Re: RFC Posted for str_begins and str_ends functions

On 2016-08-02 18:44, Yasuo Ohgaki wrote:

Hi David, On Tue, Aug 2, 2016 at 10:36 AM, David Rodrigues <david.proweb@gmail.com> wrote:

Sara Golemon wrote:

Feeling "meh" on it (neither for nor against), but I would consider consistency with other str*() functions by making case-insensitivity live in separate functions rather than as a parameter. e.g. str_begins(), str_ibegins(), str_ends(), end_iends()

I guess that "i" isn't appliable when it have slashes. In this case, functions should be: strbegins, stribegins, strends, striends. In all case, I think that is better a third parameter and keep underlined.

This is difficult issue. String function names are inconsistent currently. It is better to stick to CODING_STANDARDS naming convention for new function names. Therefore, new string functions are better to be named str_*() unless they are too strange. e.g. http://php.net/manual/en/function.str-replace.php http://php.net/manual/en/function.str-ireplace.php I would like to fix function name inconsistencies by having aliases in near future. https://wiki.php.net/rfc/consistent_function_names It might be okay to have "s" in function names, but if we want to be consistent, str_replace -> str_replaces str_ireplace -> str_ireplaces IMO, following names are better for consistency. str_begin str_ibegin str_end str_iend In addition, str_replace() has seach_value at first, so signature might be boolean str_begin(string $search_value, string $str, [boolean $case_sensitive = true]) boolean str_end(string $search_value, string $str, string $search_value [boolean $case_sensitive = true]) However, strstr() (and other str functions without "_". e.g. strpos/stripos/strrpos/strripos) has search_value as the 2nd parameter. If we follow this format, current signature is fine. It may be better sort out and fix consistency issues first, then add new functions. Otherwise, we may introduce more consistency issues. Regards, BTW, having "i" is more readable. str_ibegin("searchthis", $str); is more readable than str_begin("seachthis", $str, TRUE); as programmer does not have to know that's the TRUE means. It's small thing, but small things add up. -- Yasuo Ohgaki yohgaki@ohgaki.net

Everyone has raised important considerations. For me, the most important thing is maintaining consistency with the existing PHP string library. I do not want these functions to feel "tacked" on, as if they were haphazardly added to PHP. If these functions are added to the language, it should feel as if they have always been a part of the language (even if they haven't been). This consistency is important in order to ensure these functions ADD to PHP instead of just cluttering it up. Having separate functions for case sensitivity makes sense, that is much more consistent with the existing string library. I think the proposal should be amended to separate those two functionalities. I think like having an "s" at the end of the function names reads better, but omitting the "s" fits better with the existing function names and does not read bad. Therefore, I am in favor of dropping the "s". As far as str_begin vs strbegin, I think str_begin is more readable. Therefore, I think it would be better to implement: boolean str_begin(string $search_value, string $str) boolean str_ibegin(string $search_value, string $str) boolean str_end(string $search_value, string $str) boolean str_iend(string $search_value, string $str) This is much more consistent with the existing string library. Regards, Will