Some Database Questions

[SkullCollector]

UTStatsDB Version : 3.10 with OL Script
Mysql / MarayaDB 10.4.x
PHP 8.2.12
Apache/2.4.58
DEbian 12



First of all, I think it's great that UTStatsDB is back up and running and Panther you are doing a great Job. Thank you for this nice peace of Candy.

What I'm wondering now is why ADODB isn't being used as the database layer. It supports all the database types that the script also supports.

And the script is even faster with it. I've already partially redesigned the regular user area with it. Page loading has become even faster as a result.

It also protects against SQL injections and offers encrypted database connections.
And you don't need to close every Database with sql_close(Link) or sql_free(result) and so on. Mysql (that i'm prefer) makes it with AdoDB together
Another question about the database: Why aren't indexes used in every table and also used in queries? That would speed up page loading even more.

Currently, my slowest page is the homepage. But I think that has something to do with the server query. Because if I don't query the server, the page loads lightning fast.

Best Regards
Skully

[Panther]

I've mostly focussed on MySQL/MariaDB support, though certainly ADODB would have made supporting multiple database systems easier. It does, however, have its drawbacks. I've continued working on fixing some issues and improving performance. Some tables do not require any indexes because the entire table is read into memory such as during log processing. The specialtypes table is a good example of this. How many servers are you querying from the homepage? In the demo site, I have three UT servers being querried with a small delay:
https://utstatsdb.com/demo/

[Panther]

I should add, I've lately worked on PHP compatibility and fixing various bugs I've found. I currently have the system running smoothly on PHP 8.5.

[SkullCollector]

I've mostly focussed on MySQL/MariaDB support, though certainly ADODB would have made supporting multiple database systems easier. It does, however, have its drawbacks. I've continued working on fixing some issues and improving performance. Some tables do not require any indexes because the entire table is read into memory such as during log processing. The specialtypes table is a good example of this. How many servers are you querying from the homepage? In the demo site, I have three UT servers being querried with a small delay:
https://utstatsdb.com/demo/

Good Morning
yes i see that . But the Table Structure is MyIsam. InnoDB is much better for many OPerations. Rollbacks and something more.
The tmp Table throws sometimes a table full error if you process many logs. (was testing it with 60+ logfiles) the script crashes.

imho AdoDB is the best Choice to get all of the Databases work with one Script. Setting it up is very easy and it works fine. MySQL is the most used Database i think.MSSQL *is not a Database it's a MS Joke i think )))*

To answer your question : I query 1 Server on the Homepage. The Server runs on a Debian Linux Machine .

hmm, sometimes my site shows no grafics for frags . Got an empty space on the site. If i dont show the image i see the png as cryptic signs.
and yes gd Libary is loaded. Because on the original Script size it works well. But thats my problem i think.

So all in all, i would like to see the script with AdoDB and InnoDB Structure. *hehe*

I wait for the next release to see whats going on. Hopefully some more features like extracting the Images directly out of the Mapfile.

Regards
Skully

[SkullCollector]

By the way : Is there any chance to get a Beta Version of UTStatsDB from the next Version ?
i can test it on my Server. I have more than 200 Logs to test
Regards
Skully

[Panther]

Currently my work in progress has a couple of bugs to fix, but I'll try to get to that soon. I haven't had much time for working on it these days, but I was just looking into them the other day and corrected some of them. As for MyISAM vs InnoDB, MyISAM is significantly faster (at least for reads), InnoDB just offers transactioions for recovery and integrity. I prefer Aria myself though, which is available with MariaDB.

Extracting images from the map...do they actually contain an embedded image? Though, even if they do, it seems like excessive extra work for the web server to do, but perhaps something that automatically creates the images from the map files.

[SkullCollector]

Yes i know MyIsam is faster, but in Case of Databases Crashes an other stupid things that can happen, i prefer Security.

as far as i know and remember, in Maps can be Images .
but i will have a look in the map files and see.

[SkullCollector]

Yes i know MyIsam is faster, but in Case of Databases Crashes an other stupid things that can happen, i prefer Security.

as far as i know and remember, in Maps can be Images .
but i will have a look in the map files and see.

I found a syntax for extracting a image from Unreal Maps :

Code: Select all

@echo off if not exist temp_img mkdir temp_img if not exist img mkdir img for %%F in (maps/*.ut2) do ( title Procees %%~nF... system\ucc batchexport %%F Texture bmp ..\temp_img\ )

as far as i know it was for the Older Unrealgames. So just adjust .ut2 to the old unreal map extension to pull the Images out of the maps.

Some UT2004 Maps has Images integrated. IMHO because they converted from older UT Games.

[Panther]

That's not practical to do on the fly by the web server, so better to get your map collection together and stick them all in mapimages.

[SkullCollector]

That's not practical to do on the fly by the web server, so better to get your map collection together and stick them all in mapimages.

Yeah i know. But the most maps that i have on my Server are not Vanilla Maps and not all have made a Screenshot.

I have some Images , but not all. Anyway i have to put more work into finding map Images.

[Panther]

Many of the map images for some of the game types included in UTStatsDB are from me playing a map and taking a screenshot.