Aha, netušil jsem, že něco takového existuje. (Narazil jsem jen na 'INSERT OR IGNORE'.) Přepíšu a pošlu.

J.

Dne 26. 02. 19 v 7:08 Jiri Vlasak napsal(a):

On Mon, Feb 25, 2019 at 03:37:52PM +0100, Jaroslav Klapalek wrote:

...

In case that coffee machine loses connection during recording of a coffee, this coffee is added once again as a part of offlineQueue.

This patch resolves this issue by checking the database for the same records before adding the new one.

---

Vyřešení bugu z 18/02/2019. coffee_db.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/coffee_db.py b/coffee_db.py index 7e28f92..d21b36c 100644 --- a/coffee_db.py +++ b/coffee_db.py @@ -51,7 +51,11 @@ def add_coffee(uid, flavor, time=None): if time is None: c.execute("insert into coffees (id, flavor) values (?,?)", (uid,flavor)) else:

•    c.execute("insert into coffees (id, flavor, time) values (?,?,?)", (uid, flavor, time))
  

•    res = list(c.execute("select * from coffees where id = ? and flavor = ? and time = ?", (uid, flavor, time)))
  

•    if len(res) < 1:
  

•        c.execute("insert into coffees (id, flavor, time) values (?,?,?)", (uid, flavor, time))
  

Vyřešil bych to možná rovnou SQL dotazem viz https://stackoverflow.com/questions/19337029/insert-if-not-exists-statement-...

jiri

...

• close_db(conn)

def flavors():

2.7.4

---

Coffee mailing list Coffee@rtime.felk.cvut.cz https://rtime.felk.cvut.cz/mailman/listinfo/coffee

On Tue, Feb 26, 2019 at 01:58:42PM +0100, Jaroslav Klapalek wrote:

In case that coffee machine loses connection during recording of a coffee, this coffee is added once again as a part of offlineQueue.

This patch resolves this issue by using 'EXISTS' condition.

Podle návrhu od Jirky: Nová verze, která využívá pouze jeden dotaz.

Za me dobry.

jiri

coffee_db.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/coffee_db.py b/coffee_db.py index 7e28f92..84b5ca0 100644 --- a/coffee_db.py +++ b/coffee_db.py @@ -51,7 +51,12 @@ def add_coffee(uid, flavor, time=None): if time is None: c.execute("insert into coffees (id, flavor) values (?,?)", (uid,flavor)) else:

•    c.execute("insert into coffees (id, flavor, time) values (?,?,?)", (uid, flavor, time))
  

•    c.execute("""
  

•                insert into coffees (id, flavor, time)
  

•                select ?, ?, ?
  

•                where not exists(select * from coffees where id = ? and flavor = ? and time = ?)
  

•              """, (uid, flavor, time, uid, flavor, time))
  

• close_db(conn)

def flavors():

2.7.4

---

Coffee mailing list Coffee@rtime.felk.cvut.cz https://rtime.felk.cvut.cz/mailman/listinfo/coffee

Takže `insert or ignore` funguje.

Aktualizace databaze je třeba udělat před dump:

sqlite3 coffee.db .dump > dumped

Upravit v souboru `dumped`:

• time datetime default current_timestamp

• time datetime default current_timestamp,
• UNIQUE (id, flavor, time)

A vytvořit databázi znovu:

rm coffee.db sqlite3 coffee.db < dumped

Duplicitní záznamy se již nepřidají, ale zbytek jo.

jiri

On Wed, Feb 27, 2019 at 02:00:22PM +0100, Michal Sojka wrote:

On Tue, Feb 26 2019, Jaroslav Klapalek wrote:

...

In case that coffee machine loses connection during recording of a coffee, this coffee is added once again as a part of offlineQueue.

This patch resolves this issue by using 'EXISTS' condition.

Podle návrhu od Jirky: Nová verze, která využívá pouze jeden dotaz. coffee_db.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/coffee_db.py b/coffee_db.py index 7e28f92..84b5ca0 100644 --- a/coffee_db.py +++ b/coffee_db.py @@ -51,7 +51,12 @@ def add_coffee(uid, flavor, time=None): if time is None: c.execute("insert into coffees (id, flavor) values (?,?)", (uid,flavor)) else:

•    c.execute("insert into coffees (id, flavor, time) values (?,?,?)", (uid, flavor, time))
  

•    c.execute("""
  

•                insert into coffees (id, flavor, time)
  

•                select ?, ?, ?
  

•                where not exists(select * from coffees where id = ? and flavor = ? and time = ?)
  

•              """, (uid, flavor, time, uid, flavor, time))
  

Tohle by si ale zasloužilo aspoň komentář co to dělá. Stačilo by "insert new entry if it doesn't exit".

Ale myslím, že srozumitelnější by byla následující změna. Netestoval jsem to. Můžete to někdo zkusit?

diff --git a/coffee_db.py b/coffee_db.py index 7e28f92..c4845d9 100644 --- a/coffee_db.py +++ b/coffee_db.py @@ -51,7 +51,7 @@ def add_coffee(uid, flavor, time=None): if time is None: c.execute("insert into coffees (id, flavor) values (?,?)", (uid,flavor)) else:

•    c.execute("insert into coffees (id, flavor, time) values (?,?,?)", (uid, flavor, time))
  

•    c.execute("insert or ignore into coffees (id, flavor, time) values (?,?,?)", (uid, flavor, time))
  

  close_db(conn)

def flavors(): diff --git a/coffee_db.sql b/coffee_db.sql index d9f7d8e..380e229 100644 --- a/coffee_db.sql +++ b/coffee_db.sql @@ -20,7 +20,8 @@ create table if not exists coffees ( num integer primary key, id varchar(24) references users(id), -- id may be unknown flavor varchar(255) not null references flavors(name),

• time datetime default current_timestamp

• time datetime default current_timestamp,
• UNIQUE (id, flavor, datetime)

);

-M.

---

Coffee mailing list Coffee@rtime.felk.cvut.cz https://rtime.felk.cvut.cz/mailman/listinfo/coffee