hiplab/parser
hiplab
/
parser
1
0
Fork 0
Code Issues 4 Pull Requests Packages Projects 1 Releases 1 Wiki Activity

v2.0 upgrade #10

Merged
steve merged 26 commits from v2.0 into master 2024-02-09 16:00:08 +00:00
Conversation 0 Commits 26 Files Changed 1774 +124 -131
6 changed files with 124 additions and 131 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit 00f80d9294 - Show all commits

81
healthcareio/__main__.py
View File

@ -1,3 +1,4 @@
#!/usr/bin/env python
""" """
(c) 2019 EDI Parser Toolkit, (c) 2019 EDI Parser Toolkit,
Health Information Privacy Lab, Vanderbilt University Medical Center Health Information Privacy Lab, Vanderbilt University Medical Center
@ -10,12 +11,12 @@ This code is intended to process and parse healthcare x12 837 (claims) and x12 8
The claims/outpout can be forwarded to a NoSQL Data store like couchdb and mongodb The claims/outpout can be forwarded to a NoSQL Data store like couchdb and mongodb
Usage : Usage :
Commandline : Commandline :
python x12parser <action> # parse {x12}
healthcare-io parse <x12_folder>
# export {x12}
healthcare-io export
action:
- parser
- create.plugin
- register.plugin
- -
Embedded : Embedded :
@ -26,11 +27,12 @@ from typing import Optional
from typing_extensions import Annotated from typing_extensions import Annotated
import uuid import uuid
import os import os
import version import meta
import json import json
import time import time
from healthcareio import x12 from healthcareio import x12
from healthcareio.x12.parser import X12Parser from healthcareio.x12.parser import X12Parser
import requests
# import healthcareio # import healthcareio
# import healthcareio.x12.util # import healthcareio.x12.util
@ -39,11 +41,12 @@ app = typer.Typer()
CONFIG_FOLDER = os.sep.join([os.environ['HOME'],'.healthcareio']) CONFIG_FOLDER = os.sep.join([os.environ['HOME'],'.healthcareio'])
@app.command(name='init') @app.command(name='init')
def config(email:str,provider:str='sqlite') : def config(email:str,provider:str='sqlite') :
""" """\b
Generate configuration file needed with default data store. For supported data-store providers visit https://hiplab.mc.vanderbilt.edu/git/hiplab/data-transport.git Generate configuration file needed with default data store. For supported data-store providers visit https://hiplab.mc.vanderbilt.edu/git/hiplab/data-transport.git
:email your email :email your email
\r:provider data store provider (visit https://hiplab.mc.vanderbilt.edu/git/hiplab/data-transport.git)
:provider data store provider (visit https://hiplab.mc.vanderbilt.edu/git/hiplab/data-transport.git)
""" """
_db = "healthcareio" _db = "healthcareio"
# _PATH = os.sep.join([os.environ['HOME'],'.healthcareio']) # _PATH = os.sep.join([os.environ['HOME'],'.healthcareio'])
@ -62,27 +65,55 @@ def config(email:str,provider:str='sqlite') :
"system":{ "system":{
"uid":str(uuid.uuid4()), "uid":str(uuid.uuid4()),
"email":email, "email":email,
"version":version.__version__, "version":meta.__version__,
"copyright":version.__author__ "copyright":meta.__author__
} }
} }
# #
# Let create or retrieve a user's key/token to make sure he/she has access to features they need
# This would also allow us to allow the users to be informed of new versions
#
try:
host = "https://healthcareio.the-phi.com" if 'PARSER_HOST_URL' not in os.environ else os.environ['PARSER_HOST']
url = f"{host}/api/users/signup"
_body = {"email":email,"version":meta.__version__}
_headers = {"content-type":"application/json"}
resp = requests.post(url,headers=_headers,data=json.dumps(_body))
if resp.ok :
_config['system'] = dict(_config['system'],**resp.json())
except Exception as e:
print (e)
pass
# store this on disk # store this on disk
f = open(os.sep.join([CONFIG_FOLDER,'config.json']),'w') f = open(os.sep.join([CONFIG_FOLDER,'config.json']),'w')
f.write(json.dumps(_config)) f.write(json.dumps(_config))
f.close() f.close()
_msg = f"""
Thank you for considering using our {{x12}} parser verion {meta.__version__}
The generatted configuration file found at {CONFIG_FOLDER}
The database provider is {provider}
visit {host} to learn more about the features,
"""
print (_msg)
@app.command(name='about') @app.command(name='about')
def copyright(): def copyright():
for note in [version.__name__,version.__author__,version.__license__]: for note in [meta.__name__,meta.__author__,meta.__license__]:
print (note) print (note)
pass pass
@app.command() @app.command()
def parse (claim_folder:str,plugin_folder:str = None,config_path:str = None): def parse (claim_folder:str,plugin_folder:str = None,config_path:str = None):
""" """
This function will parse 837 and or 835 claims given a location of parsing given claim folder and/or plugin folder This function will parse 837 and or 835 claims given a location of parsing given claim folder and/or plugin folder.
plugin_folder folder containing user defined plugins (default are loaded)
config_path default configuration path
""" """
_plugins,_parents = x12.plugins.instance(path=plugin_folder) _plugins,_parents = x12.plugins.instance(path=plugin_folder)
@ -94,7 +125,7 @@ def parse (claim_folder:str,plugin_folder:str = None,config_path:str = None):
_config = json.loads(f.read()) _config = json.loads(f.read())
f.close() f.close()
_store = _config['store'] _store = _config['store']
# print (len(_files)) # # print (len(_files))
jobs = [] jobs = []
for _chunks in _files: for _chunks in _files:
pthread = X12Parser(plugins=_plugins,parents=_parents,files=_chunks, store=_store) pthread = X12Parser(plugins=_plugins,parents=_parents,files=_chunks, store=_store)
@ -103,11 +134,29 @@ def parse (claim_folder:str,plugin_folder:str = None,config_path:str = None):
while jobs : while jobs :
jobs = [pthread for pthread in jobs if pthread.is_alive()] jobs = [pthread for pthread in jobs if pthread.is_alive()]
time.sleep(1) time.sleep(1)
pass # pass
else: # else:
pass # pass
print ("...................... FINISHED .........................")
# #
# #
@app.command(name="export")
def publish (file_type:str,path:str):
"""
This function will export to a different database
file_type values are either claims or remits
path path to export configuration (data transport file)
file_type claims or remits
"""
if file_type in ['837','claims'] :
_type = 'claims'
elif file_type in ['835','remits']:
_type = 'remits'
if __name__ == '__main__' : if __name__ == '__main__' :
app() app()

107
healthcareio/x12/parser.py
View File

@ -7,6 +7,7 @@ import os
import json import json
# from healthcareio.x12.util # from healthcareio.x12.util
from healthcareio import x12 from healthcareio import x12
from healthcareio.x12.util import file, document
import numpy as np import numpy as np
import transport import transport
import copy import copy
@ -14,19 +15,21 @@ import copy
from datetime import datetime from datetime import datetime
from healthcareio.logger import X12Logger from healthcareio.logger import X12Logger
import time import time
import pandas as pd
class BasicParser (Process) : class BasicParser (Process) :
def __init__(self,**_args): def __init__(self,**_args):
super().__init__() super().__init__()
self._plugins = _args['plugins'] self._plugins = _args['plugins']
self._parents = _args['parents'] self._parents = _args['parents']
self._files = _args['files'] self._files = _args['files']
self._store = _args['store'] self._store = dict(_args['store'],**{'lock':True})
self._template = x12.util.template(plugins=self._plugins) self._template = x12.util.template(plugins=self._plugins)
# self._logger = _args['logger'] if 'logger' in _args else None
self._logger = X12Logger(store = self._store) self._logger = X12Logger(store = self._store)
if self._logger : if self._logger :
_info = { key:len(self._plugins[key].keys())for key in self._plugins} _info = { key:len(self._plugins[key].keys())for key in self._plugins}
_data = {'plugins':_info,'files': len(self._files),'model':self._template} _data = {'plugins':_info,'files': len(self._files),'model': self._template}
self._logger.log(module='BasicParser',action='init',data=_data) self._logger.log(module='BasicParser',action='init',data=_data)
def log (self,**_args): def log (self,**_args):
@ -34,6 +37,8 @@ class BasicParser (Process) :
This function logs data into a specified location in JSON format This function logs data into a specified location in JSON format
datetime,module,action,data datetime,module,action,data
""" """
if self._logger :
self._logger.log(**_args)
pass pass
def apply(self,**_args): def apply(self,**_args):
""" """
@ -44,15 +49,17 @@ class BasicParser (Process) :
_content = _args['content'] _content = _args['content']
_filetype = _args['x12'] _filetype = _args['x12']
_doc = _args['document'] #{} _doc = _args['document'] #{}
_documentHandler = x12.util.document.Builder(plugins = self._plugins,parents=self._parents)
_documentHandler = x12.util.document.Builder(plugins = self._plugins,parents=self._parents, logger=self._logger)
try: try:
_tmp = {}
for _row in _content : for _row in _content :
# _data = None # _data = None
_data,_meta = _documentHandler.bind(row=_row,x12=_filetype) _data,_meta = _documentHandler.bind(row=_row,x12=_filetype)
if _data and _meta : if _data and _meta :
_doc = _documentHandler.build(data=_data,document=_doc,meta=_meta,row=_row) _doc = _documentHandler.build(data=_data,document=_doc,meta=_meta,row=_row)
# print (['*** ',_doc]) # print (['*** ',_doc])
pass pass
@ -61,15 +68,18 @@ class BasicParser (Process) :
except Exception as e: except Exception as e:
# #
# Log something here .... # Log something here ....
print (_row) # print (_row)
print (e) print (e)
# print (_row,_doc.keys()) # print (_row,_doc.keys())
pass pass
return _doc return _doc
def run(self): def run(self):
_handleContent = x12.util.file.Content() _handleContent = file.Content() #x12.util.file.Content()
_handleDocument = x12.util.document.Builder(plugins = self._plugins,parents=self._parents) _handleDocument = document.Builder(plugins = self._plugins,parents=self._parents,logger=self._logger)
_template = self._template #x12.util.template(plugins=self._plugins) _template = self._template #x12.util.template(plugins=self._plugins)
# #
# @TODO: starting initializing parsing jobs : # @TODO: starting initializing parsing jobs :
# - number of files, plugins meta data # - number of files, plugins meta data
@ -89,10 +99,14 @@ class BasicParser (Process) :
_header = copy.deepcopy(_template[_filetype]) _header = copy.deepcopy(_template[_filetype])
_header = self.apply(content=_content[0],x12=_filetype, document=_header) _header = self.apply(content=_content[0],x12=_filetype, document=_header)
_docs = [] _docs = []
_ids = []
for _rawclaim in _content[1:] : for _rawclaim in _content[1:] :
_document = copy.deepcopy(_header) #copy.deepcopy(_template[_filetype]) _document = copy.deepcopy(_header) #copy.deepcopy(_template[_filetype])
if 'claim_id' in _document :
#
# @TODO: Have a way to get the attribute for CLP or CLM
_ids.append(_document['claim_id'])
# _document = dict(_document,**_header) # _document = dict(_document,**_header)
if type(_absolute_path) == str: if type(_absolute_path) == str:
_document['filename'] = _absolute_path _document['filename'] = _absolute_path
@ -108,7 +122,9 @@ class BasicParser (Process) :
_data = {'filename':_location, 'available':len(_content[1:]),'x12':_filetype} _data = {'filename':_location, 'available':len(_content[1:]),'x12':_filetype}
_args = {'module':'parse','action':'parse','data':_data} _args = {'module':'parse','action':'parse','data':_data}
_data['parsed'] = len(_docs) _data['parsed'] = len(_docs)
self._logger.log(**_args)
self.log(**_args)
self.log(module='parse',action='file-count', data={'file_name':_absolute_path,'file_type':_filetype,'claims':_ids, 'claim_count':len(_ids)})
# #
# Let us submit the batch we have thus far # Let us submit the batch we have thus far
# #
@ -135,78 +151,17 @@ class X12Parser(BasicParser):
_documents = _args['documents'] _documents = _args['documents']
if _documents : if _documents :
_store = copy.copy(self._store,**{}) _store = copy.deepcopy(self._store)
TABLE = 'claims' if _args['x12'] in ['837','claims'] else 'remits' TABLE = 'claims' if _args['x12'] in ['837','claims'] else 'remits'
_store['table'] = TABLE _store['table'] = TABLE
_store['cotnext'] = 'write'
_writer = transport.factory.instance(**_store) _writer = transport.factory.instance(**_store)
_writer.write(_documents) _writer.write(_documents,table=TABLE)
if getattr(_writer,'close') : if getattr(_writer,'close') :
_writer.close() _writer.close()
# #
# LOG: report what was written # LOG: report what was written
_data = {'x12':_args['x12'], 'documents':len(_documents),'filename':_args['filename']} _data = {'x12':_args['x12'], 'documents':len(_documents),'filename':_args['filename']}
self._logger.log(module='write',action='write',data=_data) # self._logger.log(module='write',action='write',data=_data)
self.log(module='parse',action='write',data=_data)
# def instance (**_args):
# """
# :path
# """
# # _files = x12.util.Files.get(_args['file'])
# # #
# # # We can split these files (multi-processing)
# # #
# # _jobCount = 1 if 'jobs' not in _args else int (_args['jobs'])
# # _files = np.array_split(_files,_jobCount)
# # PATH = os.sep.join([os.environ['HOME'],'.healthcareio','config.json'])
# # if 'config' in _args :
# # PATH = _args['config']
# # f = open(PATH)
# # _config = json.loads(f.read())
# # f.close()
# # jobs = []
# # for _batch in _files :
# # pthread = Parser(files=_batch,config=_config)
# # pthread.start()
# # jobs.append(pthread)
# # time.sleep(1)
# pass
# class parser (Process) :
# _CONFIGURATION = {}
# def __init__(self,path=None) :
# if not parser._CONFIGURATION :
# _path = path if path else os.sep.join([os.environ['HOME'],'.healthcareio/config.json'])
# #
# # @TODO: Load custom configuration just in case we need to do further processing
# config = json.loads(open(path).read())
# parser._CONFIGURATION = config['parser']
# #
# # do we have a custom configuration in this location
# #
# _custompath = _path.replace('config.json','')
# _custompath = _custompath if not _custompath.endswith(os.sep) else _custompath[:-1]
# _custompath = os.sep.join([_custompath,'custom'])
# if os.exists(_custompath) :
# files = os.listdir(_custompath)
# if files :
# _filename = os.sep.join([_custompath,files[0]])
# _customconf = json.loads(open(_filename).read())
# #
# # merge with existing configuration
# else:
# pass
# #
# #
# class getter :
# def value(self,) :
# pass
# class setter :
# def files(self,files):
# pass

2
healthcareio/x12/plugins/default/claims.py
View File

@ -1,7 +1,7 @@
import numpy as np import numpy as np
from .. import parser from .. import parser
from datetime import datetime from datetime import datetime
@parser(element='NM1',x12='*', anchor={'41':'submitter','40':'receiver','82':'rendering_provider','85':'billing_provider','87':'pay_to_provider','IL':'patient','PR':'payer','QC':'patient','DN':'referring_provider','77':'provider','2':'billing_provider'}, map={1:'type',3:'name',-1:'id'}) @parser(element='NM1',x12='*', anchor={'41':'submitter','40':'receiver','82':'rendering_provider','85':'billing_provider','87':'pay_to_provider','IL':'patient','PR':'payer','QC':'patient','DN':'referring_provider','77':'provider','2':'billing_provider'}, map={1:'type',3:'name_1',4:'name_2',-1:'id'})
def NM1 (**_args): def NM1 (**_args):
""" """
Expected Element NM1 Expected Element NM1

6
healthcareio/x12/plugins/default/remits.py
View File

@ -57,9 +57,9 @@ def SVC (**_args):
_data['paid_amount'] = np.float64(_data['paid_amount']) _data['paid_amount'] = np.float64(_data['paid_amount'])
return _data return _data
pass pass
@parser(element='N1',x12='835',anchor={'PR':'provider'},map={1:'name'}) # @parser(element='N1',x12='835',anchor={'PR':'provider'},map={1:'name'})
def N1(**_args): # def N1(**_args):
pass # pass
@parser(element='N3',x12='835',parent='N1',map={1:'address_line_1'}) @parser(element='N3',x12='835',parent='N1',map={1:'address_line_1'})
def N3(**_args): def N3(**_args):
pass pass

53
healthcareio/x12/util/document.py
View File

@ -15,7 +15,7 @@ class Builder:
self._plugins = copy.deepcopy(_args['plugins']) self._plugins = copy.deepcopy(_args['plugins'])
self._parents = copy.deepcopy(_args['parents']) self._parents = copy.deepcopy(_args['parents'])
self._loop = {} self._loop = {}
self._logger = None if 'logger' not in _args else _args['logger']
def reset (self): def reset (self):
self._last = {} self._last = {}
@ -32,34 +32,7 @@ class Builder:
if _id : if _id :
return self._last[_id] if _id in self._last else None return self._last[_id] if _id in self._last else None
return None return None
# if _id in self._parents :
# self._last[_id] =
# if 'parent' in _meta : #hasattr(_meta,'parent'):
# _hasField = 'field' in _meta
# _hasParent= _meta['element'] in self._parents
# if _hasField and _hasParent: #_meta.element in self._parents and hasattr(_meta,'field'):
# self._last = _item
# pass
# else:
# for key in self._parents :
# if _meta['element'] in self._parents[key] :
# _ikey = list(self._last.keys())[0]
# _oldinfo = self._last[_ikey]
# if type(_oldinfo) != dict :
# #
# # Only applicable against a dictionary not a list (sorry)
# pass
# else:
# _item = {_ikey: self.merge(_oldinfo,_item)}
# break
# pass
# return _item
def count(self,_element): def count(self,_element):
if _element not in self._loop : if _element not in self._loop :
self._loop[_element] = 0 self._loop[_element] = 0
@ -251,10 +224,12 @@ class Builder:
if _field : if _field :
if 'container' in _meta and type(_document[_field]) != list : if 'container' in _meta and type(_document[_field]) != list :
_document[_field] = [] _document[_field] = []
if _field and _document:
if _field and _document :
if _field not in _document : if _field not in _document :
_document[_field] =_data _document[_field] =_data
pass
else: else:
if 'container' in _meta : if 'container' in _meta :
_document[_field].append(_data) _document[_field].append(_data)
@ -263,9 +238,23 @@ class Builder:
else: else:
if not _field and 'anchor' in _meta : if not _field and 'anchor' in _meta :
# #
# This is an unusual situation ... # We should determine if the element is either a parent or has a parent
# This would allow us to avoid having runaway attributes and undermine structural integrity
#
#
# The element has NOT been specified by the plugin (alas)
# For this case we would advise writing a user-defined plugin to handle this case
#
print (self._logger)
if self._logger :
print (['....................'])
self._logger.log(action='missing-plugin',module='build',data={'element':_row[0],'anchor':_row[1]})
return _document
pass pass
# print ([_row[0],set(_data) - set(_document.keys())])
_document = self.merge(_document,_data) _document = self.merge(_document,_data)
return _document return _document

6
setup.py
View File

@ -14,13 +14,13 @@ args = {
"author":meta.__author__, "author":meta.__author__,
"author_email":"steve.l.nyemba@vumc.org", "author_email":"steve.l.nyemba@vumc.org",
"include_package_data":True, "include_package_data":True,
"license":version.__license__, "license":meta.__license__,
"packages":find_packages(), "packages":find_packages(),
"keywords":["healthcare","edi","x12","analytics","835","837","data","transport","protocol"] "keywords":["healthcare","edi","x12","analytics","835","837","data","transport","protocol"]
} }
args["install_requires"] = ['typer','flask-socketio','seaborn','jinja2','jsonmerge', 'weasyprint','data-transport@git+https://healthcareio.the-phi.com/git/code/transport.git','pymongo','numpy','cloudant','pika','boto','botocore','flask-session','smart_open','smart-top@git+https://healthcareio.the-phi.com/git/code/smart-top.git@data-collector'] args["install_requires"] = ['typer','flask-socketio','seaborn','jinja2','jsonmerge', 'weasyprint','data-transport@git+https://healthcareio.the-phi.com/git/code/transport.git','pymongo','numpy','cloudant','pika','boto','botocore','flask-session','smart_open']
args['url'] = 'https://hiplab.mc.vanderbilt.edu' args['url'] = 'https://hiplab.mc.vanderbilt.edu'
args['scripts']= ['healthcareio/healthcare-io.py'] args['scripts']= ['bin/healthcare-io']
# args['entry_points'] = { # args['entry_points'] = {
# 'console_scripts' : ['healthcareio=healthcareio:register'] # 'console_scripts' : ['healthcareio=healthcareio:register']
# } # }