About nextPutAll: $(
Locked
12
12
 
|
HI
while working on a lecture I saw that we have about 75 cases like: aStream nextPut: $(. .... aStream nextPut: $). storeOn: aStream "Store a description of the elements of the complement rather than self." aStream nextPut: $(. absent storeOn: aStream. aStream nextPut: $); space; nextPutAll: #complement. printElementsOn: aStream aStream nextPut: $(. self size > 100 ifTrue: [aStream nextPutAll: 'size '. self size printOn: aStream] ifFalse: [self keysSortedSafely do: [:key | aStream print: key; nextPutAll: '->'; print: (self at: key); space]]. aStream nextPut: $) we have surroundedBy: aString surroundedBy: aString ^ self class streamContents: [ :s| s nextPutAll: aString. s nextPutAll: self. s nextPutAll: aString ]. But only working on aString. To me it looks like it ia class for having it on stream. What do you think? |
|
|
> On 24 Jul 2015, at 22:48, stepharo <[hidden email]> wrote: > > HI > > while working on a lecture I saw that we have about 75 cases like: > > aStream nextPut: $(. > .... > aStream nextPut: $). > > > > storeOn: aStream > "Store a description of the elements of the complement rather than self." > > aStream nextPut: $(. > absent storeOn: aStream. > aStream nextPut: $); space; nextPutAll: #complement. > > > printElementsOn: aStream > aStream nextPut: $(. > self size > 100 > ifTrue: [aStream nextPutAll: 'size '. > self size printOn: aStream] > ifFalse: [self keysSortedSafely > do: [:key | aStream print: key; > nextPutAll: '->'; > print: (self at: key); > space]]. > aStream nextPut: $) > > we have surroundedBy: aString > > surroundedBy: aString > ^ self class streamContents: [ :s| > s nextPutAll: aString. > s nextPutAll: self. > s nextPutAll: aString ]. > > But only working on aString. > To me it looks like it ia class for having it on stream. > > What do you think? > |
|
|
In reply to this post by stepharo
In BioSmalltalk I use this pattern a lot: so yes, that would be a nice inclusion for me :)aStream nextPutAll: self name between: $[ -> $]. 2015-07-24 17:48 GMT-03:00 stepharo <[hidden email]>: HI |
|
|
(From my vacations) Doesn't this solves the problem? stream << $( << name << $) It is already there :) Now I comeback to watch the see...
|
Re: About nextPutAll: $(
|
|
> On 25 Jul 2015, at 19:40, Esteban Lorenzano <[hidden email]> wrote: > > (From my vacations) > > Doesn't this solves the problem? > > stream << $( << name << $) > > It is already there :) Yes, indeed: String streamContents: [ :out | out <<$( << 'test-' << 123 << $) ] (But it is not perfect, there is still confusion there) We are trying to reduce the String and Stream API, so any addition should be really important/significant. > Now I comeback to watch the see... > > > > > > On 25 Jul 2015, at 18:39, Hernán Morales Durand <[hidden email]> wrote: > >> In BioSmalltalk I use this pattern a lot: >> >> aStream >> nextPutAll: self name >> between: $[ -> $]. >> >> so yes, that would be a nice inclusion for me :) >> >> Cheers, >> >> Hernán >> >> >> >> >> 2015-07-24 17:48 GMT-03:00 stepharo <[hidden email]>: >> HI >> >> while working on a lecture I saw that we have about 75 cases like: >> >> aStream nextPut: $(. >> .... >> aStream nextPut: $). >> >> >> >> storeOn: aStream >> "Store a description of the elements of the complement rather than self." >> >> aStream nextPut: $(. >> absent storeOn: aStream. >> aStream nextPut: $); space; nextPutAll: #complement. >> >> >> printElementsOn: aStream >> aStream nextPut: $(. >> self size > 100 >> ifTrue: [aStream nextPutAll: 'size '. >> self size printOn: aStream] >> ifFalse: [self keysSortedSafely >> do: [:key | aStream print: key; >> nextPutAll: '->'; >> print: (self at: key); >> space]]. >> aStream nextPut: $) >> >> we have surroundedBy: aString >> >> surroundedBy: aString >> ^ self class streamContents: [ :s| >> s nextPutAll: aString. >> s nextPutAll: self. >> s nextPutAll: aString ]. >> >> But only working on aString. >> To me it looks like it ia class for having it on stream. >> >> What do you think? >> >> >> >> >> |
|
|
> On 25 Jul 2015, at 21:31, Sven Van Caekenberghe <[hidden email]> wrote: > > >> On 25 Jul 2015, at 19:40, Esteban Lorenzano <[hidden email]> wrote: >> >> (From my vacations) >> >> Doesn't this solves the problem? >> >> stream << $( << name << $) >> >> It is already there :) > > Yes, indeed: > > String streamContents: [ :out | > out <<$( << 'test-' << 123 << $) ] > > (But it is not perfect, there is still confusion there) I agree. I’ve never liked that operator. Apart from #// and #\\ there are pretty much no selectors that make you guess their meaning. The meaning is in all other cases transported through wording, which I feel very comfortable with. Yes, it’s convenient to write but, as Sven points out, I actually find it less readable. I like Hernan’s idea better. > > We are trying to reduce the String and Stream API, so any addition should be really important/significant. > >> Now I comeback to watch the see... >> >> >> >> >> >> On 25 Jul 2015, at 18:39, Hernán Morales Durand <[hidden email]> wrote: >> >>> In BioSmalltalk I use this pattern a lot: >>> >>> aStream >>> nextPutAll: self name >>> between: $[ -> $]. >>> >>> so yes, that would be a nice inclusion for me :) >>> >>> Cheers, >>> >>> Hernán >>> >>> >>> >>> >>> 2015-07-24 17:48 GMT-03:00 stepharo <[hidden email]>: >>> HI >>> >>> while working on a lecture I saw that we have about 75 cases like: >>> >>> aStream nextPut: $(. >>> .... >>> aStream nextPut: $). >>> >>> >>> >>> storeOn: aStream >>> "Store a description of the elements of the complement rather than self." >>> >>> aStream nextPut: $(. >>> absent storeOn: aStream. >>> aStream nextPut: $); space; nextPutAll: #complement. >>> >>> >>> printElementsOn: aStream >>> aStream nextPut: $(. >>> self size > 100 >>> ifTrue: [aStream nextPutAll: 'size '. >>> self size printOn: aStream] >>> ifFalse: [self keysSortedSafely >>> do: [:key | aStream print: key; >>> nextPutAll: '->'; >>> print: (self at: key); >>> space]]. >>> aStream nextPut: $) >>> >>> we have surroundedBy: aString >>> >>> surroundedBy: aString >>> ^ self class streamContents: [ :s| >>> s nextPutAll: aString. >>> s nextPutAll: self. >>> s nextPutAll: aString ]. >>> >>> But only working on aString. >>> To me it looks like it ia class for having it on stream. >>> >>> What do you think? >>> >>> >>> >>> >>> > > |
|
|
I too believe current #<< is perfectible, but is (IMO) a good trade off. When Camillo introduced it, he was following current conventions outside smalltalk world... And as I said, it is probably not perfect but I find it a good start.
I would not include a specific solution for a particular problem when there is a generic one that fits. Esteban, still on holidays > On 25 Jul 2015, at 22:02, Max Leske <[hidden email]> wrote: > > >> On 25 Jul 2015, at 21:31, Sven Van Caekenberghe <[hidden email]> wrote: >> >> >>> On 25 Jul 2015, at 19:40, Esteban Lorenzano <[hidden email]> wrote: >>> >>> (From my vacations) >>> >>> Doesn't this solves the problem? >>> >>> stream << $( << name << $) >>> >>> It is already there :) >> >> Yes, indeed: >> >> String streamContents: [ :out | >> out <<$( << 'test-' << 123 << $) ] >> >> (But it is not perfect, there is still confusion there) > > I agree. I’ve never liked that operator. Apart from #// and #\\ there are pretty much no selectors that make you guess their meaning. The meaning is in all other cases transported through wording, which I feel very comfortable with. Yes, it’s convenient to write but, as Sven points out, I actually find it less readable. I like Hernan’s idea better. > >> >> We are trying to reduce the String and Stream API, so any addition should be really important/significant. >> >>> Now I comeback to watch the see... >>> >>> >>> >>> >>> >>>> On 25 Jul 2015, at 18:39, Hernán Morales Durand <[hidden email]> wrote: >>>> >>>> In BioSmalltalk I use this pattern a lot: >>>> >>>> aStream >>>> nextPutAll: self name >>>> between: $[ -> $]. >>>> >>>> so yes, that would be a nice inclusion for me :) >>>> >>>> Cheers, >>>> >>>> Hernán >>>> >>>> >>>> >>>> >>>> 2015-07-24 17:48 GMT-03:00 stepharo <[hidden email]>: >>>> HI >>>> >>>> while working on a lecture I saw that we have about 75 cases like: >>>> >>>> aStream nextPut: $(. >>>> .... >>>> aStream nextPut: $). >>>> >>>> >>>> >>>> storeOn: aStream >>>> "Store a description of the elements of the complement rather than self." >>>> >>>> aStream nextPut: $(. >>>> absent storeOn: aStream. >>>> aStream nextPut: $); space; nextPutAll: #complement. >>>> >>>> >>>> printElementsOn: aStream >>>> aStream nextPut: $(. >>>> self size > 100 >>>> ifTrue: [aStream nextPutAll: 'size '. >>>> self size printOn: aStream] >>>> ifFalse: [self keysSortedSafely >>>> do: [:key | aStream print: key; >>>> nextPutAll: '->'; >>>> print: (self at: key); >>>> space]]. >>>> aStream nextPut: $) >>>> >>>> we have surroundedBy: aString >>>> >>>> surroundedBy: aString >>>> ^ self class streamContents: [ :s| >>>> s nextPutAll: aString. >>>> s nextPutAll: self. >>>> s nextPutAll: aString ]. >>>> >>>> But only working on aString. >>>> To me it looks like it ia class for having it on stream. >>>> >>>> What do you think? > > |
|
|
In reply to this post by Max Leske
I like the solution of hernan better. I think that surrounding is a
common pattern. Stef Le 25/7/15 22:02, Max Leske a écrit : >> On 25 Jul 2015, at 21:31, Sven Van Caekenberghe <[hidden email]> wrote: >> >> >>> On 25 Jul 2015, at 19:40, Esteban Lorenzano <[hidden email]> wrote: >>> >>> (From my vacations) >>> >>> Doesn't this solves the problem? >>> >>> stream << $( << name << $) >>> >>> It is already there :) >> Yes, indeed: >> >> String streamContents: [ :out | >> out <<$( << 'test-' << 123 << $) ] >> >> (But it is not perfect, there is still confusion there) > I agree. I’ve never liked that operator. Apart from #// and #\\ there are pretty much no selectors that make you guess their meaning. The meaning is in all other cases transported through wording, which I feel very comfortable with. Yes, it’s convenient to write but, as Sven points out, I actually find it less readable. I like Hernan’s idea better. > >> We are trying to reduce the String and Stream API, so any addition should be really important/significant. >> >>> Now I comeback to watch the see... >>> >>> >>> >>> >>> >>> On 25 Jul 2015, at 18:39, Hernán Morales Durand <[hidden email]> wrote: >>> >>>> In BioSmalltalk I use this pattern a lot: >>>> >>>> aStream >>>> nextPutAll: self name >>>> between: $[ -> $]. >>>> >>>> so yes, that would be a nice inclusion for me :) >>>> >>>> Cheers, >>>> >>>> Hernán >>>> >>>> >>>> >>>> >>>> 2015-07-24 17:48 GMT-03:00 stepharo <[hidden email]>: >>>> HI >>>> >>>> while working on a lecture I saw that we have about 75 cases like: >>>> >>>> aStream nextPut: $(. >>>> .... >>>> aStream nextPut: $). >>>> >>>> >>>> >>>> storeOn: aStream >>>> "Store a description of the elements of the complement rather than self." >>>> >>>> aStream nextPut: $(. >>>> absent storeOn: aStream. >>>> aStream nextPut: $); space; nextPutAll: #complement. >>>> >>>> >>>> printElementsOn: aStream >>>> aStream nextPut: $(. >>>> self size > 100 >>>> ifTrue: [aStream nextPutAll: 'size '. >>>> self size printOn: aStream] >>>> ifFalse: [self keysSortedSafely >>>> do: [:key | aStream print: key; >>>> nextPutAll: '->'; >>>> print: (self at: key); >>>> space]]. >>>> aStream nextPut: $) >>>> >>>> we have surroundedBy: aString >>>> >>>> surroundedBy: aString >>>> ^ self class streamContents: [ :s| >>>> s nextPutAll: aString. >>>> s nextPutAll: self. >>>> s nextPutAll: aString ]. >>>> >>>> But only working on aString. >>>> To me it looks like it ia class for having it on stream. >>>> >>>> What do you think? >>>> >>>> >>>> >>>> >>>> >> > > |
|
|
I also like more the Hernán proposal but I would prefer something like stream nextPutAll: self name between: $( and: $) To avoid creating an association On Jul 26, 2015 12:23 PM, "stepharo" <[hidden email]> wrote:
I like the solution of hernan better. I think that surrounding is a common pattern. |
|
|
In reply to this post by stepharo
Hi,
Well, If there wouldn't be already a way to do it, I wouldn't care which one to take. But this is not the case: we are discussing the introduction of a new vocabulary when there is already another who is fulfilling the place. In that sense, the proposed method solves a problem who is actually not there. If we do not like #<<, then we should remove it from system, but please, let's not add a more limited method when we have a more generic one who does the job... I repeat: we are not discussing the introduction of one or another method, but the introduction of one who overlaps another who is already there. Cheers, Esteban Ps: I will be back next thursday... I can discuss more then... > On 26 Jul 2015, at 17:22, stepharo <[hidden email]> wrote: > > I like the solution of hernan better. I think that surrounding is a common pattern. > > Stef > > Le 25/7/15 22:02, Max Leske a écrit : >>> On 25 Jul 2015, at 21:31, Sven Van Caekenberghe <[hidden email]> wrote: >>> >>> >>>> On 25 Jul 2015, at 19:40, Esteban Lorenzano <[hidden email]> wrote: >>>> >>>> (From my vacations) >>>> >>>> Doesn't this solves the problem? >>>> >>>> stream << $( << name << $) >>>> >>>> It is already there :) >>> Yes, indeed: >>> >>> String streamContents: [ :out | >>> out <<$( << 'test-' << 123 << $) ] >>> >>> (But it is not perfect, there is still confusion there) >> I agree. I’ve never liked that operator. Apart from #// and #\\ there are pretty much no selectors that make you guess their meaning. The meaning is in all other cases transported through wording, which I feel very comfortable with. Yes, it’s convenient to write but, as Sven points out, I actually find it less readable. I like Hernan’s idea better. >> >>> We are trying to reduce the String and Stream API, so any addition should be really important/significant. >>> >>>> Now I comeback to watch the see... >>>> >>>> >>>> >>>> >>>> >>>>> On 25 Jul 2015, at 18:39, Hernán Morales Durand <[hidden email]> wrote: >>>>> >>>>> In BioSmalltalk I use this pattern a lot: >>>>> >>>>> aStream >>>>> nextPutAll: self name >>>>> between: $[ -> $]. >>>>> >>>>> so yes, that would be a nice inclusion for me :) >>>>> >>>>> Cheers, >>>>> >>>>> Hernán >>>>> >>>>> >>>>> >>>>> >>>>> 2015-07-24 17:48 GMT-03:00 stepharo <[hidden email]>: >>>>> HI >>>>> >>>>> while working on a lecture I saw that we have about 75 cases like: >>>>> >>>>> aStream nextPut: $(. >>>>> .... >>>>> aStream nextPut: $). >>>>> >>>>> >>>>> >>>>> storeOn: aStream >>>>> "Store a description of the elements of the complement rather than self." >>>>> >>>>> aStream nextPut: $(. >>>>> absent storeOn: aStream. >>>>> aStream nextPut: $); space; nextPutAll: #complement. >>>>> >>>>> >>>>> printElementsOn: aStream >>>>> aStream nextPut: $(. >>>>> self size > 100 >>>>> ifTrue: [aStream nextPutAll: 'size '. >>>>> self size printOn: aStream] >>>>> ifFalse: [self keysSortedSafely >>>>> do: [:key | aStream print: key; >>>>> nextPutAll: '->'; >>>>> print: (self at: key); >>>>> space]]. >>>>> aStream nextPut: $) >>>>> >>>>> we have surroundedBy: aString >>>>> >>>>> surroundedBy: aString >>>>> ^ self class streamContents: [ :s| >>>>> s nextPutAll: aString. >>>>> s nextPutAll: self. >>>>> s nextPutAll: aString ]. >>>>> >>>>> But only working on aString. >>>>> To me it looks like it ia class for having it on stream. >>>>> >>>>> What do you think? > > |
Re: About nextPutAll: $(
|
|
+1
I am with Esteban, I see no real need to add this, since we try to minimise the Stream/String APIs. Also, I often add more that just a string between the parenthesis, I want to write to a stream myself, so I would need a block, and that would not be that nice. > On 26 Jul 2015, at 20:49, Esteban Lorenzano <[hidden email]> wrote: > > Hi, > > Well, If there wouldn't be already a way to do it, I wouldn't care which one to take. But this is not the case: we are discussing the introduction of a new vocabulary when there is already another who is fulfilling the place. > In that sense, the proposed method solves a problem who is actually not there. > If we do not like #<<, then we should remove it from system, but please, let's not add a more limited method when we have a more generic one who does the job... > > I repeat: we are not discussing the introduction of one or another method, but the introduction of one who overlaps another who is already there. > > Cheers, > Esteban > > Ps: I will be back next thursday... I can discuss more then... > > >> On 26 Jul 2015, at 17:22, stepharo <[hidden email]> wrote: >> >> I like the solution of hernan better. I think that surrounding is a common pattern. >> >> Stef >> >> Le 25/7/15 22:02, Max Leske a écrit : >>>> On 25 Jul 2015, at 21:31, Sven Van Caekenberghe <[hidden email]> wrote: >>>> >>>> >>>>> On 25 Jul 2015, at 19:40, Esteban Lorenzano <[hidden email]> wrote: >>>>> >>>>> (From my vacations) >>>>> >>>>> Doesn't this solves the problem? >>>>> >>>>> stream << $( << name << $) >>>>> >>>>> It is already there :) >>>> Yes, indeed: >>>> >>>> String streamContents: [ :out | >>>> out <<$( << 'test-' << 123 << $) ] >>>> >>>> (But it is not perfect, there is still confusion there) >>> I agree. I’ve never liked that operator. Apart from #// and #\\ there are pretty much no selectors that make you guess their meaning. The meaning is in all other cases transported through wording, which I feel very comfortable with. Yes, it’s convenient to write but, as Sven points out, I actually find it less readable. I like Hernan’s idea better. >>> >>>> We are trying to reduce the String and Stream API, so any addition should be really important/significant. >>>> >>>>> Now I comeback to watch the see... >>>>> >>>>> >>>>> >>>>> >>>>> >>>>>> On 25 Jul 2015, at 18:39, Hernán Morales Durand <[hidden email]> wrote: >>>>>> >>>>>> In BioSmalltalk I use this pattern a lot: >>>>>> >>>>>> aStream >>>>>> nextPutAll: self name >>>>>> between: $[ -> $]. >>>>>> >>>>>> so yes, that would be a nice inclusion for me :) >>>>>> >>>>>> Cheers, >>>>>> >>>>>> Hernán >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> 2015-07-24 17:48 GMT-03:00 stepharo <[hidden email]>: >>>>>> HI >>>>>> >>>>>> while working on a lecture I saw that we have about 75 cases like: >>>>>> >>>>>> aStream nextPut: $(. >>>>>> .... >>>>>> aStream nextPut: $). >>>>>> >>>>>> >>>>>> >>>>>> storeOn: aStream >>>>>> "Store a description of the elements of the complement rather than self." >>>>>> >>>>>> aStream nextPut: $(. >>>>>> absent storeOn: aStream. >>>>>> aStream nextPut: $); space; nextPutAll: #complement. >>>>>> >>>>>> >>>>>> printElementsOn: aStream >>>>>> aStream nextPut: $(. >>>>>> self size > 100 >>>>>> ifTrue: [aStream nextPutAll: 'size '. >>>>>> self size printOn: aStream] >>>>>> ifFalse: [self keysSortedSafely >>>>>> do: [:key | aStream print: key; >>>>>> nextPutAll: '->'; >>>>>> print: (self at: key); >>>>>> space]]. >>>>>> aStream nextPut: $) >>>>>> >>>>>> we have surroundedBy: aString >>>>>> >>>>>> surroundedBy: aString >>>>>> ^ self class streamContents: [ :s| >>>>>> s nextPutAll: aString. >>>>>> s nextPutAll: self. >>>>>> s nextPutAll: aString ]. >>>>>> >>>>>> But only working on aString. >>>>>> To me it looks like it ia class for having it on stream. >>>>>> >>>>>> What do you think? >> >> > |
|
|
In reply to this post by EstebanLM
2015-07-26 15:49 GMT-03:00 Esteban Lorenzano <[hidden email]>: HernánHi, << is horrible, it reminds me to std::cout in C++ and all those cryptic languages where you cannot write a line without checking the API reference manual. |
|
|
2015-07-27 8:26 GMT+02:00 Hernán Morales Durand <[hidden email]>:
I like "<<" because it reminds me to std::cout in C++ :)
|
|
|
In reply to this post by EstebanLM
Esteban
I'm confused. I know << and I use it. Now << is different to what I mentioned. The point is not to replace nextPutAll: by << I'm not that stupid. The point is the surrounding behavior (opening and closing). Sven of course there are more than a name so this is why I thought about stream surround: aBlock between: '( ' and: ' )' With << it means that we will have all the surrounding logic in all the code. Stef |
|
|
stream surround: aBlock between: '( ' and: ' )' This may be personal preference, but I do not like having more method arguments after a block argument... because if the block is longer it may not be visible... e.g. stream surround: [ :aStream | self size > 100 ifTrue: [ aStream nextPutAll: 'size '. self size printOn: aStream ] ifFalse: [ self keysSortedSafely do: [ :key | aStream print: key; nextPutAll: '->'; print: (self at: key); space ] ] ] between: $( and: $) vs between:and:surround: if you actually run this through formatter in Pharo, the between:and: is not even visible --- you have to scroll to see it. Of course if aBlock is a variable, than my argument falls. From visual perspective the current state actually isn't that bad because you immediately see where is beginning and where is end without even looking... with surround:between:and: not so much... Peter |
|
|
In reply to this post by stepharo
Hi. In most cases we always surround text by "()". Can we just provide method for this and not duplicate "between: '( ' and: ' )'" everywhere? Something like "stream>>printInBrackets:"2015-07-27 10:33 GMT+03:00 stepharo <[hidden email]>: Esteban |
|
|
In reply to this post by Nicolai Hess
2015-07-27 4:24 GMT-03:00 Nicolai Hess <[hidden email]>:
Maybe you're more comfortable in a C++ forum too ;) Seriously, I don't know, but whatever you choose please make it human inferrable from just by reading the selector name. Because << is arithmetic/bitwise operator for most coders in the world, and there are more developers to come and be newbies, than smalltalkers ever. Hernán |
|
|
In reply to this post by Peter Uhnak
Could be ok too.
Le 27/7/15 09:53, Peter Uhnák a écrit :
|
|
|
In reply to this post by Peter Uhnak
May be I raised a not important issue.
Stef Le 27/7/15 09:53, Peter Uhnák a écrit :
|
|
|
On 27 July 2015 at 12:23, stepharo <[hidden email]> wrote:
My impression too. :)
-- Best regards,
Igor Stasenko. |
«
Return to Pharo Smalltalk Developers
|
1 view|%1 views
| Free forum by Nabble | Edit this page |