeroes, then it assumes that the image is in 0RGB
format; otherwise it assumes it is in ARGB format
Is this behavior documented somewhere?
Is this behaviour documented somewhere?
Yes! In Raymond's post! :) That's often the way of it, no?
If you look at the foot of the page here you will find a comments box. If you raise your concerns there then MS will most likely fix their documentation. See here for an example of the process they usually follow if they consider the problem serious enough to fix.
I´ve a cropped picture in a powerpoint layer and I want to change the picture without changing the size of the cropping area.
If I change the image by clicking "change picture" the cropping area resizes too, to show the whole picture.
How can I do that without changing the cropping area?
In Apple Keynote you only have to drag the picture you want to the cropping area and you only have to align it.
From my experience, I'm afraid you have to set the crop manually by "Format Picture" -"Crop" to remain the same cropping area when you replace a picture.
This isn't much help, but Libreoffice will do this. Crop, resize, then select your image and re-insert a new image - formatting will be the same, perfect for when you want to quickly redo formatting for same sized images.
In my experience, a picture that you insert to replace an existing one, using Change Picture, will "inherit" the crop&size settings of the old picture IF the old&new images are actually of the same size (in pixels) - if not, the new one will revert to the default crop&size so you will have to redo these manually.
This changes if the image to-be-replaced has any crop setting applied to it in Powerpoint. In this case, the replacing image will not inherit those, and might also be inserted with a different size&position, leading most likely to an unwanted result & further work needed.
A workaround I found is to first remove the crop setting from the old picture, replace with new image, then reapply the crop. This is still less work than having to manually redo the size&position (or to define these based on numbers copied from the PropertiesPane of the old image).
Quick work around: if you plan to have multiple same size images and you wish to use change image to swap the content on each page (something we do with floor plans all the time); don't bother cropping them.
Put the whole image in and crop it by sticking shapes on top of it, which are the same color as the presentation background color, like a mask.
optionally crop the images only when the whole presentation is finished.
I have a view designed for printing which includes a watermark, a transparent view which draws some text atop the other content.
When printing and using the Mac OS Save as PDF feature, the watermark text is selectable. Sometimes this interferes with selecting the other content, other times it's just distracting.
How can I make the text not selectable in the generated PDF?
I tried drawing the watermark behind the other content instead of in front. It didn't prevent selecting the watermark, but kept it out of the way of the other content. However, the table view rows occluded the watermark, which of course is worse.
Commenter asked for code, so here's some code which prepares the view:
// self.view is the print view
// watermark is an instance of WatermarkBackground, an NSView
if (watermark) {
watermark.frame = self.view.frame;
[self.view addSubview:watermark positioned:NSWindowAbove relativeTo:nil];
}
And the line in [WatermarkBackground drawRect] which does the drawing:
// _message is an NSString
// textAttributes returns a dictionary with a color and font
[_message drawWithRect:textRect
options:NSLineBreakByWordWrapping
attributes:[WatermarkBackground textAttributes]];
I meant to post this screenshot originally:
One option would be to create one or multiple CGPaths from your string and draw those into the PDF instead. One way to do so would be to use CTFontCreatePathForGlyph, but it's actually quite a lot of work to do this for entire strings, Core Text does help, but it's a pretty low-level framework.
If you're always drawing the same watermark, it would be much easier to create a static PDF in some vector graphics app and use that with CGPDFPageDraw etc. Illustrator has a "Convert to Paths" command for text objects.
As far as I know, there in no way to make text unselectable in PDF. Probably the best solution would be to use an image watermark instead.
However, if it is in front of text, it can make background text difficult to select. If it is behind everything, there will be same issues with obscuring it with tables. So, possibly a better plan of action would be not to try to make text unselectable, but rather make table background transparent. Then, use image watermark.
Taking an idea from omz, instead of using CGPaths and generating them on the fly, the simplest, most elegant solution would be this:
Create a vector watermark by typing the text in a vector editor and expanding text to create outlines.
Save it as SVG or PDF.
Then, put this new vector graphic on top as a watermark. It will not be selectable, will not obscure the view, and will not be obscured by tables.
First of all I will explain my situation so you can know my problem a little better. I'm making a HTML5 app. I have a canvas, and using a color picker you can change the color of the canvas. Now i have a picture which I want to put on the canvas but that pictures color needs to be changed using a color picker. So i need to replace, lets say, black color on that picture and put it on the canvas so it dosnt screw up the background.
So that will look like this:
1st color picker- changes the color of the canvas
2nd color picker - replaces the black color on the image with the one in the color picker and puts it on the canvas
Now my problem is how to replace the color on the image without reloading the page.
My only condition is no using silverlight, flash, java or any other similar tehnology that need 3rd party software to be installed on the device.
Thanks in advance.
If you dont understand my query fully, feel free to ask.
My approach with a JS only solution could be:
Loading the image inside a canvas element. Look at the MDC canvas tutorial
Trigger the user click on the canvas and get the pixel color (see links below to know how to get the color of a pixel) and look at this answer to get the mouse position
Substitute all the colors in the canvas with the one the user pick. For some examples about pixel manipulation:
Pushing pixel with canvas at Mozilla Hacks
http://beej.us/blog/2010/02/html5s-canvas-part-ii-pixel-manipulation/
This JS at mezzoblue apply heavy filter to an image
After some canvas experiment I notice that mostly in all the browser the pixel manipulation with canvas could be very slow also with small images. So another experiment to do could be to get the pixel color and then:
pass the color information to a PHP (or another server side script) with an AJAX call
do the color manipulation with an image library like GD or imagemagik
return back your image with the Ajax response
reload your canvas with the modified version of the image