Zoomify

[Irishmaestro]

I've seen a few zoomify-based scans of things that I'm rather itching to process (Juilliard, for example, has the first edition of Bach's Partitas for Harpsichord). I've tried following the guidelines in Generoso's 'Music from Flash-Based Websites' tutorial, but for some reason I keep getting an error, which says: "c compiler cannot create executables". I'm running OS X 10.8.3 (Mountain Lion), by the way.

Anyone know what I'm doing wrong?

EDIT: I've solved it - at least, for the first two sections. On Mountain Lion, the location of the SDK is different. Just replace '/Developer/SDKs/MacOSX10.6.sdk' with 'Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.8.sdk' in every instance, and it should work!

ANOTHER EDIT: Now I'm having problems with Imaging-1.1.7. Looks like gcc-4.0 isn't there either. Time to go rooting.

[cypressdome]

Hi Irishmaestro,

I wish I could be of some help but the Mac o/s is too foreign to me. I've only used Dezoomify under Windows. As I recall I could only get it to work under Python 2.7. One issue I've run into a couple of times with Julliard's site is that of corrupt jpg tiles. If just one tile is corrupt on Julliard's server the Dezoomify script fails. Another source you might want to look at is Bach-digital which uses Zoomify. The one issue I ran into on that site is that you had to use a zoom level of 6 or the process failed.

Good luck!
Cypressdome

[Irishmaestro]

Sorry for returning to an old thread, but I have another dezoomify-related question, if someone might be able to help me. I've tried making a list of base URLs to be dezoomified, saved as a plain .txt file, but for some reason, dezoomify only processes the last image on the list. Any idea what I'm doing wrong?

[cypressdome]

No idea but perhaps a comparison might help. Here's the script and part of the text file I used for downloading Stanford's Violin Concerto No.2 from the Morgan Library.
The script:

Code: Select all

python dezoomify.py -i g:\morgan\stanford.txt -l -b -d -z 7 -q 90 -o g:\morgan\stanford\image.jpg

The first three lines of the text file:

Code: Select all

http://www.themorgan.org/music/zoom/115728/115728v_0001
http://www.themorgan.org/music/zoom/115728/115728v_0002
http://www.themorgan.org/music/zoom/115728/115728v_0003

How does the format compare with yours? Perhaps try running the above to see if you can get the three images.

Cypressdome

[Irishmaestro]

I just tried running that list, and sure enough, it only saved the third image. Perhaps it's an OS X thing, or perhaps it's because I need to run dezoomify on an older version of Python (2.7). I really can't figure it out!

[cypressdome]

Trying to run dezoomify under either Python 3.2 or 3.3 immediately generates the same error for me: File "dezoomify.py", line 216 print xmlUrl SyntaxError: invalid syntax. I've always ran it under Python 2.7 under Windows (XP and 7). Sorry I can't be of any more help.

[Irishmaestro]

More dezoomify questions! I can't figure out how to get it to work with sites like the British Library, where xml file is located somewhere other than the base directory. For instance, if you take this particular scan (the end of Mozart's Rondo in A, K.386), each page has its own separate xml file. And because of that, they're called things like 'add_ms_32181_f001r.xml' instead of ImageProperties.xml, which is what dezoomify looks for.

Has anybody found a way of downloading such pages from BL?

[cypressdome]

What's the URL for the BL site on which the score is displayed?

[Irishmaestro]

What's the URL for the BL site on which the score is displayed?

The search result itself is here: http://www.bl.uk/manuscripts/FullDispla ... 81&index=4
And this is the viewer it links to: http://www.bl.uk/manuscripts/Viewer.asp ... 2181_f250r

[cypressdome]

Irishmaestro,

I've responded to you with a private message.