I agree with Hugo as the result of TTOC is OS dependent when used without parameters. It may be better to use TRANSFORM function with the specific format instead.

>We have a MYSQL table where a datetime value is stored in a character field. How this happened is a long and very painful story - but changing it is, at the moment, not feasible. We have an interface built by a consultant using Visual Maxframe. (The consultant was also not to blame for the data type of the field). There is a piece of code

>TEXT TO qTxt NOSHOW TEXTMERGE   
>   insert into serno_method_sequence ( idmethodsheet_sequence, idserialno, idmethodsheet, 
>   date_started, idrework, rework_date, defect, action_taken, rework_fix  )
>   values (<<c_methodsheet_sequence.idmethodsheet_sequence>>, <<ocfg.nIDSerialno>>, <<ocfg.nmethodsheet>>,
>   '<<TTOC(DATETIME())>>', <<IIF(thisform.ilRework, curRework.idserno_method_sequence, 0)>>, 
>   '<<IIF(thisform.ilRework, DTOC(DATE()), "")>>', '', '', '')
>	   
>ENDTEXT 
>
>TEXT TO qTxt2 NOSHOW TEXTMERGE   
>   Select LAST_INSERT_ID() as newid
>ENDTEXT 
>
>
>nHndl = thisform.nConnHandle
>IF nHndl <1
>   nHndl = ConnOnly()
>ENDIF 
>
>cErrMsg  =GetSQL(nHndl,QTxt)
>IF !EMPTY(cErrMsg)
>   =MESSAGEBOX(cErrMsg, 16, "Connection Problem")
>   RETURN .T.
>ENDIF    
>
>cErrMsg  =GetSQL(nHndl,QTxt2, 'curGetSNID')
>IF !EMPTY(cErrMsg)
>   =MESSAGEBOX(cErrMsg, 16, "Connection Problem")
>   RETURN .T.
>ENDIF   
>   
>DO case
>CASE RECCOUNT('curGetSNID') = 0
>   nID = 0
>   
>CASE TYPE('curGetSNID.newid') = 'N'
>   nID = curGetSNID.newid
>   
>OTHERWISE
>   nID = VAL(curGetSNID.newid)
>   
>ENDCASE 
>
>IF USED('curGetSNID')   
>   USE IN curGetSNID 
>ENDIF 
>RETURN nID      
>

>
>Back in February we suddenly started experience mixed values in the results. For FEB 10, for example, we got 5098 records where the character field ("date_started") was of the expected form

>02/10/2015 07:08:59

>
>But we also got 2145 records where the result took the unexpected form of

>20150210070743

>
>which is the right value but the wrong format.
>
>I've checked the code and can't find any other place that updates the date_started field in the table.
>
>I'm sure I can build an alogrithm to convert the data into the correct format but I'm looking for some insight as to what conditions could cause this discrepancy. There are several different work stations involved, but we can't map any particular record to any particular work station. Could it be something in the drivers, or the operating system (some are Windows 7 and some are still XP machines)?
>
>Any advice will be appreciated.................Rich