插入測試資料
db.inventory.insertMany( [ { item: "journal", status: "A", size: { h: 14, w: 21, uom: "cm" }, instock: [ { warehouse: "A", qty: 5 } ] }, { item: "notebook", status: "A", size: { h: 8.5, w: 11, uom: "in" }, instock: [ { warehouse: "C", qty: 5 } ] }, { item: "paper", status: "D", size: { h: 8.5, w: 11, uom: "in" }, instock: [ { warehouse: "A", qty: 60 } ] }, { item: "planner", status: "D", size: { h: 22.85, w: 30, uom: "cm" }, instock: [ { warehouse: "A", qty: 40 } ] }, { item: "postcard", status: "A", size: { h: 10, w: 15.25, uom: "cm" }, instock: [ { warehouse: "B", qty: 15 }, { warehouse: "C", qty: 35 } ] } ]);
後面的栗子都會用到這裡的測試資料
查詢到的文件會返回所有欄位
> db.inventory.find( { status: "A" } ) { "_id" : ObjectId("60b7177a67b3da741258754b"), "item" : "journal", "status" : "A", "size" : { "h" : 14, "w" : 21, "uom" : "cm" }, "instock" : [ { "warehouse" : "A", "qty" : 5 } ] } { "_id" : ObjectId("60b7177a67b3da741258754c"), "item" : "notebook", "status" : "A", "size" : { "h" : 8.5, "w" : 11, "uom" : "in" }, "instock" : [ { "warehouse" : "C", "qty" : 5 } ] } { "_id" : ObjectId("60b7177a67b3da741258754f"), "item" : "postcard", "status" : "A", "size" : { "h" : 10, "w" : 15.25, "uom" : "cm" }, "instock" : [ { "warehouse" : "B", "qty" : 15 }, { "warehouse" : "C", "qty" : 35 } ] }
預設情況下,會返回文件的所有欄位
等價 SQL 的寫法
SELECT * from inventory WHERE status = "A"
複習下 find() 的語法格式
db.collection.find(query, projection)
- query:可選項,設定查詢操作符指定查詢條件
- projection :可選項,指定要在與 query 匹配的文件中返回的欄位,如果忽略此選項則返回所有欄位【本節重點】
僅返回指定的欄位和 _id
> db.inventory.find( { status: "A" }, { item: 1, status: 1 } ) { "_id" : ObjectId("60b7177a67b3da741258754b"), "item" : "journal", "status" : "A" } { "_id" : ObjectId("60b7177a67b3da741258754c"), "item" : "notebook", "status" : "A" } { "_id" : ObjectId("60b7177a67b3da741258754f"), "item" : "postcard", "status" : "A" }
- 即使不指定 _id 也會自動返回
- 需要返回的欄位只需要欄位值寫 1 就行 { <field>: 1 }
等價 SQL 的寫法
SELECT _id, item, status from inventory WHERE status = "A"
返回所有欄位但排除指定欄位
> db.inventory.find( { status: "A" }, { status: 0, instock: 0 } ) { "_id" : ObjectId("60b7177a67b3da741258754b"), "item" : "journal", "size" : { "h" : 14, "w" : 21, "uom" : "cm" } } { "_id" : ObjectId("60b7177a67b3da741258754c"), "item" : "notebook", "size" : { "h" : 8.5, "w" : 11, "uom" : "in" } } { "_id" : ObjectId("60b7177a67b3da741258754f"), "item" : "postcard", "size" : { "h" : 10, "w" : 15.25, "uom" : "cm" } }
不需要返回的欄位只需要欄位值填 0 就行 { <field>: 0 }
如果 _id :0 會不返回 _id 欄位嗎?
> db.inventory.find( { status: "A" }, { status: 0, instock: 0,_id : 0 } ) { "item" : "journal", "size" : { "h" : 14, "w" : 21, "uom" : "cm" } } { "item" : "notebook", "size" : { "h" : 8.5, "w" : 11, "uom" : "in" } } { "item" : "postcard", "size" : { "h" : 10, "w" : 15.25, "uom" : "cm" } }
答案是會的
能不能同時指定包含一些欄位,不包含一些欄位呢?
> db.inventory.find( { status: "A" }, { status: 0, instock: 1} ) Error: error: { "ok" : 0, "errmsg" : "Cannot do inclusion on field instock in exclusion projection", "code" : 31253, "codeName" : "Location31253" }
答案是不能的,如果想指定 <projection> 是包含欄位,那所有欄位值都得統一是 1,相反如果是不包含,也必須都是 0
那 _id 欄位有這個限制嗎?
_id 欄位同時傳了 0 和 1
> db.inventory.find( { status: "A" }, { _id : 1,_id : 0 } ) { "item" : "journal", "status" : "A", "size" : { "h" : 14, "w" : 21, "uom" : "cm" }, "instock" : [ { "warehouse" : "A", "qty" : 5 } ] } { "item" : "notebook", "status" : "A", "size" : { "h" : 8.5, "w" : 11, "uom" : "in" }, "instock" : [ { "warehouse" : "C", "qty" : 5 } ] } { "item" : "postcard", "status" : "A", "size" : { "h" : 10, "w" : 15.25, "uom" : "cm" }, "instock" : [ { "warehouse" : "B", "qty" : 15 }, { "warehouse" : "C", "qty" : 35 } ] }
答案是沒有限制,但是會以最後一個欄位值為準,比如上面的就是文件僅僅不顯示 _id 欄位
> db.inventory.find( { status: "A" }, { _id : 0,_id : 1 } ) { "_id" : ObjectId("60b7177a67b3da741258754b") } { "_id" : ObjectId("60b7177a67b3da741258754c") } { "_id" : ObjectId("60b7177a67b3da741258754f") }
像這個就是文件僅僅顯示 _id 欄位
返回巢狀文件的指定欄位
> db.inventory.find( ... { status: "A" }, ... { item: 1, status: 1, "size.uom": 1 } ... ) { "_id" : ObjectId("60b7177a67b3da741258754b"), "item" : "journal", "status" : "A", "size" : { "uom" : "cm" } } { "_id" : ObjectId("60b7177a67b3da741258754c"), "item" : "notebook", "status" : "A", "size" : { "uom" : "in" } } { "_id" : ObjectId("60b7177a67b3da741258754f"), "item" : "postcard", "status" : "A", "size" : { "uom" : "cm" } }
- 查詢條件:status 等於 A
- 返回欄位:_id、item、status、size 巢狀文件的 uom 欄位
關於指定巢狀文件的欄位,4.4 新增的新寫法
> db.inventory.find( { status: "A" }, { item: 1, status: 1, size: {uom : 1 } } ) { "_id" : ObjectId("60b7177a67b3da741258754b"), "item" : "journal", "status" : "A", "size" : { "uom" : "cm" } } { "_id" : ObjectId("60b7177a67b3da741258754c"), "item" : "notebook", "status" : "A", "size" : { "uom" : "in" } } { "_id" : ObjectId("60b7177a67b3da741258754f"), "item" : "postcard", "status" : "A", "size" : { "uom" : "cm" } }
其實就是將 "size.uom": 1 替換成 size : { uom : 1 } ,兩種寫法哪種順手用哪種
返回文件陣列中的文件的指定欄位
instock 文件陣列中的文件,只返回 qty 欄位
> db.inventory.find( { status: "A" }, { item: 1, status: 1, "instock.qty": 1 } ) { "_id" : ObjectId("60b7177a67b3da741258754b"), "item" : "journal", "status" : "A", "instock" : [ { "qty" : 5 } ] } { "_id" : ObjectId("60b7177a67b3da741258754c"), "item" : "notebook", "status" : "A", "instock" : [ { "qty" : 5 } ] } { "_id" : ObjectId("60b7177a67b3da741258754f"), "item" : "postcard", "status" : "A", "instock" : [ { "qty" : 15 }, { "qty" : 35 } ] }