👁 0
Q: What is the difference between COMP and COMP-3 in COBOL?
Answer:
COMP (COMPUTATIONAL) - Pure binary format
- Stored in binary (base-2)
- Used for subscripts and counts
- Efficient for arithmetic operations
- PIC S9(4) COMP = 2 bytes
- PIC S9(9) COMP = 4 bytes
COMP-3 (PACKED DECIMAL)
- Each digit takes 4 bits (nibble)
- Last nibble holds sign
- Used for business calculations
- PIC S9(5) COMP-3 = 3 bytes
- Formula: (n+1)/2 rounded up
When to use:
- COMP - Array subscripts, counters, loops
- COMP-3 - Money, quantities, business data
👁 0
Q: When should I use COMMIT in DB2?
Answer:
COMMIT saves all changes since the last COMMIT and releases locks.
When to COMMIT:
- After processing a logical unit of work
- Periodically in long-running batch (every N records)
- Before ending the program successfully
Frequency Guidelines:
- Too frequent: Performance overhead
- Too rare: Lock contention, large log
- Typical: Every 100-1000 records in batch
Best Practice:
MOVE 0 TO WS-COMMIT-COUNT.
PERFORM PROCESS-RECORD.
ADD 1 TO WS-COMMIT-COUNT.
IF WS-COMMIT-COUNT >= 500
EXEC SQL COMMIT END-EXEC
MOVE 0 TO WS-COMMIT-COUNT
END-IF.
Note: In CICS, syncpoint (COMMIT) happens automatically at task end. Explicit SYNCPOINT is rarely needed.
👁 0
Q: How does INSPECT work?
Answer:
INSPECT examines/modifies string contents. INSPECT field TALLYING counter FOR ALL 'X'. INSPECT field REPLACING ALL 'A' BY 'B'. INSPECT field CONVERTING 'abc' TO '123'. Useful for data validation, transformation, and counting characters.
👁 0
Q: What is GLOBAL clause?
Answer:
GLOBAL makes data available to nested programs. 01 WS-COUNTER PIC 9(5) GLOBAL. Nested programs can reference without passing as parameters. Use sparingly; explicit passing preferred. File definitions can also be GLOBAL.
👁 0
Q: What is OPTIMIZE FOR n ROWS?
Answer:
OPTIMIZE FOR n ROWS hints expected rows. SELECT * FROM t ORDER BY c OPTIMIZE FOR 1 ROW. Influences access path. Low n favors index. High n may favor scan. Use when you know actual row count.
👁 0
Q: How to use UNSTRING with TALLYING?
Answer:
UNSTRING field-1 DELIMITED BY ',' INTO field-2 field-3 TALLYING IN count-var. Count-var shows how many receiving fields got data. With COUNT IN, tracks characters per field. ALL delimiter allows multiple consecutive delimiters as one.
👁 0
Q: What is XML GENERATE?
Answer:
XML GENERATE creates XML from data. XML GENERATE output FROM data-item. Optional: COUNT IN length, WITH ENCODING, WITH XML-DECLARATION. Generates element for each field. Names from data names. COBOL 5+ feature.
👁 0
Q: How to use UNION?
Answer:
UNION combines SELECT results. UNION removes duplicates, UNION ALL keeps all. Column count and types must match. ORDER BY at end only. UNION expensive due to sort. Use UNION ALL if duplicates OK or impossible.
👁 0
Q: How to handle date arithmetic?
Answer:
DATE + n DAYS adds days. DATE - n MONTHS subtracts months. DAYS(date2) - DAYS(date1) gives day count. DATEDIFF function available. TIMESTAMPDIFF for time differences. DATE/TIME functions for extraction.
👁 0
Q: What is INSERT...SELECT?
Answer:
INSERT INTO table SELECT cols FROM other_table WHERE condition. Bulk insert from query. Column count/types must match. Faster than row-by-row. Can transform data during insert. Good for data migration.
👁 0
Q: How to use window functions?
Answer:
ROW_NUMBER() OVER (PARTITION BY col ORDER BY col2). RANK, DENSE_RANK for rankings. SUM/AVG/COUNT OVER for running totals. PARTITION BY groups within result. ORDER BY within partition. Powerful analytics.
👁 0
Q: Explain PERFORM VARYING with example
Answer:
PERFORM VARYING executes a paragraph while incrementing a counter. Syntax: PERFORM para-name VARYING WS-IDX FROM 1 BY 1 UNTIL WS-IDX > 10. The counter is initialized, tested, and incremented automatically. Can have nested VARYING with AFTER clause.
👁 0
Q: What is VOLATILE table?
Answer:
VOLATILE TABLE hints optimizer to expect different row counts. Useful for varying cardinality. CREATE TABLE ... VOLATILE CARDINALITY. Or ALTER TABLE ... VOLATILE. Helps when statistics mislead optimizer.
👁 0
Q: How to use OCCURS DEPENDING ON?
Answer:
OCCURS DEPENDING ON creates variable-length tables. 01 WS-TABLE. 05 WS-COUNT PIC 99. 05 WS-ITEM OCCURS 1 TO 100 DEPENDING ON WS-COUNT. Only variable entries allocated based on counter. Used with variable length records.
👁 0
Q: What is VALUE clause?
Answer:
VALUE initializes data items. 05 WS-FLAG PIC X VALUE 'Y'. 05 WS-COUNT PIC 9(3) VALUE ZEROS. 05 WS-TABLE OCCURS 5 VALUE 'INIT'. Figurative constants: SPACES, ZEROS, LOW-VALUES, HIGH-VALUES, QUOTES. Cannot use with REDEFINES target.
👁 0
Q: What is REPRO utility?
Answer:
REPRO copies VSAM files. IDCAMS REPRO INFILE(in) OUTFILE(out). Can REPRO between VSAM and sequential. Options: FROMKEY/TOKEY, SKIP/COUNT, REPLACE. Used for backup, conversion, extraction.
👁 0
Q: Explain BUFND and BUFNI
Answer:
BUFND is data buffer count, BUFNI is index buffer count. More buffers improve performance but use memory. JCL AMP='BUFNI=10,BUFND=20'. Default usually 2. Tune based on access pattern.
👁 0
Q: Explain IDCAMS PRINT utility
Answer:
PRINT displays VSAM contents. PRINT INFILE(dd) CHARACTER/HEX/DUMP. FROMKEY/TOKEY limits range. COUNT limits records. Useful for debugging, verification. Output to SYSPRINT.
👁 0
Q: How to handle negative numbers?
Answer:
Sign stored in PIC S9. SIGN IS LEADING/TRAILING SEPARATE CHARACTER for explicit sign byte. COMP-3 sign in low nibble. Display: S9(5)- shows trailing minus. +9(5) shows sign always. DB/CR for accounting format.
👁 0
Q: What is TITLE statement?
Answer:
TITLE 'Page Header' sets listing page headers. Appears on each page of compilation listing. Useful for documentation. Multiple TITLES change as encountered. After IDENTIFICATION DIVISION start. Compiler directive, not executable.
👁 0
Q: What is WORKING-STORAGE SECTION?
Answer:
WORKING-STORAGE SECTION declares program variables. Initialized at program load. Retains values between CALL invocations unless INITIAL program. Define all work areas, flags, counters, tables here. 01-49 levels for data, 77 for independent items.
👁 0
Q: What is TALLY special register?
Answer:
TALLY is predefined counter for EXAMINE (obsolete verb). Some code still references it. INSPECT replaced EXAMINE. Not recommended for new code. TALLY is global, can conflict. Define your own counters instead.
👁 0
Q: Explain FUNCTION LENGTH
Answer:
FUNCTION LENGTH returns byte count. FUNCTION LENGTH(field-name). For group items, includes all subordinates. For variable OCCURS, actual current length. Use in COMPUTE, MOVE, comparisons. Often used with STRING pointer initialization.
👁 0
Q: Explain JSON GENERATE
Answer:
JSON GENERATE creates JSON from data. JSON GENERATE output FROM data-item. NAME clause for custom keys. SUPPRESS clause omits fields. COUNT IN length. COBOL V6+ feature. Generates name-value pairs from structure.
👁 0
Q: How to monitor VSAM performance?
Answer:
SMF records for I/O counts. LISTCAT for statistics (READS, WRITES, SPLITS). RMF reports. Third-party tools. Track CI/CA splits, EXCP counts, buffer hit ratios.
👁 0
Q: What is DELETE command?
Answer:
DELETE removes record. EXEC CICS DELETE FILE(name) RIDFLD(key). Or after READ UPDATE: DELETE FILE(name). KEYLENGTH/GENERIC for range delete. NOTFND if key missing. NUMREC returns count deleted.
👁 0
Q: What is ACCTRG parameter?
Answer:
ACCT (accounting) provides billing information. On JOB card: ACCT=(account-number,additional-info). Installation-defined format. Passed to SMF for accounting records. May affect job processing priority.
👁 0
Q: How to handle multivolume datasets?
Answer:
VOL=(,,,n) allows n volumes. SPACE with volume count. SMS handles multivolume automatically. DISP=MOD extends to new volume. DATACLAS can specify multivolume. Large datasets need multivolume planning.
👁 0
Q: What causes S837 abend?
Answer:
S837 is end of volume and no more volumes available. Dataset needs more space. Solutions: allocate larger primary/secondary, increase volume count, extend dataset. IEFBR14 can extend with MOD DISP.
👁 0
Q: What is GROUP BY clause?
Answer:
GROUP BY aggregates rows. SELECT col, COUNT(*) FROM t GROUP BY col. All non-aggregate SELECT columns must be in GROUP BY. HAVING filters groups (vs WHERE filters rows). GROUP BY ROLLUP/CUBE for subtotals.