On Sat, Dec 15, 2018, 04:20 Richard O'Keefe <raoknz@gmail.com wrote:

Starting from the top:\

(0) It's not *wrong* to start a selector with a capital letter, but

�� it is certainly unusual.

(1) Intention-revealing names are good.�� That's why this name is bad.

�� It does not *delete* anything.

(2) The comment again says "delete" but nothing is deleted.�� Comments

�� that mislead are worse than no comments.

(3) An issue I have seen over and over in student code (in other

�� programming languages) is writing regular expressions as string

�� literals where they are used, so that every time you try a match

�� the regular expression gets recompiled, over and over again.

�� You're not exactly doing that, but you *are* rebuilding patterns

�� over and over again.�� It would make more sense to hold a

�� forbiddenPatterns collection.

(4) You are saying "(coll reject: [..test..]) notEmpty", building an

�� entire collection just to ask whether it has any members.�� What

�� you want to know is "are there any elements of coll that do not

�� satisfy the test", which can be expressed as

�� (coll allSatisfy: [..test..]) not.�� (Sadly there is no

�� #notAllSatisfy:, although #notAnySatisfy: *does* exist under the

�� name #noneSatisfy:.

(5) #match: is a highly risky way to look for a substring.�� What if

�� you want to look for a literal # or * ?�� What if you need to

�� look for 'Boring' but not 'boring'?�� ("Boring" is a perfectly

�� good surname.)

(6) You really don't need the "result" variable.

�� allowedWords

�� "Answer those words which do not contain any forbiddenWord

�� as a substring."

�� ^words select: [:each |

�� forbiddenWords noneSatisfy: [:forbidden |

�� <<each includes forbidden as a substring>>]]

(7) You can find out how to do the <<part>> by looking at the methods

�� for String.�� You will find several candidate methods.

(8) Of course, if there are W words and F forbidden words, in the

�� worst case you'll do W*F substring inclusion tests.�� This is not

�� good.�� "Multiple pattern matching" is a much-studied problem,

�� see for example Wu and Manber's paper

�� http://web.archive.org/web/20130203031412/http://www.inf.fu-berlin.de/inst/ag-bio-expired/FILES/ROOT/Teaching/Lectures/WS0506/WeiterfThemenBioinf/papers/FastAlgorithmMultiPatternSearching-WuManbers-TR94.pdf

�� There is more recent work also.

�� Does anyone know whether there's a multiple pattern algorithm for

�� Pharo?�� One could always fall back on ternary search trees...
On Fri, 14 Dec 2018 at 05:13, Roelof Wobben <r.wobben@home.nl> wrote:
Hello, 

I have this code :  

��FindAndDeleteWordsWithForbiddenParts
    "deletes all the words that contain forbidden parts"

    | result |
    result := words
        select: [ :word | 
            [ (forbiddenWords reject: [ :forbidden | '*' , forbidden , '*' match: word ]) isNotEmpty ]
                on: NotFound
                do: [ false ] ].
    ^ result


but I see warnings that I have to use a stream instead of string concentation.

Anyone hints how to do so ? 

Roelof