detabulator 0.1.0

data/README.md

@@ -0,0 +1,35 @@
+Detabulator
+===========
+
+Extract columnar data from tabulated fixed-width text.
+
+Example
+-------
+
+Given some tabulated data:
+
+    sample = <<END
+                Column 1    Column 2
+    Drinks      Beer        Whiskey
+    Not drinks  Toothpaste  Mouthwash
+    END
+
+This:
+
+    require "detabulator"
+    Detabulator.new.detabulate(sample)
+
+Will produce this:
+
+    [["", "Column 1", "Column 2"],
+     ["Drinks", "Beer", "Whiskey"],
+     ["Not drinks", "Toothpaste", "Mouthwash"]] 
+
+This is just a first release that does something useful.
+
+Limitations
+-----------
+
+* Double-width characters (e.g. Japanese) are not handled correctly.
+* Combining diacritics are not handled correctly.
+* When one cell contains much longer text and a space, an extra column is generated.

data/lib/detabulator.rb

@@ -0,0 +1,50 @@
+class Detabulator
+  SPACE = 32
+
+  def detabulate(s)
+    lines   = s.split(/\n/).map{ |line| line.unpack("U*") }
+    lengths = extract_segment_lengths(
+                collapse_space_clusters(
+                  space_mask(lines)))
+
+    offset  = nil # GC help
+    lines.map{ |line|
+      offset = 0
+      lengths.map{ |length|
+        cell = (line[offset, length] || []).pack('U*').strip
+        offset += length
+        cell }}
+  end
+
+private
+  #    aa   bb
+  #    c    dddd ee
+  # => ..TTT....T..
+  #
+  def space_mask(lines)
+    max_line_length = lines.map{ |a| a.length }.max
+    lines.inject([true] * max_line_length){ |mask, line|
+      mask.zip(line).
+           map{ |a, b| a && (!b || b == SPACE) }}
+  end
+
+  #    ..TTT....T..
+  # => ..T......T..
+  #
+  def collapse_space_clusters(mask)
+    rmask = mask.reverse
+    rmask.zip(rmask[1..-1]).map{ |a, b| a && !b }.reverse
+  end
+
+  # ..T......T.. => [3, 7, 2]
+  #
+  def extract_segment_lengths(mask)
+    mask.inject([1]){ |a,e|
+      if e
+        a << 1
+      else
+        a[-1] += 1
+        a
+      end }
+  end
+end

data/lib/detabulator/version.rb

@@ -0,0 +1,3 @@
+class Detabulator
+  VERSION = "0.1.0"
+end

metadata

@@ -0,0 +1,58 @@
+--- !ruby/object:Gem::Specification 
+name: detabulator
+version: !ruby/object:Gem::Version 
+  version: 0.1.0
+platform: ruby
+authors: 
+- Paul Battley
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2010-07-18 00:00:00 +01:00
+default_executable: 
+dependencies: []
+
+description: Extract columnar data from tabulated fixed-width text
+email: 
+- pbattley@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: []
+
+files: 
+- lib/detabulator.rb
+- lib/detabulator/version.rb
+- README.md
+has_rdoc: true
+homepage: http://github.com/threedaymonk/detabulator
+licenses: []
+
+post_install_message: 
+rdoc_options: []
+
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.5
+signing_key: 
+specification_version: 3
+summary: Extract columnar data from tabulated fixed-width text
+test_files: []
+