Hash Object Code Question

[Mixer]

Hi All,

Would someone please explain how the missing records section work? Thank you for your help.

declare hash hct(dataset: 'work.etls_compare_digest', hashexp: 10);
hct.defineKey("sourcedb_id", "case_id");
hct.defineData("sourcedb_id", "case_id", "compare_digest");
hct.defineDone();
end;

set work.W1PAEO8L end = eof;

/* create digest - version 2.1 */
source_digest = put(md5(catq(' ', contact_id, subject_id, owner_id, owner_name, case_name, Season,
case_priority, case_notes, case_state_code, case_status_code,
case_status_desc)), $hex32.);

/* source/compare match */
if hct.find() eq 0 then
do;
/* source/compare match: Unchanged records */
if source_digest eq compare_digest then
output work.WGK73ZRO;

/* source/compare match: Changed records */
else if source_digest ne compare_digest then
output work.WGK73X4D;

/* remove found key + data */
hct.remove();
end;
/* source table: New records */
else if hct.find() ne 0 then
output work.WGK73YG2;

/* Missing records */
if eof then
rc = hct.output(dataset: 'work.WGK7411D(drop = compare_digest)');

[Tom · Accepted solution]

You told SAS to set a flag when your were reading the last observation in your SET statement.

set work.W1PAEO8L end = eof;

Then when you get to the last observation you write out the remain records from the hash object. You ave added dataset options to drop one of the variables.

if eof then
  rc = hct.output(dataset: 'work.WGK7411D(drop = compare_digest)')
;