Re: RFC Idea - is_json - looking for feedback

El sáb, 30 jul 2022 a las 8:30, Dusk (<dusk@woofle.net>) escribió: > On Jul 29, 2022, at 23:05, Aleksander Machniak <alec@alec.pl> wrote: > > json_decode() has $depth argument, I think is_json() probably also > should. And I'm not sure about JSON_INVALID_UTF8_IGNORE flag. > > > > My point is that if you use these with json_decode() you might also need > to use these with is_json() > > How about if validation were a flag to json_decode() -- perhaps > JSON_VALIDATE_ONLY -- which made it return a placeholder value like true > instead of the decoded structure? (It can't return null on success; that's > used for errors if JSON_THROW_ON_ERROR isn't set.) > -- > PHP Internals - PHP Runtime Development Mailing List > To unsubscribe, visit: https://www.php.net/unsub.php > > I thought about that too, and yes can also be a valid approach that would not interfeer with my development. What I personally dont like about that approach is that a developer is able to specify flags that are related to exclusively with result generation, and when they get mixed with JSON_VALIDATE_ONLY would not make sense. for example: json_decode($jsonStringHere, false, 512, JSON_BIGINT_AS_STRING | JSON_INVALID_UTF8_SUBSTITUTE | JSON_OBJECT_AS_ARRAY | JSON_THROW_ON_ERROR | JSON_VALIDATE_ONLY); So the end result, is not clear by reading the line itself, I mean , the mechanism (of having another constant in json_decode()) is not enough to make the function self-expressive.