openscriptures / api Goto Github PK

See http://wiki.github.com/openscriptures/api/restful-uri-structure

In order to be a distributed, interoperable network of servers using this (federated) API, the servers need to know about each other. Servers should maintain a manifest of the known servers and be able to propagate and synchronize the server manifest lists with each other. Such manifests need to inclide:

• Base URL for the API on the server, e.g. http://example.com/foo/api/
• The openscriptures_api apps installed, e.g. texts, morphs, notes, etc.
• The version of the API that is being run.
• The texts and other objects that it has available.
• Any access restrictions, including rate-limiting, authentication, etc.

The federated nature of the API is critical as it allows copyright-protected works to become part of the network, and it provides redundancy and mirroring to make applications more scalable.

We currently have an importer written for Tischendorf 2.5, see http://github.com/openscriptures/api/blob/d4633d7778e24eaca6c3d74941bfee9999e81d99/openscriptures_api_texts/management/commands/load_tischendorf.py

But there were also import scripts written for several other Greek manuscript editions before the data models were refactored a bit. The old import scripts can be found here: http://github.com/openscriptures/openscriptures/tree/master/data/manuscripts/

So in once sense, the first task is to port the old import scripts to work with the updated models. After the importers are ported, however, it would be great to get Hebrew Old Testament imported as well as Bible translations, like CrossWire's parsed KJV2003 data: http://www.crosswire.org/sword/modules/ModInfo.jsp?modName=KJV
http://crosswire.org/~dmsmith/kjv2006/

Also, most importantly, we need to get the MorphHB project imported. We need an edition of the Hebrew Bible in the data models. There will likely be unique issues in doing this since there is a lot of unique punctuation marks used in Hebrew. http://github.com/openscriptures/morphhb

Refactor load_tischendorf management command so that it is a more generic load command which then takes an argument specifying which text importer to run, for example: ./manage.py load Tischendorf-2.5. If the text is not specified, then all texts are loaded. Allow multiple texts to be specified, as well as multiple osisIDs. For example:

./manage.py load Tischendorf-2.5 greek_majority_text KJV2006 | Jude John.7

Merging algorithm can be used to automatically forge these links between editions of the same text, e.g. Greek manuscript editions.

For example, given: /works/Bible.Tischendorf/tokens/2

The token object returned should include relevant information about the lemmas and morph parsings, things which are not part of the openscriptures_texts app. Therefore, the views outputting the token should have signals sprinkled throughout for the openscriptures_morphs app to insert the relevant information for the token.

jtauber: @westonruter don't call it 'type' (clashes with Python built-in) […]
http://twitter.com/jtauber/status/11521916390

Script used by older version of data models: http://github.com/openscriptures/openscriptures/blob/master/data/manuscripts/merge.py