Question

我在ruby中有一个嵌套的哈希：

{"Table1"=>{"Acct type"=>{"Expected"=>"ACC", "Actual"=>"ACC"}, "Seq No"=>{"Expected"=>"100.0", "Actual"=>#<BigDecimal:56d0b28,'0.1E3',9(18)>}, "Class"=>{"Expected"=>"AC", "Actual"=>"AC"}}, "Table 2"=>{"Date"=>{"Expected"=>"20140606", "Actual"=>"20130606"}}}

我需要以表格格式显示上面的嵌套哈希 -

Table   Field   Expected Value  Actual Value
Table 1 Acct type  ACC            ACC
Table 1 Seq No     100.0          100.0
Table 1 Class      AC             AC
Table 2 Date       20140606       20130606

任何建议/指针都会非常有用。我尝试过使用＆＃39; tableprint＆＃39;和＆＃39; text-table＆＃39;宝石但却无法获得理想的结果。上述数据是使用ActiveRecord从DB2表中提取的。

Answer 1

这是一个解决方案：

x = {
  "Table1"=>{
    "Acct type"=>{"Expected"=>"ACC", "Actual"=>"ACC"},
    "Seq No"=>{"Expected"=>"100.0", "Actual"=> 100.00},
    "Class"=>{"Expected"=>"AC", "Actual"=>"AC"}
  },
  "Table 2"=>{
    "Date"=> {"Expected"=>"20140606",
              "Actual"=>"20130606"}}}

def table_row(row_name, row_values)
  [row_name, row_values["Expected"], row_values["Actual"]]
end

def table_rows(table_hash)
  table_hash.map do |field_name, field_values|
    table_row(field_name, field_values)
  end
end

def widest_string_lengths(visual_table)
  table_columns_count = visual_table.first.size - 1
  0.upto(table_columns_count).map do |index|
    columns = visual_table.map { |row| row[index].to_s }
    columns.max_by(&:length).length + 2
  end
end

def aligned_visual_table(visual_table)
  widest_string_items = widest_string_lengths(visual_table)

  visual_table.map do |row|
    row.each_with_index.map do |cell, index|
      cell.to_s.ljust(widest_string_items[index] + 2)
    end
  end
end

table_headers = ['Table', 'Field', 'Value', 'Actual Value']

# This just turns the hash to array of arrays
visual_table = x.map do |table_name, table_hash|
   table_rows(table_hash).map do |table_rows|
      table_rows.insert(0, table_name)
    end
end.flatten(1).insert(0, table_headers)

puts "*********"
puts aligned_visual_table(visual_table).map(&:join).join("\n")

Answer 2

您的示例表明存在影响表格外观的未指定参数。例如，列标签“表”和“字段”是左调整的，而“预期值”和“实际值”则不是。此外，“预期值”与“帐户类型”重叠。下面我建议你如何构建表格，但我做出了妥协：左调整所有列标签而不是垂直重叠。

<强>代码

def print_table(h, column_labels_and_spacing, table_label)
  column_labels = column_labels_and_spacing.map(&:first)
  column_spacing = column_labels_and_spacing[0..-2].map(&:last) << 0
  rows = h.map { |k,f|
            [table_label[k]].product(f.map { |field, g|
              [[field, g.values.map(&:to_s)].flatten] }) }
          .flat_map { |e| e.map(&:flatten) }.unshift(column_labels)
  column_widths = rows.transpose.map { |c| c.max_by(&:size).size }
                      .zip(column_spacing).map { |a,b| a+b }
  rows.each do |row|
    row.zip(column_widths).each { |s, width| print s.ljust(width) }
    puts
  end  
end

示例

输入

h = {"Table1"=> {"Acct type"=>{"Expected"=>"ACC", "Actual"=>"ACC"}, "Seq No" =>{"Expected"=>"100.0", "Actual"=>100}, "Class" =>{"Expected"=>"AC", "Actual"=>"AC"}}, "Table 2"=> {"Date" =>{"Expected"=>"20140606", "Actual"=>"20130606"}}} column_labels_and_spacing = [["Table", 1], ["Field", 2], ["Expected", 3], ["Actual"]] table_label = { "Table1"=>"Table 1", "Table 2"=>"Table 2" }

为了简化说明，我将BigDecimal值转换为Fixnum。

调用方法

print_table(h, column_labels_and_spacing, table_label) #-> Table Field Expected Actual # Table 1 Acct type ACC ACC # Table 1 Seq No 100.0 100 # Table 1 Class AC AC # Table 2 Date 20140606 20130606

<强>解释

对于上面的例子：

column_labels = column_labels_and_spacing.map(&:first) #=> ["Table", "Field", "Expected", "Actual"] column_spacing = column_labels_and_spacing[0..-2].map(&:last) << 0 #=> [1, 2, 3, 0] a = h.map { |k,f| [table_label[k]].product(f.map { |field, g| [[field, g.values.map(&:to_s)].flatten] }) } # => [[["Table 1", [["Acct type", "ACC", "ACC"]]], # ["Table 1", [["Seq No", "100.0", "100"]]], # ["Table 1", [["Class", "AC", "AC"]]]], # [["Table 2", [["Date", "20140606", "20130606"]]]]]

要了解如何计算a，请考虑传递给块¹的第一个键值对h，该键被赋予块变量{{ 1}}和分配给k的值：

f

然后该块执行以下计算：

k = "Table1" f = { "Acct type"=>{ "Expected"=>"ACC", "Actual"=>"ACC" } }

现在让我们继续，计算b = f.map { |field, g| [[field, g.values.map(&:to_s)].flatten] } # => [["Acct type", "ACC", "ACC"]] c = ["Table 1"].product([[["Acct type", "ACC", "ACC"]]]) #=> [["Table 1", [["Acct type", "ACC", "ACC"]]]] d = c.flatten #=> ["Table 1", "Acct type", "ACC", "ACC"]。计算表格主体的行：

a

然后添加列标签：

e = a.flat_map { |e| e.map(&:flatten) } #=> [["Table 1", "Acct type", "ACC", "ACC"], # ["Table 1", "Seq No", "100.0", "100"], # ["Table 1", "Class", "AC", "AC"], # ["Table 2", "Date", "20140606", "20130606"]]

接下来，计算列宽：

rows = e.unshift(column_labels) #=> [["Table", "Field", "Expected", "Actual"], # ["Table 1", "Acct type", "ACC", "ACC"], # ["Table 1", "Seq No", "100.0", "100"], # ["Table 1", "Class", "AC", "AC"], # ["Table 2", "Date", "20140606", "20130606"]]

考虑打印表格的第二行（列标签后的第一行）：

f = rows.transpose #=> [["Table", "Table 1", "Table 1", "Table 1", "Table 2"], # ["Field", "Acct type", "Seq No", "Class", "Date"], # ["Expected", "ACC", "100.0", "AC", "20140606"], # ["Actual", "ACC", "100", "AC", "20130606"]] g = f.map { |c| c.max_by(&:size).size } #=> [7, 9, 8, 8] i = g.zip(column_spacing) # => [[7, 1], [9, 2], [8, 3], [8, 0]] column_widths = i.map { |a,b| a+b } #=> [8, 11, 11, 8] rows.each do |row| row.zip(column_widths).each { |s, width| print s.ljust(width) } puts end

打印第一个字段：

row = ["Table 1", "Acct type", "ACC", "ACC"] row.zip(column_widths) # => [["Table 1", 8], ["Acct type", 11], ["ACC", 11], ["ACC", 8]]

我打印冒号只是为了显示字段宽度。其他字段的印刷方式类似。

¹ Ruby 1.9+需要知道哪个键值对首先传递给块。

将Ruby中的嵌套哈希转换为表格格式

2 个答案: