I have a suggestion regarding your thoughts on generating sets of image
sizes depending on your predicted needs.

This is a fairly common thing to do and makes a lot of sense out the
outset.  However, down the road, you might run into headaches if you decide
that your presentation code needs an image size that doesn't exist.  I would
suggest looking into the idea of holding on to the original and generating
the 'called' size, the size as needed by the presentation code, rather than
bulk resizing.  You could then generate some procedures that would check on
the file system for image image1_300x200.jpg (for example) and if it cannot
find it, it would then call the image resizing proc to create one and then
look for the file again.

For instance, let's say you have two types of greeting cards.  One with the
BIG image, on with the SMALL image.  The user could view the available
images on the preview page (with thumbnails called by the preview
presentation code), and then pick the image size they want.  That way you
have generated 3 total images - one for the thumbnail, one for the BIG size,
and one for the SMALL size (assuming they previewed each).  That's three
images instead of defaulting to 5 every time and you're set up to handle x
number of sizes.  Not a big deal at first, but later, ughhh!



On 7/12/07, Dennis Daupert <[EMAIL PROTECTED]> wrote:

I am writing a Gallery application in perl/Catalyst. These are lots of
interesting choices to make!

If any one on this list has engaged in such a mad enterprise, I'd be
interested in your experience. I am interested in any advice, links to
how-tos, best practices, articles, or tips. I have questions such as:

What precautions should be taken for security?
What is a good file upload size limit, if any?
What subset of (all possible) file formats should be supported?
What gotchas might there be?

And I have a few "software architecture" questions.

I have a central gallery where images would be uploaded to. From there,
images of varying sizes could be linked to articles, stories, eCards,
thumbnail lists, etc. Because different sizes of the same image might be
needed for different uses, I thought I'd generate a whole set of image sizes
for each uploaded image.

So, I'd have subdirectories for
_orig original,
_t thumbnails,
_sm small,
_med medium, and
_lrg large file sizes.

Does this approach seem reasonable?

If the initial image file is REALLY large (such as a RAW formatted digital
image), should I convert that file to a smaller size (say, 1024x768) and
save as the "original?"

Should I convert the original image to a png to reduce lossiness of
subsequent transformations, even tho that may inflate the initial file size?

/dennis


_______________________________________________
Magick-users mailing list
[email protected]
http://studio.imagemagick.org/mailman/listinfo/magick-users

_______________________________________________
Magick-users mailing list
[email protected]
http://studio.imagemagick.org/mailman/listinfo/magick-users

Reply via email to