How to directly edit PL/SQL Scripts

I can't open PL/SQL source files directly into a PL/SQL editor - they're opened as SQL scripts. The tool still seems targeted towards database maintenance rather than developing code held in a separate version control system.
Whilst technically the PL/SQL source files are just SQL scripts this doesn't make for easy PL/SQL development.
A change to a PL/SQL source file involves:
- Open source file into SQL script tab.
- Choose connection to run script against.
- Run as a SQL script.
- Close source file.
- Navigate to package spec/body in tree browser and open read-only in a view tab.
- Click to edit object source - opens in new tab.
- Edit source and compile.
- Add 'create or replace' at top and '/' at end.
- Select 'Save As...' and navigate to original source file.
- Close script window that this automatically opens for newly saved file, and close edit source tab. Same task in PL/SQL developer:
- Open source file into a Program Window.
- Edit source and compile.
- Click save.
Making a single change is bad enough - try it when you're making substantial changes to a source file and you want to regularly check compilation and save changes.
Can anyone help make this easier ?

The tool still seems targeted towards database maintenance rather than developing code held in a separate version control system.They'll get to it eventually...
I can't open PL/SQL source files directly into a PL/SQL editor - they're opened as SQL scriptsIs the extension .sql? Change it to .pls or .pks/.pkb. They seem to use the dumb extension-based system to find out what files they are dealing with, instead of parsing the contents.
If the code doesn't get syntax coloured, open a function or procedure from database first, that seems to wake him up.
Also make sure the files are as the same as in the DB (with create or replace, without final slash), so you can compile them directly from the same file-based editor.
You still have to assign connection and navigate to the PL/SQL in the DB to view compilation errors and run/debug though. There are some requests on the announced SQL Developer Exchange for these (http://htmldb.oracle.com/pls/otn/f?p=42626:37:260789750723810::NO:::). Please vote for them to see them implemented.
K.

Similar Messages

How can I run a SQL script file...

How can I run a SQL script file from a location on my computer without providing the whole path?
Is there some way I can set a "Working folder" in SQL Plus??
Thanks!
Tom

You can create an environment variable called "SQLPATH" which is a list of directories that SQL*Plus will search for your .SQL
scripts.
I would like to use another directory than the oracle/bin...
How can I do this ??
Hello,
U can do this by this way:
Save odm_script.sql file to the default Oracle
directory i.e. Oracle-Home/bin and Run following command
through SQL Plus.
SQL>@Script_Name
I hope this will resolve ur problem.
Regards,
Omer Saeed Khan.

How to use views in sql script report?

All all
Can any one tell how to use views in sql script report?

Most of the views are based on tables (or other views which are based on tables).
The view typically shows one org at a time based on the context that is set.
If you need records for all orgs, you need to use the underlying tables. Oracle typically names the tables with a _all suffix.
e.g. PO_HEADERS will show records for one org at a time. PO_HEADERS_ALL will show records for all orgs.
Hope this answers your question,
Sandeep Gandhi

First simple report/app - how do I edit the sql in it?

I finally have some time to play with my newly installed htmldb 2.0, and created a quick sample page with a tabular report and queried a table via dblink with some sql I entered in. When I run the app I get a sql error, but for the life of me I can't find out how to get at the sql to edit it.
Where do I find (and how do I edit) the sql I just entered in?

When you create this kind of region you can edit the SQL by go to the page definition and clicking on the name you've given to the region in the Regions section of the page rendering.
This will bring you to the region definition page (you should see /f?p=4000:4651 in the URL of your browser). Then you can go to the region source and edit the SQL there.
Hope that helps.
Earl

Problem while executing script in Toad - How to use '&' in the sql script ?

I have to execute sql script in toad. Sql script has one insert query in which one insert-value is 'USA & CAN'. When I executed the script in toad by pressing F5, I got a prompt window asking for the value if 'CAN' as it is after the &.
I tried using[b] {escape '\' } .... but could not resolve the problem.
Is there any solution or workaround to overcome this problem. I have thousands of records with such values and I have to use sql script only.

There is an option in TOAD to change this behaviour.
Look in VIEW/OPTIONS/SQL Editor/
Uncheck the box for "Scan statements for bound variables before execution".
In SQL*PLUS it would be SET SCAN OFF
(desupported version is SET DEFINE OFF)
Message was edited by:
Sven Weller

How to double dereference in sql scripts

hi all,
I wanted to call a sql file(sql2) from within an another sql file(sql1). The call to sql2 from sql1 contains some parameters( in form of integer and some variables). It is possible to access integer values in sql2 but i am unable to access the value stored in the variables in file2.
How to access the parameter's(which is itself is a variable) value in the called sql script.
Thanks,
Shyam

Depending on your environment and if these are administrative (DBA) scritps or production batch then here are some ideas:
For DBA
Run the scripts via cron in a DBA privileged account
sqlplus /nolog <<EOF
connect / as sysdba
start script
exit
EOF
The above requires no password,
For production batch assuming that all production runs under one OS ID that access to is restricted then potentially you can use OS authentication so the Oracle username has no password: create user bob identified externally.
sqlplus / @script
will then do the job
HTH -- Mark D Powell --

How to hide passwords in sql scripts?

Hi!
I have got a few sql scripts for some actions. In the scripts the connection information is plain text - even the SYS password in one script.
Are there ways to solve this behaviour? Perhaps encrypt it or... What does the practice say - how do you solve this???
Thanks
Markus

Depending on your environment and if these are administrative (DBA) scritps or production batch then here are some ideas:
For DBA
Run the scripts via cron in a DBA privileged account
sqlplus /nolog <<EOF
connect / as sysdba
start script
exit
EOF
The above requires no password,
For production batch assuming that all production runs under one OS ID that access to is restricted then potentially you can use OS authentication so the Oracle username has no password: create user bob identified externally.
sqlplus / @script
will then do the job
HTH -- Mark D Powell --

How to Improve this PL/SQL script

Hi All,
I have a package/procedure that insert data into table but it takes time more than 2 days. Have any known how to modify this script to improve the performance and reduce loading time, The following code is procedure I used to insert data.....
Procedure INSERT_DATA (p_month IN DATE, p_product_id IN NUMBER ) IS
cursor c1 is select * from tab#1; --reference data
cursor c2 is select * from tab#2;
cursor c3 is select * from tab#3;
cursor c4 is select * from tab#4;
cursor c5 is select * from tab#5;
v_rec claim_table%rowtype;
Begin
for c1rec in c1 loop
exit when c1%notfound;
call procedure in package....;
open c2(c1rec.claim_no);
fetch c2 into v_location_cd ,v_claim_type_cd ;
close c2;
v_rec.location_cd := v_location_cd;
v_rec.claim_type_cd := v_claim_type_cd ;
open c3(c1rec.claim_no);
fetch c3 into v_col#3,v_col#4;
close c3;
v_rec.col#3 := v_col#3 ;
v_rec.col#4 := v_col#4 ;
open c4(c1rec.claim_no);
fetch c4 into v_col#5,v_col#6;
close c4;
v_rec.col#5 := v_col#5 ;
v_rec.col#6 := v_col#6 ;
insert into claim_table values ( v_rec.location_cd, v_rec.claim_type_cd , v_rec.col#3 , .......) ;
if (c1%rowcount/1000) = trunc(c1%rowcount /1000) then
commit;
end if;
end loop;
commit;
Exception
exception statement....
commit;
End;
Thanks All,
Mcka

A copy and paste of a reply I posted just a hour or so ago to the exact same approach used by a poster in [url http://forums.oracle.com/forums/thread.jspa?threadID=636929]this thread.
Yucky code IMO. You are using PL/SQL to drive a nested loop join. Why? SQL is by far more capable of joining tables and is not limited to a using a nested loop approach only.
Also, as you are using PL/SQL to drive it, it means that each cursor fetch in each (nested) FOR loop is a slow context switch from PL/SQL to SQL, in order to ship the row's data from the SQL engine to the PL/SQL engine. Only then to have that very same data shipped back (via yet another context switch) to the SQL engine to be inserted into table4.
This code violates one of the most rudimentary performance principles in Oracle - it is not maximizing SQL and it it not minimizing PL/SQL. It is doing the exact opposite.
As for the ad-hoc commits inside the loop - this does not make much sense. This will generate more redo and more overheads. Also, when something goes pear shape in that process, some rows would have been processed and inserted and committed. Some not.
In this case, how do you expect to restart the failed process at the very first row that failed to be committed and continue processing from there?

How to create thousands of sql scripts easily?

Hello,
I would like to create thousand of scripts as following:
insert into table tb_test('aaa');   update table tb_class( set nm = 'aaa');
insert into table tb_test('basd'); update table tb_class( set nm = 'basd');
insert into table tb_test('asdfa');update table tb_class( set nm = 'asdfs');
insert into table tb_test('xxxyy');update table tb_class( set nm = 'xxxyy');
Is there any simple way to generated those scripts automatically? I am thinking of :
    define var_str = ('aaa', 'bbb', 'asdfd',...'xxxyy');
    select 'insert into table tb_test(' || var_str || '); update table tb_class( set nm =' || var_str || ');'
    from dual;
Then how do I generate those scripts from the var_str?
Thanks,

942572 wrote:
I have more than 5000 parameters in a text file as followings, which can be opened by notepad or excel:
'aaa',
'basd',
'asdfs',
'xxxyy'
Ok, so assuming this is a 1-off event, use Excel.
Setup a quick formula to turn each into a proc call:
[edit] if the values are in column "A", then in column "B", you could do:   =concatenate("p_proc('", a1, "');")
then copy that formula down.
[/edit]
p_proc('aaa');
p_proc('basd');
etc.
Then setup a little p_proc as Frank described. (and please, use a WHERE clause on that update ... or you're going to destroy some data )
Then copy past those commands into a routine that ends up looking like:
declare
procedure p_proc ( in_value )
is
begin
      -- Frank's code goes here
end;
begin
   -- proc list goes here
p_proc('aaa');
..... etc.
end;
That's 1 very primitive way to do it.
If you need to reproduce this in future, you may want to consider loading that list of parameter values into a table, then you can do some more funky things. *shrug*

How to send output from SQL script to the specified log file (not *.sql)

## 1 -I write sql command into sql file
echo "SELECT * FROM DBA_USERS;">results.sql
echo "quit;">>results.sql
##2- RUN sqlplus, run sql file and get output/results into jo.log file
%ORACLE_HOME/bin/sqlplus / as sysdba<results.sql>>jo.log
It doesn't work please advise

$ echo "set pages 9999" >results.sql ### this is only to make the output more readable
$ echo "SELECT * FROM DBA_USERS;" >>results.sql
$ echo "quit" >>results.sql
$ cat results.sql
set pages 9999
SELECT * FROM DBA_USERS;
quit
$ sqlplus -s "/ as sysdba" @results >jo.log
$ cat jo.log
USERNAME                          USER_ID PASSWORD
ACCOUNT_STATUS                   LOCK_DATE EXPIRY_DAT
DEFAULT_TABLESPACE             TEMPORARY_TABLESPACE           CREATED
PROFILE                        INITIAL_RSRC_CONSUMER_GROUP
EXTERNAL_NAME
SYS                                     0 D4C5016086B2DC6A
OPEN
SYSTEM                         TEMP                           06/12/2003
DEFAULT                        SYS_GROUP
SYSTEM                                  5 D4DF7931AB130E37
OPEN
SYSTEM                         TEMP                           06/12/2003
DEFAULT                        SYS_GROUP
DBSNMP                                 19 E066D214D5421CCC
OPEN
SYSTEM                         TEMP                           06/12/2003
DEFAULT                        DEFAULT_CONSUMER_GROUP
SCOTT                                  60 F894844C34402B67
OPEN
USERS                          TEMP                           06/12/2003
DEFAULT                        DEFAULT_CONSUMER_GROUP
HR                                     47 4C6D73C3E8B0F0DA
OPEN
EXAMPLE                        TEMP                           06/12/2003
DEFAULT                        DEFAULT_CONSUMER_GROUPThat's only a part of the file, it's too long :-)

How to use parameters in oracle SQL script????

Right now I am writing a SQL script to create a schema and build the objects of this schema....
I use a .net winform program to run sqlplus to parse this sql script ...
The problem is that the schema name and the tablespace's location and the sys password must be input by the user, so my SQL script should use these runtime input parameters instead of const parameters....
So, how to use parameter in SQL script ...........
Are there some example scripts in oracle home directory for me to refer to????

Hi,
UNISTD wrote:
thanks .....
what's the difference between variable , define, accept in sqlplus ???VARIABLE declares (but does not assign a value to) a bind variable. Unlike substitution variables, bind variables are passed to the back end to be compiled, and they can only be values in certain data types. You can not use a bind vaiable in place of an identifier, so to do something like
CREATE USER &1 ...a bind variable won't work.
"DEFINE x = y" sets the substitution variable &x to have the value y. There is no user interaction (unless x or y happen to contain undefined substtiution variables).
"DEFINE x" shiows the value of the substitution variable &x, or, if it is undefined, raises a SQL*Plus error. I use this feature below.
ACCEPT sets a substitution variable with user interaction.
And if the user miss some parameters in “sqlplus /nolog ssss.sql par1 par2 par5 par6”, how to use default value of the miss parameters??Don't you need a @ befiore the script name, e.g.
sqlplus /nolog @ssss.sql par1 par2 par5 par6Sorry, I don't know of any good way to use default values.
The foloowing works, but, as you can see, it's ugly.
"DEFINE 1" display a message like
DEFINE 1            = "par1" (CHAR)if &1 is defined; otherwise,it will display a SQL*Plus error message like
SP2-035: symbol 1 is UNDEFINEDNotice that the former contains an '=' sign, but the latter does not.
The best way I know to use default values is to run the DEFINE command, save the output to a filee, read the file, and see if it's an error message or not.
So you can use a script like this:
--     This is DEFINE_DEFAULT.SQL
SPOOL     got_define_txt.sql
DEFINE     &dd_old
SPOOL     OFF
COLUMN     dd_new_col     NEW_VALUE     &dd_new
WITH     got_define_txt     AS
     SELECT q'[
          @got_define_txt
]'               AS define_txt
     FROM    dual
SELECT     CASE
         WHEN define_txt LIKE '%=%'
         THEN REGEXP_REPLACE ( define_txt
                           , '.+"
([^"]*)
                     , '\1'
         ELSE '&dd_default'
     END        AS dd_new_col
FROM     got_define_txt
{code}
and start your real script, ssss.sql, something like this:
{code}
DEFINE          dd_new     = sv1
DEFINE          dd_old     = 1
DEFINE          dd_default     = FOO
@DEFINE_DEFAULT
DEFINE          dd_new     = sv2
DEFINE          dd_old     = 2
DEFINE          dd_default     = "Testing spaces in value"
@DEFINE_DEFAULT
{code}
when this finishes running, the substitution variable &sv1 will either have the value you passed in &1 or, if you didn't pass anything, the default value you specified, that is FOO.
Likewise, &sw2 will have the value you passed, or, if you didn't pass anything, the 23-character string 'Testing spaces in value'.
Here's how it works:
Define_default.sql puts the output of the "DEFINE x" command into a column, define_txt, in a query. That query displays either the existing value of the substitution variable indicated by &dd_old or, if it is undefined, the default value you want to use, which is stored in the substitution variable &dd_default. The substitution variable named in &dd_new is always set to something, but that something may be its existing value.
Notice that the paramerters to define_default.sql must be passed as global varibales.
Why didn't I just use arguments, so that we could simply say:
{code}
@DEFINE_DEFAULT sv1 1 FOO
{code}
? Because that would set the substitution variables &1, &2 and &3, which are miost likely the very ones in which you're interested.
I repeat: there must be a better way, but I'm sorry, I don't know what it is.
I usually don't do the method above. Instead, I always pass the required number of parameters, but I pass dummy or plce-holder values.
For example, if I wanted to call ssss.sql, but use defulat vlaues for &1 and &3, then I would say something like:
{code}
@ssss ? par2 ?
{code}
and, inside ssss.sql, test to see if the values are the place holder '?', and, if so, replace them with some real default value. The use has to remember what the special place holder-value is, but does not need to know anything more, and only ssss.sql itself needs to change if the default values change.

Execute sql script in oracle server

hi how can i run an SQL script in oracle server in every first day of the month? i dont want to run that script manually every first day of the month.

What OS and version are you running?
You can schedule this from the OS side as well if you don't want to use the scheduler.
In linux you can make use of the crontab:
crontab -e
#then paste the following lines with the path to your scripts
#.---------------- minute (0 - 59)
#|   .------------- hour (0 - 23)
#|   |   .---------- day of month (1 - 31)
#|   |   |   .------- month (1 - 12) OR jan,feb,mar,apr ...
#|   |   |   | .----- day of week (0 - 7) (Sunday=0 or 7) OR sun,mon,tue,wed,thu,fri,sat
#|   |   |   | |
#*   *   *   * * command to be executed
00 01 1 * * /<path to your scripts>/script_name.shThe above script will run on the 1st day of every month at 01h00.
your script (the one you are calling in the cron) will have to call the sql script and should look similar to this:
#export all your environment variables here
# for example
export ORACLE_BASE=/u01/app/oracle
export ORACLE_HOME=$ORACLE_BASE/product/<your path>
export ORACLE_SID=ORCL
export PATH=$ORACLE_HOME/bin:$PATH
sqlplus user/password@DB @/path/your_sql_script.sqlHope this helps

Exceuting SQL Script through java

Hi All
I am making an installer for my application. How do i execute a SQL script through java. Its a sepeate file that has the SQL script for DB creation.
Regards
Abhinav

If you database support ";" to separate sqll, then you can use Statement.execute(sql); to run all SQLscipt once, otherwise you need to load all sqls into Vector, then reun sql step by step. Copy some code for instance:
public final static Object[] getSQLStatements(String sqlFile) throws java.
io.IOException {
java.util.Vector v = new java.util.Vector(1000);
try {
java.io.BufferedReader br = new java.io.BufferedReader(new java.io.
FileReader(sqlFile));
java.util.Vector batchs = new java.util.Vector(10);
String temp;
while ( (temp = br.readLine()) != null) {
temp = temp.trim();
if (temp.length() == 0)
continue;
switch (temp.charAt(0)) {
case '*':
case '"':
case '\'':
// System.out.println(temp);
break; //Ignore any line which begin with the above character
case '#': //Used to begin a new sql statement
if (batchs.size() > 0) {
v.addElement(getSQLStatements(batchs));
batchs.removeAllElements();
break;
case 'S':
case 's':
case '?':
if (batchs.size() > 0) {
v.addElement(getSQLStatements(batchs));
batchs.removeAllElements();
v.addElement(temp);
break;
default:
batchs.addElement(temp);
break;
if (batchs.size() > 0) {
v.addElement(getSQLStatements(batchs));
batchs.removeAllElements();
br.close();
br = null;
catch (java.io.FileNotFoundException fnfe) {
v.addElement(sqlFile); //sqlFile is a sql command, not a file Name
Object[] statements = new Object[v.size()];
for (int i = 0; i < v.size(); i++)
statements[i] = v.elementAt(i);
return statements;
for (int i = 0; i < statements.length; i++) {
if (statements[i] instanceof java.lang.String) {
String temp = (java.lang.String) statements;
switch (temp.charAt(0)) {
case 'S':
case 's':
case '?':
System.out.println(temp);
rs = stmt.executeQuery(temp);
ResultSetMetaData resultSetMetaData = rs.
getMetaData();
int iNumCols = resultSetMetaData.getColumnCount();
for (int j = 1; j <= iNumCols; j++) {
System.out.println(resultSetMetaData.getColumnLabel(j)
+ " " + resultSetMetaData.getColumnTypeName(j)
+ " " + resultSetMetaData.getColumnType(j)
+ " " + resultSetMetaData.getPrecision(j)
+ " " + resultSetMetaData.getScale(j)
Object colval;
long ncount = 0;
while (rs.next()) {
// System.out.print(rs.rowDeleted()+" ");
ncount++;
for (int j = 1; j <= iNumCols; j++) {
colval = rs.getObject(j);
System.out.print(colval + " ");
System.out.println();
rs.close(); //Let the CONCUR_UPDATABLE resultSet release its open files at once.
rs = null;
System.out.println(
"The total row number of resultset: " + ncount);
System.out.println();
break;
default:
int updateCount = stmt.executeUpdate(temp);
System.out.println(temp + " : " + updateCount);
System.out.println();
else if (statements[i] instanceof java.lang.Object[]) {
int[] updateCounts;
Object[] temp = (java.lang.Object[]) statements[i];
try {
for (int j = 0; j < temp.length; j++){
System.out.println( temp[j]);
stmt.addBatch( (java.lang.String) temp[j]);
updateCounts = stmt.executeBatch();
for (int j = 0; j < temp.length; j++)
System.out.println((j+1)+":"+temp[j]);
for (int j = 0; j < updateCounts.length; j++)
System.out.println((j+1)+":" +updateCounts[j]);
catch (java.sql.BatchUpdateException e) {
updateCounts = e.getUpdateCounts();
for (int j = 0; j < updateCounts.length; j++)
System.out.println((j+1)+":"+updateCounts[j]);
java.sql.SQLException sqle = e;
do {
System.out.println(sqle.getMessage());
System.out.println("Error Code:" +
sqle.getErrorCode());
System.out.println("SQL State:" + sqle.getSQLState());
sqle.printStackTrace();
while ( (sqle = sqle.getNextException()) != null);
catch (java.sql.SQLException sqle) {
do {
System.out.println(sqle.getMessage());
System.out.println("Error Code:" +
sqle.getErrorCode());
System.out.println("SQL State:" + sqle.getSQLState());
sqle.printStackTrace();
while ( (sqle = sqle.getNextException()) != null);
stmt.clearBatch();
System.out.println();

Generate SQL scripts from Designer 6

Hi all,
Does anyone know how to generate an SQL-script from Designer? I want to maintain the code in the repository, and also document usages against the script. I've found ways to document the script, but then I can't generate it....
Thanks in advance,
Ronald

Ronald,
How did you record the SQL script in designer? As undefined pl/sql module (type= "null")?
If so, you cannot generate the script with the ddl generator. To do so you could change the type to procedure and make post-generation changes to the script (cut of f the procedure header and trailer lines). An other way to generate it is to write your own SQL script generator as a headstart utility. You might take a look at the html help file generator utility which does a similar job (create a o/s file based on the repository contents).
Let us know if you succeed.
Regards, Marc
null

Timer - Run .sql Script automatically

How do I setup an .sql script setup in SQL Developer to run automatically as a scheduled task?
thanks

Hello,
welcome to the forum.
SQL Developer is not a tool that can be automated. When you need to execute a script automatically, then you have several options.
- Execute the script via SQL*PLUS, scheduled from OS
- Use the db scheduler to execute a procedure
- Call a OS script from a db job.
For further information about these options go to {forum:id=193}, {forum:id=144} or {forum:id=75}.
Regards
Marcus

How to directly edit PL/SQL Scripts

Similar Messages

Maybe you are looking for