Thanks seven111 for you help. it is very appreciated.   The only correction I have made is:   CONNECT TO ORACLE AS OracDB ( USER=*** orapw=*** PATH=u.data) instead of CONNECT TO ORACLE ( USER=*** orapw=*** PATH=u.data)  AS OracDB;           ... View more

Hello Patrick   I have made some progress...   As you have mentioned with the implicit approach the SAS / ACCESS engine does not push the comments.   This code is working but without the comment   proc sql; connect to oracle ( user=*** orapw=*** path=u.data); CREATE TABLE D9_CADAS_CITYN_SRC ( ID_CITYN NUM(15) NOT NULL, PREN_CITYN VARCHAR(70) NOT NULL, NOM_CITYN VARCHAR(70) NOT NULL, COD_SEX_CITYN VARCHAR(1) NOT NULL CHECK (COD_SEX_CITYN IN ('H','F','I')), COD_ETAT_CIVIL VARCHAR(1) NOT NULL CHECK (COD_ETAT_CIVIL IN ('C','M','D','S','V','J'))   /*COMMENT ON TABLE D9_CADAS_CITYN_SRC IS 'Citizen cadastre Table */ ); QUIT;   The PROC SQL Pass-Through Facility is working 　 proc sql; connect to oracle ( user=*** orapw=*** path=u.data); create table Test as select * from connection to oracle (select * from D9_CADAS_CITYN_SRC) ; disconnect from oracle; quit;   The PROC SQL Pass-Through Facility is working  but when I want to put more information about the variable, it does not take it */   Is there a way to improve the code below to make it working?   proc sql; connect to oracle ( user=*** orapw=*** path=u.data); create table Test as select * from connection to oracle (select (ID_CITYN NUMBER(15) NOT NULL)) from D9_CADAS_CITYN_SRC) ; disconnect from oracle; quit;             ... View more

Thank you for the information.   ... View more

I though it was better to add Ask the experts because I saw it on many post. If you think it is unnecessary, I won't put it anymore.   Thanks  ... View more

Thanks you  ... View more

Hello,   If I am interested to set a flag to one if there is at lease one missing value (num or string) in the dataset is there a better way to do that compare to my code below.   Regards,           ... View more

Hello Chris,   At the beginning, I though that my approach was the gone one but after some discussions with my team members, the objectives are clearer.   An XML Schema Definition file is written where some criteria regarding the table to be evaluated (SAS table or XML File) must met .   For example, imagine in Table A (provided by another service), variable sex, obs=122 is missing.  IF the fact that this observation is missing in not critical based on the criteria of the XSD file the we accept the dataset otherwise an error report must be written who contains all the information for the observation # 122.   Same thing imagine in this Table A, based on the XSD file that var10 should be numeric and contains 9 number like a phone number, and for some reason we don't know, observation 425, contains only 6 number.   Then this observation should be added to the error report with observation 122, and so on.   At the end, if the sas table is error free based on the XSD file, then we go to the next step the transformation to the XML type. Otherwise, a sas table is provided containing all the observations (complete line) who contains errors.   I hope with those clarifications you will be able to direct me toward a nice solution. Regards,       ... View more

Hello,   I will say both. First I am interested to check if there is a missing value in the dataset.  Then if so, which variable and observation? Thanks in advance for your help.     ... View more

Hello,   I already know those functions.  But how can we use those functions when we have around 100 variables. I don't want to enter the name of each variable in the code... ... View more