当我从具有多关系的表中选择一个特定字段时,我得到这样的异常:org.postgresql.util.PSQLException: ERROR: missing FROM-clause entry for table "bar"
以下是我定义的关系:
(declare foo bar)
(korma/defentity foo
(korma/pk :token)
(korma/database db)
(korma/table :foo)
(korma/has-many bar {:fk :token}))
(korma/defentity bar
(korma/pk :token)
(korma/database db)
(korma/table :bar)
(korma/belongs-to foo {:fk :token}))
这是我在sqlkorma中生成的查询:
(defn fetch []
(->
(korma/select* foo)
(korma/with bar)
(korma/fields :foo.token :bar.col)
(korma/exec)))
如果我使用 with *
选择所有字段,则查询运行没有问题。但是,我想指定要选择哪些字段,我做错了什么?
我放弃了尝试与 Korma 一起做这件事。 我认为使用基本的SQL互操作可能会更好:
(ns tst.demo.jdbc
(:use demo.core tupelo.core tupelo.test)
(:require
[clojure.java.jdbc :as jdbc]
[clojure.string :as str]
[clojure.java.io :as io] ))
(def db {:classname "org.h2.Driver"
:subprotocol "h2:mem"
;:subname "./korma.db"
:subname "demo;DB_CLOSE_DELAY=-1" ; -1 very important!!!
; http://www.h2database.com/html/features.html#in_memory_databases
; http://makble.com/using-h2-in-memory-database-in-clojure
:user "sa"
:password ""
})
(dotest
(spyx (jdbc/db-do-commands db ["drop table if exists langs"]))
(spyx (jdbc/db-do-commands db ["drop table if exists releases"]))
(spy :create
(jdbc/db-do-commands
db
(jdbc/create-table-ddl :langs
[[:id :serial]
[:lang "varchar not null"]])))
(spy :create
(jdbc/db-do-commands
db
(jdbc/create-table-ddl :releases
[[:id :serial]
[:desc "varchar not null"]
[:langId "numeric"]])))
(spy :insert
(jdbc/insert-multi! db :langs
[{:lang "Clojure"}
{:lang "Java"}]))
;; -> ({:lang "Clojure", :id 1} {:lang "Java", :id 2})
(spyx-pretty (jdbc/query db ["select * from langs"]))
(let [clj-id (grab :id (only (jdbc/query db ["select id from langs where lang='Clojure'"])))]
(spyx clj-id)
(spy :insert-rel
(jdbc/insert-multi! db :releases
[{:desc "ancients" :langId clj-id}
{:desc "1.8" :langId clj-id}
{:desc "1.9" :langId clj-id}])) )
(let [java-id (grab :id (only (jdbc/query db ["select id from langs where lang='Java'"])))]
(spyx java-id)
(spy :insert-rel
(jdbc/insert-multi! db :releases
[{:desc "dusty" :langId java-id}
{:desc "8" :langId java-id}
{:desc "9" :langId java-id}
{:desc "10" :langId java-id}])) )
(spyx-pretty
(jdbc/query db [
"select langs.lang, releases.desc
from langs inner join releases
on (langs.id = releases.langId)
where lang = 'Clojure' "]) )
)
结果:
(jdbc/db-do-commands db ["drop table if exists langs"]) => (0)
(jdbc/db-do-commands db ["drop table if exists releases"]) => (0)
:create => (0)
:create => (0)
:insert => ({:id 1} {:id 2})
(jdbc/query db ["select * from langs"]) =>
({:id 1, :lang "Clojure"} {:id 2, :lang "Java"})
clj-id => 1
:insert-rel => ({:id 1} {:id 2} {:id 3})
java-id => 2
:insert-rel => ({:id 4} {:id 5} {:id 6} {:id 7})
(jdbc/query db ["select langs.lang, releases.descn from langs inner join releasesn on (langs.id = releases.langId)n where lang = 'Clojure' "]) =>
({:lang "Clojure", :desc "ancients"}
{:lang "Clojure", :desc "1.8"}
{:lang "Clojure", :desc "1.9"})
想想,我在错误的地方通过了bar.col
字段。我想因为这是一个多对多关系,sql korma 必须离开并在这里运行 n+1 个查询。这就解释了为什么我看到一个missing FROM-clause
异常,因为我告诉sql korma在错误的查询中获取字段。
(defn fetch []
(->
(korma/select* foo)
(korma/fields :token)
(korma/with bar (korma/fields :col))
(korma/exec)))