I've trying to send a JSON file through SAS Enterprise Guide, But I believe that I'm doing some mistake on code.
Below is there my code:
>> GENERATE JSON FILE:
data teste30;
set MATABLES.TEMPJSON;
RESP=cat(CUSTOMERID,"|",RESPTRACKING_CD);
CODSMS="WXS";
MOBILE="5511111111";
DATAPARA="09/05/2019";
DATALIMI="10/05/2019";
REMETENTE="TF";
run;
filename code temp;
data _null_;
set teste30;
file code ;
put 'WRITE OPEN OBJECT;'
/ 'WRITE VALUES "TP_SMS" ' CODSMS :$quote. ';'
/ 'WRITE VALUES "NM_REMETENTESMS" ' REMETENTE :$quote. ';'
/ 'WRITE VALUES "NR_TELEFONECELULARSMS" ' MOBILE :$quote. ';'
/ 'WRITE VALUES "TX_MENSAGEMSMS" ' msgtext :$quote. ';'
/ 'WRITE VALUES "DT_PARAENVIOSMS" ' DATAPARA :$quote. ';'
/ 'WRITE VALUES "DT_LIMITEENVIOSMS" ' DATALIMI :$quote. ';'
/ 'WRITE VALUES "DS_CHAVEORIGEMSMS" ' RESP :$quote. ';'
/ 'WRITE CLOSE;'
;
run;
proc json out="%sysfunc(getoption(WORK))/TEST.json" pretty nokeys nosastags;
write open array; /* container for all the data */
%include code;
write close; /* container for all the data */
run;
My JSON seems ok. The problem occours when I try to send it calling the API:
>> CALLING API POST
FILENAME POSTA "C:\TEMP\POSTA.TXT";
FILENAME code2 "%sysfunc(getoption(WORK))/test.json";
PROC HTTP
URL="HTTPS://*********/SMS/INCLUISMS"
CT="APPLICATION/JSON"
IN=code2
METHOD="POST"
OUT=POSTA;
HEADERS
"HOST"="*****"
"AUTHORIZATION"="BEARER xxxxxxxxxxxx"
"CONTENT-TYPE"="APPLICATION/JSON"
"CONTENT-LENGTH"="xx"
RUN;
%echoFile(fn=code2);
%PUT HTTP STATUS CODE = &SYS_PROCHTTP_STATUS_CODE. : &SYS_PROCHTTP_STATUS_PHRASE.;
The log on EG return this error:
ERROR: The tcpSockRead call failed. The system error is 'The connection was reset by a peer.'.
ERROR: Connection has been closed.
I tried input the path & filename (physical) on "IN=" (example: IN="C:\TEMP\Test.json) - without success.
Has anyone had a similar experience like this? How can I send the JSON (file) using "IN=" ? Is it possible?
Tks Guys!
When I haver using the "datafiles" with only one parameter, it is working normaly. I've success on API call. Below is there the code:
/* Put the body of the JSON content in external file */
filename json_in temp;
data _null_;
file json_in;
input;
put _infile_;
datalines;
{ "Tp_Email":"CCA001" }
run;
filename posta "c:\temp\posta.txt";
filename post1 temp;
proc http
url="https://***********/*****/***/ObterEmailTexto"
method="POST"
ct="application/x-www-form-urlencoded"
in=json_in
out=posta;
headers
"Host"="****"
"Authorization"="bearer &TOKEN"
"Content-Type"="application/json"
"Content-Length"="**"
"CD_Login"="*****";
Run;
/*
data _null_;
infile posta;
input;
put _infile_;
run;
*/
%put HTTP Status code = &SYS_PROCHTTP_STATUS_CODE. : &SYS_PROCHTTP_STATUS_PHRASE.;
When I tried pass the JSON file on body, occurred the error informed.
FILENAME POSTA "C:\TEMP\POSTA.TXT";
FILENAME code2 "%sysfunc(getoption(WORK))/test.json";
PROC HTTP java_http
URL="HTTPS://**********/*****/SMS/INCLUISMS"
ct="application/x-www-form-urlencoded"
IN=code2
METHOD="POST"
OUT=POSTA;
HEADERS
"HOST"="*****"
"AUTHORIZATION"="BEARER &TOKEN"
"CONTENT-TYPE"="APPLICATION/JSON"
"CONTENT-LENGTH"="**"
"CD_LOGIN"="******";
RUN;
%PUT HTTP STATUS CODE = &SYS_PROCHTTP_STATUS_CODE. : &SYS_PROCHTTP_STATUS_PHRASE.;
Witout java_http options: ERROR: The tcpSockRead call failed. The system error is 'The connection was reset by a peer.'. ERROR: Connection has been closed.
With java_http options: ERROR: Statement HEADERS is not supported in Java execution mode.
Tks. again!!!
TF
Related
SAS Program - I am trying to unzip all .gz type files in a folder and reading them to a dataset using filename statement. However, I am not able to make it work.
I have .gz files named as follows in a folder:
EQY_US_ALL_TRADE_20210701
EQY_US_ALL_TRADE_20210702
EQY_US_ALL_TRADE_20210705
EQY_US_ALL_TRADE_20210706
EQY_US_ALL_TRADE_20210707
.....
.....
EQY_US_ALL_TRADE_20210729
EQY_US_ALL_TRADE_20210730
so on.
Note that it does NOT have files for all 31 days in a folder - files are only for business days.
See my code below:
/* Change working directory to where all the files are located */
data _null_;
rc=dlgcdir("C:\EQY_US_ALL_TRADE_202107");
put rc=;
run;
/* using filename statement unzip all files and read them into "f1" */
filename f1 zip EQY_US_ALL_TRADE_202107* gzip lrecl=500;
/* This code worked when I had the actual name of one of the file - for e.g. "EQY_US_ALL_TRADE_20210702" but does not work when I use the wildcard to run through all of them */
You can read the file names in the folder using the DREAD function, and then a dynamic INFILE statement using the FILEVAR= option to specify the gunzip stream that will be INPUTed from.
Example:
All gzipped files are presumed to be data only and do not contain a header row. The compressed files are all in a single folder and have the .gz file extension
data want(keep=filename a b c);
length folderef $8;
rc = filename (folderef, 'c:\temp\trade_data');
did = dopen(folderef);
do _n_ = 1 to dnum(did);
filename = dread(did,_n_);
if scan(filename,-1,'.') ne 'gz' then continue;
fullname = pathname(folderef) || '/' || filename;
do while(1);
infile archive zip filevar=fullname gzip dlm=',' eof=nextfile;
input a b c;
OUTPUT;
end;
nextfile:
end;
stop;
run;
I would like to prepare a manifest file using Lambda and then execute the stored procedure providing input parameter manifest_location.
Stored procedure signature:
CREATE OR REPLACE PROCEDURE stage.sp_stage_user_activity_page_events(manifest_location varchar(256))
and I would like to use this parameter as follows:
COPY stage.user_activity_event
FROM manifest_location
IAM_ROLE 'arn:aws:iam::XXX:role/redshift-s3-read-only-role'
IGNOREHEADER 1
REMOVEQUOTES
DELIMITER ','
LZOP
MANIFEST;
but Redshift is giving me ERROR:
syntax error at or near "$1" Where: SQL statement in PL/PgSQL function "sp_stage_user_activity_page_events" near line 21
How can I achieve this?
Could you try the command below?
EXECUTE 'COPY '|| parameter_for_table ||' FROM '||CHR(39)|| parameter_for_s3_path ||CHR(39)||' IAM_ROLE '||CHR(39)|| parameter_for_iam_role ||CHR(39)||' FORMAT AS PARQUET;';
Please, let me know if you have any more questions, because I created one stored_procedure, as below today:
CREATE TABLE table_for_error_log (message varchar);
CREATE OR REPLACE procedure sp_proc_name(parameter_for_table in VARCHAR, parameter_for_s3_path in VARCHAR, parameter_for_iam_role in VARCHAR)
AS
$$
BEGIN
EXECUTE 'COPY '|| parameter_for_table ||' FROM '||CHR(39)|| parameter_for_s3_path ||CHR(39)||' IAM_ROLE '||CHR(39)|| parameter_for_iam_role ||CHR(39)||' FORMAT AS PARQUET;';
EXCEPTION WHEN OTHERS THEN
RAISE INFO 'An exception occurred.';
INSERT INTO table_for_error_log VALUES ('Error message: ' || SQLERRM);
END;
$$ LANGUAGE plpgsql;
It's already tested and worked.
Just replace it with your own command.
i don't have enough reputation to upvote the above answer but it works!.
you can also use the quote_literal function eg:
DATEFORMAT as ' || quote_literal('auto') ||
'TIMEFORMAT as '|| quoteliteral('MM/DD/YYYY HH24:MI:SS') ||
etc..
I need to control the data type when reading XML data in SAS. The XML data are written and accessed using the XML libname engine in SAS.
XML File :
<Test>
<origin>YYYY</origin>
<NumToUse>50503</NumToUse>
<AcctNum>3-219HHJLJ</AcctNum>
<Status>1</Status>
<TADIG>AUSVF</TADIG>
<LocationNumber>1234567891011</LocationNumber>
<Phnumber>1234567890</Phnumber>
<ReferenceNumber>0044E71146</ReferenceNumber>
Map File :
<COLUMN name="LocationNumber">
<PATH syntax="XPath">/Test/LocationNumber</PATH>
<TYPE>character</TYPE>
<DATATYPE>string</DATATYPE>
<LENGTH>11</LENGTH>
</COLUMN>
<COLUMN name="PhNumber">
<PATH syntax="XPath">/Test/PhNumber</PATH>
<TYPE>character</TYPE>
<DATATYPE>string</DATATYPE>
<LENGTH>15</LENGTH>
</COLUMN>
<COLUMN name="ReferenceNumber">
<PATH syntax="XPath">/Test/ReferenceNumber</PATH>
<TYPE>numeric</TYPE>
<DATATYPE>double</DATATYPE>
</COLUMN>
Since the Reference Number is treated as Numeric am not able get the value for that particular column .its giving me
ERROR: Data contains invalid content for float datatype. Invalid content is 0044E71146
How to read the data into SAS dataset ? suggestion pls
The map file created by the XMLV2 library can be modified before using the libref to copy data into your SAS session.
There are many ways to process the engine generated map file (which is an xml file itself)
XSL transform (Proc XSL)
Someone (not me) well versed in XSLT language could likely write a short program to perform the modifications
Programmatic manipulation of the parsed xml document
Textual manipulation of the xml file
Hand editing
Text processing program
The map file :
defines which xml nodes are to become data set columns
defines the type, length, format, label, etc, of the column
defines which tables are to be created
defines which tables are to contain which columns
For the case of wanting to change the XMLV2 engine's interpretation of numeric columns to character columns, the mapfile needs to be modified (aka transformed).
<COLUMN> nodes have children nodes that, at a minimum, need to be changed
from
<TYPE>numeric</TYPE>
to
<TYPE>character</TYPE>
Considerations
There are other considerations you might need to make when transforming numbers to character enmasse, such as:
Is this numeric a date ?
Should the number be rendered through it's format first ?
SAS users are comfortable with the concept of single 'data set' which contains both data and metadata (in the header). For XML data, the data is in one file and the metadata (the mapfile) is in another.
When a SAS data that is exported and only the data xml is kept, the metadata gets lost. Making the round trip back to SAS means the metadata must be guessed at via automap.
Example code for "Programmatic manipulation of the parsed xml document"
This example changes all column definitions for numeric columns to character. The XPath /SXLEMAP/TABLE/COLUMN[not(#class='ORDINAL') and ./TYPE[text()='numeric'] is used to identify the column definitions that will be changed. No further special considerations are made.
Create xml data file to be processed
%macro createXmlV2(data=,folder=);
%local lib mem;
%let syslast = &data;
%let lib = %scan(&syslast,1,.);
%let mem = %scan(&syslast,2,.);
FILENAME XMLOUT "&folder./&data..xml";
LIBNAME XMLOUT XMLV2;
proc copy in=&lib out=xmlout;
select &mem;
run;
LIBNAME XMLOUT clear;
FILENAME XMLOUT clear;
%mend;
%* Something to play with;
%* Create an XMLV2 generated xml file containing the data set;
%createXmlV2 (data=sashelp.citiday, folder=/temp)
%createXmlV2 (data=sashelp.citimon, folder=/temp)
%createXmlV2 (data=sashelp.baseball, folder=/temp)
%*;
Transform automap file so numeric columns become character columns
%macro prepXmlRefsFor(file=);
FILENAME XMLFILE "&file";
FILENAME MAPOUT "&file..map";
FILENAME MAPOUT2 "&file..map.transformed";
%mend;
%prepXmlRefsFor(file=/temp/sashelp.citiday.xml)
%prepXmlRefsFor(file=/temp/sashelp.baseball.xml)
%prepXmlRefsFor(file=/temp/sashelp.citimon.xml)
%*;
%* create an automap file using XMLV2 library engine;
LIBNAME XMLFILE XMLV2 XMLTYPE=XMLMAP XMLMAP=MAPOUT AUTOMAP=REPLACE ;
%* parse and rewrite the generated map file ;
%* change ALL non-ordinal, non-character COLUMN nodes to indicate character type wanted;
proc groovy;
submit
"%sysfunc(pathname(MAPOUT))"
"%sysfunc(pathname(MAPOUT2))"
"20"
;
import javax.xml.parsers.*;
import javax.xml.xpath.*;
import javax.xml.transform.*;
import javax.xml.transform.dom.*;
import javax.xml.transform.stream.*;
// get parameter from submit line;
map_in=args[0]; // the automap
map_out=args[1]; // the automap transformed
length=args[2]; // length of character value for columns previously considered numeric
// parse mapfile;
doc = DocumentBuilderFactory.newInstance().newDocumentBuilder().parse(map_in);
xPath = XPathFactory.newInstance().newXPath();
// select the set of automap nodes that define non-ordinal, numeric columns
columns = xPath.evaluate(
"/SXLEMAP/TABLE/COLUMN" +
"[" +
"not(#class='ORDINAL')" +
" and " +
"./TYPE[text()='numeric']" +
"]", doc, XPathConstants.NODESET);
for (column in columns) {
type = xPath.evaluate("TYPE", column, XPathConstants.NODE);
dtyp = xPath.evaluate("DATATYPE", column, XPathConstants.NODE);
leng = xPath.evaluate("LENGTH", column, XPathConstants.NODE);
type.setTextContent("character");
dtyp.setTextContent("string");
if (leng == null)
column.appendChild(leng = doc.createElement("LENGTH"));
leng.setTextContent(length);
}
// rewrite mapfile with updated nodes
TransformerFactory.newInstance().newTransformer().transform(
new DOMSource(doc), new StreamResult(new File(map_out))
);
println "Programmatic transformation of mapfile completed.";
endsubmit;
quit;
* resubmit libname so libref uses transformed mapfile;
LIBNAME XMLFILE XMLV2 XMLTYPE=XMLMAP XMLMAP=MAPOUT2 AUTOMAP=REUSE;
proc copy in=xmlfile out=work;
run;
LIBNAME XMLFILE clear;
FILENAME XMLFILE clear;
FILENAME MAPOUT clear;
FILENAME MAPOUT2 clear;
One thing that became obvious after examining the 'round-trip' outcome is that xml files created by XMLV2, when re-read, will create separate column-named tables for any columns that contain missing values. These tables would have to be merged to recreate the original data set.
You might understand that the auto-mapping feature built into XMLV2 engine is choosing to define the ReferenceNumber as number, instead of as a character, is because the only one value the parser is examining is 0044E71146 and is presuming the #E# is scientific (or exponent) notation for a number.
The solution is to let the libname automap the data xml file and then update the map file xml to meet your requirements.
Example code:
XMLV2 engine creates MAPFILE, and Proc GROOVY is used to XML parse and rewrite the mapfile.
FILENAME XMLFILE "/temp/test.xml" ;
FILENAME MAPFILE "/temp/test.xml.map" ;
* parse data test.xml and write mapfile test.xml.map;
LIBNAME XMLFILE XMLV2 XMLTYPE=XMLMAP XMLMAP=MAPFILE AUTOMAP=REPLACE ;
* parse and rewrite mapfile;
* change desired column nodes to be string/character of a specified length;
proc groovy;
submit "%sysfunc(pathname(mapfile))";
import java.io.File;
import java.io.IOException;
import javax.xml.parsers.DocumentBuilder;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.parsers.ParserConfigurationException;
import javax.xml.transform.Transformer;
import javax.xml.transform.TransformerException;
import javax.xml.transform.TransformerFactory;
import javax.xml.transform.dom.DOMSource;
import javax.xml.transform.stream.StreamResult;
import javax.xml.xpath.XPathFactory;
import javax.xml.xpath.XPathConstants;
import org.w3c.dom.Document;
import org.w3c.dom.Element;
import org.w3c.dom.NamedNodeMap;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.xml.sax.SAXException;
* get parameter from submit line;
mapfile=args[0];
* parse mapfile;
doc = DocumentBuilderFactory
.newInstance()
.newDocumentBuilder()
.parse(
mapfile
)
;
xPath = XPathFactory
.newInstance()
.newXPath()
;
void setCharacter(column,length) {
* find column node and child nodes important to XMLV2 mapfile usage;
node = xPath.evaluate("/SXLEMAP/TABLE/COLUMN[#name='"+column+"']", doc, XPathConstants.NODE);
type = xPath.evaluate("TYPE", node, XPathConstants.NODE);
dtyp = xPath.evaluate("DATATYPE", node, XPathConstants.NODE);
leng = xPath.evaluate("LENGTH", node, XPathConstants.NODE);
if (type != null && !type.getTextContent().equals("character")) { type.setTextContent("character") }
if (dtyp != null && !dtyp.getTextContent().equals("string")) { dtyp.setTextContent("string") }
if (leng == null) {
leng = doc.createElement("LENGTH");
leng.setTextContent(length.toString());
node.appendChild(leng);
}
else
if (!length.getTextContent().equals(length.toString())) {
leng.setTextContent(length.toString());
}
}
// Make sure these two columns will be character, if not already
setCharacter("ReferenceNumber",25);
setCharacter("Phnumber", 20);
// rewrite mapfile with updated nodes
TransformerFactory
.newInstance()
.newTransformer()
.transform(
new DOMSource(doc),
new StreamResult(new File(mapfile))
);
endsubmit;
quit;
* resubmit libname so libref uses now updated mapfile;
LIBNAME XMLFILE XMLV2 XMLTYPE=XMLMAP XMLMAP=MAPFILE;
proc copy in=xmlfile out=work;
run;
Note: You could textually parse and rewrite the map file, however, there is a small outside chance the mapfile may not meet your 'text-parsing' expectations.
Let SAS make the map file.
FILENAME XMLFILE "/v/temp/test.xml" ;
FILENAME MAPFILE "/v/temp/test.xml.map" ;
LIBNAME XMLFILE XMLV2 XMLTYPE=XMLMAP XMLMAP=MAPFILE AUTOMAP=REUSE ;
Edit the file and fix the definition
<COLUMN name="ReferenceNumber">
<PATH syntax="XPath">/Test/ReferenceNumber</PATH>
<TYPE>character</TYPE>
<DATATYPE>string</DATATYPE>
<LENGTH>15</LENGTH>
</COLUMN>
You probably will want to save this in permanent and not a temporary location. Now use the fixed file to re-define the libref.
FILENAME XMLFILE "/v/temp/test.xml" ;
FILENAME MAPFILE "/v/permanent/fixed_xml.map" ;
LIBNAME XMLFILE XMLV2 XMLTYPE=XMLMAP XMLMAP=MAPFILE AUTOMAP=REUSE ;
So I've seen various pages that inform SAS Stored Process users creating stored process web pages how to do something like:
https://support.sas.com/documentation/cdl/en/stpug/62758/HTML/default/viewer.htm#p0q383asmm4bejn1i7z4q7iv5eke.htm
data _null_;
file _webout;
put '<HTML>';
put '<HEAD><TITLE>Hello World!</TITLE></HEAD>';
put '<BODY>';
put '<H1>Hello World!</H1>';
put '</BODY>';
put '</HTML>';
run;
The put HTML statements are nice and a quick way to get HTML to the page, but I'm noticing my HTML using put is getting quite messy and not maintainable.
How can I reference an external HTML file instead? I do have some SAS macros that I reference in the HTML put statements so I'm not sure if those would resolve in the external html but still interested to see a working example of referencing an external HTML file in a stored process.
UPDATE - My Example Program
In this put HTML I am just writing a form to the web page with some data populated in a drop down. I would just like to call this as raw html from an external HTML file while still resolving the SAS code embedded in the HTML like macro variable references.
%macro test;
proc sql;
create table eqmtid_name as
select distinct EQMT_ID
from library.joined_data;
%let neqmtid = &sqlobs;
quit;
data _null_;
set eqmtid_name;
suffix=put(_n_,8.);
call symput(cats('EQMTID',suffix), EQMT_ID);
run;
data _null_;
file _webout;
put '<script type="text/javascript">document.body.innerHTML = "";</script>';
put '<HTML>';
put '<head>';
put '<script type="text/javascript">';
put 'document.addEventListener("DOMContentLoaded", function initialize() {';
put ' var svgs = document.getElementsByTagName("svg");';
put ' for (i = 0; i < svgs.length; ++i) {';
put ' e = svgs[i];';
put ' e.removeAttribute("height");';
put ' e.removeAttribute("width");';
put ' e.setAttribute("width","80%");';
put ' }';
put ' ';
put '});';
put '</script>';
put '</head>';
put '<BODY onload="initialize()">';
put '<HEADER class="pageheader">';
put '<TABLE>';
put '<tr>';
put '<td>';
put '<div class="divcenter">Header Text</div>';
put '</td>';
put '</tr>';
put '</TABLE>';
put '</HEADER>';
put '<FORM NAME="sub" METHOD="GET" ACTION="/SASStoredProcess/do">';
put '<INPUT TYPE="HIDDEN" NAME="_program" SIZE="100" VALUE="/Company/IT/Development/APP/stp_program">';
put '<INPUT TYPE="CHECKBOX" NAME="_debug" VALUE="log">Show SAS Log <br>';
put '<TABLE>';
put '<tr>';
put '<td>';
put '<label class="formitem">Equipment ID:</label>';
put '</td>';
put '<td>';
put '<select class="formitem styleinput" name="EQMTID">' /
'<option value=""></option>' /
%do i = 1 %to &neqmtid; /
"<option value = %bquote(")%left(%trim(&&EQMTID&i))%bquote(") >&&EQMTID&i</option>" /
%end; /
'</select>';
put '</td>';
put '</TABLE>';
put '<br>';
put '<TABLE>';
put '<tr>';
put '<td>';
put '<INPUT class="button" class="formitem" TYPE="SUBMIT" VALUE="Search"><br>';
put '</td>';
put '</FORM>';
put '<FORM NAME="download" TARGET="_blank" METHOD="GET" ACTION="/SASStoredProcess/do">';
put '<INPUT TYPE="HIDDEN" NAME="_program" SIZE="100" VALUE="/Company/IT/Development/APP/export_data_stp">';
put '<td>';
put '<INPUT class="button" class="formitem" TYPE="SUBMIT" VALUE="Download"><br>';
put '</td>';
put '</FORM>';
put '</tr>';
put '</TABLE>';
put '</BODY>';
put '</HTML>';
run;
%mend;
%test;
UPDATE SOLVED
Thanks to the example Joe outlined below, this is what I had to code to make this work... in SAS Code:
filename htmlfile "/sasdata/IT/APP/file.html";
proc stream outfile=_webout;
begin
&streamdelim; %include htmlfile;
;;;;
run;
Note: in your html file if you have macro statements like %DO loops, you might get an error like "Error: The %DO statement is not valid in open code". In order to resolve that error I included a macro wrapper in the html file around that statement like this:
<select class="formitem styleinput" name="EQMTID">
<option value=""></option>
%macro doeqmt;
%do i = 1 %to &neqmtid;
"<option value = %bquote(")%left(%trim(&&EQMTID&i))%bquote(") >&&EQMTID&i</option>"
%end;
%mend;
%doeqmt;
</select>
Depending on your exact details, there are a few ways to do this, but the simplest two are either read the HTML into a data step (in a character variable, one per line) and then putting it back out again (this is better if you need to perform some modifications on it), or as Tom suggested you can use PROC STREAM if you don't need to modify it much or if you can insert macro variables (or macros) in the raw HTML pre-file.
PROC STREAM is pretty simple. Imagine I have a hello world type file, in c:\temp\helloworld.html
<html>
<header>
<title>My Example Page</title>
</header>
<body>
<h1> Hello, World </h1>
</body>
</html>
Then I could write this short program:
filename htmlfile "c:\temp\helloworld.html";
filename outfile "c:\temp\stream.html";
proc stream outfile=outfile;
begin
&streamdelim readfile htmlfile;
;;;;
run;
And I'd now have a new file, stream.html, with the same contents. And with any macro variables or macros in it resolved.
For example, let's change it just slightly:
Input:
<html>
<header>
<title>My Example Page</title>
</header>
<body>
<h1> &mystring. </h1>
</body>
</html>
Output, noting the new %let:
filename htmlfile "c:\temp\helloworld.html";
filename outfile "c:\temp\stream.html";
%let mystring = Hello, World;
proc stream outfile=outfile;
begin
&streamdelim readfile htmlfile;
;;;;
run;
Note that it resolves the macro variable for us.
You can just change the outfile to _WEBOUT and you should get what you're hoping for, if I understand you right.
The way we do this in the SAS Apps team is to compile the raw HTML into SAS programs prior to deployment.
We call this approach "streaming apps" and it can work on both SAS 9 (metadata) AND Viya. Here's a breakdown: https://sasapps.io/sas-streamed-apps
I am writing a deployment script, and would like to programmatically register a simple (and empty) BASE library, such as the one below, in Metadata.
libname MYLIB 'C:\temp';
Sample XML syntax can be found here.. Am just not sure how to combine that with proc metadata to perform the update (eg how do the metadata ID's get generated?)
#user2173800 Did u ever recieve a solution to the question above?
Here is what i came up with :
The below Code creates a SAS Library called BASE_Metalib under the Metadata
folder :/Shared Data/Libraries/BASE_Metalib (this folder is assumed to already exist in Metadata). The code also resgisters all tables under this Directory defined for this Library. The below code uses Metadata Datastep functions to Interface with metadata.
/*Creating a Metadata Library with BASE Engine and register all the tables under it */
options metaserver="taasasf2"
metaport=8561
metauser="testuser"
metapass="test123"
metarepository="Foundation";
%Let MetaLibName=BASE_Metalib; /* Name of the SAS Library with BASE Engine to be created */
data _null_;
length luri uri muri $256;
rc=0;
Call missing(luri,uri,muri);
/* Create a SASLibrary object in the Shared Data folder. */
rc=metadata_newobj("SASLibrary",
luri,
"&MetaLibname.",
"Foundation",
"omsobj:Tree?#Name=%bquote('&Metalibname.')",
"Members");
put rc=;
put luri=;
/* Add PublicType,UsageVersion,Engine,Libref,IsDBMSLibname attribute values. */
rc=metadata_setattr(luri,
"PublicType",
"Library");
put rc=;
put luri=;
rc=metadata_setattr(luri,
"UsageVersion",
"1000000.0");
put rc=;
put luri=;
rc=metadata_setattr(luri,
"Engine",
"BASE");
put rc=;
put luri=;
rc=metadata_setattr(luri,
"Libref",
"SASTEST");
put rc=;
put luri=;
rc=metadata_setattr(luri,
"IsDBMSLibname",
"0");
put rc=;
put luri=;
/* Set Directory Object via UsingPackages Association for the SAS Library Object */
rc=metadata_newobj("Directory",
uri,
"");
put uri=;
rc=metadata_setassn(luri,
"UsingPackages",
"Replace",
uri);
put rc=;
rc=metadata_setattr(uri,"DirectoryName","/shrproj/files/ANA_AR2_UWCRQ/data");
put rc=;
/* Set Server Context Object via DeployedComponents Association for the SAS Library Object */
rc=metadata_getnobj("omsobj:ServerContext?#Name='SASApp'",1,muri);
put muri=;
rc=metadata_setassn(luri,
"DeployedComponents",
"Append",
muri);
put rc=;
Run;
proc metalib;
omr (library="&Metalibname.");
report;
run;
I finally got around to this - there are a few things to consider!
1) Making sure all the necessary objects exist (to avoid orphan metadata data)
2) Checking to ensure that objects are successfully created
3) Checking to avoid creating the library twice (idempotence)
4) General preference to avoid data step metadata functions and the corresponding risk of infinite loops
The XML part of the program looks like this:
/**
* Prepare the XML and create the library
*/
data _null_;
file &frefin;
treeuri=quote(symget('treeuri'));
serveruri=quote(symget('serveruri'));
directoryuri=quote(symget('directoryuri'));
libname=quote(symget('libname'));
libref=quote(symget('libref'));
IsPreassigned=quote(symget('IsPreassigned'));
prototypeuri=quote(symget('prototypeuri'));
/* escape description so it can be stored as XML */
libdesc=tranwrd(symget('libdesc'),'&','&');
libdesc=tranwrd(libdesc,'<','<');
libdesc=tranwrd(libdesc,'>','>');
libdesc=tranwrd(libdesc,"'",''');
libdesc=tranwrd(libdesc,'"','"');
libdesc=tranwrd(libdesc,'0A'x,'
');
libdesc=tranwrd(libdesc,'0D'x,'
');
libdesc=quote(trim(libdesc));
put "<AddMetadata><Reposid>$METAREPOSITORY</Reposid><Metadata> "/
'<SASLibrary Desc=' libdesc ' Engine="BASE" IsDBMSLibname="0" '/
' IsHidden="0" IsPreassigned=' IsPreassigned ' Libref=' libref /
' UsageVersion="1000000" PublicType="Library" name=' libname '>'/
' <DeployedComponents>'/
' <ServerContext ObjRef=' serveruri "/>"/
' </DeployedComponents>'/
' <PropertySets>'/
' <PropertySet Name="ModifiedByProductPropertySet" '/
' SetRole="ModifiedByProductPropertySet" UsageVersion="0" />'/
' </PropertySets>'/
" <Trees><Tree ObjRef=" treeuri "/></Trees>"/
' <UsingPackages> '/
' <Directory ObjRef=' directoryuri ' />'/
' </UsingPackages>'/
' <UsingPrototype>'/
' <Prototype ObjRef=' prototypeuri '/>'/
' </UsingPrototype>'/
'</SASLibrary></Metadata><NS>SAS</NS>'/
'<Flags>268435456</Flags></AddMetadata>';
run;
For full code, check out the github repo.