Oracle Built-in Packages
by Steven Feuerstein, Charles Dye and John Beresniewicz


Following are the changes made in the 5/99 reprint:


(30) the dynconst script: the last 2 FUNCTION statements in the
package body should not have semicolons after INTEGER.

Here is how the code now looks:

    FUNCTION v6 RETURN INTEGER
       IS BEGIN RETURN DBMS_SQL.V6; END;

    FUNCTION v7 RETURN INTEGER
       IS BEGIN RETURN DBMS_SQL.V7; END;

    FUNCTION native RETURN INTEGER
       IS BEGIN RETURN DBMS_SQL.NATIVE; END;


{33} Half way down the page, the code line beginning:

        PRAGMA RESTRICT_REFERENCES

is now in bold.


(35) 1st Paragraph, 3rd sentence reads:

         "Here is the program for the GETLENGTH function:"

now reads:

         "Here is the pragma for the ... "

and

In the SQL statement towards the bottom of the page there is now a
closing single quote after 'FEUERSTEIN:

        "WHERE family = 'FEUERSTEIN' AND ..."


{55} Figure 2-1:  under the "Executing PL/SQL block?" diamond, there
was a Yes line extending to the right of the line pointing down. It has
been removed.  The Yes goes with the line going down.


{61} code 3, line 3:

        "DBMS_SQL.BIND_VARIABLE(cur, 'varname', 'a', 100);"

now reads:

        "DBMS_SQL.BIND_VARIABLE(cur, 'varname', var_name_in);"


{68} Last code sample (list of "more specific variants") now
also includes:

        PROCEDURE DBMS_SQL.DEFINE_COLUMN_LONG
           (c IN INTEGER
            ,position IN INTEGER);


(71) last sentence: "So hat" now reads "so that"


(88) para 2, 1st line: "prodecure" now reads "procedure".


(235) first table, table headings "Parameter" and "Description"
are now in boldface


{450} function book_file :

         return clob

now reads:

          return bfile
and,
         return book_rec.chapter_file

now reads:

         return book_rec.book_file

and,

        SELECT chapter_file

now reads:

        SELECT book_file


{451} The second line on the page now reads:

        "...using the following statements (stored in the
                                compare_text.sql file):

The next 3 lines of code have been deleted. The rest of the code now
reads:

        SELECT chapter_descr, chapter_text
          FROM my_book_text
         WHERE chapter_descr = '&1'

        ROLLBACK;

        EXEC DBMS_OUTPUT.PUT_LINE ('Rollback completed');

        SELECT chapter_descr, chapter_text
          FROM my_book_text
         WHERE chapter_descr = '&1'
      END;
      /


{485} Last line, the code now begins:

        @compare_text


{487} Last line, the code now begins:

        @compare_text


{489} Last line, the code now begins:

        @compare_text


{491} Middle of the page, the code now begins:

        @compare_text


{493} Middle of the page, the code now begins:

        @compare_text


{500} Line -6 now reads:

        FUNCTION DBMS_ROWID.ROWID_ROW_NUMBER (row_id IN ROWID)


{504} The last bit of the last line of code now reads:

        ... (rowid_value, NULL, NULL, 0) = 1;


(526) The section head said UTL_REF is Oracle8.1 only.  This is wrong;
UTL_REF was introduced in 8.0.4.


(527) The same error listed for page 526 occurred in the middle
paragraph. It too now reads "Oracle8.0.4."


{538 through 577} in the following sections "NULL" has been replaced
with "COMPUTE":

    p538, 1/3 down
    p538, table of parameters (method)
    p539, table of parameters (method)
    p539, bottom
    p577, table of parms (method)
    p577, 2/3 down page


{571} Under "Getting Started with DBMS_DDL," the first line begins:

        This DBMS_DDL.....


(591) last section near bottom, for consistency:

heading "DBMS_SESSION.SET_MLS_LABEL_FORMAT" now reads, like
all the other headings:

        "The DBMS_SESSION.SET_MLS_LABEL_FORMAT procedure"


{657} Replaced the text under "JOB_QUEUE_KEEP_CONNECTIONS" with:

The third INIT.ORA parameter that supposedly affects the
behavior of the SNP processes is JOB_QUEUE_KEEP_CONNECTIONS.
This parameter has been made obsolete in Oracle8i. In fact,
it apparently never actually had any effect under previous releases,
although it has been documented as having various effects. Some sources 
say that it controlled the database sessions held by the SNP background
processes; others say that it controlled the sessions in remote databases
for jobs using database links.

Although setting this parameter appears to do no harm,
it is best to leave it alone.


{662} Replaced with Warning with:

When you are using the ISUBMIT procedure, a subtle problem
can arise from the fact that next_date has datatype VARCHAR2
instead of DATE (as in SUBMIT).  When DATE values are assigned to
next_date in calls to ISUBMIT, an implicit conversion to VARCHAR2 is
done using the session's current NLS_DATE_FORMAT as the date mask.  This can
result in unexpected truncation of next_date. For instance, if the
session NLS_DATE_FORMAT is 'YY-MM-DD', dates assigned to next_date will be
truncated to 12:00 a.m. To avoid this problem, either make sure that the 
session NLS_DATE_FORMAT includes all date components or supply next_date
as a fully specified VARCHAR2.


{670-671} table:  all of the expressions in the right column
("Interval Value") are now enclosed in single quotes as in the table
on page 670

and also, row 2,

        "Every day at 8:00 p.m."

now reads

        "Every day at 8:00 a.m."


{690} second PROCEDURE example, 2nd line

    (ab IN OUT dbms_utility.uncl_array,

now reads

    (tab IN OUT dbms_utility.uncl_array,

("ab" should be "tab")


{728} Line 6 now reads:

        Table 15-2 lists the programs in DBMS_REPCAT_ADMIN

and, under "DBMS_REPCAT_ADMIN exceptions" toward the bottom of the
page, the line begins:

        The DBMS_REPCAT_ADMIN package may ......