If the first parameter for TableUpdate() is .T., this means that changes to all records will be saved. IMO, you can't expect a command that processes multiple records to stay on the current record. For comparison, a REPLACE ALL will move the record pointer to EOF.

You can use TableUpdate() only for the current record, with appropriate parameters.

If you need to TableUpdate() all records, and still stay on the current record, you will have to save recno() to a variable, and go back to it later.

>Yes, I issue TableUpdate for each tables, (.t.,.f.,[tables])
>
>It does not return .f. to tableupdate command. However, it still jumps record pointer.
>
>I did not noticed this behaviour earlier.
>
>Why it is changing record pointer.
>
>I was able to reproduce the error;
>
>

>1. Open a Record (open a form) (e.g. recno() = 12)
>2. mark the record as lock (repl rec_lock with .t.)
>3. tableupdate(.t.,.f.,"tableA")
>4. show form
>5. change text under memo field (Editbox)
>6. from within editbox, press ESC key
>7. ESC fires exit without save procedure
>8. tablerevert(.t.,"tableA")
>9. unmark record (repl rec_lock with .f.)
>10. tableupdate(.t.,.f.,"tableA")
>11. hide form
>12. move to another record (e.g. recno() = 15)
>13. Open the record (recno() = 15)
>14. lock record (repl rec_lock with .t.) (recno() = 15)
>15. tableupdate(.t.,.f.,"tableA")
>16. Record pointer changes to recno() = 12
>

>
>I don;t have any clue why this is happening??
>
>Regds/Shaishav
>
>
>>>Re: TableUpdate()
>>>Can you pls explain why it will not save records in TableB, what is the work around?
>>
>>Some things to check:
>>
>>If you have several tables, you must do TableUpdate() for each table. (I think you did this.)
>>
>>See if TableUpdate() returns .T. or .F.; .F. means that it could not save.
>>
>>If TableUpdate() returns .F., use aerror() to find out what was the problem.
>>
>>

>>if not TableUpdate(...)
>>  aerror(MyErrorArray)
>>  MessageBox("Error #: " + trans(MyErrorArray(1) + " message: ";
>>    + MyErrorArray(2))
>>endif
>>

>>
>>>
>>>Re: CHange of record pointer
>>>For I am only updating one record in TableA, in fact the problem comes when I do the following
>>>
>>>

** I am on record 10
>>>TableRevert(.t.,"TableA")
>>>** still on record 10
>>>do something
>>>TableUpdate(1,.f.,"TableA")
>>>** still on record 10
>>>** move to record 15
>>>do something
>>>TableUpdate(1,.f.,"TableA")
>>>** record pointer moves back to 10
>>>

>>>
>>>Why?? Any Answer???
>>>
>>>
>>>>Hi,
>>>>
>>>>>>
>>>>I have a form with two tables, both are using "5-Optimistic Table Buffering".
>>>>
>>>>TableA has one-to-many relationship with TableB. When a record is changed in TableA, there are more than one Records affected in TableB.
>>>>
>>>>When I want to issue Table Update, what should I do?
>>>>
>>>>Currently I am using
>>>>TableUpdate(1,.f.,"tableA") and
>>>>TableUpdate(1,.f.,"tableB")
>>>>
>>>>Is it correct?
>>>>>>
>>>>
>>>>There is a danger that the update to tableA will succeed but the update to tableB will fail. Use a transaction to wrap the two updates:
>>>>

>>>>BEGIN TRANSACTION
>>>>IF TABLEUPDATE(1,.f.,"tableA") .AND. TABLEUPDATE(1,.f.,"tableB")
>>>>  END TRANSACTION
>>>>ELSE
>>>>  ROLLBACK
>>>>  * Handle the failure
>>>>ENDIF
>>>>

>>>>
>>>>>>
>>>>I have another problem where when after I issue TableRevert(.t.,"TableA") while on record number X, if I go to Record number Y and issue TableUpdate(1,.f.,"TableA"), the record number changes to X automatically.
>>>>>>
>>>>
>>>>From TableUpdate Help:
>>>>If table buffering is used and multiple records are updated, TABLEUPDATE( ) moves the record pointer to the last record updated.
>>>>
>>>>HTH,
>>>>Viv