locushandler 0.2.4.1

No project description provided

LocusHandler

This package was created to automate Locus coded data processing steps. It provides functions to parse and validate Locus codes. It also allows to process entire data files that use the Locus classification system or other common economic classification systems such as NAICS or SOC.

Getting Started

Installation

The package can be downloaded and installed using pip.

pip install locushandler

You will then be able to import the package in your Python scripts.

import locushandler as lh

You can also import specific parts of the package.

import locushandler.string_parser as sp
import locushandler.file_handler as fh

Prerequisites

The LocusHandler is coded in Python 3. Using pip install, all Python libraries required to use the LocusHandler should be automatically installed on your machine.

How to use the LocusHandler

Granularity

The granularity parameters takes in a string with the format <a>x<r> with <a> in [1,4,12,36] and <r> in [1,6,6x4,6x4x3]

Activity

• '1' : no activity
• '4' : 4 phases activty cycle
• '12' : 12 phases activity cycle
• '36' : 36 phases activity cycle

Resource

• '1' : no resource
• '6' : resource categories
• '6x4' : resource categories staged
• '6x4x3' : resource categories sub-staged

Examples

• Work Locus : - 4x6 : 1 B, 3 F - 12x6x4 : (B4) 1.2 B4, 2.2 C3 - 36x6x4x3 : 3.3.1 E3ii, (B4ii) 3.1.2 Div
• Resource Locus : - 1x6 : B, F - 12x6x4 : B4 2.2 B4, A4 1.3 C3 - 36x6x4x3 : B4ii 1.2.2 E3ii, B4ii 2.2.2 Div

Input syntax

• The LocusHandler assumes that all work loci contain parentheses around the Distinguishing Resources and the Information Outputs. Please make sure that the input data follows that syntax.

Parsing a Locus code

The string_parser module allows you to parse any work or resource Loci, in a string format, such as

'(B4ii) 1.2.2 B4i'
'B4ii 2.2.2 F'

and get a list, a dictionary or a string of that code at the required granularity.

Parsing to a dictionary

sp.string_parser('(B4ii) 1.2.2 B4Div', 'dict', '36x6x4x3', show_dr=True, show_io=True))

will return the following dictionary

{'dr': {'r1': 'B', 'r2': '4', 'r3': 'ii'},
 'act': {'a1': '1', 'a2': '2', 'a3': '2'},
 'obj': {'r1': 'B',  'r2': '4', 'r3': 'V'},
 'io':  {'r1': '',  'r2': '', 'r3': ''}
 }

make use of the merge_fields parameter to return (field, field value) pairs:

sp.string_parser('(B4ii) 1.2.2 B4Div', 'dict', '36x6x4x3', show_dr=True, show_io=True))

will return the following dictionary

{'dr': 'B4ii',
 'act': '1.2.2',
 'obj': 'B4V',
 'io': ''
}

Parsing to a list

sp.string_parser('(B4ii) 1.2.2 B4Div', 'list', '4x6', show_dr=True, show_io=True))

will return the following list

['B','2','B','']

Parsing to a string

sp.string_parser('(B4ii) 1.2.2 B4Div', 'string', '12x6x4', show_dr=False, show_io=False))

will return the following string

'1.2 B4'

Parsing a Locus code column in a file

The file_handler module allows you to parse an entire column of a .csv file. If the input file contains Locus code as strings the function parse_file can parse it and return a Dataframe or a name of a new saved .csv file with columns containing each element the Locus code at the required granularity. An example of input data would be the following table.

Enterprise_Locus	Employment level	...	Area	Year
'(B4ii) 1.2.2 B4i'	1366	...	NY	2010
(A4iii) 3.3.2 B3ii	235	...	VA	2008
...	...	...	...	...
'(B4ii) 1.3.2 F'	78	...	KS	2010

Parsing to a Pandas dataframe

fh.parse_file('file.csv', 'Enterprise_Locus', 'df', '12x6x4x3', show_dr=False, show_io=False)

would return a Pandas dataframe with the following information

a1	a2	r1	r2	r3	Employment level	...	Area	Year
1	2	B	4	i	1366	...	NY	2010
3	3	B	3	ii	235	...	VA	2008
...	...	...	...	...	...	...	...	...
1	3	F			78	...	KS	2010

Parsing to a new .csv file

fh.parse_file('file.csv', 'Enterprise_Locus', 'path', '12x6x4x3', show_dr=False, show_io=False)

would return the path of the new .csv file that contain the same information as above. In this case, the path would be 'file_parsed_12x6x4x3_Enterprise_Locus.csv'

Mapping to Locus codes

If the input data uses a common classification system, the LocusHandler provide a function to map the input data to Locus code at the desired granularity. An example of input data would be the following table.

Naics	Employment level	...	Area	Year
13589	3468735	...	USA	2010
78621	87685	...	FRANCE	2008
...	...	...	...	...
34697	34786	...	NIGERIA	2010

To call the map_to_locus you need the path of your input file (file_naics.csv), the name of the column with the classification system you map from (Naics) , the name of the column of the classification sytem in the crosswalk file (NAICS5), the path to the crosswalk file (naics2locus.csv), the barcode field you are interested in (enterprise_locus), the granularity (4x6) and the need for dr and io.

fh.map_to_locus('file_naics.csv', 'Naics', 'NAICS5', 'naics2locus.csv', 'entreprise_locus','4x6', show_dr=True, show_io=True)

would return a Pandas dataframe with the following information*.

dr1	a1	r1	io1	Employment level	...	Area	Year
B	1	B		3468735	...	USA	2010
	2	A	C	87685	...	FRANCE	2008
...	...	...	...	...	...	...	...
A	3	B		34786	...	NIGERIA	2010

_{*Naics to Locus mapping not accurate here}

Documentation

• Please read the Technical Brief for details on the project.
• Please read the Specs for details on the functions available.

Authors

• Phillip Nguyen - String parsing
• Olivia Dalglish - Code validation
• Emeline Floc'h - File handling

Contributors

• Vinharng Chew - Stefanie Bourland - Chris Haack - Aaron Lee - Atul Prasad - Reviewers