MongoDB 索引文本搜索仅适用于完全匹配

Posted 2023-03-16

【中文标题】MongoDB 索引文本搜索仅适用于完全匹配

【英文标题】：MongoDB indexed text search only works for exact match

【发布时间】：2021-05-19 06:55:43

【问题描述】：

我的字段“用户名”填充了数据。 这段代码没有给我任何结果：

history = db.history
history.create_index([('user_name', 'text')])
history.find('$text' : '$search' : 'a')

但是当我指定确切的名称时，它会起作用

history.find('$text' : '$search' : 'exact name')

这里是 'a' 搜索的 explain() 输出：

    "executionSuccess": true,
    "nReturned": 0,
    "executionTimeMillis": 0,
    "totalKeysExamined": 0,
    "totalDocsExamined": 0,
    "executionStages": 
        "stage": "TEXT",
        "nReturned": 0,
        "executionTimeMillisEstimate": 0,
        "works": 1,
        "advanced": 0,
        "needTime": 0,
        "needYield": 0,
        "saveState": 0,
        "restoreState": 0,
        "isEOF": 1,
        "indexPrefix": ,
        "indexName": "user_name_text",
        "parsedTextQuery":  "terms": [], "negatedTerms": [], "phrases": [], "negatedPhrases": [] ,
        "textIndexVersion": 3,
        "inputStage": 
            "stage": "TEXT_MATCH",
            "nReturned": 0,
            "executionTimeMillisEstimate": 0,
            "works": 0,
            "advanced": 0,
            "needTime": 0,
            "needYield": 0,
            "saveState": 0,
            "restoreState": 0,
            "isEOF": 1,
            "docsRejected": 0,
            "inputStage": 
                "stage": "FETCH",
                "nReturned": 0,
                "executionTimeMillisEstimate": 0,
                "works": 0,
                "advanced": 0,
                "needTime": 0,
                "needYield": 0,
                "saveState": 0,
                "restoreState": 0,
                "isEOF": 1,
                "docsExamined": 0,
                "alreadyHasObj": 0,
                "inputStage":  "stage": "OR", "nReturned": 0, "executionTimeMillisEstimate": 0, "works": 0, "advanced": 0, "needTime": 0, "needYield": 0, "saveState": 0, "restoreState": 0, "isEOF": 1, "dupsTested": 0, "dupsDropped": 0 
            
        
    ,
    "allPlansExecution": []

这里是用户名 ('akkcess') 完全匹配的 explain() 输出：

    "executionSuccess": true,
    "nReturned": 39,
    "executionTimeMillis": 1,
    "totalKeysExamined": 39,
    "totalDocsExamined": 39,
    "executionStages": 
        "stage": "TEXT",
        "nReturned": 39,
        "executionTimeMillisEstimate": 0,
        "works": 40,
        "advanced": 39,
        "needTime": 0,
        "needYield": 0,
        "saveState": 0,
        "restoreState": 0,
        "isEOF": 1,
        "indexPrefix": ,
        "indexName": "user_name_text",
        "parsedTextQuery":  "terms": ["akkcess"], "negatedTerms": [], "phrases": [], "negatedPhrases": [] ,
        "textIndexVersion": 3,
        "inputStage": 
            "stage": "TEXT_MATCH",
            "nReturned": 39,
            "executionTimeMillisEstimate": 0,
            "works": 40,
            "advanced": 39,
            "needTime": 0,
            "needYield": 0,
            "saveState": 0,
            "restoreState": 0,
            "isEOF": 1,
            "docsRejected": 0,
            "inputStage": 
                "stage": "FETCH",
                "nReturned": 39,
                "executionTimeMillisEstimate": 0,
                "works": 40,
                "advanced": 39,
                "needTime": 0,
                "needYield": 0,
                "saveState": 0,
                "restoreState": 0,
                "isEOF": 1,
                "docsExamined": 39,
                "alreadyHasObj": 0,
                "inputStage": 
                    "stage": "OR",
                    "nReturned": 39,
                    "executionTimeMillisEstimate": 0,
                    "works": 40,
                    "advanced": 39,
                    "needTime": 0,
                    "needYield": 0,
                    "saveState": 0,
                    "restoreState": 0,
                    "isEOF": 1,
                    "dupsTested": 39,
                    "dupsDropped": 0,
                    "inputStage": 
                        "stage": "IXSCAN",
                        "nReturned": 39,
                        "executionTimeMillisEstimate": 0,
                        "works": 40,
                        "advanced": 39,
                        "needTime": 0,
                        "needYield": 0,
                        "saveState": 0,
                        "restoreState": 0,
                        "isEOF": 1,
                        "keyPattern":  "_fts": "text", "_ftsx": 1 ,
                        "indexName": "user_name_text",
                        "isMultiKey": false,
                        "isUnique": false,
                        "isSparse": false,
                        "isPartial": false,
                        "indexVersion": 2,
                        "direction": "backward",
                        "indexBounds": ,
                        "keysExamined": 39,
                        "seeks": 1,
                        "dupsTested": 0,
                        "dupsDropped": 0
                    
                
            
        
    ,
    "allPlansExecution": []

你知道它为什么会这样吗？ 根据文档和教程，这应该可以工作。

【问题讨论】：

根据哪个文档？

【参考方案1】：

“a”几乎可以肯定是stop word。几乎所有自然语言文本都会包含它。因此，如果搜索它，您将获得结果集中的每一个文档。由于这不是很有用，文本搜索会从查询中删除像“a”这样的停用词。

另外，MongoDB 文本搜索确实包含精确匹配功能，但它需要引用您尚未完成的查询，因此您使用的是常规词干匹配，而不是您发布的精确匹配查询。

【讨论】：

我不想要完全匹配。我的数据库中有名为“berry”的用户。我寻找“berr”，它返回 0 个结果。但是当我搜索 berry 时，它可以工作。

“berr” 不是一个词，因此如果它不是源于“berry”，我不会感到惊讶。

那么我如何找到所有包含“berr”的用户名？

你可以试试正则表达式。 MongoDB 没有文本子字符串提取运算符。

我已经研究过正则表达式。我想也许索引会更有效。无论如何，谢谢你的帮助。