SASPAC/SAS_PACKAGES
1
0
Fork 0
mirror of https://github.com/yabwon/SAS_PACKAGES.git synced 2025-12-11 03:04:35 +00:00
Code Issues Packages Projects Releases Wiki Activity

The BasePlus package [ver. 1.15]

Browse Source 
The **BasePlus** package [ver. 1.15]

- 3 new macros added: `%LDSN()`, `%LVarNm()`, and `%LVarNmLab()`

hash: 0331C673052D5221DA98C5CC93295634D8A0BC62C7D2FEF9D0D85B0B2DEDE4E9
This commit is contained in:
Bart Jablonski (yabwon)
2022-05-08 19:10:55 +02:00
parent 6d10426474
commit 642bc5ba72
5 changed files with 357 additions and 55 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@@ -146,7 +146,7 @@ SHA256 digest for macroArray: DF63B0E027827A82038F1C8422787A0BC569BA93104BA1778D
[Documentation for macroArray](https://github.com/yabwon/SAS_PACKAGES/blob/main/packages/macroarray.md "Documentation for macroArray")
- **BasePlus**\[1.14\] adds a bunch of functionalities I am missing in BASE SAS, such as:
- **BasePlus**\[1.15\] adds a bunch of functionalities I am missing in BASE SAS, such as:
```sas
call arrMissToRight(myArray);
call arrFillMiss(17, myArray);
@@ -164,7 +164,7 @@ format x bool.;
%zipLibrary(sashelp,libOut=work)
```
SHA256 digest for BasePlus: 12A9A2155D9C6F969DF0A66E3A18D0938B8194FA3AA9D70EDDE1CB71DEF9691E
SHA256 digest for BasePlus: 0331C673052D5221DA98C5CC93295634D8A0BC62C7D2FEF9D0D85B0B2DEDE4E9
[Documentation for BasePlus](https://github.com/yabwon/SAS_PACKAGES/blob/main/packages/baseplus.md "Documentation for BasePlus")

4
packages/README.md
View File

@@ -90,7 +90,7 @@ SHA256 digest for macroArray: DF63B0E027827A82038F1C8422787A0BC569BA93104BA1778D
---
- **BasePlus**\[1.14\] adds a bunch of functionalities I am missing in BASE SAS, such as:
- **BasePlus**\[1.15\] adds a bunch of functionalities I am missing in BASE SAS, such as:
```sas
call arrMissToRight(myArray);
call arrFillMiss(17, myArray);
@@ -108,7 +108,7 @@ format x bool.;
%zipLibrary(sashelp,libOut=work)
```
SHA256 digest for BasePlus: 12A9A2155D9C6F969DF0A66E3A18D0938B8194FA3AA9D70EDDE1CB71DEF9691E
SHA256 digest for BasePlus: 0331C673052D5221DA98C5CC93295634D8A0BC62C7D2FEF9D0D85B0B2DEDE4E9
[Documentation for BasePlus](https://github.com/yabwon/SAS_PACKAGES/blob/main/packages/baseplus.md "Documentation for BasePlus")

3
packages/SHA256_for_packages.txt
View File

@@ -1,3 +1,6 @@
/* 20220508 */
BasePlus: 0331C673052D5221DA98C5CC93295634D8A0BC62C7D2FEF9D0D85B0B2DEDE4E9
/* 20220420 */
BasePlus: 12A9A2155D9C6F969DF0A66E3A18D0938B8194FA3AA9D70EDDE1CB71DEF9691E
DFA: 6B3FB0B06B47A7EF1BB004B483B0F39B8F553F7F16D02A7E24CDE388BBA704EA

401
packages/baseplus.md
View File

@@ -45,12 +45,17 @@
* [`%RainCloudPlot()` macro](#raincloudplot-macro)
* [`%zipLibrary()` macro](#ziplibrary-macro)
* [`%unzipLibrary()` macro](#unziplibrary-macro)
* [`%LDSN()` macro](#ldsn-macro)
* [`%LVarNm()` macro](#lvarnm-macro)
* [`%LVarNmLab()` macro](#lvarnmlab-macro)
* [License](#license)
---
# The BasePlus package [ver. 1.14] <a name="baseplus-package"></a> ###############################################
# The BasePlus package [ver. 1.15] <a name="baseplus-package"></a> ###############################################
The **BasePlus** package implements useful
functions and functionalities I miss in the BASE SAS.
@@ -67,7 +72,9 @@ Kudos to all who inspired me to generate this package:
*Richard DeVenezia*,
*Christian Graffeuille*,
*Allan Bowe*,
*Anamaria Calai*.
*Anamaria Calai*,
*Michal Ludwicki*,
*Quentin McMullen*.
---
@@ -187,7 +194,6 @@ Kudos to all who inspired me to generate this package:
%rainCloudPlot(sashelp.cars,DriveTrain,Invoice)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
![Rain Cloud Plot](./baseplus_RainCloudPlot_Ex0_9.png)
**Example 10**: Zip SAS library.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~sas
@@ -196,64 +202,77 @@ Kudos to all who inspired me to generate this package:
%unzipLibrary(%sysfunc(pathname(work)), zip=sashelp, mode=S, clean=1)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
**Example 11**: Long dataset names.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~sas
data %LDSN( work. peanut butter & jelly time with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s (drop = sex rename=(name=first_name) where = (age in (12,13,14))) );
set sashelp.class;
run;
proc print data = %LDSN( work. peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s );
run;
data MyNextDataset;
set %LDSN( work. peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s );
where age > 12;
run;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
---
Package contains:
1. macro deduplistc
2. macro deduplistp
3. macro deduplists
4. macro deduplistx
5. macro functionexists
6. macro getvars
7. macro qdeduplistx
8. macro qgetvars
9. macro qzipevalf
10. macro raincloudplot
11. macro symdelglobal
12. macro unziplibrary
13. macro zipevalf
14. macro ziplibrary
15. format bool
16. format boolz
17. format ceil
18. format floor
19. format int
20. function arrfill
21. function arrfillc
22. function arrmissfill
23. function arrmissfillc
24. function arrmisstoleft
25. function arrmisstoleftc
26. function arrmisstoright
27. function arrmisstorightc
28. function bracketsc
29. function bracketsn
30. function catxfc
31. function catxfi
32. function catxfj
33. function catxfn
34. function deldataset
35. function semicolonc
36. function semicolonn
37. format brackets
38. format semicolon
39. proto qsortincbyprocproto
40. function frommissingtonumberbs
41. function fromnumbertomissing
42. function quicksort4notmiss
43. function quicksorthash
44. function quicksorthashsddv
45. function quicksortlight
1. macro deduplistc
2. macro deduplistp
3. macro deduplists
4. macro deduplistx
5. macro functionexists
6. macro getvars
7. macro ldsn
8. macro lvarnm
9. macro lvarnmlab
10. macro qdeduplistx
11. macro qgetvars
12. macro qzipevalf
13. macro raincloudplot
14. macro symdelglobal
15. macro unziplibrary
16. macro zipevalf
17. macro ziplibrary
18. format bool
19. format boolz
20. format ceil
21. format floor
22. format int
23. functions arrfill
24. functions arrfillc
25. functions arrmissfill
26. functions arrmissfillc
27. functions arrmisstoleft
28. functions arrmisstoleftc
29. functions arrmisstoright
30. functions arrmisstorightc
31. functions bracketsc
32. functions bracketsn
33. functions catxfc
34. functions catxfi
35. functions catxfj
36. functions catxfn
37. functions deldataset
38. functions semicolonc
39. functions semicolonn
40. format brackets
41. format semicolon
42. proto qsortincbyprocproto
43. functions frommissingtonumberbs
44. functions fromnumbertomissing
45. functions quicksort4notmiss
46. functions quicksorthash
47. functions quicksorthashsddv
48. functions quicksortlight
*SAS package generated by generatePackage, version 20220420*
The SHA256 hash digest for package BasePlus:
`12A9A2155D9C6F969DF0A66E3A18D0938B8194FA3AA9D70EDDE1CB71DEF9691E`
`0331C673052D5221DA98C5CC93295634D8A0BC62C7D2FEF9D0D85B0B2DEDE4E9`
---
# Content description ############################################################################################
@@ -3497,6 +3516,286 @@ run;
---
## >>> `%LDSN()` macro: <<< <a name="ldsn-macro"></a> #######################
The LDSN (Long DataSet Names) macro function
allows to use an "arbitrary" text string to name a dataset.
---
The idea for the macro came from the following story:
Good friend of mine, who didn't use SAS for quite some time,
told me that he lost a few hours for debugging because
he forgot that the SAS dataset name limitation is 32 bytes.
I replied that it shouldn't be a problem to do a workaround
for this inconvenience with a macro and the `MD5()` hashing function.
I said: *The macro should take an "arbitrary string" for a dataset
name, convert it, with help of `MD5()`, to a hash digest, and
create a dataset with an "artificial" `hex16.` formated name.*
Starting with something like this:
~~~~~~~~~~~~~~~~~~~~~~~sas
data %LDSN(work. peanut butter & jelly with a hot-dog in [a box] and s*t*a*r*s (drop = sex rename=(name=first_name) where = (age in (12,13,14))) );
set sashelp.class;
run;
~~~~~~~~~~~~~~~~~~~~~~~
the macro would do:
~~~~~~~~~~~~~~~~~~~~~~~sas
%sysfunc(MD5(peanut butter & jelly with a hot-dog in [a box] and s*t*a*r*s), hex16.)
~~~~~~~~~~~~~~~~~~~~~~~
and (under the hood) return and execute the following code:
~~~~~~~~~~~~~~~~~~~~~~~sas
data work.DSN_41D599EF51FBA58_(drop = sex rename=(name=first_name) where = (age in (12,13,14))) ;
set sashelp.class;
run;
~~~~~~~~~~~~~~~~~~~~~~~
Also in the next data step user should be able to do:
~~~~~~~~~~~~~~~~~~~~~~~sas
data my_next_data_step;
set %DSN(work. peanut butter & jelly with a hot-dog in [a box] and s*t*a*r*s);
run;
~~~~~~~~~~~~~~~~~~~~~~~
and work without the "dataset-name-length-limitation" issue.
---
See examples below for the details.
The `%LDSN()` macro executes like a pure macro code.
**Known "Limitations":**
- dataset name _cannot_ contain dots (`.`) since they are used as separators!
- dataset name _cannot_ contain round brackets(`(` and `)`) since they are used as separators
(but `[]` and `{}` are allowed)!
- dataset name _cannot_ contain unpaired quotes (`'` and `"`),
text: `a "hot-dog"` is ok, but `John's dog` is not!
**Behaviour:**
- dataset name text is *converted to upcase*
- dataset name text *leading and trailing spaces are ignored*,
e.g. the following will give the same hash digest:
`%ldsn(work.test)`, `%ldsn( work.test)`, `%ldsn(work.test )`,
`%ldsn(work .test)`, `%ldsn(work. test)`, `%ldsn(work . test)`.
- if the user want to add an extra suffix to the dataset the `%ldsn()` macro
**has** to be wrapped inside the `%unquote()` macro function.
~~~~~~~~~~~~~~~~~~~~~~~sas
data
%unquote(%LDSN(WORK.some strange! DS name))_1
%unquote(%LDSN(WORK.some strange! DS name))_2
;
set sashelp.class;
if SEX="M" then output %unquote(%LDSN(WORK.some strange! DS name))_1;
else output %unquote(%LDSN(WORK.some strange! DS name))_2;
run;
~~~~~~~~~~~~~~~~~~~~~~~
The reason for this is a "bug" like behaviour of SAS tokenizer/macroprocesor.
See the following SAS-L discussion thread:
`https://listserv.uga.edu/scripts/wa-UGA.exe?A2=SAS-L;4b2bcf80.2205A&S=`
- macro calls of the form:
`data %LDSN(); run;`, `data %LDSN( ); run;`, `data %LDSN( . ); run;` or even
`data %LDSN( . (keep=x)); run;` are resolved to empty string, so the result is
equivalent to `data; run;`
### SYNTAX: ###################################################################
The basic syntax is the following, the `<...>` means optional parameters:
~~~~~~~~~~~~~~~~~~~~~~~sas
%LDSN(
arbitrary text string (in line with limitations)
)
~~~~~~~~~~~~~~~~~~~~~~~
### EXAMPLES AND USECASES: ####################################################
**EXAMPLE 1.**
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~sas
options nomprint source nomlogic nosymbolgen ls = max ps = max;
data %LDSN( work. peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s (drop = sex rename=(name=first_name) where = (age in (12,13,14))) );
set sashelp.class;
run;
proc print data = %LDSN( work. peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s );
run;
data MyNextDataset;
set %LDSN( work. peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s );
where age > 12;
run;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
---
## >>> `%LVarNm()` macro: <<< <a name="lvarnm-macro"></a> #######################
The LVarNm() macro function works like the LDSN() macro function, but for variables.
Supported by LVarNmLab() macro function which allows to remember "user names" in labels.
The motivation for the macro was similar one as for the LDSN() macro.
---
See examples below for the details.
The `%LVarNm()` macro executes like a pure macro code.
**Known "Limitations":**
- variable name _cannot_ contain unpaired quotes (`'` and `"`),
text: `a "hot-dog"` is ok, but `John's dog` is not!
**Behaviour:**
- variable name text is *converted to upcase*
- variable name text *leading and trailing spaces are ignored*,
e.g. the following will give the same hash digest:
`%LVarNm(test)`, `%LVarNm( test)`, `%LVarNm(test )`.
- if the user want to add an extra suffix to the variable,
e.g. to get a numerical suffix, the `%LVarNm()` macro
**has** to be wrapped inside the `%unquote()` macro function.
~~~~~~~~~~~~~~~~~~~~~~~sas
data test4;
array X[*] %unquote(%LVarNm(some strange! name))_0 - %unquote(%LVarNm(some strange! name))_10;
do i = lbound(X) to hbound(X);
X[i] = 2**(i-1);
put X[i]=;
end;
run;
~~~~~~~~~~~~~~~~~~~~~~~
The reason for this is a "bug" like behaviour of SAS tokenizer/macroprocesor.
See the following SAS-L discussion thread:
`https://listserv.uga.edu/scripts/wa-UGA.exe?A2=SAS-L;4b2bcf80.2205A&S=`
### SYNTAX: ###################################################################
The basic syntax is the following, the `<...>` means optional parameters:
~~~~~~~~~~~~~~~~~~~~~~~sas
%LVarNm(
arbitrary text string (in line with limitations)
)
~~~~~~~~~~~~~~~~~~~~~~~
---
### EXAMPLES AND USECASES: ####################################################
**EXAMPLE 1.**
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~sas
options ls=max;
data test;
%LVarNmLab( peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s )
do %LVarNm( peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s ) = 1 to 10;
y = 5 + %LVarNm( peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s ) * 17;
output;
end;
run;
data test2;
set test;
where %LVarNm( peanut butter & jelly with a "Hot-Dog" in [a box], popcorn, and s*t*a*r*s ) < 5;
run;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
**EXAMPLE 2.**
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~sas
data test3;
%LVarNmLab() = 17;
%LVarNm() = 17;
%LVarNm( ) = 42;
%LVarNm( ) = 303;
run;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
**EXAMPLE 3.**
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~sas
data test3;
%LVarNm(test) = 1;
%LVarNm( test) = 2;
%LVarNm(test ) = 3;
run;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
**EXAMPLE 4.**
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~sas
data test4;
array X[*] %unquote(%LVarNm(some strange! name))_0 - %unquote(%LVarNm(some strange! name))_10;
do i = lbound(X) to hbound(X);
X[i] = 2**(i-1);
put X[i]=;
end;
run;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## >>> `%LVarNmLab()` macro: <<< <a name="lvarnmlab-macro"></a> #######################
The LVarNmLab() macro function supports LVarNm() and allows to remember "user names" in labels.
The motivation for the macro was similar one as for the LDSN() macro.
---
See examples in LVarNm() documentation for the details.
The `%LVarNmLab()` macro executes like a pure macro code.
**Known "Limitations":**
- variable name _cannot_ contain unpaired quotes (`'` and `"`),
text: `a "hot-dog"` is ok, but `John's dog` is not!
**Behaviour:**
- variable name text is *converted to upcase*
- variable name text *leading and trailing spaces are ignored*,
e.g. the following will give the same hash digest:
`%LVarNmLab(test)`, `%LVarNmLab( test)`, `%LVarNmLab(test )`.
### SYNTAX: ###################################################################
The basic syntax is the following, the `<...>` means optional parameters:
~~~~~~~~~~~~~~~~~~~~~~~sas
%LVarNmLab(
arbitrary text string (in line with limitations)
)
~~~~~~~~~~~~~~~~~~~~~~~
---
## License ####################################################################
Copyright (c) 2020 Bartosz Jablonski

BIN
packages/baseplus.zip
View File

Binary file not shown.