Concept: Utils\DateTime deprecation & replacement#82
Concept: Utils\DateTime deprecation & replacement#82Majkl578 wants to merge 2 commits intonette:masterfrom
Conversation
244b179 to
3356dc4
Compare
|
I definitely like the idea. 👍 |
|
Good job 👍 |
|
createMutable() vs createImmutable() makes users think… |
|
👍 For immutable only. The only good use-case for mutable DateTime is compatibility with old libraries which accepts |
Agreed, maybe it could be
Exactly. And I think that is very good reason for supporting it. As an example, whole Doctrine stack still does not natively support
Actually, I really thought it was in (I used it in this PR, didn't even realize), but it was reverted afterwards, because Derick was against it. :( php/php-src#1145 |
3356dc4 to
0165413
Compare
6c03b1f to
a4bddc8
Compare
2998062 to
f6303c6
Compare
46f5434 to
ecbbdfc
Compare
d830ff4 to
580569d
Compare
ba38669 to
bba41ff
Compare
f6984a9 to
eed59a3
Compare
|
What to do with this? Still doable for 3.0 where DateTime with inheritance sadly still exists, or should I just close it? |
8b993d6 to
98975bf
Compare
8aa61b9 to
fd48510
Compare
e2a373b to
a316b52
Compare
Hi,
this is an attempt to phase out
Nette\Utils\DateTimeand replace it by a new helper class.The current DateTime makes date & time manipulation easier since it enhances
\DateTime, yet it has some major flaws: mutability and inheritance.Why should we want the new helper class?
\DateTimeand immutable\DateTimeImmutablevariants, through\DateTimeInterface;\DateTime;This PR is a draft, not a final proposal yet. It consists of new tests for existing DateTIme class as well as the actual replacement with tests as well.
I'd like to hear some feedback -- whether is it really what we want (I hope so) and what & how could be eventually changed.
(This depends on #79 which bumps composer.json & Travis versions. Hence Travis fails.)