Staging table means all my efforts to clean the data have to be migrated from Perl to SQL. This is particularly awkward for location. To take an example, the registered office at Companies House might be as daft as:

Country of Origin: Channel Islands

RegAddressCountry: GY1 4PW

RegAddressCounty: St. Peterport

PostTown:

Address Line1:MR FRED BLOGGS

Address Line 2: The Company Name

CompanyName: A variation of the company name

Care of:

Postcode:

So the algorithm to clean the data is broadly: delete or translate some countries (Channel Islands is deleted because it isn't a country. But postcode GY1 implies the country is Guernsey. For my model, the county is also Guernsey as opposed to Alderney. If more than one level is on the same line they have to be split. If data is at too high a level, e.g. the registered country was mistakenly put in Country of Origin and then everything moved up one from where it should be, defaults have to be pushed on the stack moving the location lines down (largest location items being higher in the list, the way I am doing it). So I have hashes of common translations e.g. if Curacao has a sedilla whereas my country table does not have sedillas and BVI gets unabbreviated to British Virgin Islands. The number of ways Ireland gets spelled is particularly astounding ROI, Rep of. etc.. Just about every Republic of something needs different translations there are even several ways for China. Mis-spellings like United Kinmod etc. Scotnadl, Isalnds, Isles-> Islands all have to be corrected, there's a method that splits location lines where pattern matching is needed, a list of deletable countries including Channel Islands and British West Indies to force the next level to be promoted to a country, a list of pushable items where the country is missing already, like Leicester being entered as a country in the register and the facility to assume registered country is the same as the country of origin for some cases. I could go on forwever, but it's about 200 lines of OO Perl and would be about 100 pages in SQL if I load it in before processing.

Update: Yes, I could call the Perl from Postgres after staging, at least I can do that in my dev. env., but I can't expect that to hold true when it gets ported to the hosting env.

---

---

• Are you posting in the right place? Check out Where do I post X? to know for sure.
• Posts may use any of the Perl Monks Approved HTML tags. Currently these include the following:

  <code> <a> <b> <big> <blockquote> <br /> <dd> <dl> <dt> <em> <font> <h1> <h2> <h3> <h4> <h5> <h6> <hr /> <i> <li> <nbsp> <ol> <p> <small> <strike> <strong> <sub> <sup> <table> <td> <th> <tr> <tt> <u> <ul>

• Snippets of code should be wrapped in <code> tags not <pre> tags. In fact, <pre> tags should generally be avoided. If they must be used, extreme care should be taken to ensure that their contents do not have long lines (<70 chars), in order to prevent horizontal scrolling (and possible janitor intervention).
• Want more info? How to link or How to display code and escape characters are good places to start.