Input Data Flow

Uploading Scans

One or more people, probably based here in the US, will be responsible for uploading scans for input operators in India to type in. Those who have this task will run custom software on their PC/Mac/Linux box and select a directory full of scans to upload. The core of this software is finished but more work needs to be done to make it user-friendly. Here is the source code: org.asianclassics.center.input.upload

Directory Structure

The upload app can upload multiple sets of pages at once. This "page set" (called a "book") could be a title or a volume. The "base directory" is specified in the upload app, and the content of this directory determines what happens. If the base directory contains only images, then the software assumes that only a single book is being uploaded. If the base directory contains multiple sub directories with images in each one, then each sub directory is considered a book, and they are all uploaded. The sub directories should be named such that they appear in the file system in the proper order. Numbers with leading zeros (001, 002, 003, etc.) works well for this. Likewise, the images inside the directories should be named so that they sort alphabetically in order by page number. Assuming these guidelines are followed, the system will feed these pages to the input operators in order by page and by book.

Dispatch

Data is exchanged between each input center (IC) and HQ. Scans of pecha and instructions about what to do with them are uploaded from HQ to each IC. Pages that have been typed in and corrected are downloaded from each IC to HQ. The software that handles this task will run on the HQ server. Here is the source code: org.asianclassics.center.input.dispatch