Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-03-03 Thread Mesut Öncel
Does anyone have a different opinion or will this topic be closed? Mesut Öncel , 1 Mar 2021 Pzt, 13:27 tarihinde şunu yazdı: > Thank you for your detailed explanation. You are right, they have tried to > shape the database suitable for this structure for a long time, but > removing the

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-03-01 Thread Mesut Öncel
Thank you for your detailed explanation. You are right, they have tried to shape the database suitable for this structure for a long time, but removing the expression indexes will not cause a crisis. People and products using the database created by Django will already have to create a standard

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-03-01 Thread Hannes Ljungberg
I kind of agree that using `UPPER` instead of `ILIKE` for `__icontains` on PostgreSQL isn’t optimal. But it is quite easy to create a functional trigram GIN-index which use `UPPER` to allow these lookups to use an index. This will be even easier in Django 3.2 where you can create functional

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Mesut Öncel
Hi James, Thanks for your explanations. However, I wanted to explain the disadvantage of using "UPPER like" instead of "ilike" in icontains, istartswith and iendswith. The performance problem should not be ignored. James Bennett , 1 Mar 2021 Pzt, 04:05 tarihinde şunu yazdı: > On Sun, Feb 28,

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread James Bennett
On Sun, Feb 28, 2021 at 2:39 AM Tom Forbes wrote: > > Thank you for the clarification! I think the biggest argument for this change > is the fact that uppercasing Unicode can cause incorrect results to be > returned. > > Given that we now have much better support for custom index types, perhaps

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Mesut Öncel
The ticket I posted in the first email does not belong to me. The ticket I opened is https://code.djangoproject.com/ticket/32485. Sky Lord , 28 Şub 2021 Paz, 14:59 tarihinde şunu yazdı: > Guys, am noob anyone gonna guide? > > On Sun, Feb 28, 2021, 11:39 AM Tom Forbes wrote: > >> Thank you for

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Sky Lord
Guys, am noob anyone gonna guide? On Sun, Feb 28, 2021, 11:39 AM Tom Forbes wrote: > Thank you for the clarification! I think the biggest argument for this > change is the fact that uppercasing Unicode can cause incorrect results to > be returned. > > Given that we now have much better support

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Mesut Öncel
A discussion was requested from here to make changes on this issue. Do you think it would be appropriate if I open a pull request for this change? Mesut Öncel , 28 Şub 2021 Paz, 14:06 tarihinde şunu yazdı: > You're welcome. Yes, database administrators can create an index as > desired. However,

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Mesut Öncel
You're welcome. Yes, database administrators can create an index as desired. However, creating multiple cin indexes in a single column with different methods will cause performance issues. I don't think using expressions when using like and ilike is a correct use for query performance. It doesn't

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Tom Forbes
Thank you for the clarification! I think the biggest argument for this change is the fact that uppercasing Unicode can cause incorrect results to be returned. Given that we now have much better support for custom index types, perhaps we should change this? We need a custom expression index

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Tom Forbes
Thank you for the clarification! On Sun, 28 Feb 2021 at 09:23, Mesut Öncel wrote: > Yes, 7 years ago I reviewed the ticket for the updated code, but there are > problems there.https: //code.djangoproject.com/ticket/32485 The ticket I > opened has the necessary details. If you want more details,

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Mesut Öncel
Yes, 7 years ago I reviewed the ticket for the updated code, but there are problems there.https: //code.djangoproject.com/ticket/32485 The ticket I opened has the necessary details. If you want more details, I can make the necessary explanations. Tom Forbes , 28 Şub 2021 Paz, 12:36 tarihinde şunu

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Tom Forbes
Unless I’m missing something, the ticket you linked was last updated 7 years ago with two now-broken images. Which ticket did you open with the examples? Tom On Sun, 28 Feb 2021 at 08:31, Mesut Öncel wrote: > The reason I opened all the problems in a single ticket was to show the >

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-28 Thread Mesut Öncel
The reason I opened all the problems in a single ticket was to show the disadvantages of the "UPPER" function used in the "icontains" operator.I think that instead of using "UPPER (column) like" used in the Django ORM filter method, the "icontains" operator should use the "ilike" operator and this

Re: About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-27 Thread Tom Forbes
Hey, I think it would be best if you opened a new ticket explaining your issue and the performance problems you’ve found (especially around Turkish characters). I’m not entirely clear what you wish to be changed: you wish for an expression index to be created automatically? Tom On Sat, 27 Feb

About the ORM icontains operator's disadvantage on PostgreSQL performance and query results.

2021-02-27 Thread Mesut Öncel
Hi, I opened the ticket https://code.djangoproject.com/ticket/3575. In this ticket, I wanted to prove the effect of UPPER function used in icontains operator on indexes and query result. UPPER function causes performance problems regarding indexing in PostgreSQL. Another issue is that it does