xsv 0.1.2 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 2eed2bd4654d89a75f817365a919f2ca4c8f4f9bee17e63f5fc938b574826696
-  data.tar.gz: ddfa4e09f2d9378a79943cef340b1af20ae49764b64bcb28478b672d676e4565
+  metadata.gz: 47e4ee16a95b100a1c1bbc526912235bbd1386601a33e3c3449320fe4ea8bc52
+  data.tar.gz: 5b0f8320ff29a3dd036cf4396052cbcee00556c2f116dce63c81eb6f4bb69e2a
 SHA512:
-  metadata.gz: 6b79348311d076d9397b7cd0629ac4502dc8a3410835b68682e4410af64dea2603d6ea2a68157e0a0bd933f9cdc3b3780034e645d7df9036d7696bda927b858d
-  data.tar.gz: e40fa7237b12c4d83639f797ba38c9928e6d412bb658b0fa6855dc2bbca6004d89f36a306c74b207e453d2a5e3597b7e45fd76b464c1c5b3a45a716d09d876f0
+  metadata.gz: 40bde712c1df13d4fd330b24fdd3cb9e40e5983b4271c172b6f77245aa4d57b8aadc470be5faffb5c759e807c6bf5a507f87139279b86395cddf0fc70b5446fb
+  data.tar.gz: 943f5436b416f226f8bcfa2a856dc8079f061be20ca8cc534a9e147692906a280f85f6eb300b7c4e4681ce70a6c5bb81029c6abb70deab87950919e21f03104a

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    xsv (0.1.1)
+    xsv (0.1.2)
       nokogiri (~> 1.10)
       rubyzip (~> 2.2)
 

data/README.md

@@ -1,8 +1,10 @@
 # Xsv .xlsx reader
 
 Xsv is a very basic parser for Excel files in the .xlsx format that strives to
-provide feature parity with common CSV readers and nothing more. This should
-allow for fast parsing of large worksheets with minimal RAM and CPU consumption.
+provide feature parity with common CSV readers and nothing more. This means
+it only parses values to basic Ruby types and does not deal with formatting
+or more advanced functionality. The goal is to allow for fast parsing of large
+worksheets with minimal RAM and CPU consumption.
 
 Xsv stands for 'Excel Separated Values' because Excel just gets in the way.
 
@@ -24,18 +26,48 @@ Or install it yourself as:
 
 ## Usage
 
+Xsv has two modes of operation. By default it returns an array for
+each row in the sheet:
+
 ```ruby
-x = Xsv::File.new("sheet.xlsx")
+x = Xsv::Workbook.open("sheet.xlsx")
+
+sheet = x.sheets[0]
 
-x.sheets[0].each_row(read_headers: true) do |row|
-  row # => { "header1" => "value1", "header2", "value2" }
+# Iterate over rows
+sheet.each_row do |row|
+  row # => ["header1", "header2"], etc.
 end
-j
-x.sheets[0].each_row do |row|
-  row # => ["header1", "header2"]
+
+# Access row by index (zero-based)
+sheet[1] # => ["value1", "value2"]
+```
+
+Alternatively, it can load the headers from the first row and return a hash
+for every row:
+
+```ruby
+x = Xsv::Workbook.open("sheet.xlsx")
+
+sheet = x.sheets[0]
+
+sheet.mode # => :array
+
+# Parse headers and switch to hash mode
+sheet.parse_headers!
+
+sheet.mode # => :hash
+
+sheet.each_row do |row|
+  row # => {"header1" => "value1", "header2" => "value2"}, etc.
 end
+
+sheet[1] # => {"header1" => "value1", "header2" => "value2"}
 ```
 
+Be aware that hash mode will lead to unpredictable results if you have multiple
+columns with the same name!
+
 ## Development
 
 After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake test` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.

data/lib/xsv/helpers.rb

@@ -1,5 +1,39 @@
 module Xsv
   module Helpers
+    BUILT_IN_NUMBER_FORMATS = {
+      1 => "0",
+      2 => "0.00",
+      3 => "#, ##0",
+      4 => "#, ##0.00",
+      5 => "$#, ##0_);($#, ##0)",
+      6 => "$#, ##0_);[Red]($#, ##0)",
+      7 => "$#, ##0.00_);($#, ##0.00)",
+      8 => "$#, ##0.00_);[Red]($#, ##0.00)",
+      9 => "0%",
+      10 => "0.00%",
+      11 => "0.00E+00",
+      12 => "# ?/?",
+      13 => "# ??/??",
+      14 => "m/d/yyyy",
+      15 => "d-mmm-yy",
+      16 => "d-mmm",
+      17 => "mmm-yy",
+      18 => "h:mm AM/PM",
+      19 => "h:mm:ss AM/PM",
+      20 => "h:mm",
+      21 => "h:mm:ss",
+      22 => "m/d/yyyy h:mm",
+      37 => "#, ##0_);(#, ##0)",
+      38 => "#, ##0_);[Red](#, ##0)",
+      39 => "#, ##0.00_);(#, ##0.00)",
+      40 => "#, ##0.00_);[Red](#, ##0.00)",
+      45 => "mm:ss",
+      46 => "[h]:mm:ss",
+      47 => "mm:ss.0",
+      48 => "##0.0E+0",
+      49 => "@",
+    }
+
     # Return the index number for the given Excel column name
     def column_index(col)
       val = 0
@@ -10,5 +44,42 @@ module Xsv
       end
       return val - 1
     end
+
+    # Return a Date for the given Excel date value
+    def parse_date(number)
+      Date.new(1899, 12, 30) + number
+    end
+
+    # Return a time as a string for the given Excel time value
+    def parse_time(number)
+      base = number * 24
+
+      hours = base.truncate
+      minutes = (base - hours) * 60
+
+      "%02d:%02d" % [base, minutes.round]
+    end
+
+    def parse_number(string)
+      if string.include? "."
+        string.to_f
+      else
+        string.to_i
+      end
+    end
+
+    # Tests if the given format string is a date
+    def is_date_format?(format)
+      return false if format.nil?
+      # If it contains at least 2 sequences of d's, m's or y's it's a date!
+      format.scan(/[dmy]+/).length > 1
+    end
+
+    # Tests if the given format string is a time
+    def is_time_format?(format)
+      return false if format.nil?
+      # If it contains at least 2 sequences of h's, m's or s's it's a time!
+      format.scan(/[hms]+/).length > 1
+    end
   end
 end

data/lib/xsv/sheet.rb

@@ -2,25 +2,37 @@ module Xsv
   class Sheet
     include Xsv::Helpers
 
-    attr_reader :xml
+    attr_reader :xml, :mode
 
     def initialize(workbook, xml)
       @workbook = workbook
       @xml = xml
       @headers = []
+
+      # Determine number of columns
+      bounds = @xml.css("cols col").map { |c| [c["min"].to_i, c["max"].to_i] }.flatten
+      @column_count = (bounds.max - bounds.min) + 1
+
+      @mode = :array
     end
 
     def inspect
       "#<#{self.class.name}:#{self.object_id}>"
     end
 
-    # Iterate over rows. Returns an array if read_headers is false, or a hash
-    # with first row values as keys if read_headers is true
-    def each_row(read_headers: false)
-      @parse_headers if read_headers
+    # Iterate over rows
+    def each_row
+      row_index = 0
+      @xml.css("sheetData row").each do |row_xml|
+        row_index += 1
 
-      @xml.css("sheetData row").each_with_index do |row_xml, i|
-        next if i == 0 && @headers.any?
+        next if row_index == 1 && @mode == :hash
+
+        # pad empty rows
+        while row_index < row_xml["r"].to_i do
+          yield(empty_row)
+          row_index += 1
+        end
 
         yield(parse_row(row_xml))
       end
@@ -30,14 +42,23 @@ module Xsv
 
     # Get row by number, starting at 0
     def [](number)
-      parse_row(@xml.css("sheetData row:nth-child(#{number + 1})").first)
+      row_xml = xml.css("sheetData row[r=#{number + 1}]").first
+
+      if row_xml
+        parse_row(row_xml)
+      else
+        empty_row
+      end
     end
 
     # Load headers in the top row of the worksheet. After parsing of headers
     # all methods return hashes instead of arrays
     def parse_headers!
+      @mode = :array
       parse_headers
 
+      @mode = :hash
+
       true
     end
 
@@ -47,14 +68,17 @@ module Xsv
       @headers = parse_row(@xml.css("sheetData row").first)
     end
 
-    def parse_row(xml)
-      if @headers.any?
-        row = {}
-      else
-        row = []
+    def empty_row
+      case @mode
+      when :array
+        [nil] * @column_count
+      when :hash
+        @headers.zip([]).to_h
       end
+    end
 
-      next_index = 0
+    def parse_row(xml)
+      row = empty_row
 
       xml.css("c").each do |c_xml|
         value = case c_xml["t"]
@@ -65,7 +89,23 @@ module Xsv
           when "e" # N/A
             nil
           when nil
-            c_xml.css("v").inner_text.to_i
+            value = parse_number(c_xml.css("v").inner_text)
+
+            if c_xml["s"]
+              style = @workbook.xfs[c_xml["s"].to_i]
+              numFmtId = style[:numFmtId].to_i
+              if numFmtId == 0
+                value
+              elsif is_date_format?(@workbook.numFmts[numFmtId])
+                parse_date(value)
+              elsif is_time_format?(@workbook.numFmts[numFmtId])
+                parse_time(value)
+              else
+                value
+              end
+            else
+              value
+            end
           else
             raise Xsv::Error, "Encountered unknown column type #{c_xml["t"]}"
           end
@@ -73,22 +113,12 @@ module Xsv
         # Determine column position and pad row with nil values
         col_index = column_index(c_xml["r"].scan(/^[A-Z]+/).first)
 
-        (col_index - next_index).times do
-          if @headers.any?
-            row[@headers[next_index]] = nil
-          else
-            row << nil
-          end
-          next_index += 1
-        end
-
-        if @headers.any?
-          row[@headers[next_index]] = value
-        else
-          row << value
+        case @mode
+        when :array
+          row[col_index] = value
+        when :hash
+          row[@headers[col_index]] = value
         end
-
-        next_index += 1
       end
 
       row

data/lib/xsv/version.rb

@@ -1,3 +1,3 @@
 module Xsv
-  VERSION = "0.1.2"
+  VERSION = "0.2.0"
 end

data/lib/xsv/{file.rb → workbook.rb}

@@ -2,14 +2,25 @@ require 'nokogiri'
 require 'zip'
 
 module Xsv
-  class File
+  class Workbook
 
-    attr_reader :sheets, :shared_strings
+    attr_reader :sheets, :shared_strings, :xfs, :numFmts
+
+    # Open the workbook of the given filename
+    def self.open(file)
+      @workbook = self.new(Zip::File.open(file))
+    end
+
+    # Open a workbook from an instance of Zip::File
+    def initialize(zip)
+      @zip = zip
 
-    def initialize(file)
-      @zip = Zip::File.open(file)
       @sheets = []
+      @xfs = []
+      @numFmts = Xsv::Helpers::BUILT_IN_NUMBER_FORMATS
+
       fetch_shared_strings
+      fetch_styles
       fetch_sheets
     end
 
@@ -17,11 +28,6 @@ module Xsv
       "#<#{self.class.name}:#{self.object_id}>"
     end
 
-    def close
-      # FIXME @sheets.each { |s| s.xml.close }
-      @zip.close
-    end
-
     private
 
     def fetch_shared_strings
@@ -37,6 +43,19 @@ module Xsv
       stream.close
     end
 
+    def fetch_styles
+      stream = @zip.glob("xl/styles.xml").first.get_input_stream
+      xml = Nokogiri::XML(stream)
+
+      xml.css("cellXfs xf").each do |xf|
+        @xfs << xf.attributes.map { |k, v| [k.to_sym, v.value] }.to_h
+      end
+
+      xml.css("numFmts numFmt").each do |numFmt|
+        @numFmts[numFmt["numFmtId"].to_i] = numFmt["formatCode"]
+      end
+    end
+
     def fetch_sheets
       @zip.glob("xl/worksheets/sheet*.xml").sort do |entry|
         entry.name.scan(/\d+/).first.to_i

data/lib/xsv.rb

@@ -1,7 +1,9 @@
-require "xsv/file"
+require "date"
+
 require "xsv/helpers"
 require "xsv/sheet"
 require "xsv/version"
+require "xsv/workbook"
 
 module Xsv
   class Error < StandardError; end

data/test.sh

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+ruby -Ilib:test test/*_test.rb

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: xsv
 version: !ruby/object:Gem::Version
-  version: 0.1.2
+  version: 0.2.0
 platform: ruby
 authors:
 - Martijn Storck
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2020-02-18 00:00:00.000000000 Z
+date: 2020-02-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rubyzip
@@ -97,10 +97,11 @@ files:
 - bin/console
 - bin/setup
 - lib/xsv.rb
-- lib/xsv/file.rb
 - lib/xsv/helpers.rb
 - lib/xsv/sheet.rb
 - lib/xsv/version.rb
+- lib/xsv/workbook.rb
+- test.sh
 - xsv.gemspec
 homepage: https://github.com/martijn/xsv
 licenses: