read_exo_suggestion_pseudo.py

#you'll need to easy_install jsonschema and couchdbkit
#
# -- download the schema description by 
#    running this from your terminal
# 
# wget https://raw.github.com/gadamc/persephone/schema/_attachments/schema/data.schema.ietf_draft_v3.aarm_spec_v1.01.json -O schema.json
#

import json, cPickle, jsonschema, couchdbkit


doclist =[]
schemadoc = json.load( open('schema.json') )

#for loop over each line in latex file  # as james has shown

  #parse line in latex file to extract data
  #'line' is a list of values from the table.
  
  #now store the values in the 'line' list in a python dictionary.
  #that matches our schema.json

  doc={}
  doc['type'] ='measurement'
  #...
  doc_sample={}
  doc_sample['m_name'] = line[1]
  #...
  doc['sample'] = doc_sample

  #... measurement, data source,
  
  #use a python schema validator tool here to ensure docs are in correct format.
  #note, this is version 1.01 which requires doc['measurement']['m_date'] to be
  #a list of one or two date strings in the format of YYYY-MM-DD
  #and doc['measurement]['m_result][X]['error'] to be a list of one or two numbers to specify 
  #a symmetric or asymmetric error bar
  #
  #also, any extra user fields can be added to 'data_source', 'sample' and 'measurement' 
  #as an object containing only simple values stored within 'u_extra'. For example:
  #doc['sample']['u_extra'] = { 'a':'string', 'b':5    }
  #
  #this is not allowed
  # doc['sample']['u_extra'] = { 'a':'string', 'b': [5, 4, 3]    }
  #
  
  
  jsonschema(doc, schemadoc) #raises an Excpetion if doc format fails.
  
  doclist.append(doc) #store data in list for later


#three ways to store to disk
json.dump(doclist, open('/path/to/filename.json', 'w') , indent=1)  #you don't need indent, but it prints pretty json

cPickle.dump(doclist, open('/path/to/filename.pickle', 'w'))

db = couchdbkit.Server('http://localhost:5984').get_or_create_db('exodata')
db.bulk_save(doclist)


# To read the data out

doclist = json.load(open('/path/to/filename.json'))

doclist = cPickle.load( open('/path/to/filename.pickle') )

db = couchdbkit.Server('http://localhost:5984').get_or_create_db('exodata')
vr = db.all_docs(include_docs=True)
doclist = []
for row in vr:
  doclist.append(row['doc'])