gregrahn

--------------
alter table call_center add constraint cc_d1 foreign key  (cc_closed_date_sk) references date_dim (d_date_sk)
--------------

Query OK, 6 rows affected (0.03 sec)
Records: 6  Duplicates: 0  Warnings: 0

--------------
alter table call_center add constraint cc_d2 foreign key  (cc_open_date_sk) references date_dim (d_date_sk)
--------------

gregrahn

create unique index person_pk on person(person_id);
create index person_n1 on person(date_id);
create unique index calendar_pk on calendar(date_id) ;

gregrahn

hive> set hive.execution.engine=tez;
hive> select split('I crash Tez',' ');
Query ID = hue_20140718150505_1eafc45c-6e2b-49f1-9d72-0ec737a72377
Total jobs = 1
Launching Job 1 out of 1


Status: Running (application id: application_1405217363240_0016)

Map 1: -/-	

gregrahn

# 1) get your inmap from http://inmaps.linkedinlabs.com/
# 2) zoom all the way in and note the url for the individual tiles (I used Charles Proxy to do this, 
#    but right/control-click, "Inspect Element", "Network" tab, then reload page, also works.)
# 3) download your tiles
# 4) merge the tiles into column-wise strips
# 5) merge the column-wise strips into the final high resolution image


# download the individual tiles
# my inmap happended to be 15 tiles wide and 12 tiles tall

gregrahn

--
-- triangle_counting.sql
--
-- get flat file from http://www.vertica.com/benchmark/TriangleCounting/edges.txt.gz
-- gunzip & place in /tmp
--

create or replace directory GRAPH_DATA as '/tmp';

DROP TABLE ET_EDGES;

gregrahn

WITH RecursiveCTE (rnum) AS
(
    SELECT 1 AS rnum 
    FROM dual
    UNION ALL
    SELECT rnum + 1
    FROM RecursiveCTE
    WHERE rnum < 24
)
SELECT rnum

gregrahn

col CURSOR_TYPE for a15
col PROGRAM for a36

select 
  s.program, 
  o.sql_id, 
  o.sql_exec_id, 
  o.cursor_type, 
  o.sql_text
from 

gregrahn

--
-- open EMP2 tab & browse DATA tab, notice there are 2 open cursors, noting the SQL_EXEC_ID numbers
--

PROGRAM                              SQL_ID        SQL_EXEC_ID CURSOR_TYPE     SQL_TEXT
------------------------------------ ------------- ----------- --------------- ------------------------------------------------------------
SQL Developer                        apdcq2s3sxn9y    16777216 OPEN            SELECT ROWID "ROWID", ORA_ROWSCN "ORA_ROWSCN", EMPNO EMPNO,
oracle@localhost.localdomain (P000)  apdcq2s3sxn9y    16777216 OPEN-RECURSIVE  SELECT ROWID "ROWID", ORA_ROWSCN "ORA_ROWSCN", EMPNO EMPNO,
oracle@localhost.localdomain (P001)  apdcq2s3sxn9y    16777216 OPEN-RECURSIVE  SELECT ROWID "ROWID", ORA_ROWSCN "ORA_ROWSCN", EMPNO EMPNO,
SQL Developer                        apdcq2s3sxn9y    16777217 OPEN            SELECT ROWID "ROWID", ORA_ROWSCN "ORA_ROWSCN", EMPNO EMPNO,

gregrahn

create table emp2 as select * from emp;
insert into emp2 select * from emp2;
insert into emp2 select * from emp2;
insert into emp2 select * from emp2;
insert into emp2 select * from emp2;
insert into emp2 select * from emp2;
insert into emp2 select * from emp2;
commit;
alter table emp2 parallel 2;

gregrahn

library(RJDBC)

drv <-JDBC("oracle.jdbc.driver.OracleDriver","/Users/grahn/code/jdbc/ojdbc6.jar")

conn<-dbConnect(drv,"jdbc:oracle:thin:@grahn-dev.us.oracle.com:1521:orcl","scott","tiger")

data <-dbGetQuery(conn, "select * from emp")

write.table(data, file="emp.csv", sep = ",", row.names=FALSE)