About craig159753

craig159753 · ‎10-22-2024

Thanks for doing this, I will accept this as the solution, but once you get an update about the ticket and a patch, could you let me know please.

craig159753 · ‎10-22-2024

Hi Patrick, Thank you after further investigation, I came to the same resolution as you, my SAS ODA server's source code for XPT2LOC was ever so slightly different to my server's source code for XPT2LOC. My Server's Source Code process_data_record:; file metadata; if v6comp or nobs > 0 then do; file metadata; put '/* INPUT STATEMENT HERE */'; file inptstmt; put '/* IN DATA_RECORD */'; put '@@;'; if v6comp then put 'if not &obs_mem' memnum z4. ' then stop;'; put ' output; if _n_=&obs_mem' memnum z4. ' then stop;'; put 'run;'; end; else do; file metadata; put '/* INPUT STATEMENT HERE */'; put 'run;'; file inptstmt; put ';stop;'; put 'run;'; end; firstobs=recnum+1; retain firstobs; return; run; The ODA Server's Source Code process_data_record:; file metadata; put '/* INPUT STATEMENT HERE */'; file inptstmt; put '@@;'; put ' output; if _n_=&obs_mem' memnum z4. ' then stop;'; put 'run;'; firstobs=recnum+1; retain firstobs; return; run; The quickest and easiest solution is to edit our XPT2LOC.sas macro and replace that single line: put ';stop;'; With this line: put ';if _n_=&obs_mem' memnum z4. ' then stop;'; But I believe the way NOBS is being calculated in the data step above is incorrect and the empty dataset code is executing by mistake. To help, I have attached the three files created by XPT2LOC macro - INPTSTMT, METADATA and SASCODE.

craig159753 · ‎10-22-2024

Yes, the file is unzipped (the only reason it is attached as a zip here is because the formats xpt and xptv8 are not supported on this website).

craig159753 · ‎10-21-2024

Hi, I have a version 8 transport file (xptv8) and I am using the XPT2LOC autocall macro to convert the transport file into a SAS dataset. Now, previously I was calling this XPT2LOC macro inside SAS Enterprise Guide 8.3 on the SAS On Demand for Academics (ODA) server, which is running SAS 9.4 M8. This was working fine. However, since then, I have setup my own server, also running SAS 9.4 M8 but running a slightly newer version of SAS Enterprise Guide, version 8.4 (not that it likely matters but worth mentioning). Now on this new system, the code which ran without issues on the ODA server, is still running without any issues appearing in the SAS log but the dataset being output is empty. Below is the code I am running on both systems: options mprint mlogic symbolgen; libname source 'C:\Users\myuser\Documents'; %* Convert the XPT file to a SAS dataset; %xpt2loc( libref=source, memlist=_all_, filespec='C:\Users\myuser\Documents\ae.xptv8'); Now, I have done some testing and looking at the logs for both the system on the ODA server and the system on my new server, and I noticed one minor detail,. Within the XPT2LOC macro there is a dataset which reads in the XPT file using an INFILE statement. This data step processes the variables, defines their length, applies labels, etc. and then it outputs the dataset. The code from my new system (see sas_log_1.txt attached), the data step ends like this stop; run; But in the ODA system (see sas_log_2.txt attached) the same data step ends like this if _n_=29 then stop; run; This 29 is the number of records / observations in my XPT file. Now, this line is determined by the macro variable OBS_MEM0001, which is created in the first and very large data step. In the old ODA system this macro variable revolves to 29, but in my new system it resolves to 0. I believe something is making my new system not see the number of records in my XPT file and I do not know why. Is this a bug? I have attached the xpt I am working with and the SAS logs from my new system. Any insight would be greatly appreciated.

craig159753 · ‎09-28-2022

This does help, I can also use the log axis wit the regression plot too. Thank you.

craig159753 · ‎09-27-2022

Hi, I am trying to plot a SCATTERPLOT with a slope, via LINEPARM. This works perfectly for a linear axis, but does not like it when I change the Y axis to a logarithmic axis. Can anyone help? proc template; define statgraph layoutoverlay; %* Define Dynamic labels; dynamic xlabel ylabel ; %* Begin the graph template; begingraph / border = false designwidth = 900px designheight = 530px; %* Title; entrytitle "My title " / textattrs = (family = "Courier new" weight=bold size=10); layout overlay / /* X axis */ xaxisopts = (label = xlabel labelattrs = (family = "arial" weight=bold size=10) labelfitpolicy = splitalways labelsplitchar = "|" display = (label ticks tickvalues) type = linear linearopts = (viewmin=0 viewmax=50) tickvalueattrs = (family = "arial" size=9) offsetmin = 0.05 offsetmax = 0.05) /* Y axis */ yaxisopts = (label = ylabel labelattrs = (family = "arial" weight=bold size=10) labelfitpolicy = splitalways labelsplitchar = "|" display = (label ticks tickvalues) type = log logopts=(base=10 tickintervalstyle=logexpand thresholdmin=1 thresholdmax=1) tickvalueattrs = (family = "arial" size=9) offsetmin = 0.05 offsetmax = 0.05); %* Line Plot; scatterplot x=dose y=result / group = mygroup groupdisplay = overlay name = "scatter"; %* Diagonal line; lineparm x=0 y=intercept_log slope=slope_log / clip=true; %* Legend; discretelegend "scatter" / title = "Treatment: " titleattrs = (family = "arial" weight=bold size=10) valueattrs = (family = "arial" size=9) location = outside halign = center valign = bottom; endlayout; endgraph; end; run; %* Produce output; proc sgrender data = my_data template = layoutoverlay; dynamic xlabel = "X Axis" ylabel = "Y Axist"; run; When running this with the Y axis defined as log I get WARNING: LINEPARM statement has a conflict with the axis type. After some googling I could only find this, which talks about "working with transformed values", but I am merely the programmer not the statistician. As a note, my INTERCEPT=5.05 and SLOPE=4.17 which come directly from a PROC REG which used values converted using the LOG10 function, and ALPHA=0.1. Any help would be greatly appreciated.

craig159753 · ‎09-13-2022

Hi, I have a trivial question, considering the following dataset (MY_DATASET): /* Variable1 | Variable2 | Variable3 | Variable4 | Variable5 | Page | PageNumbering | ---------------------------------------------------------------------------------- xxxx1 | AAAAA | YY | LLL | W | 1 | Page 1 of 8 | xxxx2 | AA | YYY | LLLLLL | WWWWWWWW | 1 | Page 1 of 8 | xxxx3 | AAA | | L | WWW | 1 | Page 1 of 8 | xxxx4 | AAAAA | YY | LLL | W | 2 | Page 2 of 8 | xxxx5 | AA | YYY | LLLLLL | WWWWWWWW | 2 | Page 2 of 8 | xxxx6 | AAA | | L | WWW | 2 | Page 2 of 8 | ... */ Consider the following PROC REPORT code: proc report data=my_dataset headline nowd split="@" missing spacing=1; column variable1 variable2 variable3 variable4 variable5; by page; %* ID Variable(s); define variable1 / order order=internal id; %* Page 1; define variable2 / display; define variable3 / display; %* Page 2; define variable4 / display; define variable5 / display; compute before _page_; line @1 pagenumbering $200.; encomp; run; Note: the code will produce an ODS RTF output. I have left this code off for simplicity. As you can see, the code above will produce a "wide" PROC REPORT. What I mean by this, the output will produce two pages for every value of the PAGE variable: Page 1 of x Page=1 PageNumbering=Page 1 of 8 Variable1 Variable2 Variable3 Page 2 of x Page=1 PageNumbering=Page 1 of 8 Variable1 Variable4 Variable5 Variable1 being the ID variable appearing on all pages. Variable2 and Variable3 appearing on the first page. Variable4 and Variable5 appearing on the second page. The Issue: The current code will print a line of text above the table on each page via the COMPUTE block. However, as the output is wide the page numbering is repeating, and I know why, but I was curious is there a nice solution, that I do not know about, which will give me the correct page numbering per page. Page 1 of 16 Variable1 Variable2 Variable3 Page 2 of 16 Variable1 Variable4 Variable5 Notice the Pagenumbering variable has changed 8 to 16 (due to there are technically 2 pages per page as the output is wide) and that the page number is counting correctly. Note, I have created two new PageNumbering variables, so I have these at my disposal too: /* Variable1 | Variable2 | Variable3 | Variable4 | Variable5 | Page | PageNumbering | PageNumbering1 | PageNumbering2 | ------------------------------------------------------------------------------------------------------------------------- xxxx1 | AAAAA | YY | LLL | W | 1 | Page 1 of 8 | Page 1 of 16 | Page 2 of 16 | xxxx2 | AA | YYY | LLLLLL | WWWWWWWW | 1 | Page 1 of 8 | Page 1 of 16 | Page 2 of 16 | xxxx3 | AAA | | L | WWW | 1 | Page 1 of 8 | Page 1 of 16 | Page 2 of 16 | xxxx4 | AAAAA | YY | LLL | W | 2 | Page 2 of 8 | Page 3 of 16 | Page 4 of 16 | xxxx5 | AA | YYY | LLLLLL | WWWWWWWW | 2 | Page 2 of 8 | Page 3 of 16 | Page 4 of 16 | xxxx6 | AAA | | L | WWW | 2 | Page 2 of 8 | Page 3 of 16 | Page 4 of 16 | ... */ So Variable2 and Variable3 would use PageNumbering1 and Variable4 and Variable5 would use PageNumbering2. Things I have already tried (some work): Looping the PROC REPORT per page (works but very slow on large datasets) Using the ODS RTF ^{this page} stuff (works but when I combine outputs the page numbering is linked to the entire document and not the output itself). Any thoughts welcome.

craig159753 · ‎01-10-2022

Thanks for commenting, please see my response to Cynthia above.

craig159753 · ‎01-10-2022

Hi Cynthia, These reason I have opted for TAGESET.RTF over RTF is a trivial reason, notice in the images above, my first title is split into two titles, one of the left and the other on the right. As I require the TITLES and FOOTNOTES in the body of the output I am also using the BODYTITLE option. I have tried several "solutions" none of which have worked. Solution 1 (using RTF code): title1 j=left "^R'\ql 'Some text^R'\qr 'Some other text"; Soltuin 2 (using JUST twice in the TITLE statement): title1 j=left "some text" j=right "some other text"; Both of which did not work, with solution 1 having the text on the left and solution 2 having the text on the right. Any ideas?

craig159753 · ‎01-07-2022

Hi Ksharp, Thanks for replying, I should have added, currently I am using the following options on the ODS line options(vspace="off" tables_off="off") Toggling these has no effect on my issue.

craig159753 · ‎01-07-2022

Hi All, I hope someone can help me, I am producing some RTF outputs destined to be opened in Microsoft Word, I have opted to use the ODS TAGSETS.RTF method. It works as in tended but I have started to notice odd inconsistencies between some of the outputs. Now the titles and footnotes are printed within the body of the word document and not in the header and footer space found in Word, this is exactly what I want. The issue I have is with the space after the footnotes of my output and the footer space in word. There seems to be a large white gap in some outputs but not in others. Let me explain, take this output: Example 1 As you can see the is a normal-ish gap after the table. But consider this output: Example 2 - spilling The gap between the table and footer is a lot bigger, and the table is spilling as denoted by the "Continued" text. It is worth noting the next page is a single line, which looking at the first example would fit on the first page without spilling. My question is, why is the second output spilling when there is clearly space? Is there an option which controls this white space? I am using margins: leftmargin=3.05cm rightmargin=2.84cm topmargin=2.54cm bottommargin=2.54cm Both outputs use the same style. I cannot post the full code, but both programs use near identical PROC REPORT (with the only difference being the column used and the labels. Below is some sample code, I have had to censor aspects of it. proc report data=final split = "@" nowd missing headline; column xxx0 xxx1-xxx8; define xxx0 / order order=internal noprint; define xxx1 / order order=internal noprint; define xxx2 / order order=internal "XXXXXXX" style(header)=[just=l cellwidth=20%] style(column)=[just=l] id; define xxx3 / order order=internal noprint; define xxx4 / order order=internal "XXXXXXX" style(header)=[just=l cellwidth=14.5%] style(column)=[just=l] id; define xxx5 / display "XXXXXXX" style(header)=[just=c cellwidth=16%] style(column)=[just=c vjust=top asis=on pretext="^R'\ql\li150 '"]; define xxx6 / display "XXXXXXX" style(header)=[just=c cellwidth=16%] style(column)=[just=c vjust=top asis=on pretext="^R'\ql\li150 '"]; define xxx7 / display "XXXXXXX" style(header)=[just=c cellwidth=16%] style(column)=[just=c vjust=top asis=on pretext="^R'\ql\li150 '"]; define xxx8 / display "XXXXXXX" style(header)=[just=c cellwidth=16%] style(column)=[just=c vjust=top asis=on pretext="^R'\ql\li150 '"]; compute before xxx0 / style = [just=l font=("Courier New", 4pt)]; line " "; endcomp; compute after xxx1 / style = [just=l font=("Courier New", 4pt)]; line " "; endcomp; break after xxx1 / page; run; Note: I am using a custom style, so I am guessing there could be an option which controls this space.

craig159753 · ‎08-02-2021

Hi Paige and Tom, Thank you for your help, using the SYSMACDELETE it works, so I made a rough macro to delete specifically the work macros, as I do not need any of them if I am changing the SASAUTOS option. %* Reset / Remove any work macro(s); %macro m_reset(); %local m_work_macros m_work_macro; %let m_work_macros=; %let m_work_macro=; %* Fetch work macro(s) previously compiled; proc sql noprint; select distinct strip(objname) into :m_work_macros separated by " " from sashelp.vcatalg where upcase(objtype)="MACRO" and upcase(libname)="WORK" and compress(upcase(objname)) ne "M_RESET"; quit; %* Loop through work macro(s) previously compiled and remove them; %if %length(&m_work_macros.) > 0 %then %do; %do i = 1 %to %sysfunc(countw(&m_work_macros.)) %by 1; %* Define macro; %let m_work_macro = %scan(&m_work_macros., &i.); %* Ignore current macro / null terms; %if %upcase(&m_work_macro.) ne M_RESET and %length(&m_work_macro.)>0 %then %do; %put NO%STR(TE:) Following work macro removed - %upcase(&m_work_macro.); %sysmacdelete &m_work_macro. / nowarn; %end; %end; %end; %mend m_reset; %m_reset;

craig159753 · ‎08-02-2021

Hi Paige, So you are suggesting the following: options mrecall mautosource sasautos=("!sasroot\core\sasmacro" "C:\M_LIB1"); %put %sysfunc(getoption(sasautos)); %mymacro; options mrecall mautosource sasautos=("!sasroot\core\sasmacro" "C:\M_LIB2"); %put %sysfunc(getoption(sasautos)); %mymacro; Which yes does show two different SASAUTOS options, however the macro called is from M_LIB1 exclusively, it seems the second location does not take priority even when SASAUTOS is redefined. I have also tried the following %sysmstoreclear; proc catalog catalog = work.sasmacr kill force; quit; Which would clear the work areas, and still no luck

craig159753 · ‎08-02-2021

Hi, I am using the SASAUTOS option to recall macros, without having to use the %INCLUDE statements and this working perfectly, but when I decide to change the SASAUTOS option mid-session it does not change. Now, this causes an issue because imagine there were two macro locations, call them M_LIB1 and M_LIB2, which both contain the macro MYMACRO. When I open SAS and define the SASAUTOS to look at the MYMACRO in M_LIB1, it works as expected. Then I decide to change the SASAUTOS location to M_LIB2, during my current session, (i.e., I do not close SAS), the SASAUTOS does not seem to change and when I call MYMACRO expecting to be recalled from M_LIB2, it is actually being recalled from M_LIB1. Below is some basic an example code, where I define the SASAUTOS location, call the macro then print the SASAUTOS locations to the log, then change the SASAUTOS location and repeat: options mrecall mautosource set = sasautos("!sasroot\core\sasmacro" "C:\m_lib1"); %mymacro; filename sasautos list; options mrecall mautosource set = sasautos("!sasroot\core\sasmacro" "C:\m_lib2"); %mymacro; filename sasautos list; The issue is that the FILENAME code with the LIST option displays the same results for both which implies SASAUTOS is not changing, I was expecting the second statement to show the M_LIB2 path but it shows the M_LIB1 path. Can anyone help?

craig159753 · ‎07-19-2021

Hi Tom, Yeah this method does work, and for the definitions to take over a macro say in both a catalogue and in a SAS program the entry would need to be removed from the catalogue first, or by using the %INCLUDE statement like you suggested. After a reboot it is working now. Turns out the %sysget(sasautos) was causing an issue it returns with parenthesis which meaning the original SASAUTOS option was not correctly defined, once I resolved this I first had to reboot SAS. Then it worked.

Online Status	Offline
Date Last Visited	‎10-23-2024 11:57 AM

Re: XPT2LOC Producing Empty Dataset (V8)

Re: XPT2LOC Producing Empty Dataset (V8)

Re: XPT2LOC Producing Empty Dataset (V8)

XPT2LOC Producing Empty Dataset (V8)

Re: GTL: LOGPARM with Logarithmic Axis

GTL: LOGPARM with Logarithmic Axis

Wide PROC REPORT and Page Numbering

Re: TAGSETS.RTF Space After Table Inconsistent

Re: TAGSETS.RTF Space After Table Inconsistent

Re: TAGSETS.RTF Space After Table Inconsistent

Re: XPT2LOC Producing Empty Dataset (V8)

Re: Boxplot Statement in SAS GTL and the output data set

Re: XPT2LOC Producing Empty Dataset (V8)

Re: XPT2LOC Producing Empty Dataset (V8)

Re: XPT2LOC Producing Empty Dataset (V8)

XPT2LOC Producing Empty Dataset (V8)

Re: GTL: LOGPARM with Logarithmic Axis

GTL: LOGPARM with Logarithmic Axis

Wide PROC REPORT and Page Numbering

Re: TAGSETS.RTF Space After Table Inconsistent

Re: TAGSETS.RTF Space After Table Inconsistent

Re: TAGSETS.RTF Space After Table Inconsistent

TAGSETS.RTF Space After Table Inconsistent

Re: SAS Macros: Changing/Updating SASAUTOS During SAS Session

Re: SAS Macros: Changing/Updating SASAUTOS During SAS Session

SAS Macros: Changing/Updating SASAUTOS During SAS Session

Re: SAS Macro Not Resolving with MAUTOSOURCE / MRECALL / SASAUTOS Defi...