Re: Double click

Thanks for the ideas. I am gonna try a few things, using what you said, and
using some ideas I have. I would be more than happy to let you know how
things go if you want me to.
Thanks again for your time.


"Mark J. McGinty" <mmcginty@xxxxxxxxxxxxxxx> wrote in message
news:V58re.9899$tr.3551@xxxxxxxxxxxxx
>
> "Jeff" <gig_bam@xxxxxxxxxxxx> wrote in message
> news:eNidnWaWS5aEdDHfRVn-vA@xxxxxxxxxxxxxxx
>> No, your assumptions are right on. I have a form that folks use to report
>> matches (golf), and what is happening, is some not being careful, and
>> clicking twice, or clicking a mouse button that is programed for double
>> click.
>
> If you had a way to tell before further processing that the submission was
> likely a duplicate, you should be able to just discard the duplicate, and
> leave the first intact. Like maybe a separate table containing just user
> id and time of day (and/or maybe even some identifying facet of the match
> record they're submitting.) Keep it external to the rest of your tables,
> and use it only for a short historical view. The viability would depend
> on what asumptions might be made as far as how close together a given user
> should be able to submit an entry.
>
> If you can pick-out the duplicates by hand, then it may be possible to
> identify them programattically before they happen; that, I'd think, would
> be best case... of course it would also add some overhead, but then again,
> so does duplicate processing and subsequent cleanup.
>
> A db constraint-based solution might be possible as well, but impossible
> to say without knowledge of the schema.
>
> Almost lastly, I seem to recall that the client-side problems I had with
> Netscape had to do specifically with manipulating elements in other
> frames, I think hiding the button worked just fine, I just couldn't show
> it again when the frame I posted to completed. If the posting page is
> reloaded as part of the process, that wouldn't apply.
>
> And finally lastly, the main reason I gave up was because all that was at
> stake was some server time, when clueless users submitted a lengthy search
> more than once because they were impatient. (That doesn't help anything,
> it just adds more load.) And I did leave the IE solution in place, with
> the rationale, at least 98% of them won't be able to submit twice... Point
> being that I didn't mean to imply you should give up, by mentioning the
> fact that I did. :-)
>
>
> -Mark
>
>
>> I could time stamp, then delete, but lots of stuff happens when someone
>> reports. I have to go to different tables and adjust totals and a bunch
>> of other things. Which I have to do manually anyway. I was just hoping
>> there was something that could be done.
>>
>> Thanks for your time, and for your response.
>>
>>
>> "Mark J. McGinty" <mmcginty@xxxxxxxxxxxxxxx> wrote in message
>> news:uswJkR6bFHA.3048@xxxxxxxxxxxxxxxxxxxxxxx
>>>
>>> "Jeff" <gig_bam@xxxxxxxxxxxx> wrote in message
>>> news:KIWdnT2ZLriTKzHfRVn-sw@xxxxxxxxxxxxxxx
>>>> Thanks for your NOT SO FRIENDLY response. I thought, since the form is
>>>> written with asp, and it submits to an asp page, and the results are
>>>> displayed with asp, that this was the place to ask. I didn't see a NG
>>>> labled submit buttons.
>>>> I know that most in here, if there was a solution, would give it.
>>>> Therefore, I do not accept your response, and will wait patiently to
>>>> see if someone can help me.
>>>
>>> Don't mind him, he's just the resident know-nothing ***-head. Whatever
>>> glimpse of programming he may have once had is now eclipsed by a bad
>>> personality and a tendency to be wrong much of the time. (I suspect
>>> there may be a language barrier issue as well, but his level of
>>> arrogance isn't usually present in such cases, so the jury is still out
>>> on that score.)
>>>
>>> I'm going to assume that by "double click" you mean double submit, where
>>> the user clicks the submit button a second time, before the first
>>> request has been delivered to the user's browser, but after the first
>>> request has been received by the server.
>>>
>>> Unfortunately there isn't a bunch you can do about this on the server
>>> side. You might try time-stamping the requests as they come in, and
>>> discarding ant that are received within a certain proximity, but this
>>> can be difficult to do, depending on the processing that's involved.
>>>
>>> There is of course the widely used "social engineering" scheme (i.e.,
>>> the 'do not press this button more than once' warning) but sadly it's
>>> less than effective.
>>>
>>> There are some client-side remedys, such as hiding the button after it
>>> has been clicked, but before the request has been submitted (process
>>> onclick) but I had 7 kinds of hell trying to make that work in non-IE
>>> browsers, and finally gave up.
>>>
>>> You might want to try clarifying you post (particularly if my
>>> assumptions were incorrect) and resubmitting it to
>>> microsoft.public.inetsdk.programming.scripting.jscript. There is a
>>> definite server-side slant to your issue, and doubly-submitted requests
>>> surely can cause database issues, but your most effective solutions are
>>> likely to reside on the client end.
>>>
>>>
>>> -Mark
>>>
>>>
>>> btw, to any of you that might care to flame me for taking a shot at that
>>> guy, first ask yourself, "does he have a valid point?" And second, be
>>> sure to quote back the "offensive" part because that guy likely has me
>>> kill-filtered, and I'd hate for him to have to wonder about what was
>>> said.
>>>
>>>
>>>> "Evertjan." <exjxw.hannivoort@xxxxxxxxxxxx> wrote in message
>>>> news:Xns9673F29B1D72Eeejj99@xxxxxxxxxxxxxxxxxx
>>>>> Jeff wrote on 12 jun 2005 in microsoft.public.inetserver.asp.db:
>>>>>
>>>>>> This has probably been asked 100's of times, but I couldn't find
>>>>>> anything on it.
>>>>>>
>>>>>> Is there a small script of some sort, that would disable the ability
>>>>>> to double click a submit button>?
>>>>>
>>>>> ASP runs serverside and does know noting about clicking.
>>>>>
>>>>> Databases under ASP know even less.
>>>>>
>>>>> Please ask a apopriate clientside NG.
>>>>>
>>>>> --
>>>>> Evertjan.
>>>>> The Netherlands.
>>>>> (Replace all crosses with dots in my emailaddress)
>>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>
>


.

Loading