DOC Performance improvements for changelog #333
Conversation
| - RelationList::foreignIDFilter() | ||
| - HasManyList::foreignIDFilter() - 108% faster (0.1584s vs 0.3304s) | ||
| - ManyManyList::foreignIDFilter() - 108% faster (0.1529s vs 0.3119s) | ||
| - ManyManyThroughList::foreignIDFilter() - 27% faster (0.6901s vs (0.8766s) |
There was a problem hiding this comment.
How did you test this? I'm guessing you didn't just call foreignIDFilter()?
There was a problem hiding this comment.
I've linked to the test setup in the description - silverstripe/silverstripe-framework#10904 (comment)
There was a problem hiding this comment.
Might be worth being explicit about the test, something like "This was tested by passing {x number} IDs into the forForeignID() method of each list class, then executing the query"
There was a problem hiding this comment.
Updated, I omitted the 'then executing the query' bit cos I'm "sharing" the text with byIDs() - however I don't think we need to be that pedantic here, I think the message of "made the method run faster" is clear
2adea21 to
f590481
Compare
|
As discussed offline, we'll keep the % faster language since it's more commonly used and easier to say and easiest to understand that it's a "good thing", though it's less "grammatically correct". Saying that query time has been reduced could be misinterpreted as being a "bad" thing since the word "reduced" is generally negative. Using the word "faster" make it more immediately obvious that it's positive. e.g. DataList::byIDs() - 198% faster - (0.0608s vs 0.1812s) vs DataList::byIDs() - query time has been reduced by 66% - (0.0608s vs 0.1812s) |
Issue #330
Performance measurements