Use chained methods for scrape params #15

5 년 전 · 09b289d1fe
--- a/README.md
+++ b/README.md
@@ -1,12 +1,19 @@
 # UK Planning Scraper

 **PRE-ALPHA: Only works with Idox and Northgate sites and spews a lot of stuff to STDOUT. Not for production use.**
 **PRE-ALPHA: Only works with Idox and Northgate sites and spews a lot of stuff
 to STDOUT. Not for production use.**

 This gem scrapes planning applications data from UK local planning authority websites, eg Westminster City Council. Data is returned as an array of hashes, one hash for each planning application.
 This gem scrapes planning applications data from UK local planning authority
 websites, eg Westminster City Council. Data is returned as an array of hashes,
 one hash for each planning application.

 This scraper gem doesn't use a database. Storing the output is up to you. It's just a convenient way to get the data.
 This scraper gem doesn't use a database. Storing the output is up to you. It's
 just a convenient way to get the data.

 Currently this only works for Idox and Northgate sites. The ultimate aim is to provide a consistent interface in a single gem for all variants of all planning systems: Idox Public Access, Northgate Planning Explorer, OcellaWeb, Agile Planning and all the one-off systems.
 Currently this only works for Idox and Northgate sites. The ultimate aim is to
 provide a consistent interface in a single gem for all variants of all planning
 systems: Idox Public Access, Northgate Planning Explorer, OcellaWeb, Agile
 Planning and all the one-off systems.

 This project is not affiliated with any organisation.

@@ -15,7 +22,8 @@ This project is not affiliated with any organisation.
 Add this line to your application's Gemfile:

 ```ruby
 gem 'uk_planning_scraper', :git => 'https://github.com/adrianshort/uk_planning_scraper/'
 gem 'uk_planning_scraper', \
  git: 'https://github.com/adrianshort/uk_planning_scraper/'
 ```

 And then execute:
@@ -38,66 +46,87 @@ require 'pp'

 ### Scrape from a council

 Applications in Westminster decided in the last seven days:

 ```ruby
 apps = UKPlanningScraper::Authority.named('Westminster').scrape({ decided_days: 7 })
 pp apps
 pp UKPlanningScraper::Authority.named('Westminster').decided_days(7).scrape
 ```

 ### Scrape from a bunch of councils

 Scrape the last week's planning decisions across the whole of
 London (actually 23 of the 35 authorities right now):

 ```ruby
 auths = UKPlanningScraper::Authority.tagged('london')
 authorities = UKPlanningScraper::Authority.tagged('london')

 auths.each do |auth|
  apps = auth.scrape({ decided_days: 7 })
  pp apps # You'll probably want to save `apps` to your database here
 authorities.each do |authority|
  applications = authority.decided_days(7).scrape
  pp applications
  # You'll probably want to save `applications` to your database here
 end
 ```

 Yes, we just scraped the last week's planning decisions across the whole of London (actually 23 of the 35 authorities right now) with five lines of code.

 ### Satisfy your niche interests

 Launderette applications validated in the last seven days in Scotland:

 ```ruby
 auths = UKPlanningScraper::Authority.tagged('scotland')
 authorities = UKPlanningScraper::Authority.tagged('scotland')

 auths.each do |auth|
  apps = auth.scrape({ validated_days: 7, keywords: 'launderette' })
  pp apps # You'll probably want to save `apps` to your database here
 authorities.each do |authority|
  applications = authority.validated_days(7).keywords('launderette').scrape
  pp applications # You'll probably want to save `apps` to your database here
 end
 ```

 ### More search parameters
 ### More scrape parameter methods

 Chain as many scrape parameter methods on a `UKPlanningScraper::Authority`
 object as you like, making sure that `scrape` comes last.

 ```ruby
 received_from(Date.parse("1 Jan 2016"))
 received_to(Date.parse("31 Dec 2016"))

 # Received in the last n days (including today)
 # Use instead of received_to, received_from
 received_days(7) 

 validated_to(Date.today)
 validated_from(Date.today - 30)
 validated_days(7) # instead of validated_to, validated_from

 decided_to(Date.today)
 decided_from(Date.today - 30)
 decided_days(7) # instead of decided_to, decided_from

 # Check that the systems you're scraping return the
 # results you expect for multiple keywords (AND or OR?)
 keywords("hip gable") 

 applicant_name("Mr and Mrs Smith") # Currently Idox only
 application_type("Householder") # Currently Idox only
 development_type("") # Currently Idox only

 # Don't try these all at once
 params = {
  received_to: Date.today,
  received_from: Date.today - 30,
  received_days: 7, # instead of received_to, received_from
  validated_to: Date.today,
  validated_from: Date.today - 30,
  validated_days: 7, # instead of validated_to, validated_from
  decided_to: Date.today,
  decided_from: Date.today - 30,
  decided_days: 7 # instead of decided_to, decided_from
  keywords: "hip gable", # Check that the systems you're scraping return the results you expect for multiple keywords (AND or OR?)
 }

 apps = UKPlanningScraper::Authority.named('Camden').scrape(params)
 scrape # runs the scraper
 ```

 ### Save to a SQLite database

 This gem has no interest whatsoever in persistence. What you do with the data it outputs is up to you: relational databases, document stores, VHS and clay tablets are all blissfully none of its business. But using the [ScraperWiki](https://github.com/openaustralia/scraperwiki-ruby) gem is a really easy way to store your data:
 This gem has no interest whatsoever in persistence. What you do with the data it
 outputs is up to you: relational databases, document stores, VHS and clay
 tablets are all blissfully none of its business. But using the
 [ScraperWiki](https://github.com/openaustralia/scraperwiki-ruby) gem is a really
 easy way to store your data:

 ```ruby
 require 'scraperwiki' # Must be installed, of course
 ScraperWiki.save_sqlite([:authority_name, :council_reference], apps)
 ScraperWiki.save_sqlite([:authority_name, :council_reference], applications)
 ```

 That `apps` param can be a hash or an array of hashes, which is what gets returned by our `Authority.scrape`.
 That `applications` param can be a hash or an array of hashes, which is what
 gets returned by our `Authority.scrape`.

 ### Find authorities by tag

@@ -130,11 +159,18 @@ and whatever you'd like to add that would be useful to others.

 ### WTF is up with London?

 London has got 32 London Boroughs, tagged `londonboroughs`. These are the councils under the authority of the Mayor of London and the Greater London Authority.
 London has got 32 London Boroughs, tagged `londonboroughs`. These are the
 councils under the authority of the Mayor of London and the Greater London
 Authority.

 It has 33 councils: the London Boroughs plus the City of London (named `City of London`). We don't currently have a tag for this, but if you want to add `londoncouncils` please go ahead.
 It has 33 councils: the London Boroughs plus the City of London (named `City of
 London`). We don't currently have a tag for this, but if you want to add
 `londoncouncils` please go ahead.

 And it's got 35 local planning authorities: the 33 councils plus the two `londondevelopmentcorporations`, named `London Legacy Development Corporation` and `Old Oak and Park Royal Development Corporation`. The tag `london` covers all (and only) the 35 local planning authorities in London.
 And it's got 35 local planning authorities: the 33 councils plus the two
 `londondevelopmentcorporations`, named `London Legacy Development Corporation`
 and `Old Oak and Park Royal Development Corporation`. The tag `london` covers
 all (and only) the 35 local planning authorities in London.

 ```ruby
 UKPlanningScraper::Authority.tagged('londonboroughs').size
@@ -151,13 +187,13 @@ UKPlanningScraper::Authority.tagged('london').size

 ```ruby
 UKPlanningScraper::Authority.named('Merton').tags
 # => ["england", "london", "londonboroughs", "northgate", "outerlondon", "southlondon"]
 # => ["england", "london", "londonboroughs", "northgate", "outerlondon", "southlondon"]

 UKPlanningScraper::Authority.not_tagged('london')
 # => [...]
 # => [...]

 UKPlanningScraper::Authority.named('Islington').tagged?('southlondon')
 # => false
 # => false
 ```

 ### List all authorities
@@ -177,15 +213,20 @@ The list of authorities is in a CSV file in `/lib/uk_planning_scraper`:

 https://github.com/adrianshort/uk_planning_scraper/blob/master/lib/uk_planning_scraper/authorities.csv

 The easiest way to add to or edit this list is to edit within GitHub (use the pencil icon) and create a new pull request for your changes. If accepted, your changes will be available to everyone with the next version of the gem.
 The easiest way to add to or edit this list is to edit within GitHub (use the
  pencil icon) and create a new pull request for your changes. If accepted, your
  changes will be available to everyone with the next version of the gem.

 The file format is one line per authority, with comma-separated:

 - Name (omit "the", "council", "borough of", "city of", etc. and write "and" not "&", except for `City of London` which is a special case)
 - Name (omit "the", "council", "borough of", "city of", etc. and write "and" not
  "&", except for `City of London` which is a special case)
 - URL of the search form (use the advanced search URL if there is one)
 - Tags (use as many comma-separated tags as is reasonable, lowercase and all one word.)
 - Tags (use as many comma-separated tags as is reasonable, lowercase and all one
  word.)

 There's no need to manually add tags to the `authorities.csv` file for the software systems like `idox`, `northgate` etc as these are added automatically.
 There's no need to manually add tags to the `authorities.csv` file for the
 software systems like `idox`, `northgate` etc as these are added automatically.

 Please check the tag list before you change anything:

@@ -195,10 +236,17 @@ pp UKPlanningScraper::Authority.tags

 ## Development

 After checking out the repo, run `bin/setup` to install dependencies. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
 After checking out the repo, run `bin/setup` to install dependencies. You can
 also run `bin/console` for an interactive prompt that will allow you to
 experiment.

 To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
 To install this gem onto your local machine, run `bundle exec rake install`. To
 release a new version, update the version number in `version.rb`, and then run
 `bundle exec rake release`, which will create a git tag for the version, push
 git commits and tags, and push the `.gem` file to
 [rubygems.org](https://rubygems.org).

 ## Contributing

 Bug reports and pull requests are welcome on GitHub at https://github.com/adrianshort/uk_planning_scraper.
 Bug reports and pull requests are welcome on GitHub at
 https://github.com/adrianshort/uk_planning_scraper.
--- a/lib/uk_planning_scraper.rb
+++ b/lib/uk_planning_scraper.rb
@@ -1,5 +1,6 @@
 require "uk_planning_scraper/version"
 require "uk_planning_scraper/authority"
 require "uk_planning_scraper/authority_scrape_params"
 require "uk_planning_scraper/application"
 require 'uk_planning_scraper/idox'
 require 'uk_planning_scraper/northgate'
--- a/lib/uk_planning_scraper/authority.rb
+++ b/lib/uk_planning_scraper/authority.rb
@@ -3,6 +3,7 @@ require 'csv'
 module UKPlanningScraper
  class Authority
    attr_reader :name, :url
    
    @@authorities = []

    def initialize(name, url)
@@ -10,43 +11,25 @@ module UKPlanningScraper
      @url = url.strip
      @tags = [] # Strings in arbitrary order
      @applications = [] # Application objects
      @scrape_params = {}
    end

    def scrape(params, options = {})
    def scrape(options = {})
      default_options = {
        delay: 10,
      }
      options = default_options.merge(options) # The user-supplied options override the defaults
      
      # Validated within the last n days
      # Assumes that every scraper/system can do a date range search
      if params[:validated_days]
        params[:validated_to] = Date.today
        params[:validated_from] = Date.today - (params[:validated_days] - 1)
      end
        
      # Received within the last n days
      # Assumes that every scraper/system can do a date range search
      if params[:received_days]
        params[:received_to] = Date.today
        params[:received_from] = Date.today - (params[:received_days] - 1)
      end
      
      # Decided within the last n days
      # Assumes that every scraper/system can do a date range search
      if params[:decided_days]
        params[:decided_to] = Date.today
        params[:decided_from] = Date.today - (params[:decided_days] - 1)
      end
      
      # The user-supplied options override the defaults
      options = default_options.merge(options)

      # Select which scraper to use
      case system
      when 'idox'
        @applications = scrape_idox(params, options)
        @applications = scrape_idox(@scrape_params, options)
      when 'northgate'
        @applications = scrape_northgate(params, options)
        @applications = scrape_northgate(@scrape_params, options)
      else
        raise SystemNotSupported.new("Planning system not supported for #{@name} at URL: #{@url}")
        raise SystemNotSupported.new("Planning system not supported for \
          #{@name} at URL: #{@url}")
      end
      
      # Post processing
@@ -58,6 +41,10 @@ module UKPlanningScraper
      output = []
      # FIXME - silently ignores invalid apps. How should we handle them?
      @applications.each { |app| output << app.to_hash if app.valid? }
      
      # Reset so that old params don't get used for new scrapes
      clear_scrape_params
      
      output  # Single point of successful exit
    end
    
@@ -82,15 +69,15 @@ module UKPlanningScraper
    
    def system
      if @url.match(/search\.do\?action=advanced/i)
        s = 'idox'
        'idox'
      elsif @url.match(/generalsearch\.aspx/i)
        s = 'northgate'
        'northgate'
      elsif @url.match(/ocellaweb/i)
        s = 'ocellaweb'
        'ocellaweb'
      elsif @url.match(/\/apas\//)
        s = 'agileplanning'
        'agileplanning'
      else
        s = 'unknownsystem'
        'unknownsystem'
      end
    end

@@ -135,7 +122,8 @@ module UKPlanningScraper
    def self.load
      # Don't run this method more than once
      return unless @@authorities.empty?
      CSV.foreach(File.join(File.dirname(__dir__), 'uk_planning_scraper', 'authorities.csv')) do |line|
      CSV.foreach(File.join(File.dirname(__dir__), 'uk_planning_scraper', \
        'authorities.csv')) do |line|
        auth = Authority.new(line[0], line[1])
        auth.add_tags(line[2..-1])
        auth.add_tag(auth.system)
--- a/lib/uk_planning_scraper/authority_scrape_params.rb
+++ b/lib/uk_planning_scraper/authority_scrape_params.rb
@@ -0,0 +1,134 @@
 require 'date'

 module UKPlanningScraper
  class Authority
    # Parameter methods for Authority#scrape
    # Desgined to be method chained, eg:
    # 
    # applications = UKPlanningScraper::Authority.named("Barnet"). \
    # development_type("Q22").keywords("illuminat"). \
    # validated_days(30).scrape

    def validated_days(n)
      # Validated within the last n days
      # Assumes that every scraper/system can do a date range search
      check_class(n, Fixnum)

      unless n > 0
        raise ArgumentError.new("validated_days must be greater than 0")
      end
      
      validated_from(Date.today - (n - 1))
      validated_to(Date.today)
      self
    end

    def received_days(n)
      # received within the last n days
      # Assumes that every scraper/system can do a date range search
      check_class(n, Fixnum)

      unless n > 0
        raise ArgumentError.new("received_days must be greater than 0")
      end
      
      received_from(Date.today - (n - 1))
      received_to(Date.today)
      self
    end

    def decided_days(n)
      # decided within the last n days
      # Assumes that every scraper/system can do a date range search
      check_class(n, Fixnum)

      unless n > 0
        raise ArgumentError.new("decided_days must be greater than 0")
      end
      
      decided_from(Date.today - (n - 1))
      decided_to(Date.today)
      self
    end
    
    def applicant_name(s)
      unless system == 'idox'
        raise NoMethodError.new("applicant_name is only implemented for Idox. \
          This authority (#{@name}) is #{system.capitalize}.")
      end
      
      check_class(s, String)
      @scrape_params[:applicant_name] = s.strip
      self
    end

    def application_type(s)
      unless system == 'idox'
        raise NoMethodError.new("application_type is only implemented for \
          Idox. This authority (#{@name}) is #{system.capitalize}.")
      end
      
      check_class(s, String)
      @scrape_params[:application_type] = s.strip
      self
    end

    def development_type(s)
      unless system == 'idox'
        raise NoMethodError.new("development_type is only implemented for \
          Idox. This authority (#{@name}) is #{system.capitalize}.")
      end
      
      check_class(s, String)
      @scrape_params[:development_type] = s.strip
      self
    end

    private
    
    # Handle the simple params with this
    def method_missing(method_name, *args)
      sc_params = {
        validated_from: Date,
        validated_to: Date,
        received_from: Date,
        received_to: Date,
        decided_from: Date,
        decided_to: Date,
        keywords: String
      }
      
      value = args[0]
      
      if sc_params[method_name]
        check_class(value, sc_params[method_name], method_name.to_s)
        value.strip! if value.class == String
        
        if value.class == Date && value > Date.today
          raise ArgumentError.new("#{method_name} can't be a date in the " + \
            "future (#{value.to_s})")
        end
        
        @scrape_params[method_name] = value
        self
      else
        raise NoMethodError.new(method_name.to_s)
      end
    end

    def clear_scrape_params
      @scrape_params = {}
    end
    
    # https://stackoverflow.com/questions/5100299/how-to-get-the-name-of-the-calling-method
    def check_class(
      param_value,
      expected_class,
      param_name = caller_locations(1, 1)[0].label) # name of calling method
      unless param_value.class == expected_class
        raise TypeError.new("#{param_name} must be a " \
          "#{expected_class} not a #{param_value.class.to_s}")
      end
    end
  end
 end
--- a/lib/uk_planning_scraper/version.rb
+++ b/lib/uk_planning_scraper/version.rb
@@ -1,3 +1,3 @@
 module UKPlanningScraper
  VERSION = "0.3.2"
  VERSION = "0.4.0"
 end
--- a/spec/brighton_spec.rb
+++ b/spec/brighton_spec.rb
@@ -11,7 +11,7 @@ describe UKPlanningScraper::Authority do

      it 'returns apps' do
        apps = VCR.use_cassette("#{self.class.description}") {
          scraper.scrape({ decided_days: 4 }, { delay: 0 })
          scraper.decided_days(4).scrape({ delay: 0 })
        }
        pp apps
 #        expect(authority).to be_a(UKPlanningScraper::Authority)