Ruby 选择由不同子字符串覆盖的字符串

我想选择一组具有以下属性的子字符串所覆盖的字符串部分：

它们属于原始字符串。 它们可能有不同的长度和位置。 它们可以重叠。 它们的顺序可能与原始字符串中的顺序不同。 例如：

string = "MGLSDGEWQQVLNVWGKVEADIAGHGQEVLIHSKHPGDFGADAQGAMTKALELFRNDIAAKYKELGFQG"
substring1 = "HPGDFGADAQGAMTKALELFR"
substring2 = "GEWQQVLNVWGK"
substringn = "ALELFRNDIAAKYK"

我想得到：

coverage = "MGLSD<b>GEWQQVLNVWGK</b>VEADIAGHGQEVLIHSK<b>HPGDFGADAQGAMTKALELFRNDIAAKYK</b>ELGFQG"

---

这样，我就得到了每个子串的开始和结束位置。我如何将它们全部合并，尤其是考虑到它们可能重叠并以不同的顺序出现？这是一个好的策略吗？

这应该很管用，但它确实管用：

string = "MGLSDGEWQQVLNVWGKVEADIAGHGQEVLIHSKHPGDFGADAQGAMTKALELFRNDIAAKYKELGFQG"
substring1 = "HPGDFGADAQGAMTKALELFR"
substring2 = "GEWQQVLNVWGK"
substring3 = "ALELFRNDIAAKYK"

substrings = [substring1, substring2, substring3]

overlapping_indexes = substrings.map do |substring|
  start_pos = string.index substring
  end_pos = start_pos + substring.length
  (start_pos..end_pos)
end

# the following 3 methods are from Wayne Conrad in this question: http://stackoverflow.com/questions/6017523/how-to-combine-overlapping-time-ranges-time-ranges-union

def ranges_overlap?(a, b)
  a.include?(b.begin) || b.include?(a.begin)
end

def merge_ranges(a, b)
  [a.begin, b.begin].min..[a.end, b.end].max
end

def merge_overlapping_ranges(ranges)
  ranges.sort_by(&:begin).inject([]) do |ranges, range|
    if !ranges.empty? && ranges_overlap?(ranges.last, range)
      ranges[0...-1] + [merge_ranges(ranges.last, range)]
    else
      ranges + [range]
    end
  end
end

indexes = merge_overlapping_ranges(overlapping_indexes)

x = "<b>"
y = "</b>"
offset = 0

indexes.each do |index|
  string.insert(index.begin + offset, x)
  offset += x.length
  string.insert(index.end + offset, y)
  offset += y.length
end

p string

---

这看起来是一个很好的开始策略，尤其是当您将.each转换为.map并返回需要标记的开始/结束位置列表时。然后你的问题变成了如何合并这些范围，以便重叠的范围合并成单个更大的范围。我会说你已经完成了80%，所以你要做的是在开始插入和结束插入时，这项工作似乎完成了，你可以克隆字符串并修改克隆的字符串，然后返回克隆的字符串作为你的答案，假设new_string=string，那么在获得start_pos和end_pos之后，您希望在新的_字符串中插入标记，用于插入的ruby文档是这样的，这将不起作用。插入的标记将破坏原始字符串，避免与另一个与当前子字符串重叠的子字符串进一步匹配。@sawa-你说得对。我更新了我的答案。谢谢

string = "MGLSDGEWQQVLNVWGKVEADIAGHGQEVLIHSKHPGDFGADAQGAMTKALELFRNDIAAKYKELGFQG"
substring1 = "HPGDFGADAQGAMTKALELFR"
substring2 = "GEWQQVLNVWGK"
substring3 = "ALELFRNDIAAKYK"

substrings = [substring1, substring2, substring3]

overlapping_indexes = substrings.map do |substring|
  start_pos = string.index substring
  end_pos = start_pos + substring.length
  (start_pos..end_pos)
end

# the following 3 methods are from Wayne Conrad in this question: http://stackoverflow.com/questions/6017523/how-to-combine-overlapping-time-ranges-time-ranges-union

def ranges_overlap?(a, b)
  a.include?(b.begin) || b.include?(a.begin)
end

def merge_ranges(a, b)
  [a.begin, b.begin].min..[a.end, b.end].max
end

def merge_overlapping_ranges(ranges)
  ranges.sort_by(&:begin).inject([]) do |ranges, range|
    if !ranges.empty? && ranges_overlap?(ranges.last, range)
      ranges[0...-1] + [merge_ranges(ranges.last, range)]
    else
      ranges + [range]
    end
  end
end

indexes = merge_overlapping_ranges(overlapping_indexes)

x = "<b>"
y = "</b>"
offset = 0

indexes.each do |index|
  string.insert(index.begin + offset, x)
  offset += x.length
  string.insert(index.end + offset, y)
  offset += y.length
end

p string