On سه‌شنبه ۱۸ اوت ۲۰۲۶، ۰۲:۳۳ mickmackusa <[email protected]> wrote:

> As a demonstration, this is new "top posted" text because the earlier
> posted conversation in the thread exists below the newly written reply
> text. [BAD]
>
>
> On Tue, 18 Aug 2026, 08:36 سپهر محمودی, <[email protected]> wrote:
>
>> Hi Larry,
>>
>> [...]
>>
>> Regards,
>>
>> Sepehr
>>
>
> This is new "bottom posted" text because the earlier posted conversation
> in the thread exists above the newly written reply text. [GOOD]
>
> In the PHP Internals mailing list, you are expected to only "bottom post".
> This rule is to allow easy human comprehension of the conversation. By
> always bottom posting, the discussion in the thread of emails always reads
> as oldest communication at the top and most recent at the bottom. Sepehr,
> many of your recent emails have violated this simple rule. Please
> permanently stop writing your replies with the historic content below your
> new reply.
>
> Also, I am not a voting member of this organization, but I can say that
> I've never had a professional project that required your proposed function
> nor do I recall ever encountering a Stack Overflow question which would
> benefit from your proposed function (and I have been a heavy curator of
> php&arrays tagged content for many years). That doesn't mean that your
> proposed function is worthless, I just haven't encountered a use case
> personally.
>
> If a PHP array needed a pagination-style search function, should perhaps
> the data structure be reconsidered? Is the data structure purpose-built or
> is the function overcoming a suboptimal data structure?
>
> mickmackusa
>

Hi mickmackusa,

Thank you for the feedback.

First, I sincerely apologize for the bottom-posting mistake.

I am new to the PHP Internals mailing list, and I’m still learning

the rules. I will make sure to follow bottom posting correctly from

now on.

Regarding the use case:

I agree that this function is not needed for small arrays, where

array_slice() is perfectly fine. However, for very large arrays

that are kept in memory (such as cached database results, log

files read line by line, streaming data chunks, or paginated

queues), array_slice() creates a full copy of the array. This copy

costs both time and memory, and it grows with the size of the

array.

array_search_range() avoids this copy entirely, because it only

searches within a range of the existing array without creating a

new one.

So while the use case may not be common on Stack Overflow, it

does exist in real systems that work with large in-memory arrays.

I’d be happy to hear more of your thoughts.

Regards,

Sepehr

>
>

Reply via email to