为sql server 增加 parseJSON 和 ToJSON 函数

在SqlServer中增加Json处理的方法

Sql Server 存储非结构话数据可以使用xml类型，使用xpath方式查询，以前写过一篇随笔:Sql Server xml 类型字段的增删改查

除了xml类型也可以使用文本类型（char、vchar等）存储json格式的数据，如何在sql语句中解析json数据，这里有一篇博客 [转]在SqlServer 中解析JSON数据，它的来源是Consuming JSON Strings in SQL Server

针对json解析需要一个自定义类型Hierarchy、一个表值函数parseJSON、一个标量值函数ToJSON。语句如下：

 /****** Object:  UserDefinedTableType [dbo].[Hierarchy]    Script Date: 2016/5/6 17:24:48 ******/
 CREATE TYPE [dbo].[Hierarchy] AS TABLE(
     [element_id] [INT] NOT NULL,
     [sequenceNo] [INT] NULL,
     [parent_ID] [INT] NULL,
     [Object_ID] [INT] NULL,
     [NAME] [NVARCHAR]() NULL,
     [StringValue] [NVARCHAR](MAX) NOT NULL,
     [ValueType] [VARCHAR]() NOT NULL,
     PRIMARY KEY CLUSTERED
 (
     [element_id] ASC
 )WITH (IGNORE_DUP_KEY = OFF)
 )
 GO

Hierarchy

CREATE FUNCTION [dbo].[parseJSON]( @JSON NVARCHAR(MAX))
RETURNS @hierarchy TABLE
  (
   element_id INT IDENTITY(, ) NOT NULL, /* internal surrogate primary key gives the order of parsing and the list order */
   sequenceNo [int] NULL, /* the place in the sequence for the element */
   parent_ID INT,/* if the element has a parent then it is in this column. The document is the ultimate parent, so you can get the structure from recursing from the document */
   Object_ID INT,/* each list or object has an object id. This ties all elements to a parent. Lists are treated as objects here */
   NAME NVARCHAR(),/* the name of the object */
   StringValue NVARCHAR(MAX) NOT NULL,/*the string representation of the value of the element. */
   ValueType VARCHAR() NOT null /* the declared type of the value represented as a string in StringValue*/
  )
AS
BEGIN
  DECLARE
    @FirstObject INT, --the index of the first open bracket found in the JSON string
    @OpenDelimiter INT,--the index of the next open bracket found in the JSON string
    @NextOpenDelimiter INT,--the index of subsequent open bracket found in the JSON string
    @NextCloseDelimiter INT,--the index of subsequent close bracket found in the JSON string
    @Type NVARCHAR(),--whether it denotes an object or an array
    @NextCloseDelimiterChar CHAR(),--either a '}' or a ']'
    @Contents NVARCHAR(MAX), --the unparsed contents of the bracketed expression
    @Start INT, --index of the start of the token that you are parsing
    @end INT,--index of the end of the token that you are parsing
    @param INT,--the parameter at the end of the next Object/Array token
    @EndOfName INT,--the index of the start of the parameter at end of Object/Array token
    @token NVARCHAR(),--either a string or object
    @value NVARCHAR(MAX), -- the value as a string
    @SequenceNo int, -- the sequence number within a list
    @name NVARCHAR(), --the name as a string
    @parent_ID INT,--the next parent ID to allocate
    @lenJSON INT,--the current length of the JSON String
    @characters NCHAR(),--used to convert hex to decimal
    @result BIGINT,--the value of the hex symbol being parsed
    @index SMALLINT,--used for parsing the hex value
    @Escape INT --the index of the next escape character

  DECLARE @Strings TABLE /* in this temporary table we keep all strings, even the names of the elements, since they are 'escaped' in a different way, and may contain, unescaped, brackets denoting objects or lists. These are replaced in the JSON string by tokens representing the string */
    (
     String_ID INT IDENTITY(, ),
     StringValue NVARCHAR(MAX)
    )
  SELECT--initialise the characters to convert hex to ascii
    @characters='0123456789abcdefghijklmnopqrstuvwxyz',
    @SequenceNo=, --set the sequence no. to something sensible.
  /* firstly we process all strings. This is done because [{} and ] aren't escaped in strings, which complicates an iterative parse. */
    @parent_ID=;
  WHILE = --forever until there is nothing more to do
    BEGIN
      SELECT
        @start=PATINDEX('%[^a-zA-Z]["]%', @json collate SQL_Latin1_General_CP850_Bin);--next delimited string
      IF @start= BREAK --no more so drop through the WHILE loop
      IF SUBSTRING(@json, @start+, )='"'
        BEGIN --Delimited Name
          SET @start=@Start+;
          SET @end=PATINDEX('%[^\]["]%', RIGHT(@json, LEN(@json+'|')-@start) collate SQL_Latin1_General_CP850_Bin);
        END
      IF @end= --no end delimiter to last string
        BREAK --no more
      SELECT @token=SUBSTRING(@json, @start+, @end-)
      --now put in the escaped control characters
      SELECT @token=REPLACE(@token, FROMString, TOString)
      FROM
        (SELECT
          '\"' AS FromString, '"' AS ToString
         UNION ALL SELECT '\\', '\'
         UNION ALL SELECT '\/', '/'
         UNION ALL SELECT '\b', CHAR()
         UNION ALL SELECT '\f', CHAR()
         UNION ALL SELECT '\n', CHAR()
         UNION ALL SELECT '\r', CHAR()
         UNION ALL SELECT '\t', CHAR()
        ) substitutions
      SELECT @result=, @escape=
  --Begin to take out any hex escape codes
      WHILE @escape>
        BEGIN
          SELECT @index=,
          --find the next hex escape sequence
          @escape=PATINDEX('%\x[0-9a-f][0-9a-f][0-9a-f][0-9a-f]%', @token collate SQL_Latin1_General_CP850_Bin)
          IF @escape> --if there is one
            BEGIN
              WHILE @index< --there are always four digits to a \x sequence
                BEGIN
                  SELECT --determine its value
                    @result=@result+POWER(, @index)
                    *(CHARINDEX(SUBSTRING(@token, @escape++-@index, ),
                                @characters)-), @index=@index+ ;

                END
                -- and replace the hex sequence by its unicode value
              SELECT @token=STUFF(@token, @escape, , NCHAR(@result))
            END
        END
      --now store the string away
      INSERT INTO @Strings (StringValue) SELECT @token
      -- and replace the string with a token
      SELECT @JSON=STUFF(@json, @start, @end+,
                    '@string'+CONVERT(NVARCHAR(), @@identity))
    END
  -- all strings are now removed. Now we find the first leaf.
  WHILE =  --forever until there is nothing more to do
  BEGIN

  SELECT @parent_ID=@parent_ID+
  --find the first object or list by looking for the open bracket
  SELECT @FirstObject=PATINDEX('%[{[[]%', @json collate SQL_Latin1_General_CP850_Bin)--object or array
  IF @FirstObject =  BREAK
  IF (SUBSTRING(@json, @FirstObject, )='{')
    SELECT @NextCloseDelimiterChar='}', @type='object'
  ELSE
    SELECT @NextCloseDelimiterChar=']', @type='array'
  SELECT @OpenDelimiter=@firstObject

  WHILE = --find the innermost object or list...
    BEGIN
      SELECT
        @lenJSON=LEN(@JSON+'|')-
  --find the matching close-delimiter proceeding after the open-delimiter
      SELECT
        @NextCloseDelimiter=CHARINDEX(@NextCloseDelimiterChar, @json,
                                      @OpenDelimiter+)
  --is there an intervening open-delimiter of either type
      SELECT @NextOpenDelimiter=PATINDEX('%[{[[]%',
             RIGHT(@json, @lenJSON-@OpenDelimiter)COLLATE SQL_Latin1_General_CP850_Bin)--object
      IF @NextOpenDelimiter=
        BREAK
      SELECT @NextOpenDelimiter=@NextOpenDelimiter+@OpenDelimiter
      IF @NextCloseDelimiter<@NextOpenDelimiter
        BREAK
      IF SUBSTRING(@json, @NextOpenDelimiter, )='{'
        SELECT @NextCloseDelimiterChar='}', @type='object'
      ELSE
        SELECT @NextCloseDelimiterChar=']', @type='array'
      SELECT @OpenDelimiter=@NextOpenDelimiter
    END
  ---and parse out the list or name/value pairs
  SELECT
    @contents=SUBSTRING(@json, @OpenDelimiter+,
                        @NextCloseDelimiter-@OpenDelimiter-)
  SELECT
    @JSON=STUFF(@json, @OpenDelimiter,
                @NextCloseDelimiter-@OpenDelimiter+,
                '@'+@type+CONVERT(NVARCHAR(), @parent_ID))
  WHILE (PATINDEX('%[A-Za-z0-9@+.e]%', @contents COLLATE SQL_Latin1_General_CP850_Bin))<>
    BEGIN
      IF @Type='Object' --it will be a -n list containing a string followed by a string, number,boolean, or null
        BEGIN
          SELECT
            @SequenceNo=,@end=CHARINDEX(':', ' '+@contents)--if there is anything, it will be a string-based name.
          SELECT  @start=PATINDEX('%[^A-Za-z@][@]%', ' '+@contents COLLATE SQL_Latin1_General_CP850_Bin)--AAAAAAAA
          SELECT @token=SUBSTRING(' '+@contents, @start+, @End-@Start-),
            @endofname=PATINDEX('%[0-9]%', @token COLLATE SQL_Latin1_General_CP850_Bin),
            @param=RIGHT(@token, LEN(@token)-@endofname+)
          SELECT
            @token=LEFT(@token, @endofname-),
            @Contents=RIGHT(' '+@contents, LEN(' '+@contents+'|')-@end-)
          SELECT  @name=stringvalue FROM @strings
            WHERE string_id=@param --fetch the name
        END
      ELSE
        SELECT @Name=NULL,@SequenceNo=@SequenceNo+
      SELECT
        @end=CHARINDEX(',', @contents)-- a string-token, object-token, list-token, number,boolean, or null
      IF @end=
        SELECT  @end=PATINDEX('%[A-Za-z0-9@+.e][^A-Za-z0-9@+.e]%', @Contents+' ' COLLATE SQL_Latin1_General_CP850_Bin)
          +
       SELECT
        @start=PATINDEX('%[^A-Za-z0-9@+.e][A-Za-z0-9@+.e]%', ' '+@contents COLLATE SQL_Latin1_General_CP850_Bin)
      --select @start,@end, LEN(@contents+'|'), @contents
      SELECT
        @Value=RTRIM(SUBSTRING(@contents, @start, @End-@Start)),
        @Contents=RIGHT(@contents+' ', LEN(@contents+'|')-@end)
      IF SUBSTRING(@value, , )='@object'
        INSERT INTO @hierarchy
          (NAME, SequenceNo, parent_ID, StringValue, Object_ID, ValueType)
          SELECT @name, @SequenceNo, @parent_ID, SUBSTRING(@value, , ),
            SUBSTRING(@value, , ), 'object'
      ELSE
        IF SUBSTRING(@value, , )='@array'
          INSERT INTO @hierarchy
            (NAME, SequenceNo, parent_ID, StringValue, Object_ID, ValueType)
            SELECT @name, @SequenceNo, @parent_ID, SUBSTRING(@value, , ),
              SUBSTRING(@value, , ), 'array'
        ELSE
          IF SUBSTRING(@value, , )='@string'
            INSERT INTO @hierarchy
              (NAME, SequenceNo, parent_ID, StringValue, ValueType)
              SELECT @name, @SequenceNo, @parent_ID, stringvalue, 'string'
              FROM @strings
              WHERE string_id=SUBSTRING(@value, , )
          ELSE
            IF @value IN ('true', 'false')
              INSERT INTO @hierarchy
                (NAME, SequenceNo, parent_ID, StringValue, ValueType)
                SELECT @name, @SequenceNo, @parent_ID, @value, 'boolean'
            ELSE
              IF @value='null'
                INSERT INTO @hierarchy
                  (NAME, SequenceNo, parent_ID, StringValue, ValueType)
                  SELECT @name, @SequenceNo, @parent_ID, @value, 'null'
              ELSE
                IF PATINDEX('%[^0-9]%', @value COLLATE SQL_Latin1_General_CP850_Bin)>
                  INSERT INTO @hierarchy
                    (NAME, SequenceNo, parent_ID, StringValue, ValueType)
                    SELECT @name, @SequenceNo, @parent_ID, @value, 'real'
                ELSE
                  INSERT INTO @hierarchy
                    (NAME, SequenceNo, parent_ID, StringValue, ValueType)
                    SELECT @name, @SequenceNo, @parent_ID, @value, 'int'
      IF @Contents=' ' SELECT @SequenceNo=
    END
  END
INSERT INTO @hierarchy (NAME, SequenceNo, parent_ID, StringValue, Object_ID, ValueType)
  SELECT '-',, NULL, '', @parent_id-, @type
--
   RETURN
END

parseJSON

 /****** Object:  UserDefinedFunction [dbo].[ToJSON]    Script Date: 2016/5/6 17:25:49 ******/
 SET ANSI_NULLS ON
 GO

 SET QUOTED_IDENTIFIER ON
 GO

 CREATE FUNCTION [dbo].[ToJSON]
 (
       @Hierarchy Hierarchy READONLY
 )

 /*
 the function that takes a Hierarchy table and converts it to a JSON string

 Author: Phil Factor
 Revision: 1.5
 date: 1 May 2014
 why: Added a fix to add a name for a list.
 example:

 Declare @XMLSample XML
 Select @XMLSample='
   <glossary><title>example glossary</title>
   <GlossDiv><title>S</title>
    <GlossList>
     <GlossEntry ID="SGML" SortAs="SGML">
      <GlossTerm>Standard Generalized Markup Language</GlossTerm>
      <Acronym>SGML</Acronym>
      <Abbrev>ISO 8879:1986</Abbrev>
      <GlossDef>
       <para>A meta-markup language, used to create markup languages such as DocBook.</para>
       <GlossSeeAlso OtherTerm="GML" />
       <GlossSeeAlso OtherTerm="XML" />
      </GlossDef>
      <GlossSee OtherTerm="markup" />
     </GlossEntry>
    </GlossList>
   </GlossDiv>
  </glossary>'

 DECLARE @MyHierarchy Hierarchy -- to pass the hierarchy table around
 insert into @MyHierarchy select * from dbo.ParseXML(@XMLSample)
 SELECT dbo.ToJSON(@MyHierarchy)

        */
 RETURNS NVARCHAR(MAX)--JSON documents are always unicode.
 AS
 BEGIN
   DECLARE
     @JSON NVARCHAR(MAX),
     @NewJSON NVARCHAR(MAX),
     @Where INT,
     @ANumber INT,
     @notNumber INT,
     @indent INT,
     @ii INT,
     @CrLf CHAR()--just a simple utility to save typing!

   --firstly get the root token into place
   SELECT @CrLf=CHAR()+CHAR(),--just CHAR() in UNIX
          @JSON = CASE ValueType WHEN 'array' THEN
          +COALESCE('{'+@CrLf+'  "'+NAME+'" : ','')+'['
          ELSE '{' END
             +@CrLf
             + CASE WHEN ValueType='array' AND NAME IS NOT NULL THEN '  ' ELSE '' END
             + '@Object'+CONVERT(VARCHAR(),OBJECT_ID)
             +@CrLf+CASE ValueType WHEN 'array' THEN
             CASE WHEN NAME IS NULL THEN ']' ELSE '  ]'+@CrLf+'}'+@CrLf END
                 ELSE '}' END
   FROM @Hierarchy
     WHERE parent_id IS NULL AND valueType IN ('object','document','array') --get the root element
 /* now we simply iterat from the root token growing each branch and leaf in each iteration. This won't be enormously quick, but it is simple to do. All values, or name/value pairs withing a structure can be created in one SQL Statement*/
   SELECT @ii=
   WHILE @ii>
     BEGIN
     SELECT @where= PATINDEX('%[^[a-zA-Z0-9]@Object%',@json)--find NEXT token
     IF @where= BREAK
     /* this is slightly painful. we get the indent of the object we've found by looking backwards up the string */
     SET @indent=CHARINDEX(CHAR()+CHAR(),REVERSE(LEFT(@json,@where))+CHAR()+CHAR())-
     SET @NotNumber= PATINDEX('%[^0-9]%', RIGHT(@json,LEN(@JSON+'|')-@Where-)+' ')--find NEXT token
     SET @NewJSON=NULL --this contains the structure in its JSON form
     SELECT
         @NewJSON=COALESCE(@NewJSON+','+@CrLf+SPACE(@indent),'')
         +CASE WHEN parent.ValueType='array' THEN '' ELSE COALESCE('"'+TheRow.NAME+'" : ','') END
         +CASE TheRow.valuetype
         WHEN 'array' THEN '  ['+@CrLf+SPACE(@indent+)
            +'@Object'+CONVERT(VARCHAR(),TheRow.[OBJECT_ID])+@CrLf+SPACE(@indent+)+']'
         WHEN 'object' THEN '  {'+@CrLf+SPACE(@indent+)
            +'@Object'+CONVERT(VARCHAR(),TheRow.[OBJECT_ID])+@CrLf+SPACE(@indent+)+'}'
         WHEN 'string' THEN '"'+dbo.JSONEscaped(TheRow.StringValue)+'"'
         ELSE TheRow.StringValue
        END
      FROM @Hierarchy TheRow
      INNER JOIN @hierarchy Parent
      ON parent.element_ID=TheRow.parent_ID
       WHERE TheRow.parent_id= SUBSTRING(@JSON,@where+, @Notnumber-)
      /* basically, we just lookup the structure based on the ID that is appended to the @Object token. Simple eh? */
     --now we replace the token with the structure, maybe with more tokens in it.
     SELECT @JSON=STUFF (@JSON, @where+, +@NotNumber-, @NewJSON),@ii=@ii-
     END
   RETURN @JSON
 END

 GO

toJson

在Sql查询中使用parseJSON和toJson方法

Sql Server xml 类型字段的增删改查 文章中介绍了几种用法，我这里非常简单。有一张表 Candidate_Ext，CVParseJson 字段存储了json格式数据，其中有个节点是SkillName存储了技能列表，这里使用sql语句把技能名称查出来逗号分隔。

    "SkillList": [
        {
            "SkillId": ,
            "CandidateId": ,
            "TenantId": ,
            "SkillName": "Oracle",
            "SkillLevel": ,
            "SkillLevelName": "熟练",
            "SkillLevelName1": null
        },
        {
            "SkillId": ,
            "CandidateId": ,
            "TenantId": ,
            "SkillName": "TCP/IP",
            "SkillLevel": ,
            "SkillLevelName": "精通",
            "SkillLevelName1": null
        }
    ],

sql语句：

SELECT TOP  l.CandidateId ,l.CVParseJson,
        ( SELECT    StringValue+','
          FROM      parseJSON(l.CVParseJson) json
          WHERE  json.NAME='SkillName'
        FOR
          XML PATH('')
        ) 专业技能
FROM    dbo.Candidate_Ext l
WHERE   CVParseJson LIKE '{"candi%'

结果：

CandidateId    专业技能
              Oracle,TCP/IP,        

如果多个节点都有SkillName 这个属性处理起来就比较麻烦了，性能也不见得好，所以小数据出来使用这个方法还是比较方便的。