i want to develop an iPhone optimized website, which includes an generated image. To generate the image i call an external website (aspx). Example:
https://url/filedownload.aspx?documentid=1234&mimetype=image/jpeg
This is no problem on any standard browser on the pc, the images shows up after 2 seconds. But on the iphone there is only the blue question mark showing.
The resultion is ~ 170 * 100 72dpi.
This is certainly a server-side issue, as suggested - a working link or a sample of the source is needed to help resolve. In the meantime, some other things to check; does the image type created match that of the content type specified?
I would then try adding the exact same image (as a static physical file) to the website, load this in an iPhone (to prove there is nothing fundamentally wrong with the image file data) and then compare response headers from your dynamic image code to ensure they are the same - do you require attachment-filename for example?
Related
I am wondering if someone can help me with a problem I am encountering with iPhone/iPad. I have an email the generates a report as an HTML attachment. In that HTML attachment, there are two images. One is a static image that pulls down a logo using a normal HTML img tag.
The other is dynamic. It also uses the normal HTML img tag, however it calls out to a web api with some information identifying the user.
<img src="http://MyApiServerName.com/api/User/{id}/{OtherParm}" >
The Web API uses information on the user to return appropriate images to them. For some reason the dynamic image does not show up with the attachment is opened on an iPad/iPhone.
The image works just fine when the attachment is opened from a computer. I thought maybe something might be blocking images, however then I would expect the static image not to display as well. I verified that the "Load Remote Images" setting is turned on.
I am at a loss here. Is there something with iOS that prevents dynamically generated images from showing in HTML attachments?
Thanks
Ok, I can't explain why the above is not working, but I did find something that does. Instead of using the following URL
<img src="http://MyApiServerName.com/api/User/{id}/{OtherParm}" >
use
<img src="http://MyApiServerName.com/api/User/{id}/{OtherParm.jpg}/" >
The final param needs to end with ".jpg/". If the slash is not there, you will get a 404 error.
Then in the API method, you simply strip off the .jpg part of "OtherParm" to get the intended value for "OtherParm".
My image thumbnail links to the full size image. Now only the image is showing up as some sort of code. I have been using the same HTML editor for years, all the pages are the same. It is only this group of 5 pages that have thumbnails. Take this link for example:
ÿØÿáExifII*ÿìDucky2ÿáyhttp://ns.adobe.com/xap/1.0/ ÿíHPhotoshop 3.08BIMZ%G8BIM%üá‰È·Éx/4b4XwëÿîAdobedÀÿÛ
I also discovered that if I get this code an then go up to the history and click on the name of the photo it pops right up as it should have been. I have no idea how to fix this. I have uploaded the photos several times with two file managers, used other photo software aswell. My editor shows that these pages are fine, but IE shows that the pages have no style and all browsers show the linked image problem.
I know this doesn't make much sense to you professionals, but this is my business website and a customer told me about the problem. After 8 hours of trying to fix this I am hoping you could help.
The problem:
This problem occurs when an IMG changes or got touched (by adobe, in your case) AFTER it was uploaded to the server and the name of the new (or modified) IMG remained the same. The image shown does not get refreshed. The old image is still shown, even though the database holds the right image. I have narrowed it down to the fact that the IMAGE IS CACHED in the web browser. If we hit the RELOAD button in Firefox/Explorer/Safari, everything gets refreshed fine and the correct image just appears.
I verified that on your site running the following:
function is_cached(img_url){
var imgEle = document.createElement("img");
imgEle.src = img_url;
return imgEle.complete || (imgEle.width+imgEle.height) > 0;
}
And then I called this function which returns true or false depending on cached or not:
is_cached("photos/back_to_school_pr5_tn.jpg");
And the result came back as:
<- true
Remember: When uploading an image, its filename is not kept in the database. It is renamed as Image.jpg (to simply things out when using it). When replacing the existing image with a new one, the name doesn't change either. Just the content of the image file changes.
Solution one - No code required:
Since you are not in the coding business (as you claim), all you need to do is to rename the IMG after it's been touched by adobe.
Whenever you modify a photo in adobe or you upload a new IMG you should give it a new name since the older version of the IMG IS STILL IN CACHE.
Solution two - Coding required:
If touching the IMG with adobe is a constant thing that you do after you upload the original picture, then the above solution can be an hassle, so you might want to look into ways to force the web browser to NOT cache images from this page.
NOTE: This solution is only good for the future. In your case, the IMG is already cached.
Besides, this will only work if the actual IMG is inside your HTML, but on your site the link takes you to the actual IMG.
Solution three - Coding required:
An important addition to the above solution is that you can never force a browser to do anything. All you can do is make friendly suggestions. It's up to the browser and the user to actually follow those suggestions. A browser is free to ignore this, or a user could override the defaults.
So the best long-term solution will be to save the filename with the database. This way, if the image is changed, the src of the IMG tag will also change.
For example:
<img src="picture.jpg?1222259157.415" alt="">
where "1222259157.415" is the current time on the server. (Note: I used python's time.time() to generate that)
I've recently come across a situation to work with progressive images. My understanding is, image has to be stored as progressive and browsers will do the loading job.
To prove this, I've created an image as progressive image (with 3 parses and 5 parses) and used it in a sample HTML file with <img> tag. I've uploaded the HTML and image the server. And I don't see any progressive loading.
Is there anything we should do to enable it? As I said, I thought browser will do the loading part automatically.
<<Update>>
Here the example HTML - https://dl.dropboxusercontent.com/u/2456662/Progressive/progressive-demo.html
First two images are exported as progressive images with 3 parses and 5 parses respectivey. Thrid image is the base line image.
When I access the HTML, all 3 are downloaded like base line (regular model). Progressive should load the entire picture in the first shot followed by other pixels - right?
Is there source code (or a browser plugin) to convert the contents of an HTML 5 web page to an image file? This would not just include the visible contents, but the hidden contents as well (assuming there were scroll bars in the page). If there isn't, any advice on how to approach this particular functionality would be appreciated, and I can look into it.
I found this...
html to jpg with c#
However...
I think they just had text in the page, so it doesn't have any dynamic images on the page. My page specifically uses the HTML 5 canvas functionality to draw images. So that must be part of the image file.
It looks like you should be able to do it using javascript with this technique:
http://www.html5canvastutorials.com/advanced/html5-canvas-save-drawing-as-an-image/
Make sure to take note of the following caveat however:
Note: The toDataURL() method requires that any images drawn onto the canvas are hosted on a web server with the same domain as the code executing it. If this condition is not met, a SECURITY_ERR exception is thrown.
EDIT: You may also want to check out these related questions:
Save HTML5 canvas contents, including dragged-upon images
How to save a HTML5 Canvas as Image on a server
I am having trouble getting the proper image to show up when I a link to my website in a status update. It is either grabbing an image from the front page which I don't want it to use (links to the front page) or no image at all (links to specific pages).
I found several tutorials that all gave the same advice about using a meta tag to specify which image to use, which I have done. Example Tutorial. Example:
<link rel="image_src" href="http://URL-TO-IMAGE" />
This had no effect. The article mentions that Facebook caches these lookups and provided a link to a URL Debugger, which was supposed to scrub that cache for me. I used it to verify that my meta tag was inserted properly (it was), but it had no effect on new status updates. Am I missing something? The way the tutorials talk gives me the indication that using this debugger will clear the Facebook cache for the page so that the next lookup will load and re-cache the proper image. Instead it seems that all this does is bypass the cache this one time for the purposes of testing.
Is there a way to actually force clear this cache, or do I just need to wait? It's been several days. How long does this cache take to expire?
Steps to Reproduce:
Visit Facebook News Feed or Timeline
Enter URL of my website in status update (can provide if neeeded)
Expected Outcome
The Favicon of my website is loaded as the icon for the status update
Actual Outcome
An image of a person (appears in the sidebar of our site) is used as the icon (for links to the front page. Links to the inner pages of the website load no icon at all.)
Further:
Visit the URL Debugger
Enter URL for website
Verify proper icon is loading
Post link in new status update.
Expected Outcome
FB's cache will be updated and the Favicon will now be used.
Actual Outcome
Nope... still the person from the sidebar on front page links and no icon at all for inner page links.
UPDATE 2/22/2013:
The image that loads when I post my URL to Facebook has changed! But it's still wrong :(.
I went to test it this morning and I now have an additional image as an option, which means Facebook did update what it's loading from the site, but it's still not the image that I specified in my tag. It's just grabbing another image from one of my other side bars, and I'm still not getting an image at all for my inner page.
It's weird.... the URL Debugger tool grabs the correct image, so I don't think the problem is my Meta tags. That's what the URL Debugger is supposed to help me identify. I think there's some disconnect between the lookup and what Facebook actually posts. I think this is a problem with Facebook, unless I'm missing something huge, but I don't see it....
UPDATE 2/25/2013:
I've made progress, but there's still something weird going on. Martey set me straight on the using the og: meta tags rather than just the tag. The tutorial I had been following said to ignore the Open Graph warnings, but once I started paying attention to them, I got some good information. Like, it actually said that my image is too small and that it will use another image instead. Guess I should have paid attention :)
Anyway, so the issue was that the favicon is too small. According to the Open Graph warning, it needs to be 200 pixels in both directions. So I grabbed the actual header logo and tried to use that. It's 340 pixels wide. Oops, it was only 164 pixels tall.
So I used The GIMP to set it onto a transparent background that was 200 pixels tall and tried again, but it's still telling me it's too small. It's no longer telling me that it needs to be 200 pixels. It just says that it's too small.
I'm stumped again...
Update 2/25/2013: Issue Resolved
The problem was transparency. I filled the background in white and tried again and now it's loading fine.
Thanks for your help, Martey!
Instead of using a third party tutorial, you should refer to Facebook's official OpenGraph documentation. They recommend using an og:image metatag to refer to document images.
As I noted in my comment, issues with data not updating on Facebook is likely to be the result of misformatted or wrong OpenGraph metatags. Without the URL of the troublesome page, or information on the URL Debugger's warnings, it is difficult to recommend solutions.