修复 PinyinTokenFilter 排序标志未重置的问题

[Ryan19929]

修复 PinyinTokenFilter 排序标志未重置的问题

问题

使用 keyword + pinyin filter 组合时(包含pinyin filter即可)，第一次分词结果与之后执行的结果顺序不一致。

• 对 [GT40] 分词

  "ignore_pinyin_offset": "true",
  "keep_first_letter": "false",
  "keep_none_chinese_in_joined_full_pinyin": "false",
  "keep_none_chinese_together": "true",
  "keep_original": "true",
  "limit_first_letter_length": 16,
  "lowercase": "true",
  "remove_duplicated_term": "false",
  "type": "pinyin"

• 第一次检索

{
  "tokens": [
    {
      "token": "g",
      "start_offset": 0,
      "end_offset": 4,
      "type": "word",
      "position": 0
    },
    {
      "token": "gt40",
      "start_offset": 0,
      "end_offset": 4,
      "type": "word",
      "position": 0
    },
    {
      "token": "t",
      "start_offset": 0,
      "end_offset": 4,
      "type": "word",
      "position": 1
    },
    {
      "token": "40",
      "start_offset": 0,
      "end_offset": 4,
      "type": "word",
      "position": 2
    }
  ]
}

• 之后的测试

{
  "tokens": [
    {
      "token": "g",
      "start_offset": 0,
      "end_offset": 4,
      "type": "word",
      "position": 0
    },
    {
      "token": "t",
      "start_offset": 0,
      "end_offset": 4,
      "type": "word",
      "position": 1
    },
    {
      "token": "40",
      "start_offset": 0,
      "end_offset": 4,
      "type": "word",
      "position": 2
    },
    {
      "token": "gt40",
      "start_offset": 0,
      "end_offset": 4,
      "type": "word",
      "position": 2
    }
  ]
}

原因

PinyinTokenFilter.resetVariable() 中没有重置 processedSortCandidate。

修复

在 resetVariable() 中添加 this.processedSortCandidate = false;

[medcl]

@kin122 please help to review this PR, thanks :)