case_scraper 0.0.4 → 0.1.0

checksums.yaml

@@ -1,7 +1,15 @@
 ---
-SHA1:
-  metadata.gz: 79b5532947888d98823625fd8d4dce3dad260cab
-  data.tar.gz: 6d43160afde9bbd261b7f3d1dfca49af3b913496
-SHA512:
-  metadata.gz: 3f0e434c0fd38294e9ee47080483e4df2428e7f640de3c6742798d2649bbbcf73b165e6f52e5b2b8bba0a930dbd8e997594395564e39039f1c2126e82b5cff5b
-  data.tar.gz: 3625f884841744b71d5f29f3a6b1adfee0b4882c1846116f37953d1ea1fa57dad04e5cd7151ea59d1506f7738270e3b29231883d600e3bedd36538fd6653868c
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    NGQ0ODk4YjUzYTJjNWE0YTU0MGY3ZDAyMjE4ZGY3MTM3MjY1ODE1Ng==
+  data.tar.gz: !binary |-
+    YTQ4MmMxNDk2MmEwOWE2N2YzOTFhYThlMTQ0NTY0YWRlZTVjYzQ2MQ==
+!binary "U0hBNTEy":
+  metadata.gz: !binary |-
+    ZGFkODBjYTU2NGE0YWU0OTg0OTcxYmVlNWNjNGI4Y2I5ODQwMzI5ZDQ5Yjdj
+    MDVjMDkyZWRiZDVhMzAwNWJkYmY1ZjlmZGRhMWZlMjIzYTlmMjk4MWFiNDgz
+    OGMyODAyYjM1OTE2ZDk3ZDQ4M2RiMmI2NTllZTNiYjc2MGY2ZWM=
+  data.tar.gz: !binary |-
+    YTNjYzI1YmVkOWVlMjVmNDMwYTczMDRhODJkZDQzYjUwZTI4N2EwYjBlZDVk
+    NTA1NWQ1NzhiZWYzNmJmYmYwMDY0OTMwYzQ1OGVhNzU3OGYwYmMyZWFkZjQ5
+    YjQwODVlYjhlZTBhZDJjNjBhOThiN2ZiMzg1MWVjNWQzZTMxN2M=

data/lib/case_scraper.rb

@@ -2,13 +2,22 @@
 #
 # This script uses the Yomu library 
 # to extract the text from a legal
-# brief in RTF format. It then uses
-# regular expressions to search the
-# document for case citations and
-# extract the volume, reporter, and
-# starting page for each case. The case
-# citations are then output to a text
-# file, with one citation per line.
+# brief that is in RTF format. The
+# script then applies regular
+# expressions to search the text of
+# the document for case citations.
+
+# You have the option of extracting
+# only the volume, reporter, and starting
+# page number for each case, or extracting
+# the entire citation (though this is
+# imperfect) along with the organizational
+# headings from the brief, with the case
+# citations each placed under the
+# appropriate section or subsection. The
+# case citations and heading text are
+# then output to a text file, with one
+# citation or heading per line.
 #
 # This is an experimental script. Use 
 # at your own risk and please be sure 
@@ -37,16 +46,57 @@
 # Require the external Yomu library.
 
 require 'yomu'
+require 'optparse'
+
+# Define options for toggling citation
+# type and inclusion of headings in the
+# brief.
+
+options = {}
+
+optparse = OptionParser.new do|opts|
+   # Help banner.
+   opts.banner = "Usage: case_scraper.rb [options] InputFile.rtf OutputFile.txt ..."
+ 
+   # Define options.
+   options[:fullcite] = false
+   opts.on( '-f', '--full', 'Scrape complete case citations with headings.' ) do
+     options[:fullcite] = true
+   end
+ 
+   opts.on( '-h', '--help', 'Display help.' ) do
+     puts opts
+     exit
+   end
+ end
+ 
+# Parse the command-line and remove options from ARGV array.
+optparse.parse!
 
 # Test that input and output files 
 # are specified.
 
 unless ARGV.length == 2
   puts "You must include an input filename (RTF format) and an output filename (TXT format)."
-  puts "Usage: ruby MyScript.rb InputFile.rtf OutputFile.txt\n"
+  puts "Usage: ruby case_scraper.rb InputFile.rtf OutputFile.txt\n"
   exit
 end
 
+if options[:fullcite]
+
+# Regexp for scraping the entire case citation
+
+  citationregex = "(((([a-z]{3}(\\.|\\?|!)\"?)|see also|see|accord)\\s)|^[A-Z][a-z]*\\s([Vv][\\.]*\\s))[A-Z].{1,125}(\\,|\\s(\\[\\d{4,4}\\])|\\s\\(\\d{4,4}\\)|\\s)\\s\\d{1,4}\\s[A-Za-z0-9\\.\\s]{3,17}\\d{1,8}(\\s?|\\,|\\.)((\\(|\\[)([A-Za-z0-9\\.\\,\\s]{1,20})*\\d{4,4}(\\]|\\))*)*|^\\s*[A-Za-z0-9]{1,4}\\.\\s.{1,150}$"
+else
+
+  # Regexp for scraping the just the volume,
+  # reporter, and first page number.
+
+  citationregex = "\\d{1,4}\\s\[^\\[\\(]{1,16}\\s\\d{1,7}"
+end
+
+# Regexp for scraping the brief headings.
+
 # Set the two command line arguments 
 # as variables.
 
@@ -82,22 +132,67 @@ else
 end
   
 # Using regular expressions, extract 
-# the case volume, the reporter 
-# abbreviation, and the starting page 
-# number for (hopefully . . . ) each 
-# full case citation in the brief.
-#
+# the citations and headings, per
+# the specified options.
 # The regular expressions should
 # catch common variations, including 
 # spacing or lack of spacing in the 
 # reporter abbreviation, 
 # state-specific citation formats for 
 # New York and California, and
-# Westlaw and LEXIS citations for 
-# unpublished rulings.
+# WestLaw and LEXIS citations for 
+# unpublished rulings. As you may see
+# if you run the example file, the script
+# is imperfect and may over-select when
+# dealing with short sentences or
+# certain citations at the beginning
+# of a new line.
 
 puts "Scraping case citations from #{input}."
-text.scan(/\d+\s\i*\w+\.+\s*\w*\.*\s*\w*\.*\s*\w*\.*\s\d+/) { |w| target.write "#{w}\n" }
+
+# Alternative processing procedures
+# depending on whether the script is
+# extracting full or partial citations.
+# Slightly more processing is required
+# for full citations because the regexp
+# has a lot of nested captures and the
+# resulting array is more complex.
+
+if options[:fullcite]
+	allcapture_array = text.scan(/(#{citationregex})/)
+
+	fullcase_array = allcapture_array.transpose
+
+	fullcases_only_array = fullcase_array.first
+
+	fullcases_only_array.flatten
+
+	fullcases_only_array.each do |a|
+	  a.gsub!(/^[a-z]{3,3}\."?\s/, "")
+	end
+
+	fullcases_only_array.reject! { |c| c.empty? }
+	
+	puts "Writing case citations to #{output}."
+		
+	File.open(target, "w+") do |f|
+	  f.puts(fullcases_only_array)
+	end
+
+else
+	allcapture_array = text.scan(/(#{citationregex})/)
+	fullcase_array = allcapture_array.transpose
+
+	fullcases_only_array = fullcase_array.first
+
+	fullcases_only_array.reject! { |c| c.empty? }
+
+	
+	fullcases_only_array.flatten
+		File.open(target, "w+") do |f|
+	  f.puts(fullcases_only_array)
+	end
+end
 
 puts "Writing case citations to #{output}."
 
@@ -108,12 +203,19 @@ target.close()
 # Open the output file in the user's 
 # default text editor.
 #
-# The citation list, one per line, is 
-# in the appropriate format for 
+# If the chosen output is a bare,
+# partial citation list, then the
+# output, with one citation per line, 
+# is in the appropriate format for 
 # cutting and pasting directly into 
 # Westlaw "Find & Print." (The 
 # maximum number of lines for a 
 # single query is 99.)
+#
+# For full citations with headings, 
+# the output should save you time in
+# structuring your opposition / reply
+# brief.
 
 %x{ call #{output} }
 

metadata

@@ -1,31 +1,31 @@
 --- !ruby/object:Gem::Specification
 name: case_scraper
 version: !ruby/object:Gem::Version
-  version: 0.0.4
+  version: 0.1.0
 platform: ruby
 authors:
 - Greg Kochansky
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-10-03 00:00:00.000000000 Z
+date: 2013-10-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: yomu
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
-description: Scrape case citations - volume, reporter, and starting page number -
-  from legal briefs that are in rtf format.
+description: Scrape case citations and section headings from legal briefs in *.rtf
+  format.
 email: greg@greg-k.com
 executables: []
 extensions: []
@@ -42,12 +42,12 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
@@ -55,6 +55,5 @@ rubyforge_project:
 rubygems_version: 2.0.7
 signing_key: 
 specification_version: 4
-summary: Scrape case citations - volume, reporter, and starting page number - from
-  legal briefs that are in rtf format.
+summary: Scrape case citations and section headings from legal briefs in *.rtf format.
 test_files: []