diff --git a/packages/README.md b/packages/README.md index 6c682b7..e6818ba 100644 --- a/packages/README.md +++ b/packages/README.md @@ -49,7 +49,7 @@ run; SHA256 digest for DFA: BB8768E977D62429368CFF2E5338A6553C35C998AEC09AF24088BA713BB54DDA -- **macroArray**\[0.4\], implementation of an array concept in a macrolanguage, e.g. +- **macroArray**\[0.5\], implementation of an array concept in a macrolanguage, e.g. ``` %array(ABC[17] (111:127), macarray=Y); @@ -68,7 +68,7 @@ SHA256 digest for DFA: BB8768E977D62429368CFF2E5338A6553C35C998AEC09AF24088BA713 which = 1:H:2 ); ``` -SHA256 digest for macroArray: 5C9208ADD091E354794C24FA830F527D17CFC758C24CB77BF2154949059F7E6F +SHA256 digest for macroArray: 69F6CF496F921D0E21F3524FC7FD130B4B8290C1E0B9BB3ABEA212B734EBE8A1 - **BasePlus**\[0.62\] adds a bunch of functionalities I am missing in BASE SAS, such as: diff --git a/packages/SHA256_for_packages.txt b/packages/SHA256_for_packages.txt index d32174d..cf1646e 100644 --- a/packages/SHA256_for_packages.txt +++ b/packages/SHA256_for_packages.txt @@ -1,5 +1,7 @@ /* 20200914 */ SQLinDS: DD5E319EB5AA29C7054EC428072F987E77C29D36874DED1AE5C62E4B300845EB +macroArray: 69F6CF496F921D0E21F3524FC7FD130B4B8290C1E0B9BB3ABEA212B734EBE8A1 + /* 20200911 */ sqlindsdemo: CCA3CB51587E30D1A4338EAF732EF03E0922918AAA21C3ECF85CABE93CD2FB15 macroArray: 5C9208ADD091E354794C24FA830F527D17CFC758C24CB77BF2154949059F7E6F diff --git a/packages/macroarray.md b/packages/macroarray.md new file mode 100644 index 0000000..d95f55c --- /dev/null +++ b/packages/macroarray.md @@ -0,0 +1,1121 @@ +# The macroArray package [ver. 0.5] ############################################### + +The **macroArray** package implements a macro array facility: +- `%array()`, +- `%do_over()`, +- `%make_do_over()`, +- `%deletemacarray()`, +- `%concatarrays()`, +- `%appendcell()`. +- etc. + +The set of macros, which emulates classic +data-step-array functionality on the macro +programming level, is provided. + +*Note:* +If you are working with BIG macroarrays do not +forget to verify your session setting for macro +memory limits. Run: +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + proc options group = macro; + run; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +to verify the following options: + +| option | description | +|-------------:|:-----------------------------------------------------------------------------------------------| +|`MEXECSIZE=` | specifies the maximum macro size that can be executed in memory. | +|`MSYMTABMAX=` | specifies the maximum amount of memory available to the macro variable symbol table or tables. | +|`MVARSIZE=` | specifies the maximum size for a macro variable that is stored in memory. | + + +Package contains: +1. macro appendarray, +2. macro appendcell, +3. macro array, +4. macro concatarrays, +5. macro deletemacarray, +6. macro do_over, +7. macro do_over2, +8. macro do_over3, +9. macro make_do_over, + +Required SAS Components: + *Base SAS Software* + +*SAS package generated by generatePackage, version 20200911* + +The SHA256 hash digest for package macroArray +`69F6CF496F921D0E21F3524FC7FD130B4B8290C1E0B9BB3ABEA212B734EBE8A1` + +### Content description ####################################################### + +## >>> `%appendArray()` macro: <<< ############################################## + +The `%appendArray()` macro is a macrowrapper +which allows to concatenate two macroarrays +created by `%array()` macro. + +By default values of the second macroarray are *not* removed. + +Dimensions of the first macroarray are extended. + +The `%appendArray()` macro executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%appendArray( + first + ,second +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +1. first - *Required*, a name of a macroarray created by the `%array()` macro. + +2. second - *Required*, a name of a macroarray created by the `%array()` macro. + + +### EXAMPLES AND USECASES: #################################################### + +**EXAMPLE 1.** Append macroarrays LL and MM. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(ll[2:4] $ 12, + function = quote(put(today() + 10*_I_, yymmdd10.)), + macarray=Y + ) + %array(mm[10:13] $ 1200, + function = quote(repeat("A",12*_I_)), + macarray=Y + ) + %put *%ll(2)*%ll(3)*%ll(4)*; + + %appendArray(ll, mm); + %put *%ll(2)*%ll(3)*%ll(4)*%ll(5)*%ll(6)**%ll(7)*%ll(8)*; + + %put *%mm(10)**%mm(11)*%mm(12)*%mm(13)*; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 2.** Error handling. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %appendArray(ll, ) + %appendArray(, mm) + + %appendArray(noExistA, noExistB) +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +## >>> `%appendCell()` macro: <<< ############################################### + +The `%appendCell()` macro allows to append +a macrovariable to a macroarray created by the `%array()` macro. + +Dimensions of the macroarray are extended. + +The `%appendCell()` macro executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%appendCell( + first + ,second + ,hilo +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +1. first - *required*, a name of a macroarray created by the `%array()` macro. + +2. second - *required*, a name of a macrovariable to be append to the macroarray. + +3. hilo - *required*, if `H` macrovariable is appended at the end + if `L` macrovariable is appended at the beginning +); + + +### EXAMPLES AND USECASES: #################################################### + +**EXAMPLE 1.** Create two macro wrappers. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %* Macro wrapper to append a macrovariable to the end of a macroarray; + %macro appendHC(array,cell); + %appendCell(&array.,&cell.,H) + %mend appendHC; + + %* macro wrapper to append a macrovariable to the beginning of a macroarray; + %macro appendLC(array,cell); + %appendCell(&array.,&cell.,L) + %mend appendLC; + + + %* create macroarrays X and variables W,Y,Z; + + %array(X[2:4] $ ("AAA", "BBB", "CCC"), macarray=Y) + %let W=1; + %let Y=2; + %let Z=3; + %put *%do_over(X)*&=W*&=Y*&=Z*; + + %put BEFORE *%do_over(X)**&=xLBOUND*&=xHBOUND*&=xN*; + %appendCell(X,Y,H) + %put AFTER1 *%do_over(X)**&=xLBOUND*&=xHBOUND*&=xN*; + + %appendLC(X,W) + %put AFTER2 *%do_over(X)**&=xLBOUND*&=xHBOUND*&=xN*; + + %appendHC(X,Z) + %put AFTER3 *%do_over(X)**&=xLBOUND*&=xHBOUND*&=xN*; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 2.** Error handling +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %appendCell(X,Y,blahblah) + + %appendCell(X,,H) + %appendCell(,Y,H) +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +**EXAMPLE 3.** Adding variable below lower bound. +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(zero[0:2] $ ("AAA", "BBB", "CCC"), macarray=Y) + %let belowzero=zzz; + + %put BEFORE *%do_over(zero)**&=zeroLBOUND*&=zeroHBOUND*&=zeroN*; + %appendCell(zero,belowzero,L) + %put AFTER *%do_over(zero)**&=zeroLBOUND*&=zeroHBOUND*&=zeroN*; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +## >>> `%array()` macro: <<< #################################################### + +The code of a macro was inspired by +*Ted Clay's* and *David Katz's* macro `%array()`. + +The `%array()` macro version provided in the package +is designed to facilitate +the idea of macro array concept, i.e. *a list of +macrovariables with common prefix and numerical suffixes*. +Usually such construction is then resolved by +double ampersand syntax, e.g. `&&perfix&i` or similar one. + +What is new/extension to the `%array()` macro concept are: + +0. The syntax is closer to the data step one. +1. It is a pure macro code (it can be executed in any place + of 4GL code), this includes generating macro arrays out + of datasets. +2. When a macroarrray is created it allows also to generate + a new macro (named the same as the array name) and replace + the double ampersand syntax with more array looking one, + i.e. for array ABC user can have `%ABC(1)`, `%ABC(2)`, or `%ABC(&i)` + constructions. +3. The array macro allows to use data step functions to generate + array's entries. + +The `%array()` macro executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%array( + array + <,function=> + <,before=> + <,after=> + <,vnames=N> + <,macarray=N> + <,ds=> + <,vars=> +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +1. `array` - *Mandatory*, an array name and a declaration/definition of an array,
+ e.g. `myArr[*] x1-x3 (4:6)`
+ or `myBrr[*] $ y1-y3 ("a" "b" "c")`
+ or `myCrr[3] $ ("d d d" "e,e,e" "f;f;f")`
+ or `myDrr p q r s`.
+ Macrovariables created by the macro are *global*. + If an array name is `_` (single underscore) then attached variables + list names are used, a call of the form: + `%array(_[*] p1 q2 r3 s4 (-1 -2 -3 -4))` + will create macrovariables: `p1`, `q2`, `r3`, and `s4` with respective + values: `-1`, `-2`, `-3`, and `-4`.
+ Three additional *global* macrovariables: + `LBOUND`, `HBOUND`, and `N` + are generated with the macroarray. See examples for more use-cases. + +* `function=` - *Optional*, a function or an expression to be applied to all array cells, + `_I_` is as array iterator, e.g. `_I_ + rand("uniform")`. + +* `before=` - *Optional*, a function or an expression to be added before looping through + array, e.g. `call streaminit(123)`. + +* `after=` - *Optional*, a function or an expression to be added after looping through + array, e.g. `call sortn(ABC)`. + +* `vnames=N` - *Optional*, Default value `N`, if set to `Y`/`YES` then macroarray is built based + on variables names instead values, e.g. + `%array(myArr[*] x1-x3 (4:6), vnames=Y)` + will use `x1`, `x2`, and `x3` as values instead `4`, `5`, and `6`. + +* `macarray=N` - *Optional*, Default value `N`, if set to `Y`/`YES` then macro named with array + name is compiled to create convenient envelope for multiple ampersands, e.g. + `%array(myArr[*] x1-x3 (4:6), macarray=Y)` + will create `%myArr(J)` macro which will allow to extract "data" + from macroarray like: + `%let x = %myArr(1);` + or when used with second parameter equal `I` (insert) allow to overwrite macroarrays + value: + `%let %myArr(17,i) = 42;` + +* `ds=` - *Optional*, Use a dataset as a basis for a macroarray data, + if used by default overwrites use of the `array` parameter, honors `macarray=` + argument, dataset options are allowed, e.g. `sashelp.class(obs=5)` + +* `vars=` - *Optional*, a list of variables used to create macroarrays from a dataset, + the list format can be as follows (`<...>` means optional): + `variable1 <... variableN>` + delimiters are hash(`#`) and pipe(`|`), currently only space + is supported as separator, the meaning of `#` and `|` delimiters + will be explained in the following example: + if the `vars = height#h weight weight|w age|` value is provided + then the following macroarrays will be created:
+ 1) macroarray "H" with ALL(`#`) values of variable "height"
+ 2) macroarray "WEIGHT" with ALL(no separator is equivalent to #) + values of variable "weight"
+ 3) macroarray "W" with UNIQUE(|) values of variable "weight" and
+ 4) macroarray "AGE" with UNIQUE(|) values of variable "age", + + + +### EXAMPLES AND USECASES: #################################################### + + +**EXAMPLE 1.** Basic use-case. + Creating macroarray like in the array statement; + values are used by default; + different types of brackets are allowed; + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(a[*] x1-x5 (1:5)) + + %array(b{5} (5*17)) + + %* Mind the $ since it is a character array! + %array(c(3) $ 10 ("a A" "b,B" "c;C")) + + %array(d x1-x5 (5 4 3 2 1)) + %put _user_; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 2.** Index ranges. + If range starts < 0 then it is shifted to 0. + In case when range is from `1` to `M` + then macrovariable `N` is set to `M` + In case when range is different + the `N` returns number of + elements in the array `(Hbound - Lbound + 1)`. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(d[-2:2] $ ("a" "b" "c" "d" "e")) + %put &=dLBOUND. &=dHBOUND. &=dN.; + %put &=d0. &=d1. &=d2. &=d3. &=d4.; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 3.** Functions. + It is possible to assign value of a function + or an expression to a cell of the array, + e.g. `array[_I_] = function(...)`. + You can use an iterator in a function. + As in case of usual arrays it is `_I_`. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(e[-3:3] $, function = "A" ) + %put &=eLBOUND. &=eHBOUND. &=eN.; + %put &=e0. &=e1. &=e2. &=e3. &=e4. &=e5. &=e6.; + + %array(f[-3:3], function = (2**_I_) ) + %put &=fLBOUND. &=fHBOUND. &=fN.; + %put &=f0. &=f1. &=f2. &=f3. &=f4. &=f5. &=f6.; + + %array(g[0:2], function = ranuni(123) ) + %put &=gLBOUND. &=gHBOUND. &=gN.; + %put &=g0. &=g1. &=g2.; + + %* Or something more complex; + %array(gg[0:11] $ 11, function = put(intnx("MONTH", '1jun2018'd, _I_, "E"), yymmn.)) + %put &=ggLBOUND. &=ggHBOUND. &=ggN.; + %put &=gg0 &=gg1 &=gg2 ... &=gg11; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 4.** Functions cont. + If there is need for set-up something *before* or *after*: + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(h[10:12] + ,function = rand('Uniform') + ,before = call streaminit(123) + ,after = call sortn(of h[*]) + ) + %put &=h10. &=h11. &=h12.; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 5.** Fibonacci series. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(i[1:10] (10*0) + ,function = ifn(_I_ < 2, 1, sum(i[max(_I_-2,1)], i[max(_I_-1,2)]) ) ) + %put &=i1 &=i2 &=i3 &=i4 &=i5 &=i6 &=i7 &=i8 &=i9 &=i10; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 6a.** "Uppercas Letters" + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(UL[26] $, function = byte(rank("A")+_I_-1) ) + %put &=UL1 &=UL2 ... &=UL25 &=UL26; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 6b.** "Lowercase Letters" + Extended by `macarray=Y` option and + the input mode support (with `I`). + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(ll[26] $, function = byte(rank("a")+_I_-1), macarray=Y) + %put *%ll(&llLBOUND.)*%ll(3)*%ll(4)*%ll(5)*...*%ll(25)*%ll(&llHBOUND.)*; + + %* The range handling, warning; + %put *%ll(265)*; + + %* The input mode; + %put *before:*%ll(2)*; + %let %ll(2,I) = bbbbb; + %put *after: *%ll(2)*; + + %* The range handling, error; + %let %ll(265,I) = bbb; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 7.** The use of `vnames=Y` + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(R R1978-R1982) + %put &=R1 &=R2 &=R3 &=R4 &=R5; + + %array(R R1978-R1982 (78:82)) + %put &=R1 &=R2 &=R3 &=R4 &=R5; + + %array(R R1978-R1982 (78:82), vnames=Y) + %put &=R1 &=R2 &=R3 &=R4 &=R5; + + %array(R R1978-R1982, vnames=Y) + %put &=R1 &=R2 &=R3 &=R4 &=R5; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 8.** A "no name" array i.e. the `_[*]` array + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(_[*] x1-x5 (1:5)) + %put _user_; + + %array(_[*] p q r s (4*42)) + %put _user_; + + %* If no variables named than use _1 _2 ... _N; + %array(_[4] (-1 -2 -3 -4)) + %put &=_1 &=_2 &=_3 &=_4; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 9.** Pure macro code can be used in a data step. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + data test1; + set sashelp.class; + %array(ds[*] d1-d4 (4*17)) + a1 = &ds1.; + a2 = &ds2.; + a3 = &ds3.; + a4 = &ds4.; + run; + + data test2; + set sashelp.class; + %array(_[*] j k l m (4*17)) + a1 = &j.; + a2 = &k.; + a3 = &l.; + a4 = &m.; + run; + + data test3; + set sashelp.class; + %array(alpha[*] j k l m (101 102 103 104), macarray=Y) + a1 = %alpha(1); + a2 = %alpha(2); + a3 = %alpha(3); + a4 = %alpha(4); + a5 = %alpha(555); + run; + + data test4; + set sashelp.class; + %array(beta[*] j k l m (101 102 103 104), vnames=Y, macarray=Y) + a1 = "%beta(1)"; + a2 = "%beta(2)"; + a3 = "%beta(3)"; + a4 = "%beta(4)"; + a5 = "%beta(555)"; + run; + + data test5; + set sashelp.class; + %array(gamma[4] $ 12 ("101" "102" "103" "104"), macarray=Y) + a1 = "%gamma(1)"; + a2 = "%gamma(2)"; + a3 = "%gamma(3)"; + a4 = "%gamma(4)"; + a5 = "%gamma(555)"; + run; + + data test6; + set sashelp.class; + %array(ds = sashelp.cars, vars = Cylinders|, macarray=Y) + a0 = %Cylinders(0); + a1 = %Cylinders(1); + a2 = %Cylinders(2); + a3 = %Cylinders(3); + a4 = %Cylinders(4); + a5 = %Cylinders(555); + run; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 10.** Creating an array from a dataset, basic case. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(ds = sashelp.class, vars = height weight age) + %put _user_; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 11. Creating an array from a dataset, advanced. + If: `vars = height#h weight weight|w age|` + then create: + 1. macroarray "h" with ALL(#) values of variable "height" + 2. macroarray "weight" with ALL(no separator is equivalent to #) values of variable "weight" + 3. macroarray "w" with UNIQUE(|) values of variable "weight" + 4. macroarray "age" with UNIQUE(|) values of variable "age" + Currently the only separator in VARS is a space. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(ds = sashelp.class, vars = height#h weight weight|w age|) + %put _user_; + + %array(ds = sashelp.class, vars = height#hght weight weight|wght age|, macarray=Y) + %put *%hght(&hghtLBOUND.)**%weight(2)**%wght(&wghtHBOUND.)**%age(3)*; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 12.** Creating an array from a dataset with dataset options + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(ds = sashelp.cars(obs=100 where=(Cylinders=6)), vars = Make| Type| Model, macarray=Y) + %put *%make(&makeLBOUND.)*%Model(2)*%Model(3)*%Model(4)*%type(&typeHBOUND.)*; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +## >>> `%concatArrays()` macro: <<< ############################################# + +The `%concatArrays()` macro allows to concatenate +two macroarrays created by the `%array()` macro. + +By default values of the second macroarray are removed. + +Dimensions of the first macroarray are extended. + +The `%concatArrays()` macro executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%concatArrays( + first + ,second + <,removeSecond=Y> +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +1. first - *Required*, a name of a macroarray created by the `%array()` macro. + +2. second - *Required*, a name of a macroarray created by the `%array()` macro. + +* removeSecond=Y - *Optional*, Default value `Y`, if set to `Y` then + the second array is removed. + + +### EXAMPLES AND USECASES: #################################################### + +**EXAMPLE 1.** Concatenate macroarrays LL and MM. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(ll[2:4] $ 12, + function = quote(put(today() + 10*_I_, yymmdd10.)), + macarray=Y + ) + %array(mm[10:13] $ 12000, + function = quote(repeat("A",123*_I_)), + macarray=Y + ) + %put *%ll(2)*%ll(3)*%ll(4)*; + + %concatArrays(ll, mm); + %put *%ll(2)*%ll(3)*%ll(4)*%ll(5)*%ll(6)**%ll(7)*%ll(8)*; + + %put *%mm(10)**%mm(11)*%mm(12)*%mm(13)*; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 2.** Error handling. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %concatArrays(ll, ) + %concatArrays(, mm) + + %concatArrays(noExistA, noExistB) +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +## >>> `%deleteMacArray()` macro: <<< ########################################### + +The `%deleteMacArray()` macro allows to delete +macroarrays created by the `%array()` macro. + +The `%deleteMacArray()` macro executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%deleteMacArray( + arrs + <,macarray=N> +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +1. arrs - *Required*, a space separated list of manes + of macroarray created by the `%array()` macro. + +* macarray=N - *Optional*, indicator should a macro + associated with macroarray to be deleted? + If `Y` or `YES` then the associated macro is deleted. + + + +## >>> `%do_over()` macro: <<< ################################################ + +The code of the macro was inspired by +*Ted Clay's* and *David Katz's* macro `%do_over()`. + +The `%DO_OVER()` macro allows to iterate over macroarray created with the `macarray=Y` +parameter of the `%ARRAY()` macro. + +The `%do_over()` macro executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%do_over( + array + <,phrase=%nrstr(%&array(&_I_.))> + <,between=%str( )> + <,which = > +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +1. array - Required, indicates a macroarray which metadata (Lbound, Hbouns) + are to be used to loop in the `%do_over()` + +* phrase= - *Optional*, Default value `%nrstr(%&array(&_I_.))`, + a statement to be called in each iteration + of the internal do_over's loop. Loop iterator is `_I_`, + if you want to use `_I_` or array name + [e.g. `%myArr(&_I_.)`] *enclose it* in the `%NRSTR()` + macro quoting function. + +* between= - *Optional*, Default value `%str( )` (space), + a statement to be called in between each + iteration of the internal do_over loop. + If macroquoted (e.g. `%str( + )`) then the `%unquote()` + function is automatically applied. + +* which = - *Optional*, a _SPACE_ separated list of indexes which + should be used to iterate over selected macroarray. + Possible special characters are `H` and `L` which means + *high* and *low* bound of an array, list could be set with + colons(`:`) in form of `start:end:by` (*no spaces between!*), + if `by` is omitted the default is `1`. If possible use + `1:5` rather `1 2 3 4 5` since the firs works faster. + + +### EXAMPLES AND USECASES: #################################################### + +**EXAMPLE 1.** Simple looping. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(beta[*] j k l m (101 102 103 104), vnames=Y, macarray=Y) + + %put #%do_over(beta)#; + + %put #%do_over(beta, phrase=%nrstr("%beta(&_I_.)"), between=%str(,))#; + + data test1; + %array(beta[*] j k l m (101 102 103 104), vnames=Y, macarray=Y) + %do_over(beta, phrase=%nrstr(a&_I_. = "%beta(&_I_.)";)) + put _all_; + run; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 2.** Multiple arrays looping. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(alpha[*] j k l m n, vnames=Y, macarray=Y) + %array( beta[5] $ , function = "a", macarray=Y) + %array(gamma[4] (101 102 103 104), macarray=Y) + + data test2; + call streaminit(123); + %do_over(beta + , phrase = %nrstr(%beta(&_I_.) = %gamma(&_I_.) * rand('Uniform'); output;) + , between = put _all_; + ); + put _all_; + run; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 3.** Multiple arrays looping, cont. + Create multiple datasets. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %do_over(beta + , phrase = %nrstr( + data %alpha(&_I_.)2; + call streaminit(123); + %beta(&_I_.)x = %gamma(&_I_.) * rand('Uniform'); + output; + run; + ) + ) +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 4.** Multiple arrays looping, cont. + Create multiple datasets using a macro. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %macro doit(ds, var=a, val=1); + data &ds.; + call streaminit(123); + &var. = &val. * rand('Uniform'); + output; + run; + %mend doit; + + %do_over(beta + , phrase = %nrstr( + %DOIT(%alpha(&_I_.)1, var = %beta(&_I_.), val = %gamma(&_I_.)) + ) + ) +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 5.** `%do_over()` inside `%array()` + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(test[*] x1-x12 (1:12), macarray=Y) + + %put **%test(1)**%test(12)**; + + %put #%do_over(test)#; + + %array(abc[*] x1-x12 (%do_over(test,phrase=%nrstr(%eval(100-%test(&_I_.))))), macarray=Y) + + %put **%abc(1)**%abc(12)**; + + %put #%do_over(abc)#; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 6.** Looping over array with *macroquoted* separator. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(alpha[11] (5:15), macarray=Y) + + %let x = %do_over(alpha + , phrase = %NRSTR(%alpha(&_I_.)) + , between= %str( + ) + ); + %put &=x.; + %put %sysevalf(&x.); +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 7.** Working with the `WHICH=` optional parameter + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(test[*] x01-x12, vnames= Y, macarray=Y) + + %put #%do_over(test)#; + + %put #%do_over(test, which= 1 3 5)#; + + %put #%do_over(test, which= 1:5)#; + + %put #%do_over(test, which= 1:5:2 7 8)#; + + %put #%do_over(test, which= L:H l:h)#; + + %put #%do_over(test, which= L:3 10:h)#; + + %put #%do_over(test, which= L:H h:l:-1 13 14)#; + + %put #%do_over(test, which= %eval(1+1):%eval(5+1))#; + + %put #%do_over(test, which= L:H h:l:-1 13 14, between=%str(,))#; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +## >>> `%do_over2()` macro: <<< ################################################ + +The code of the macro was inspired by +*Ted Clay's* and *David Katz's* macro `%do_over()`. + +The `%DO_OVER2()` macro allows to iterate over *two* macroarray created with the `macarray=Y` +parameter of the `%ARRAY()` macro. + +The `%do_over2()` macro executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%do_over2( + arrayI + ,arrayJ + <,phrase=%nrstr(%&arrayI(&_I_.) %&arrayJ(&_J_.))> + <,between=%str( )> +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +1. arrayI - Required, indicates the first macroarray which metadata (Lbound, Hbouns) + are to be used in the outer loop in the `%do_over2()` + +2. arrayJ - Required, indicates the second macroarray which metadata (Lbound, Hbouns) + are to be used in the inner loop in the `%do_over2()` + +* phrase= - *Optional*, Default value `%nrstr(%&arrayI(&_I_.) %&arrayJ(&_J_.))`, + a statement to be called in each iteration + of the *inner* loop. The outer loop iterator is `_I_`, + the inner loop iterator is `_J_`, + if you want to use `_I_`, `_J_`, or arrays names + [e.g. `%myArr(&_I_.)`] *enclose them* in the `%NRSTR()` + macro quoting function. + +* between= - *Optional*, Default value `%str( )` (space), + a statement to be called in between each + iteration of the internal do_over2 loop. + If macroquoted (e.g. `%str( + )`) then the `%unquote()` + function is automatically applied. + +*/ +/* +### EXAMPLES AND USECASES: #################################################### + +**EXAMPLE 1.** Looping over two arrays. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(alpha[*] j k l m n, vnames=Y, macarray=Y) + %array( beta[4] (101 102 103 104), macarray=Y) + + %put *%do_over2(alpha, beta + , phrase = %NRSTR((%alpha(&_I_.), %beta(&_J_))) + )*; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 2.** Looping over two arrays with a separator. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(alpha[11] (5:15), macarray=Y) + %array( beta[ 4] (101 102 103 104), macarray=Y) + + %let x = %do_over2(alpha, beta + , phrase = %NRSTR((%alpha(&_I_.) * %beta(&_J_))) + , between= + + ); + %put &=x.; + %put %sysevalf(&x.); +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 3.** Looping over two arrays with *macroquoted* separator. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(alpha[11] (5:15), macarray=Y) + %array( beta[ 4] (101 102 103 104), macarray=Y) + + %let x = %do_over2(alpha, beta + , phrase = %NRSTR((%alpha(&_I_.) * %beta(&_J_))) + , between= %str( + ) + ); + %put &=x.; + %put %sysevalf(&x.); +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +## >>> `%do_over3()` macro: <<< ################################################ + +The code of the macro was inspired by +*Ted Clay's* and *David Katz's* macro `%do_over()`. + +The `%DO_OVER3()` macro allows to iterate over *three* macroarray created with the `macarray=Y` +parameter of the `%ARRAY()` macro. + +The `%do_over3()` macro executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%do_over2( + arrayI + ,arrayJ + ,arrayK + <,phrase=%nrstr(%&arrayI(&_I_.) %&arrayJ(&_J_.) %&arrayK(&_K_.))> + <,between=%str( )> +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +1. arrayI - Required, indicates the first macroarray which metadata (Lbound, Hbouns) + are to be used in the outer loop in the `%do_over3()` + +2. arrayJ - Required, indicates the second macroarray which metadata (Lbound, Hbouns) + are to be used in the inner loop in the `%do_over3()` + +3. arrayK - Required, indicates the third macroarray which metadata (Lbound, Hbouns) + are to be used in the inner loop in the `%do_over3()` + +* phrase= - *Optional*, Default value `%nrstr(%&arrayI(&_I_.) %&arrayJ(&_J_.) %&arrayK(&_K_.))`, + a statement to be called in each iteration + of the *inner* loop. The *outer* loop iterator is `_I_`, + the *middle* loop iterator is `_J_`, the *inner* loop iterator is `_K_`, + if you want to use `_I_`, `_J_`, `_K_`, or arrays names + [e.g. `%myArr(&_I_.)`] *enclose them* in the `%NRSTR()` + macro quoting function. + +* between= - *Optional*, Default value `%str( )` (space), + a statement to be called in between each + iteration of the internal do_over2 loop. + If macroquoted (e.g. `%str( + )`) then the `%unquote()` + function is automatically applied. + + + +### EXAMPLES AND USECASES: #################################################### + +**EXAMPLE 1.** Looping over 3 macroarrays. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(a1_[2] (0 1), macarray=Y) + %array(a2_[2] (2 3), macarray=Y) + %array(a3_[2] (4 5), macarray=Y) + + %do_over3(a1_, a2_, a3_ + , phrase = %NRSTR(%put (%a1_(&_I_.), %a2_(&_J_), %a3_(&_K_));) + ) +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +**EXAMPLE 2.** Looping 3 times over a macroarray. + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(a[0:2] (0 1 2), macarray=Y) + + %do_over3(a, a, a + , phrase = %NRSTR(%put (%a(&_I_.), %a(&_J_), %a(&_K_));) + ) +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +/* +## >>> `%make_do_over()` macro: <<< ########################################### + +The code of the macro was inspired by +*Ted Clay's* and *David Katz's* macro `%do_over()`. + +The `%make_do_over()` macro allows to generate +the `%DO_OVER()` macros. It works *only* for *n>3*! + +The `%make_do_over()` macro does *not* executes like a pure macro code. + +### SYNTAX: ################################################################### + +The basic syntax is the following, the `<...>` means optional parameters: +~~~~~~~~~~~~~~~~~~~~~~~{.sas} +%make_do_over( + size +) +~~~~~~~~~~~~~~~~~~~~~~~ + +**Arguments description**: + +size - Required, indicates the number of dimensions + (i.e. inner loops) of the `%DO_OVER()` macro. + + + +### EXAMPLES AND USECASES: #################################################### + +**EXAMPLE 1.** Code of created "4-loop" %DO_OVER4() macro + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %macro do_over4( + arrayI1, + arrayI2, + arrayI3, + arrayI4, + phrase=%nrstr( + %&arrayI1(&_I1_.) + %&arrayI2(&_I2_.) + %&arrayI3(&_I3_.) + %&arrayI3(&_I4_.) + ), + between=%str( ) + ); + %local _I1_ _I2_ _I3_ _I4_; + %do _I1_ = &&&arrayI1.LBOUND %to &&&arrayI1.HBOUND; + %do _I2_ = &&&arrayI2.LBOUND %to &&&arrayI2.HBOUND; + %do _I3_ = &&&arrayI3.LBOUND %to &&&arrayI3.HBOUND; + %do _I4_ = &&&arrayI4.LBOUND %to &&&arrayI4.HBOUND; + %if not ( + &_I1_. = &&&arrayI1.LBOUND + AND &_I2_. = &&&arrayI2.LBOUND + AND &_I3_. = &&&arrayI3.LBOUND + AND &_I4_. = &&&arrayI4.LBOUND + ) + %then %do;%unquote(&between.)%end;%unquote(%unquote(&phrase.)) + %end; + %end; + %end; + %end; + %mend do_over4; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 3.** Create a 4-loop `%DO_OVER4()` macro + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %make_do_over(4); + + %array(a1_[2] (0 1), macarray=Y) + + %do_over4(a1_, a1_, a1_, a1_ + , phrase = %NRSTR(%put (%a1_(&_I1_.), %a1_(&_I2_), %a1_(&_I3_), %a1_(&_I4_));) + ) + + %put *%do_over4(a1_, a1_, a1_, a1_ + , between = * + )*; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +**EXAMPLE 3.** Create a 5-loop `%DO_OVER5()` macro + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %make_do_over(5); + + %array(a1_[2] (0 1), macarray=Y) + + %do_over5(a1_, a1_, a1_, a1_, a1_ + , phrase = %NRSTR(%put (%a1_(&_I1_.), %a1_(&_I2_), %a1_(&_I3_), %a1_(&_I4_), %a1_(&_I5_));) + ) + + %put *%do_over5(a1_, a1_, a1_, a1_, a1_ + , between = * + )* + ; + + options nomprint; + data test2; + %do_over5(a1_, a1_, a1_, a1_, a1_ + , phrase = %NRSTR(x1 = %a1_(&_I1_.); x2 = %a1_(&_I2_); x3 = %a1_(&_I3_); x4 = %a1_(&_I4_); x5 = %a1_(&_I5_);) + , between = output; + ) + output; + run; +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + +**EXAMPLE 4.** Create all from 6 to 10 "do_overs" + +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.sas} + %array(loop[6:10] (6:10), macarray=Y) + %do_over(loop + , phrase = %nrstr( + %make_do_over(%loop(&_I_.)) + ) + ); +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +## License #################################################################### +Copyright (c) Bartosz Jablonski, since January 2019 + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/packages/macroarray.zip b/packages/macroarray.zip index 0325ac7..50d13c5 100644 Binary files a/packages/macroarray.zip and b/packages/macroarray.zip differ