Timeline for Using MATCH, AGAINST multiple times in MySQL fulltext search
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 15, 2019 at 15:38 | comment | added | Rick James | @SDeaV - My Plans do not preserve your formula exactly, but they have similar effect. | |
| Jan 15, 2019 at 9:13 | comment | added | sdeav |
And one last question @RickJames. I did something like Plan C. I can get relevance of two colums seperately. But what I can use in WHERE clause? Should I use OR or there is somethin that i can do?
|
|
| Jan 14, 2019 at 8:13 | vote | accept | sdeav | ||
| Jan 13, 2019 at 1:40 | comment | added | Rick James | @SDeaV - Indexes are automatically updated as you do updates/inserts/deletes. Just create the index and don't worry about it. | |
| Jan 13, 2019 at 1:39 | comment | added | Rick James | @SDeaV - The Search will be the toughest problem you need to solve. Put aside the concepts of brands, options, etc, and focus on "what will the user want to search on?" and "How can I keep that simple and efficient?" Putting all the likely names, tags, model numbers, etc, into a single table with a single FULLTEXT index is probably the best. | |
| Jan 12, 2019 at 20:01 | comment | added | sdeav |
I thought about joining all in one table. But there are two problems. 1) What if user selects specific brands or models?. I'll have to use brands ids in with OR. But i can add id of brand or model to 'new table' as brand_id. Then I can use this table without joining tables. 2) This is a biggest problem. Tables options, option_values and product_option_values. If user selects specific options values. Like screen_size = FHD or HD or just all available sizes. I think I'll have to use JOIN with OR. What do you think, @Rick James?
|
|
| Jan 12, 2019 at 18:03 | comment | added | sdeav |
thank you a lot. But I'm new in SQL queries. And to be fair I don't understand your Plan D. And little confused if I understand plan A. I like Plan B and Plan C. But it will look like denormalization of my database. Will it take much time in index'ing that table for every update? And I'm thinking of adding another synonyms column for each name column. It will be something like tags.
|
|
| Jan 11, 2019 at 23:41 | history | answered | Rick James | CC BY-SA 4.0 |