Sql Server 2016 and JSON
Download as PPTX, PDF1 like233 views
The document introduces JSON (JavaScript Object Notation) in the context of MS SQL Server, explaining its structure as name-value pairs and its supported data types. It highlights the JSON support in SQL Server starting from compatibility level 130, detailing various functions such as isjson, json_value, json_query, json_modify, and openjson, each with examples of usage. The document also emphasizes the importance of understanding JSON paths, validation, and additional resources for further learning.
![ISJSON
• https://msdn.microsoft.com/en-us/library/dn921896.aspx
• Returns 1 if the string contains valid JSON
• Examples:
• SELECT ISJSON('[1,2,3]')
• SELECT ISJSON('<tag>Not JSON</tag>')
• SELECT Id, Fragment FROM DocumentsWHERE ISJSON(Fragment) = 1](https://image.slidesharecdn.com/sqlserver2016andjsonfull-170620162105/85/Sql-Server-2016-and-JSON-10-320.jpg)
![JSON_QUERY
• https://msdn.microsoft.com/en-us/library/dn921884.aspx
• Extracts an object or an array from a JSON string
• JSON_VALUE returns scalar values; JSON_QUERY returns objects & arrays
• Examples:
• SELECT JSON_QUERY('{"a":[1,2,3]}', '$.a')
• SELECT JSON_QUERY('{"Me":{"Name":"Greg"}}', '$.Me')](https://image.slidesharecdn.com/sqlserver2016andjsonfull-170620162105/85/Sql-Server-2016-and-JSON-12-320.jpg)
![JSON_MODIFY
• https://msdn.microsoft.com/en-us/library/dn921892.aspx
• Updates the value of a property in a JSON string and returns the new JSON
• Benefit: performs some validation and can safely replaces values
• Examples:
• SELECT JSON_MODIFY('{"Name":"Greg"}', '$.Name', 'Gregory')
• SELECT JSON_MODIFY('{"a":[1,2]}', 'append $.a', 3)](https://image.slidesharecdn.com/sqlserver2016andjsonfull-170620162105/85/Sql-Server-2016-and-JSON-13-320.jpg)
![OPENJSON
• https://msdn.microsoft.com/en-us/library/dn921885.aspx
• Table-value function that parses JSON text and returns objects and properties as
rows and columns
• Requires COMPATIBILITY_LEVEL >= 130
• Supports defining a schema for JSON document
• Examples:
• SELECT * FROM OPENJSON('{"a":null,"b":1,"c":"2","d":[1,2],"e":{"f":6}}')](https://image.slidesharecdn.com/sqlserver2016andjsonfull-170620162105/85/Sql-Server-2016-and-JSON-14-320.jpg)
![Computed Columns
• Non-persisted data in the tables
• ProTip:You can build an index on a computed column
• This can be very useful for indexing into your JSON document
• Example:
ALTERTABLE [Table] ADD ColAS JSON_VALUE(JsonData, '$.Field')](https://image.slidesharecdn.com/sqlserver2016andjsonfull-170620162105/85/Sql-Server-2016-and-JSON-16-320.jpg)
Sql Server 2016 and JSON
- 1. Introducing JSON in MS SQL Server Greg McMurray Business Solutions Developer Western Electricity Coordinating Council
- 2. Western Electricity Coordinating Council • Non-profit to ensure the reliability of the western interconnection • Approved Regional Entity by Federal Energy Regulatory Commission • Create, monitor & enforce reliability standards • Publish various models and independent perspective • Covers 14 western states, 2 Canadian provinces and Baja Mexico • We are hiring! https://www.wecc.biz/careers
- 3. Greg McMurray • Worked in Aerospace, Branding & Marketing, Energy, Healthcare, Software • Currently Business Solutions Developer atWECC • Manager at Aritus Computer Services, L.L.C. for 18 years • Active in many local user groups • Find me online • @goyuix • https://www.linkedin.com/in/goyuix • https://stackoverflow.com/cv/goyuix - top 3% of users Hi! I am Greg – not Jason ↓
- 4. JSON? Who is Jason? • “Discovered” by Douglas Crockford in 2001 • “JSON is XML for young people” Scott Hanselman DEVIntersection EU 2015 • Sample JSON: {“User Group”:”Utah SQL Server Group”} ↑This is Scott - not Jason ↑
- 5. JavaScript Object Notation (JSON) • What is JSON? • Name-Value pairs • Only 7 values: string, number, object, array, true, false, null • Scalar values, objects and arrays - you can actually model a table or dataset quite nicely • Number type: Supports both integer and floating point, but how they are implemented is up to the parser. • It is usually UTF-8 encoded – recommended to use NVARCHAR Douglas Crockford – also not Jason↓
- 6. Want to Learn More About JSON? • JSON.org is a great quick reference and starting place • There are RFCs (4627 & 7159) and ECMA standards (404 & partly 262) as published standards for data exchange • Validation and schemas exist, not commonly used but are gaining in popularity
- 7. Some JSON Gotchas & Quirks • So what happens with dates or other complex types? • Usually a convention of using one of the available types • Date object might be represented by the ISO 8601 string • How you do know if it was a string or date? Do you care? • SQL Server JSON parser defaults to returning NVARCHAR typically • JSON paths default to lax mode; strict mode changes behaviors
- 8. JSON Support in SQL Server • Requires Compatibility Level 130 (SQL 2016) • Some functions available at lower levels, but still require Server 2016 engine • 120 might be the default even in new Azure SQL databases – you really need to check • Available Functions • ISJSON - more like IsValid JSON • JSON_VALUE - for extracting scalar values • JSON_MODIFY - like find / replace, but structured • JSON_QUERY - useful for preventing SQL server from double escaping • OPENJSON - treat JSON documents/objects/arrays as a table • FOR JSON AUTO / PATH - convert query results to JSON document - similar to FOR XML • Many functions can use / require a JSON path to identify data • ProTip:You can use OPENROWSET to read saved JSON files
- 9. JSON Paths • https://msdn.microsoft.com/en-us/library/mt577087.aspx • String used in when calling various JSON functions • Has a lax and strict mode; lax is the default • Specify strict by including it at the start of your path, e.g. ‘strict $.Name’ • Strict mode changes: • Generally raises errors on any possible parsing issues • Error out on case-sensitivity issues rather than returning NULL • Error out on missing keys for JSON_MODIFY, rather than adding them • There is also an append mode used to add values to a JSON array
- 10. ISJSON • https://msdn.microsoft.com/en-us/library/dn921896.aspx • Returns 1 if the string contains valid JSON • Examples: • SELECT ISJSON('[1,2,3]') • SELECT ISJSON('<tag>Not JSON</tag>') • SELECT Id, Fragment FROM DocumentsWHERE ISJSON(Fragment) = 1
- 11. JSON_VALUE • https://msdn.microsoft.com/en-us/library/dn921898.aspx • Extracts a scalar value from a JSON string • Examples: • SELECT JSON_VALUE('{"Group":"SLC SQL UG"}', '$.Group') • SELECT JSON_VALUE('{"Me":{"Name":"Greg"}}', '$.Me.Name')
- 12. JSON_QUERY • https://msdn.microsoft.com/en-us/library/dn921884.aspx • Extracts an object or an array from a JSON string • JSON_VALUE returns scalar values; JSON_QUERY returns objects & arrays • Examples: • SELECT JSON_QUERY('{"a":[1,2,3]}', '$.a') • SELECT JSON_QUERY('{"Me":{"Name":"Greg"}}', '$.Me')
- 13. JSON_MODIFY • https://msdn.microsoft.com/en-us/library/dn921892.aspx • Updates the value of a property in a JSON string and returns the new JSON • Benefit: performs some validation and can safely replaces values • Examples: • SELECT JSON_MODIFY('{"Name":"Greg"}', '$.Name', 'Gregory') • SELECT JSON_MODIFY('{"a":[1,2]}', 'append $.a', 3)
- 14. OPENJSON • https://msdn.microsoft.com/en-us/library/dn921885.aspx • Table-value function that parses JSON text and returns objects and properties as rows and columns • Requires COMPATIBILITY_LEVEL >= 130 • Supports defining a schema for JSON document • Examples: • SELECT * FROM OPENJSON('{"a":null,"b":1,"c":"2","d":[1,2],"e":{"f":6}}')
- 15. FOR JSON • https://msdn.microsoft.com/en-us/library/ms173812.aspx • Converts rows and columns to a JSON document • Uses the FOR JSON syntax (similar to XML) • Examples: • SELECT name,object_id from sys.tables FOR JSON AUTO
- 16. Computed Columns • Non-persisted data in the tables • ProTip:You can build an index on a computed column • This can be very useful for indexing into your JSON document • Example: ALTERTABLE [Table] ADD ColAS JSON_VALUE(JsonData, '$.Field')
- 17. Peeking Inside Pages –Table & Index • Trace flags are your friend again: DBCCTRACEON(3604); • DBCC IND('Database','dbo.Table',-1); • DBCC PAGE('Database',FileNum,PagePID,3) WITHTABLERESULTS • https://www.sqlskills.com/blogs/paul/inside-the-storage-engine-using- dbcc-page-and-dbcc-ind-to-find-out-if-page-splits-ever-roll-back/
- 18. Want to learn more? • BertWagner https://blog.bertwagner.com/ • SQL Server 2016 support for JSON https://msdn.microsoft.com/en-us/library/dn921897.aspx • JSON in SQL Server (4 Parts) https://blogs.technet.microsoft.com/dataplatforminsider/2016/01/05/json-in-sql- server-2016-part-1-of-4/ • JSON on MSDN Channel 9 https://channel9.msdn.com/Search?term=JSON
- 19. Questions and Discussion • Did I introduce everything you hoped to learn today? • Would you like any clarifications on something we covered? • Share with us: How are you going to apply this is your world? • Feel free to reach out to me online: • Twitter: @goyuix • LinkedIn: https://www.linkedin.com/in/goyuix