Age | Commit message (Collapse) | Author | Lines |
|
|
|
|
|
The first { $_->raw_id eq $id } @{ $self->{results} } calls effectively
handle the results as a linked list, while in both cases only a single
element (with a unique ID, too) is looked up. A hashmap makes this about
30% faster (580ms total -> 490ms total)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Note that this does NOT include stations outside of germany, even if they
have Deutsche Bahn trains. This may be fixed in another commit.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* fixes undef warnings in recent DateTime::Format::Strptime versions
* Improves parser performance quite a bit
|
|
|
|
|
|
|
|
|
|
draft.
|
|
|
|
|