NHSTechnicalTest

Language and environment

The software is written in pure Python 3 and although written and intended for use in a Linux environment should run in pretty much any Python >= 3.4 environment.

To check what version of Python/Python3 you are currently running execute

$ python -V

or

$ python3 -V

from the shell e.g.

tim@merlin:~/metatron/Projects/NHSTechnicalTest$ python -V

Python 3.6.0 :: Anaconda 4.3.0 (64-bit)

It is assumed that a suitable file containing the input data is available. By default this would be the file import_data.csv in the current working directory, but a different location may be specified either on the command line or passed as a parameter to the appropriate scripts.

Files

The software consists of the following source files

1. NHSPostCode.py Fundamental definitions (principally the PostCode class and status codes)
2. NHSTechnicalTestPart1.py Part 1 tests
3. NHSTechnicalTestPart2.py Part 2 tests
4. NHSTechnicalTestPart3.py Part 3 tests

Running the software

Part 1

From the bash shell run

$ python3 NHSTechnicalTestPart1.py

The command line supports the various options available in the Python unittest library (e.g. --verbose)

Part 2

From the bash shell run

$ python3 NHSTechnicalTestPart2.py

this assumes that the import_data.csv file is in the current directory and by default will write output of failed records to failed_validation.csv in the current directory as specified.

These defaults can be overridden using the --input and --unmatched options respectively e.g.

$ python3 NHSTechnicalTestPart2.py --input /home/fred/myfile.csv --unmatched /tmp/foo.csv

Part 3

From the bash shell run

$ python3 NHSTechnicalTestPart3.py

this assumes that the import_data.csv file is in the current directory and by default will write output to failed_validation.csv and matched records to successeded_validation.csv both in the current directory as specified.

These defaults can be overridden using the --input, --matched and --unmatched options respectively e.g.

$ python3 NHSTechnicalTestPart3.py --input /home/fred/myfile.csv --unmatched /tmp/foo.csv --matched /tmp/bar.csv

Validation and Status Codes

In the event that a PostCode does not validate an analysis can optionally be run. Objects of the PostCode class have an attribute status.

The following codes are defined

###OK The postcode has matched correctly against the RE.

###UNKNOWN The postcode has failed to match against the RE, but the cause is unknown (the analyse flag was either not set or set to False)

###OUTWARD_MALFORMED The outward group of the postcode (i.e. to the left of the space) does not fit the known/allowed patterns e.g. AA9A

###OUTWARD_AA9A_MALFORMED The outward group of an AA9A postcode is malformed. This is usually because it contains an illegal character (e.g. Q)

###OUTWARD_AA9_MALFORMED
The outward group of an AA9 postcode is malformed. This is usually because it contains an illegal character (e.g. Q). Note that errors in single and double digit areas have separate codes, SINGLE_DIGIT_DISTRICT and DOUBLE_DIGIT_DISTRICT.

###OUTWARD_A9_MALFORMED
The outward group of an A9 postcode is malformed. This is usually because it contains an illegal character (e.g. Q)

###INWARD_MALFORMED
The inward group (to the right of the space) of any postcode is malformed, usually because it contains an illegal character (e.g. I)

###INCORRECT_GROUPING The number of groups (i.e. sets of alphanumeric characters separated by a single whitespace) is not exactly two.

###JUNK The postcode contains non alphanumeric/whitespace characters.

###SINGLE_DIGIT_DISTRICT The postcode area should only have single digit districts.

###DOUBLE_DIGIT_DISTRICT The postcode area should only have double digit districts.