|
Is it possible to SEO-style normalise a URI in a MySQL stored function? I'm thinking of a function that converts "Lingeries sets" into "lingerie-sets", and an appropriate method of querying the database for reverse lookup. Normalising is easy on the command line: code:
code:
code:
MrMoo fucked around with this message at 20:05 on Jan 30, 2009 |
# ¿ Jan 30, 2009 18:37 |
|
|
# ¿ May 17, 2024 17:30 |
|
moostaffa posted:I currently have all this data in one table (No expensive queries will be needed on this, only lookups on the single index), would it give any performance boost to split this table into, say, 10 or 100 smaller tables? On high end databases no, it will make it worse. The database' job is to manage indexes and large tables. However I think only Oracle is good at this. Sybase used to be terrible unless things have changed.
|
# ¿ Feb 26, 2009 02:48 |
|
fletcher posted:This was with mysqld running on my localhost and /tmp/test_data.csv is also on my local machine, shouldn't the FILE permission cover that? It is covered in the docs: quote:Using LOCAL is a bit slower than letting the server access the files directly, because the contents of the file must be sent over the connection by the client to the server. On the other hand, you do not need the FILE privilege to load local files. And it has its own page on security issues.
|
# ¿ Dec 2, 2014 04:40 |
|
How does MariaDB manage variables? I'm seeing this unexpected case that a key is not being used when a specifying a local variable.SQL code:
MrMoo fucked around with this message at 20:03 on Dec 4, 2014 |
# ¿ Dec 4, 2014 18:55 |
|
Ugh, apparently character set hell. An outstanding bug, workaround is to force the same character set everywhere. http://stackoverflow.com/q/16752922/175849 The TokuDB storage engine is nice though, 100GB source data down to 10GB table data and only 2GB compressed. MrMoo fucked around with this message at 23:18 on Dec 5, 2014 |
# ¿ Dec 5, 2014 23:14 |
|
There is clearly something special with the backup process, find out what and how it as affecting the SELECT query. It is not normal, but you may have a bug in your version of MySQL too
|
# ¿ Dec 12, 2014 17:42 |
|
Well for starters you can clean up that query using the IN expression, secondly it looks like you simply want a ranking function but this is not standard SQL and thus depends upon your database.
|
# ¿ Feb 20, 2015 23:43 |
|
Can you update this SQL Fiddle with your intent?
|
# ¿ Feb 21, 2015 02:00 |
|
Is there a standard solution for supporting a dynamic query of form a OR b AND (c OR d) from some proprietary interface to SQL? It seems that some limits are required to stop forcing full table scans. Table along the lines of:code:
edit: It looks like using the MySQL family I can use the MATCH() ... AGAINST() function suite if I prepare all the codes into a FULLTEXT index per row (article_id above). Then that provides the full boolean mode used in text searches. edit 2: Is this only possible with a "search engine"? I'm looking at Sphinx and the problem appears to come down to performing boolean matching on what they term multi-valued attributes (MVAs). Sphinx appears cute as can integrate with MariaDB to be driven by SQL. http://stackoverflow.com/questions/13643114/how-do-i-search-for-mvas-with-sphinxse https://mariadb.com/kb/en/mariadb/sphinx-storage-engine/ MrMoo fucked around with this message at 13:58 on Jul 22, 2015 |
# ¿ Jul 22, 2015 02:02 |
|
I think Percona is using the functionality from TokuDB which supports a lot of online changes with background rebuilds. I certainly recommend looking at MariaDB + TokuDB.
|
# ¿ Jan 7, 2016 02:02 |
|
Couldn't you summon a super view over the set of tables? I would be surprised if Oracle does not have something nice for this configuration, the MySQL family have been slowly adding multi-partition and multi-table engines so a single query is replicated across each.
|
# ¿ Mar 7, 2016 17:01 |
|
I want to say w3m and a website but I would hope there is something native that works with ncurses and is less retarded.
|
# ¿ Mar 20, 2016 00:53 |
|
For data loading have a gander at MemSQL, it's something crazy like 1 million rows per second loading data. MariaDB + TokuDB has some good optimizations on empty tables too.
|
# ¿ May 28, 2016 21:19 |
|
Stupid question ahoy. If I have a simple soft delete query, i.e.SQL code:
SQL code:
SQL code:
Currently targeting SQLite, open to learn for other vendor dialects. The obvious answer is to not bother making the distinction, but it's nice to see if it is possible and practical. MrMoo fucked around with this message at 04:11 on Oct 7, 2023 |
# ¿ Oct 7, 2023 03:59 |
|
The problem appears to be more about reading a column value before an UPDATE is applied. With Postgres this is possible, but Sqlite likes to only return the updated value. https://stackoverflow.com/questions/7923237/return-pre-update-column-values-using-sql-only i.e. the following is fine in PGSQL: SQL code:
|
# ¿ Oct 7, 2023 04:45 |
|
Alas, sqlite doesn't like that for some reason, need to check why, https://www.sqlite.org/lang_update.htmlSQL code:
Raises "no such column: y.name" 🤷♀️ Doesn't seem to understand table alias in the RETURNING clause. There are three conditions:
I think that just leaves this: SQL code:
SQL code:
quote:The first prototype of the RETURNING clause returned values as they were generated. That approach used less memory, but it had other problems: MrMoo fucked around with this message at 18:28 on Oct 7, 2023 |
# ¿ Oct 7, 2023 17:50 |
|
It's in England though? https://sqlbits.com/news/sqlbits-2024-unveiled/
|
# ¿ Nov 15, 2023 02:46 |
|
|
# ¿ May 17, 2024 17:30 |
|
GigaPeon posted:SQL Prompt Flyway is pretty nice, although you have to do the work to make it powerful. That is an impressive choice of styling.
|
# ¿ Nov 21, 2023 18:47 |