Example on how to read a CSV file using DictReader and then upload them to couchdb (with couchdbkit)

csvtocouch.py

#!/usr/bin/env python
from couchdbkit import Server
from csv import DictReader
import time, sys, subprocess, math, os, datetime, pytz, calendar, json

#______________
# parseDoc
def parseDoc(doc):
  for k,v in doc.items():
    
    #strips off any spaces found in the keys.
    del doc[k]
    kk = k.strip(' ') 
    doc[kk] = v  

    # see if this string is really an int or a float
    if (isinstance(v,str) or isinstance(v, unicode)):

      if v.isdigit()==True: #int
        doc[kk] = int(v)
      else: #try a float
        try:
          if math.isnan(float(v))==False:
            doc[kk] = float(v) 
        except:
          pass  

  return doc
  

#______________
# uploadFile
def uploadFile(fname, uri, dbname):
     
  print 'Upload contents of %s to %s/%s' % (fname, uri, dbname)
  
  # #connect to the db
  theServer = Server(uri)
  db = theServer[dbname]
  
  #loop on file for upload
  reader = DictReader(open(fname, 'rU'), dialect = 'excel')
  
  #store a list of docs for upload to db
  docs = []

  for doc in reader:

    newdoc = parseDoc(doc)
  
    #enforce some sort of schema. that is, require the existence of a set of keys in the database documents
    
    #requiredSet = set(['muonmodule', 'end', 'year', 'month', 'day', 'HV channel', 'ADC channel', 'ADC card', 'TDC channel'])
    #if (requiredSet < set(newdoc.keys())) is False:
    #  print 'Quitting! Your file MUST have the following columns'
    #  print ', '.join([x for x in requiredSet])
    #  sys.exit(1)


    #Do any extra formatting of 'newdoc' or add new fields here.
  
  
    #append to list for bulk upload when ready
    docs.append(newdoc)
  
  
  db.bulk_save(docs)
  

if __name__=='__main__':
  
  csvfile = sys.argv[1]
  uri = sys.argv[2]
  dbname = sys.argv[3]
  
  uploadFile(csvfile, uri, dbname)