Comments on: Reading out EXIF data via Python

By: Bob Wiliiams

Bob Wiliiams — Wed, 18 Mar 2009 16:36:28 +0000

Personally I like EXIF.py with is totally self contained and works well. It’s easy to use. It is a sourceforge project, but the official release didn’t work for me and I used a bug-fix version I found here:

http://www.dudek.org/blog/154

I think the bug-fix version is mostly needed of the unusual EXIF formatting you get from some versions of iphoto.

By: Nare

Nare — Fri, 11 Jul 2008 08:17:26 +0000

Hi Towi,

Can u provide me \"photoshopy.py\"

By: Wolfram

Wolfram — Mon, 10 Apr 2006 11:16:17 +0000

btw Pierre, actually it’s a little bit the same as it was a couple years ago with Mac OS and Windows, I switched to Mac OS because I wanted less trouble with the computer caused by viruses etc. (a computer causes enough trouble besides, so avoid those at least).
Now with PHP and Python, actually the feeling was the same, just use the (less common) language that has less _known_ issues, just to ship around those problems by default. Yes, that was one of my main reasons why to switch to Python … that it didn’t quite work out yet is written on another piece of paper …

By: towi

towi — Fri, 07 Apr 2006 15:03:07 +0000

Hi,

I use Photoshop to read nearly all EXIF data. I have a photoshopy.py module and use the COM+ typelib to remote control photoshop. If you are interested I can send you more detail. It works something like this:

app = photoshopy.Application()
doc = app.Load(”MyImg.jpg”)
print doc.EXIF[u"Author"]
print doc.EXIF[u"Aparture"]

I don’t have the script right here, but its quite easy…

tschau, towi.

By: Stephen Vermeulen

Stephen Vermeulen — Mon, 03 Apr 2006 18:19:22 +0000

You might also want to look at:

http://www.emilas.com/jpeg/

By: Kevin Horton

Kevin Horton — Mon, 03 Apr 2006 12:23:06 +0000

This comment has nothing to do about EXIF parsing, but I couldn’t find another obvious way to get in contact with Wolfram.

I wanted to pick up the RSS feed for this blog, but found that the two feeds listed (Entries and Comments), are identical. They both actually contain only the comments. This looks like a blunder, and I figured you might want to know about it.

By: Jos Yule

Jos Yule — Fri, 31 Mar 2006 16:25:30 +0000

Nice one - i’ll take a look at the Parser class too.

By: Fredrik

Fredrik — Fri, 31 Mar 2006 15:56:37 +0000

“Unfortunately the file-like object urllib returns doesn’t have a ’seek’ method which Image.open requires.”

To avoid temporary files, you can load the data into a StringIO instance:

import urllib, cStringIO

file = cStringIO.StringIO(urllib.urlopen(URL).read())
img = Image.open(file)

A more advanced approach is to use the ImageFile.Parser class, and “feed” incoming data directly into the parser object (see the documentation for an example).

Hope this helps! /F

By: Wolfram

Wolfram — Fri, 31 Mar 2006 15:52:36 +0000

Jos, thanks a lot.

By: Jos Yule

Jos Yule — Fri, 31 Mar 2006 15:30:08 +0000

I just tried PIL’s Image.open when using a urllib.urlopen()’ed file. Unfortunately the file-like object urllib returns doesn’t have a ’seek’ method which Image.open requires.

However, you can use urllib.urlretrieve(), which saves the object the url points to in a local file (see the docs). So you can do something like the following:

## CODE START

import Image
import urllib

dl_img = urllib.urlretrieve(”URL-to-image”)
img = Image.open(dl_img[0])

## CODE END

dl_img is a tuple, the first part being the path to the locally saved file. Again, see the docs for urllib.