Reply to this post | Go Back

View Post [edit]

Poster:	Branko Collin	Date:	Sep 13, 2007 8:22am
Forum:	opensource	Subject:	How to upload book scans

I tend to upload the scans of the books that I want to run through Distributed Proofreaders, and so far I have done that in DJVU format.

But now I notice that if I upload in other formats, I get all kinds of conversions for free, so I would like to start uploading my sets of scans in these other formats.

How should I upload page scans saved as a single TIFF or JPEG per scan? What is the optimal format to scan as?

And do you have some documentation as to what the names of the formats in the Derivatives tables mean? The only place on the internet some of these names turn up on is the Internet Archive, according to Google.

Reply to this post

Reply [edit]

Poster:	Branko Collin	Date:	Sep 16, 2007 3:52am
Forum:	opensource	Subject:	Re: How to upload book scans

OK, so I experimented a little with existing books I had up, and it seems the following works:

- upload a ZIP containing cropped and deskewed TIFF, JPEG or JPEG2000 files.

- the ZIP must be named identifier_tif.zip for a collection of TIFFs

- the ZIP must contain a directory called identifier_tif

- the directory must contain files called identifier_0001.tif, identifier_0002.tif and so on

(Obviously, if you upload a different format like JPEG2000, these names differ.)

- the rest is simple: edit item -> item manager -> derive

If old files do not strictly conform to the identifier.extension, they won't get overwritten.

Reply to this post

Reply [edit]

Poster:	Branko Collin	Date:	Sep 27, 2007 6:10am
Forum:	opensource	Subject:	Re: How to upload book scans

Just to make sure, the phrase "identifier" should of course be replaced by the actual indentifier of your project.