RFR: JDK-8200436 - String::isBlank

[Jim Laskey]

“isBlank" was primarily chosen because isBlank has been the name used in Apache StringUtils for the last fifteen years. The fact that Excel and Kotlin use isBlank as well might have influenced the choice. Not many other languages have the isBlank concept, usually requiring several method calls to gather the same effect. Secondarily, isBlank was chosen because of brevity, in the same vein as isEmpty.

The primary use case I see for the predicate is to filter out “blank” lines in streams of strings; eliminate non-informational data. As a predicate, isBlank seems very apropos. The name reflects its purpose and not focused on how it is implemented.

Cheers,

— Jim


> On May 16, 2018, at 1:43 AM, Jeremy Manson <jeremymanson at google.com> wrote:
> 
> Blank can even mean U+2422, if it comes to that.
> 
> Jeremy
> 
> On Tue, May 15, 2018 at 12:58 PM Louis Wasserman <lowasser at google.com>
> wrote:
> 
>> Does this method offer significant benefits over the more general approach
>> of String.codePoints().allMatch(Character::isWhitespace)?
>> 
>> " there is a notion of all characters in > > isBlank that you do not have
>> with isWhitespace."
>> I'm not sure I follow.  I do tend to interpret "is this string whitespace?"
>> as implying that it consists only of whitespace, whereas "blank" could mean
>> "" or maybe even "___" but not obviously "   ".
>> 
>> On Tue, May 15, 2018 at 11:19 AM Jonathan Bluett-Duncan <
>> jbluettduncan at gmail.com> wrote:
>> 
>>> How about String.isWhitespaceOrEmpty? I think that accurately describes
>> the
>>> method.
>>> 
>>> Cheers,
>>> Jonathan
>>> 
>>> On 15 May 2018 at 19:15, Jeremy Manson <jeremymanson at google.com> wrote:
>>> 
>>>> Seems to me that consistency matters here: for Character to call it
>>>> whitespace and for String to call it blank is a little weird.
>>>> 
>>>> Is there a well-understood definition of "blank string", or did this
>> just
>>>> seem like a good name choice?
>>>> 
>>>> Jeremy
>>>> 
>>>> On Mon, May 14, 2018 at 11:14 PM Remi Forax <forax at univ-mlv.fr> wrote:
>>>> 
>>>>> Hi Louis,
>>>>> I prefer isBlank to isWhitespace, there is a notion of all characters
>>> in
>>>>> isBlank that you do not have with isWhitespace.
>>>>> 
>>>>> Rémi
>>>>> 
>>>>> ----- Mail original -----
>>>>>> De: "Louis Wasserman" <lowasser at google.com>
>>>>>> À: "Xueming Shen" <xueming.shen at oracle.com>
>>>>>> Cc: "core-libs-dev" <core-libs-dev at openjdk.java.net>
>>>>>> Envoyé: Lundi 14 Mai 2018 22:15:50
>>>>>> Objet: Re: RFR: JDK-8200436 - String::isBlank
>>>>> 
>>>>>> It's not clear to me that "isBlank" is a good name for this method.
>>>>>> "isWhitespace" might be more appropriate, perhaps.
>>>>>> 
>>>>>> On Mon, May 14, 2018 at 12:48 PM Xueming Shen <
>>> xueming.shen at oracle.com
>>>>> 
>>>>>> wrote:
>>>>>> 
>>>>>>> +1
>>>>>>> 
>>>>>>> On 5/14/18, 8:25 AM, Jim Laskey wrote:
>>>>>>>> New string instance method that returns true if the string is
>>> empty
>>>> or
>>>>>>> contains only white space, where white space is defined as any
>>>> codepoint
>>>>>>> returns true when passed to Character::isWhitespace.
>>>>>>>> 
>>>>>>>> webrev:
>> http://cr.openjdk.java.net/~jlaskey/8200436/webrev/index.
>>>> html
>>>>>>>> jbs: https://bugs.openjdk.java.net/browse/JDK-8200436
>>>>>>>> csr: https://bugs.openjdk.java.net/browse/JDK-8200437
>>>>>>>> 
>>>>>>> 
>>>>> 
>>>> 
>>> 
>>