Hi Tore,

"Nothing but bugs" is surely an exaggeration.
I admit that I get upset at the tone some people take as regards FabioL's posts regarding "bugs".

I have no idea as to why some people get red hot at the conjoined words "bug" and "VFP", but it sure is clear that this is a hot-button for many, and especially some of our more esteemed colleagues.

I have no ideas why, even at this point in FabioL's writings, people cannot seem to make an allowance for FabioL's limited command of English.

FabioL has exposed numerous inconsistencies and bugs in VFP. That many of the inconsistencies are long known and well accepted does not mean that they do not deserve questioning, or at the very least some constructive debate.

I see, just off the top of my head, 2 additions to VFP 9 that help demonstrate. One was vociferously argued by many that it was not at all needed - the option to supress the useles open table dialog. One was simply a long accepted deficiency - the ability for TYPE() to return the fact that something is an array.
Both of these will now be a part of the language! If no one had said anything then they would remain only dreams for some people.

I find (it requires a good attitude to do so) that the huge majority of FabioL's "bugs" are interesting and certainly deserving of both exposure and discussion.

One final note regarding this type of issue.
That perhaps many of the things FabioL exposes are long-term "features" of the product matters not one wit *IF* they remain unmentioned in the product documentation, and I mean the product documentation here and NOT some well obscured MSKB article or some third party white paper. As long as things he raises remain unexplained in the product documentation then they merit the label "bug". Those things that are documented properly but which are disagreeable to FabioL (often for good and legitimate reason, I might add) then I'd say he should use "poor design" or "questionable design" to characterize them.

One thing is sure, however, and that is that FabioL does not merit the level of abuse shown to him here. I bet that in many cases he has taken the time to expose things that we ourselves just haven't taken the time to pursue. I know that's the case for me.

Jim


>According to Fabio, VFP is nothing but bugs, so please have that in mind when you read his messages.
>
>The way he has 'cried wolf' over the last year, I have learnt that whenever things does not work the way Fabio expect them to work, they are bugs in his mind. Fabio is a remarkable programmer, but he needs to be more realistic. Microsoft has to decide whether they shall keep up on making new versions of VFP compatible with older versions, or do it the Fabio way. I am happy that compatibility is most important.
>
>
>>Fabio told me that VCX and SCX object are just vfp tables that can be edited with a simple browse command, he also told me that you cannot trust abiut VFP calculation and i asked him to tell me more about this "bug"
>>
>>E ora te speri di cavartela così ?
>>più che una pulce mi hai piazzato un cammello nell'orecchio.... che è 'sta storia dei calcoli ?
>>Comunque sapevo che le classi sono nient'altro che tabelle, ma conitnuo a non capire, ammesso che esista, il metodo per ovviare al piccolo inconveniente di cui all'oggetto :-)
>>Comunque una cosa l'ho capita... è un gran bel casino !
>>Alla luce di tutto ciò...secondo te conviene continuare a programmare in VFP o conviene passare ad un altro linguaggio (ammesso che ne esista uno affidabile) ?
>>Tu cosa usi ? e soprattutto dormi sonno tranquilli ? :-)
>>
>>
>>
>>>Nonchalance, mi fai sorridere, se questo fosse il peggio, sarei ben felice.
>>>( lo sai che non puoi fidarti di come fa i calcoli ?)
>>>
>>>Comunque, soprassediamo.
>>>
>>>VCX e SCX sono tabelle VFP. Se le apri con USE puoi facilmente vederlo.
>>>Rapidamente,
>>>Apri una classe, chiudi, sul command ti resta MODIFI .... OF ...,
>>>riprendi la riga e cambia MODIFY ... OF con USE, enter.
>>>BROWSE.
>>>
>>>Non è semplice ed intuitivo, ma cerco di spiegartelo:
>>>VFP mette il codice ( p-code ) compilato in un campo memo dei VCX/SCX di nome ObjCode ( è il codice degli fxp),
>>>quando tu apri una Classe/Form, VFP non analizza il codice ASCII ( come qualunque essere pensante farebbe ),
>>>no!, lui prende il p-code, lo analizza, estrae i puntatori al codice sorgente che ha generato il p-code
>>>e usa questi per estrarre le procedure dal campo methods ( dove c'è il programma ASCII ).
>>>
>>>Quindi, se per qualche motivo le due cose non vanno d'accordo,
>>>oppure ( come penso ) la cosa è inutilmente complessa e VFP ha un bug che si può presentare ARANDOM,
>>>i puntatori vanno per conto loro e ti estraggono o routine diverse ( come ti è capitato )
>>>oppure ( come è capitato a me ) pezzi sparsi di routine diverse.
>>>
>>>Un vero gusto!
>>>
>>>Ciao.
>>>Fabio
>>>>Cavolo.... e me lo dici con questa nonchalance o..come diavolo si scrive ?
>>>>'to fatto mi ha fatto usire pazzo per mesi e mesi... anni ormai ma come ho scritto interessava solo gli eventi close e open del dataenvironment, così mi risultava piuttosto semplice mterci una "pezza manuale".
>>>>Cosa intendo per ObjCode... spiegami melio ch non ho capito bne cosa intendi dire.
>>>>Grazie comunque
>>>>Alessio
>>>>
>>>>
>>>>What do you mean with the method code of ObjCode ?
>>>>Thank you anyway
>>>>Alessio
>>>>
>>>>
>>>>>>Sometimes happens, and it happened since the VFP6 version that the code i put inside a method migrate in another one.
>>>>>>Till today it happened only in the Dataenvironment. After some ununderstandable errors i go to see the code and what I wrote inside the open table event went inside the close table.. till today it was a little bit nasty but it was limited o these two events.
>>>>>>Today the same thing happened in two method of a textbox object.
>>>>>>Anyone knows anything about this ? how can i get rid of this problem ?
>>>>>>Thank you
>>>>>>Alessio
>>>>>
>>>>>Confermo, sia su VCX che SCX.
>>>>>Quello che ho osservato è che il codice si scambia a coppie due metodi.
>>>>>
>>>>>Il problema nasce dal fatto che VFP usa sia il codice del campo Methods che il codice ObjCode
>>>>>( se cambi il codice di Methods e ricompili senza prima cancellare ObjCode, non ottieni alcun effetto ).
>>>>>Penso che ci sia un bug nei codice che correla Methods con ObjCode ( objCode è una sequenza di link a 16 bits )
>>>>>Impossibile da riprodurre.
>>>>>
>>>>>I confirm, is on VCX that SCX.
>>>>>What I have observed is that the code exchanges to braces two methods.
>>>>>
>>>>>I think, that the problem is born from the fact that VFP uses is the code of the Methods field that the ObjCode code
>>>>>(if changes the code of Methods and compile without before to cancel ObjCode, does not obtain some effect).
>>>>>Probably there is a bug in the code that correlates Methods with ObjCode (objCode is one sequence of link to 16 bits)
>>>>>Impossible to reproduce.