rebol [ title: "Engroup" file: %engroup.r version: 0.2 date: 12-4-2009 author: "Christian Ensel" purpose: "Mezzanine function to arrange values into groups of all equal values." comment: { What's with the superfluous WORD! argument in the simple envocation? Having to use a word here is really pointless, compare >> engroup i [1 1 2 2 2 3 4 4 4 4 5 6 7 7] == [[1 1] [2 2 2] [3] [4 4 4 4] [5] [6] [7 7]] vs. >> engroup [1 1 2 2 2 3 4 4 4 4 5 6 7 7] == [[1 1] [2 2 2] [3] [4 4 4 4] [5] [6] [7 7]] For now, I'm going with the WORD! argument, anyway. } usage: { Simple usage of ENGROUP arranges values into groups of values all equal: >> engroup i [1 1 2 2 2 3 4 4 4 4 5 6 7 7] == [[1 1] [2 2 2] [3] [4 4 4 4] [5] [6] [7 7]] The /OVER refinement specifies over wich value such groups are build. This is particular useful when dealing with objects. Compare the different grouping in the following examples: >> engroup i [#1 #11 #12 #2 #31 #32 #4 #41 #411 #412 #42 #43] == [[#1] [#11] [#12] [#2] [#31] [#32] [#4] [#41] [#411] [#412] [#42] [#43]] >> engroup/over i [#1 #11 #12 #2 #31 #32 #4 #41 #411 #412 #42 #43] [first i] == [[#1 #11 #12] [#2] [#31 #32] [#4 #41 #411 #412 #42 #43]] By default, ENGROUP ignores include NONE values. But beware, NONE values by design still have an group separating effect: >> engroup i reduce [1 1 2 none 2 3 3] == [[1 1] [2] [2] [3 3]] If by /OVER-ing NONE values are introduced, the by default will be ignored, too: >> engroup/over i [1 1 2 2 2 2 3 4 4 4 5 5 5] [if odd? i [i]] == [[1 1] [3] [5 5 5]] Use /ANY to include NONE values: >> engroup/any i reduce [1 1 2 2 2 3 none none none none 5 6 7 7] == [[1 1] [2 2 2] [3] [none none none none] [5] [6] [7 7]] Use /ONLY to group block values as blocks: >> engroup i [[1] [1] [2] [2]] == [[1 1] [2 2]] >> engroup/only i [[1] [1] [2] [2]] == [[[1] [1]] [[2] [2]]] For convenience, the /AS refinement allows to modify the value "in place". >> engroup/as/over i [#1 #1.1 #1.2 #2 #3.1 #3.2 #4 #4.1 #4.1.1 #4.1.2 #4.2 #4.3] [join "" [<item> i </item>]] [first i] == [["<item>1</item>" "<item>1.1</item>" "<item>1.2</item>"] ["<item>2</item>"] ["<item>3.1</item>" "<item>3.2</item>"] ["<item>4</... But beware, different input values that are equal after modification still will belong to different groups (grouping is applied after /OVER but before /AS, so to say): >> engroup/as i [1 2 3] [max i 4] == [[4] [4] [4]] >> engroup/over i [1 2 3 4] [round/to i 2] == [[1 2] [3 4]] >> engroup/over/as i [1 2 3 4] [round/to i 2] [round/to i 2] == [[2 2] [4 4]] Use /ALL to include NONE values introduced by /AS modifications (usually these are not part of the resulting groups): >> engroup/as i reduce [1 2 2 3 3 3 4 4 4 4] [all [even? i i]] == [[2 2] [4 4 4 4]] >> engroup/all/as i reduce [1 2 2 3 3 3 4 4 4 4] [all [even? i i]] == [[none] [2 2] [none none none] [4 4 4 4]] } library: [ level: 'intermediate Platform: 'all type: [function idiom] code: 'function domain: 'dialects license: 'public-domain support: none see-also: none tested-under: [view 2.7.6.3.1 on [WinXP] "CHE"] ] ] engroup: func [ "Arranges values into groups of all equal values." 'word [word!] "Variable to hold current value" data [series!] "The series to traverse" /over input [block!] "Value to compare" /as output [block!] "Modify values before grouping" /any "Build groups over NONE values, too." /all "Don't exclude NONE values from groups." /only "Group block values as blocks." /into collector "Where to collect results" /local any* all* do-input value do-output group groups val args prev ][ any*: any any: get in system/words 'any all*: all all: get in system/words 'all do-input: func reduce [[throw] word] any [input reduce [word]] do-output: func reduce [[throw] word] any [output reduce [word]] groups: any [all [into collector] copy []] until [ group: copy [] args: 0 prev: none foreach input data [ value: do-input input either all [args > 0 previous <> value] [break] [ args: args + 1 previous: value output: do-output input if all [ any [any* value] any [all* output] ][ either only [append/only group output] [append group output] ] ] ] unless empty? group [append/only groups group] tail? data: skip data args ] groups ]