Thanks @Patrick and @Ksharp  That works. Annoying that the documentation page I linked to does not mention DBSASTYPE. For what it's worth, this is what the Import statement in the macro looks like now, and it works. So that has spared me rewriting the macro to use the LIBNAME with PROC SQL. PROC IMPORT OUT=WORK.D_&newtable._&n. TABLE="&table." DBMS=ACCESS REPLACE; %IF &newtable. EQ DDI_Info %THEN %DO; DBDSOPTS="DBSASTYPE=('Drug_Factors'='CHAR(255)' 'Patient_Factors'='CHAR(255)' 'Clinical_Factors'='CHAR(255)')"; %END; DATABASE=&dfile.; RUN; ... View more

Tom, the documentation notes: Restrictions SCANMEMO= applies only to MEMO data type columns; it does not apply to TEXT type columns. ... View more

I'd still be interested to hear suggestions, but I realised I could use this approach for my macro instead: Create a SAS/ACCESS LIBNAME to the database. Read the table (using SAS name literals) and write it to a SAS dataset using PROC SQL.   A quick test shows that these variables are now reading properly from at least one dataset when using the LIBNAME.   I may post some basic code when I am done. ... View more

Why, though, does linestyle work with the CNS names, and not with the RGB values? ... View more

Hi @ballardw    I have been struggling to express what I want. So ignore the Subject of the  thread (in fact, I will rename it to be less confusing). Also, thanks for spotting the CXRRGGBB mistake, that was, of course, a column header that got copied in. The first row, of course, should be CX000000.   So let me try to state the problem all over again. I have multiple plots for which I want to use an attribute map, this one is an example: DATA ORs; INPUT YLabel $ Level E L U IndentWt GrpText GrpLine Ref _N; DATALINES; Death 1 . . . 0 0 0 1 1 Intervention 2 0.55913 0.47636 0.6563 1 1 1 . 2 Days 2 2.80867 2.56724 3.0728 1 2 2 . 3 Sex 2 0.76004 0.64817 0.8912 1 3 3 . 4 Age 2 6.26008 3.89077 10.0722 1 4 4 . 5 Ward 2 1.67351 1.33596 2.0963 1 5 5 . 6 ED 2 4.11632 3.2314 5.2436 1 6 6 . 7 Lactate 1 . . . 0 0 0 8 8 Intervention 2 0.87583 0.70427 1.0892 1 1 1 . 9 Days 2 3.31815 2.99353 3.678 1 2 2 . 10 Sex 2 0.41019 0.33098 0.5084 1 3 3 . 11 Age 2 0.66652 0.51794 0.8577 1 4 4 . 12 Ward 2 0.44343 0.35548 0.5532 1 5 5 . 13 ED 2 1.23311 0.98223 1.5481 1 6 6 . 14 BP 1 . . . 0 0 0 15 15 Intervention 2 0.69487 0.62944 0.7671 1 1 1 . 16 Days 2 2.49906 2.37615 2.6283 1 2 2 . 17 Sex 2 0.7858 0.71351 0.8654 1 3 3 . 18 Age 2 1.14297 1.00717 1.2971 1 4 4 . 19 Ward 2 0.59325 0.53268 0.6607 1 5 5 . 20 ED 2 1.2314 1.10566 1.3714 1 6 6 . 21 eGFR 1 . . . 0 0 0 22 22 Intervention 2 0.82326 0.76245 0.8889 1 1 1 . 23 Days 2 3.0386 2.91297 3.1697 1 2 2 . 24 Sex 2 0.98662 0.91651 1.0621 1 3 3 . 25 Age 2 4.25326 3.68021 4.9155 1 4 4 . 26 Ward 2 0.80556 0.73838 0.8789 1 5 5 . 27 ED 2 2.3231 2.13229 2.531 1 6 6 . 28 ; RUN; PROC SGPLOT NOAUTOLEGEND NOBORDER NOCYCLEATTRS NOWALL DATTRMAP=_AttrMap DATA=ORs; /* It works to use GB hex values in STYLEATTRS */ /* STYLEATTRS DATACONTRASTCOLORS=(CXF78874 CXDE90BD CX7CAFD7 CX3FBDA5 CX92B85B CXE39F4F) DATALINEPATTERNS=(1);*/ XAXIS DISPLAY=(NOLABEL) LABELATTRS=(SIZE=8) VALUES=(0.25 0.5 1 to 12 by 1) VALUEATTRS=(SIZE=6) TYPE=Log; YAXIS REVERSE TYPE=DISCRETE DISPLAY=NONE OFFSETMIN=0.01; YAXISTABLE YLabel /LOCATION=INSIDE POSITION=LEFT TEXTGROUP=GrpText TEXTGROUPID=text NOLABEL INDENTWEIGHT=IndentWt SEPARATOR; REFLINE 1 /AXIS=X LINEATTRS=(COLOR=BLACK); REFLINE Ref /AXIS=Y LINEATTRS=(COLOR=LIGHTBLUE) TRANSPARENCY=0.8 DISCRETETHICKNESS=0.8; SCATTER Y=_N X=E /MARKERATTRS=(SIZE=6 SYMBOL=SQUARE) ATTRID=line GROUP=GrpLine; HIGHLOW Y=_N LOW=L HIGH=U /ATTRID=line GROUP=GrpLine; TITLE; RUN; If I define the colours using the STYLEATTRS statement (commented out above), then it works, but this defeats the purpose of using an attribute map to simplify the creation of numerous charts using a consistent colour scheme. If I create an attribute map using CNS colour names, the SGPLOT statement works: DATA _AttrMap; INPUT ID $ Value TextColor $ TextSize TextWeight $ TextStyle $ linecolor $ fillcolor $ linepattern; DATALINES; text 0 black 7 bold normal . . 1 text 1 red 7 normal italic . . 1 text 2 pink 7 normal italic . . 1 text 3 purple 7 normal italic . . 1 text 4 yellow 7 normal italic . . 1 text 5 blue 7 normal italic . . 1 text 6 green 7 normal italic . . 1 line 0 black . . . black black 1 line 1 red . . . red red 1 line 2 pink . . . pink pink 1 line 3 purple . . . purple purple 1 line 4 yellow . . . yellow yellow 1 line 5 blue . . . blue blue 1 line 6 green . . . green green 1 ; RUN; However, if I recreate the attribute map using RGB colours (this time using arbitrary colour codes from the list created by @Ksharp , just to ensure that I have only valid codes): DATA _AttrMap; INPUT ID $ Value TextColor $ TextSize TextWeight $ TextStyle $ linecolor $ fillcolor $ linestyle; DATALINES; text 0 CX000000 7 bold normal . . 1 text 1 CX0000FF 7 normal italic . . 1 text 2 CXD2691E 7 normal italic . . 1 text 3 CX7FFF00 7 normal italic . . 1 text 4 CX5F9EA0 7 normal italic . . 1 text 5 CXDC143C 7 normal italic . . 1 text 6 CX00FFFF 7 normal italic . . 1 line 0 CX000000 . . . CX000000 CX000000 1 line 1 CX0000FF . . . CX0000FF CX0000FF 1 line 2 CXD2691E . . . CXD2691E CXD2691E 1 line 3 CX7FFF00 . . . CX7FFF00 CX7FFF00 1 line 4 CX5F9EA0 . . . CX5F9EA0 CX5F9EA0 1 line 5 CXDC143C . . . CXDC143C CXDC143C 1 line 6 CX00FFFF . . . CX00FFFF CX00FFFF 1 ; RUN;   Then the full log is just a repetition of what I copied in the original post, but to satisfy everyone's curiosity: ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. ERROR 22-322: Syntax error, expecting one of the following: ;, (, FILLATTRS, LINEATTRS, MARKERATTRS, TEXTATTRS. ERROR 200-322: The symbol is not recognized and will be ignored. WARNING: Object will not be saved. NOTE: PROCEDURE SGPLOT used (Total process time): real time 0.01 seconds cpu time 0.01 seconds NOTE: The SAS System stopped processing this step because of errors. NOTE: There were 14 observations read from the data set WORK._ATTRMAP. @ballardw I find it interesting that your example works, but the plot is slightly different, it uses textcolor, whereas I am using textcolor. linecolor and fillcolor. Why does my example attribute map work when those three are defined with CNS names, but not with RGB codes? ... View more

Hi Jos As I said in my original post, the error is generated when I run an SGPLOT step. So, for example, here is some data: data ORs; INPUT YLabel $ Level E L U IndentWt GrpText GrpLine Ref _N; DATALINES; Death 1 . . . 0 0 0 1 1 Intervention 2 0.55913 0.47636 0.6563 1 1 1 . 2 Days 2 2.80867 2.56724 3.0728 1 2 2 . 3 Sex 2 0.76004 0.64817 0.8912 1 3 3 . 4 Age 2 6.26008 3.89077 10.0722 1 4 4 . 5 Ward 2 1.67351 1.33596 2.0963 1 5 5 . 6 ED 2 4.11632 3.2314 5.2436 1 6 6 . 7 Lactate 1 . . . 0 0 0 8 8 Intervention 2 0.87583 0.70427 1.0892 1 1 1 . 9 Days 2 3.31815 2.99353 3.678 1 2 2 . 10 Sex 2 0.41019 0.33098 0.5084 1 3 3 . 11 Age 2 0.66652 0.51794 0.8577 1 4 4 . 12 Ward 2 0.44343 0.35548 0.5532 1 5 5 . 13 ED 2 1.23311 0.98223 1.5481 1 6 6 . 14 BP 1 . . . 0 0 0 15 15 Intervention 2 0.69487 0.62944 0.7671 1 1 1 . 16 Days 2 2.49906 2.37615 2.6283 1 2 2 . 17 Sex 2 0.7858 0.71351 0.8654 1 3 3 . 18 Age 2 1.14297 1.00717 1.2971 1 4 4 . 19 Ward 2 0.59325 0.53268 0.6607 1 5 5 . 20 ED 2 1.2314 1.10566 1.3714 1 6 6 . 21 eGFR 1 . . . 0 0 0 22 22 Intervention 2 0.82326 0.76245 0.8889 1 1 1 . 23 Days 2 3.0386 2.91297 3.1697 1 2 2 . 24 Sex 2 0.98662 0.91651 1.0621 1 3 3 . 25 Age 2 4.25326 3.68021 4.9155 1 4 4 . 26 Ward 2 0.80556 0.73838 0.8789 1 5 5 . 27 ED 2 2.3231 2.13229 2.531 1 6 6 . 28 ; RUN; And here is the plot: PROC SGPLOT NOAUTOLEGEND NOBORDER NOCYCLEATTRS NOWALL DATTRMAP=_AttrMap DATA=ORs; /* It works to use GB hex values in STYLEATTRS */ /* STYLEATTRS DATACONTRASTCOLORS=(CXF78874 CXDE90BD CX7CAFD7 CX3FBDA5 CX92B85B CXE39F4F) DATALINEPATTERNS=(1);*/ XAXIS DISPLAY=(NOLABEL) LABELATTRS=(SIZE=8) VALUES=(0.25 0.5 1 to 12 by 1) VALUEATTRS=(SIZE=6) TYPE=Log; YAXIS REVERSE TYPE=DISCRETE DISPLAY=NONE OFFSETMIN=0.01; YAXISTABLE YLabel /LOCATION=INSIDE POSITION=LEFT TEXTGROUP=GrpText TEXTGROUPID=text NOLABEL INDENTWEIGHT=IndentWt SEPARATOR; REFLINE 1 /AXIS=X LINEATTRS=(COLOR=BLACK); REFLINE Ref /AXIS=Y LINEATTRS=(COLOR=LIGHTBLUE) TRANSPARENCY=0.8 DISCRETETHICKNESS=0.8; SCATTER Y=_N X=E /MARKERATTRS=(SIZE=6 SYMBOL=SQUARE) ATTRID=line GROUP=GrpLine; HIGHLOW Y=_N LOW=L HIGH=U /ATTRID=line GROUP=GrpLine; TITLE; RUN; ... View more

Thanks for that, but you will notice that the first hyperlink I provided was https://support.sas.com/documentation/cdl/en/graphref/63022/HTML/default/viewer.htm#colors-specify-color.htm#colors-define-list  Which is the web page that lists those macros, and yes, I had looked at them, but there is no macro to convert RGB to CNS. Even if there were, that was not my question, as I pointed out in my previous reply. My question is this: I can create an attribute map, but when I use it in an SGPLOT step, it generates an error. It seems as if attribute maps either can only contain CNS colour names when used in plot statements, or the RGB (or HLS) colours need to specified in some way that I cannot figure out. I have not been able to find any documentation explicitly stating that attribute maps must use CNS colour names, but all the examples I could find only use CNS colour names. ... View more

Thank you, that looks useful, but on my system, I am only getting 151 colours, and lists like this are actually available on the web. It does not, for example, contain the list of colours I want to use. In the end, I was asking how to use the RGB codes in the attribute map, not how to find CNS equivalents for them. ... View more

Thanks for your response. That helped me see the blindingly obvious. I was using a format to make the missing values "visible" and so I just dropped the format, and hard-coded an actual value in the place of missing values if missing(gender) then gender='_missing_'; Without the format, it works as it should. I should have seen that, but I think I got myself thinking in circles and not seeing the obvious. ... View more