Re: WAS Strange non-display problem, now BAD MATH

This WebDNA talk-list message is from

2002


It keeps the original formatting.
numero = 45980
interpreted = N
texte = Hi Larry,Thanks for that, but probably sorry I asked ;-) lolRegards, Tim> From: Larry Goodhew > Reply-To: (WebCatalog Talk) > Date: Wed, 4 Dec 2002 17:58:26 -0800 > To: (WebCatalog Talk) > Subject: Re: WAS Strange non-display problem, now BAD MATH > > Hi Tim, > > It is kind of involved but what I meant is that if you have a bunch of real > numbers (Real number are numbers with decimal points like: 1001.1, or 1002.1 > or > 1002.3 etc.) not all of those numbers can be stored in binary representation. > That is why, in many financial programs, the numbers are not stored in binary > representation, because people are not amused by rounding errors when they > balance their checkbook. > > When you use a math function on a real number in WebCatalog you are doing math > with the binary representation of the number, which is just a bunch of zeros > and ones, not the decimal number you entered. There is a finite number of > real > numbers that you can store in a binary 32 or 64 bits. There is an infinite > number of real numbers between any two real numbers you choose. So you can > see > that there is no way to represent an infinite number of real numbers in a > finite number of bits. The representations become approximations. Some are > better than others but few are perfectly represented. > > Hope that explains it better! > > Larry Goodhew > >> >> Hi Larry, >> >> I don't really get what you mean? A number's a number isn't it? And this >> subtraction couldn't be more simple. I mean, it works on my onscreen Apple >> calculator.... >> >> Tim >> >> >>> From: Larry Goodhew >>> Reply-To: (WebCatalog Talk) >>> Date: Wed, 4 Dec 2002 17:10:48 -0800 >>> To: (WebCatalog Talk) >>> Subject: Re: WAS Strange non-display problem, now BAD MATH >>> >>> I imagine all the decimal places are caused by the fact that very few real >>> numbers can be represented correctly in base two. Once you start doing math >>> with them the slight errors in representation begin to add up and cause the >>> type of error you are seeing here. >>> >>> Larry Goodhew >>> >>>> >>>> >>>> It seems that the problem has been fixed by making [Format thousands] into >>>> [Format thousands .2f]. It is now working consistently. But the problems >>>> seems to have been caused by bad math!!! >>>> >>>> For example, here's a few of the variables. >>>> >>>> 2002_SALES_2=125360.1 >>>> SAP_SALES_2=123000 >>>> but >>>> [math][SAP_SALES_2]-[2002_SALES_2][/math]= -2360.10000000001 >>>> >>>> Where did all the decimals places come from?!?! It terms of a [math] >>>> calculation, this is pretty bloody easy. I could probably come closer to >>>> that >>>> in my head! (well at least in terms of decimal places...) >>>> >>>> What's going on here?!?! And now i find when you add the [Format thousands >>>> .2f] to a value, it can't display the 'minus' of the negative value..... >>>> >>>> Any ideas on this bad math? >>>> >>>> Regards, >>>> Tim >>>> >>>> >>>> >>>>> From: Gary Krockover >>>>> Reply-To: (WebCatalog Talk) >>>>> Date: Wed, 4 Dec 2002 17:24:13 -0600 >>>>> To: (WebCatalog Talk) >>>>> Subject: Re: Strange non-display problem >>>>> >>>>> Are some of the cells coming back empty and thus collapsing? Add a >>>>> non-breaking space to all the cells:   >>>>> >>>>> That will at least keep the cells from collapsing. >>>>> >>>>> And not sure, but maybe the other dates in this search need to be id'd as >>>>> dates: >>>>> >>>>> From: >>>>> [Search >>>>> db=data/directsales.db&rnSALEDATEdata=[DATE_RANGE1a],[DATE_RANGE1b]&SALEDA >>>>> TE >>>>> type=date&saledatewbrk=,] [text show=F][START_YEAR]_SALES_1=[Sum >>>>> field=sale_total][/text] [/Search] >>>>> >>>>> To: >>>>> [Search >>>>> db=data/directsales.db&rnSALEDATEdata=[DATE_RANGE1a],[DATE_RANGE1b]&SALEDA >>>>> TE >>>>> type=date&DATE_RANGE1atype=date&DATE_RANGE1btype=date&saledatewbrk=,] >>>>> [text >>>>> show=F][START_YEAR]_SALES_1=[Sum field=sale_total][/text] [/Search] >>>>> >>>>> On your lookups, substitue a high value, like 9999 for TextIfNotFound >>>>> and >>>>> see if that is getting added to the value. Perhaps the lookups are >>>>> failing >>>>> and the math isn't working because the 'not found' is a text string. >>>>> >>>>> GK >>>>> >>>>> >>>>> >>>>> | Hi everyone, >>>>> | >>>>> | I'm getting a very strange behaviour in a financial reporting table only >>>>> 7 >>>>> | columns wide, and 12 rows high (one row for each month). TD cells are >>>>> filled | with a variety of information - a search, two lookups, and >>>>> calculations on | those values. | | The table displays something different >>>>> every time.... sometimes there is | nothing in the cell, sometimes the >>>>> correct value, and sometimes the WRONG | value. But when I do a list >>>>> variables at the bottom of the page, ALL | variables have the correct >>>>> value!! | | I initially had the 12 month rows working as a loop, but the >>>>> results were | even worse! Having manually coded the 12 months, it's >>>>> better >>>>> but still | inconsistent. The loop solution should've worked properly, and >>>>> so should | this. | | The values that seems to drop out most often are >>>>> [SAP_SALES_1] and | [BUDGET_SALES_1]. But to prove I'm not mad, the last >>>>> two >>>>> columns ALWAYS | work, even though they use these two fields for the >>>>> calculations.... | | Does anyone have any ideas? All help greatly >>>>> appreciated as always. I've | copied relevant code below. | | Regards, | >>>>> Tim >>>>> | -- | Tim Robinson | IDFK Web Developments | tim@idfk.com.au | 114a/40 >>>>> Yeo >>>>> Street | Neutral Bay 2089 | Australia | Phone +612 9908 2134 | Fax >>>>> +612 9908 4837 | | | | | [TEXT show=f]START_YEAR=2002[/TEXT] | | >>>>> >>>> border=0 cellpadding=6 cellspacing=2 id=text> | >>>> id=whitetext> | Month Ending | >>>> align=right>COMS >>>>> Sales | SAP Sales | >>>> align=right>Sys Var | Budget | >>>>> >>>> align=center colspan=2>Budget Var | | | [Search | >>>>> db=data/directsales.db&rnSALEDATEdata=[DATE_RANGE1a],[DATE_RANGE1b]&SALEDA >>>>> TE >>>>> | type=date&saledatewbrk=,] | [text show=F][START_YEAR]_SALES_1=[Sum >>>>> field=sale_total][/text] | [/Search] | | [TEXT >>>>> show=F]SAP_SALES_1=[LOOKUP | >>>>> db=data/SAP_sales.db&value=[START_YEAR]_1&lookInField=SKU&returnField=SAP_ >>>>> SA >>>>> | LES¬Found=TextIfNotFound][/TEXT] | | [TEXT | >>>>> show=F]SYS_VAR_1=[math][SAP_SALES_1]-[interpret][[START_YEAR]_SALES_1][/in >>>>> te >>>>> | rpret][/math][/TEXT] | | [TEXT show=F]BUDGET_SALES_1=[LOOKUP | >>>>> db=data/SAP_sales.db&value=[START_YEAR]_1&lookInField=SKU&returnField=BUDG >>>>> ET >>>>> | _SALES¬Found=TextIfNotFound][/TEXT] | | [TEXT >>>>> show=F]BUDGET_VAR_1=[math][SAP_SALES_1]-[BUDGET_SALES_1][/math][/TEXT] | | >>>>> [TEXT | >>>>> show=F]VAR_PERCENT_1=[math][BUDGET_VAR_1]/[BUDGET_SALES_1]*100[/math][/TEX >>>>> T] >>>>> | | | [DATE_RANGE1b] | >>>> align=right>[Format thousands | >>>>> .2f][interpret][[START_YEAR]_SALES_1][/interpret][/format] | >>>> align=right>[Format thousands][SAP_SALES_1][/format] | >>>> align=right>[Format thousands][SYS_VAR_1][/format] | >>>> align=right>[Format thousands][BUDGET_SALES_1][/format] | >>>> align=right>[Format thousands][BUDGET_VAR_1][/format] | >>>> align=right>[Format .2f][VAR_PERCENT_1][/format]% | | | | >>>>> Regards, | Tim | -- | Tim Robinson | IDFK Web Developments | >>>>> tim@idfk.com.au >>>>> | 114a/40 Yeo Street | Neutral Bay 2089 | Australia | Phone +612 9908 2134 >>>>> | >>>>> Fax >>>>> +612 9908 4837 | | | | >>>>> ------------------------------------------------------------- | This >>>>> message >>>>> is sent to you because you are subscribed to | the mailing list >>>>> . | To unsubscribe, E-mail to: >>>>> | To switch to the DIGEST mode, >>>>> E-mail >>>>> to | Web Archive of this list is >>>>> at: http://webdna.smithmicro.com/ >>>>> >>>>> >>>>> ------------------------------------------------------------- >>>>> This message is sent to you because you are subscribed to >>>>> the mailing list . >>>>> To unsubscribe, E-mail to: >>>>> To switch to the DIGEST mode, E-mail to >>>>> >>>>> Web Archive of this list is at: http://webdna.smithmicro.com/ >>>>> >>>> >>>> >>>> >>>> ------------------------------------------------------------- >>>> This message is sent to you because you are subscribed to >>>> the mailing list . >>>> To unsubscribe, E-mail to: >>>> To switch to the DIGEST mode, E-mail to >>>> >>>> Web Archive of this list is at: http://webdna.smithmicro.com/ >>>> >>> >>> >>> >>> ------------------------------------------------------------- >>> This message is sent to you because you are subscribed to >>> the mailing list . >>> To unsubscribe, E-mail to: >>> To switch to the DIGEST mode, E-mail to >>> >>> Web Archive of this list is at: http://webdna.smithmicro.com/ >>> >> >> >> >> ------------------------------------------------------------- >> This message is sent to you because you are subscribed to >> the mailing list . >> To unsubscribe, E-mail to: >> To switch to the DIGEST mode, E-mail to >> >> Web Archive of this list is at: http://webdna.smithmicro.com/ >> > > > > ------------------------------------------------------------- > This message is sent to you because you are subscribed to > the mailing list . > To unsubscribe, E-mail to: > To switch to the DIGEST mode, E-mail to > > Web Archive of this list is at: http://webdna.smithmicro.com/ > ------------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://webdna.smithmicro.com/ Associated Messages, from the most recent to the oldest:

    
  1. Re: WAS Strange non-display problem, now BAD MATH (Tim Robinson 2002)
  2. Re: WAS Strange non-display problem, now BAD MATH (Larry Goodhew 2002)
  3. Re: WAS Strange non-display problem, now BAD MATH (Tim Robinson 2002)
  4. Re: WAS Strange non-display problem, now BAD MATH (Larry Goodhew 2002)
  5. WAS Strange non-display problem, now BAD MATH (Tim Robinson 2002)
Hi Larry,Thanks for that, but probably sorry I asked ;-) lolRegards, Tim> From: Larry Goodhew > Reply-To: (WebCatalog Talk) > Date: Wed, 4 Dec 2002 17:58:26 -0800 > To: (WebCatalog Talk) > Subject: Re: WAS Strange non-display problem, now BAD MATH > > Hi Tim, > > It is kind of involved but what I meant is that if you have a bunch of real > numbers (Real number are numbers with decimal points like: 1001.1, or 1002.1 > or > 1002.3 etc.) not all of those numbers can be stored in binary representation. > That is why, in many financial programs, the numbers are not stored in binary > representation, because people are not amused by rounding errors when they > balance their checkbook. > > When you use a math function on a real number in WebCatalog you are doing math > with the binary representation of the number, which is just a bunch of zeros > and ones, not the decimal number you entered. There is a finite number of > real > numbers that you can store in a binary 32 or 64 bits. There is an infinite > number of real numbers between any two real numbers you choose. So you can > see > that there is no way to represent an infinite number of real numbers in a > finite number of bits. The representations become approximations. Some are > better than others but few are perfectly represented. > > Hope that explains it better! > > Larry Goodhew > >> >> Hi Larry, >> >> I don't really get what you mean? A number's a number isn't it? And this >> subtraction couldn't be more simple. I mean, it works on my onscreen Apple >> calculator.... >> >> Tim >> >> >>> From: Larry Goodhew >>> Reply-To: (WebCatalog Talk) >>> Date: Wed, 4 Dec 2002 17:10:48 -0800 >>> To: (WebCatalog Talk) >>> Subject: Re: WAS Strange non-display problem, now BAD MATH >>> >>> I imagine all the decimal places are caused by the fact that very few real >>> numbers can be represented correctly in base two. Once you start doing math >>> with them the slight errors in representation begin to add up and cause the >>> type of error you are seeing here. >>> >>> Larry Goodhew >>> >>>> >>>> >>>> It seems that the problem has been fixed by making [Format thousands] into >>>> [Format thousands .2f]. It is now working consistently. But the problems >>>> seems to have been caused by bad math!!! >>>> >>>> For example, here's a few of the variables. >>>> >>>> 2002_SALES_2=125360.1 >>>> SAP_SALES_2=123000 >>>> but >>>> [math][SAP_SALES_2]-[2002_SALES_2][/math]= -2360.10000000001 >>>> >>>> Where did all the decimals places come from?!?! It terms of a [math] >>>> calculation, this is pretty bloody easy. I could probably come closer to >>>> that >>>> in my head! (well at least in terms of decimal places...) >>>> >>>> What's going on here?!?! And now i find when you add the [Format thousands >>>> .2f] to a value, it can't display the 'minus' of the negative value..... >>>> >>>> Any ideas on this bad math? >>>> >>>> Regards, >>>> Tim >>>> >>>> >>>> >>>>> From: Gary Krockover >>>>> Reply-To: (WebCatalog Talk) >>>>> Date: Wed, 4 Dec 2002 17:24:13 -0600 >>>>> To: (WebCatalog Talk) >>>>> Subject: Re: Strange non-display problem >>>>> >>>>> Are some of the cells coming back empty and thus collapsing? Add a >>>>> non-breaking space to all the cells:   >>>>> >>>>> That will at least keep the cells from collapsing. >>>>> >>>>> And not sure, but maybe the other dates in this search need to be id'd as >>>>> dates: >>>>> >>>>> From: >>>>> [Search >>>>> db=data/directsales.db&rnSALEDATEdata=[DATE_RANGE1a],[DATE_RANGE1b]&SALEDA >>>>> TE >>>>> type=date&saledatewbrk=,] [text show=F][START_YEAR]_SALES_1=[Sum >>>>> field=sale_total][/text] [/Search] >>>>> >>>>> To: >>>>> [Search >>>>> db=data/directsales.db&rnSALEDATEdata=[DATE_RANGE1a],[DATE_RANGE1b]&SALEDA >>>>> TE >>>>> type=date&DATE_RANGE1atype=date&DATE_RANGE1btype=date&saledatewbrk=,] >>>>> [text >>>>> show=F][START_YEAR]_SALES_1=[Sum field=sale_total][/text] [/Search] >>>>> >>>>> On your lookups, substitue a high value, like 9999 for TextIfNotFound >>>>> and >>>>> see if that is getting added to the value. Perhaps the lookups are >>>>> failing >>>>> and the math isn't working because the 'not found' is a text string. >>>>> >>>>> GK >>>>> >>>>> >>>>> >>>>> | Hi everyone, >>>>> | >>>>> | I'm getting a very strange behaviour in a financial reporting table only >>>>> 7 >>>>> | columns wide, and 12 rows high (one row for each month). TD cells are >>>>> filled | with a variety of information - a search, two lookups, and >>>>> calculations on | those values. | | The table displays something different >>>>> every time.... sometimes there is | nothing in the cell, sometimes the >>>>> correct value, and sometimes the WRONG | value. But when I do a list >>>>> variables at the bottom of the page, ALL | variables have the correct >>>>> value!! | | I initially had the 12 month rows working as a loop, but the >>>>> results were | even worse! Having manually coded the 12 months, it's >>>>> better >>>>> but still | inconsistent. The loop solution should've worked properly, and >>>>> so should | this. | | The values that seems to drop out most often are >>>>> [SAP_SALES_1] and | [BUDGET_SALES_1]. But to prove I'm not mad, the last >>>>> two >>>>> columns ALWAYS | work, even though they use these two fields for the >>>>> calculations.... | | Does anyone have any ideas? All help greatly >>>>> appreciated as always. I've | copied relevant code below. | | Regards, | >>>>> Tim >>>>> | -- | Tim Robinson | IDFK Web Developments | tim@idfk.com.au | 114a/40 >>>>> Yeo >>>>> Street | Neutral Bay 2089 | Australia | Phone +612 9908 2134 | Fax >>>>> +612 9908 4837 | | | | | [TEXT show=f]START_YEAR=2002[/TEXT] | | >>>>> >>>> border=0 cellpadding=6 cellspacing=2 id=text> | >>>> id=whitetext> | Month Ending | >>>> align=right>COMS >>>>> Sales | SAP Sales | >>>> align=right>Sys Var | Budget | >>>>> >>>> align=center colspan=2>Budget Var | | | [Search | >>>>> db=data/directsales.db&rnSALEDATEdata=[DATE_RANGE1a],[DATE_RANGE1b]&SALEDA >>>>> TE >>>>> | type=date&saledatewbrk=,] | [text show=F][START_YEAR]_SALES_1=[Sum >>>>> field=sale_total][/text] | [/Search] | | [TEXT >>>>> show=F]SAP_SALES_1=[LOOKUP | >>>>> db=data/SAP_sales.db&value=[START_YEAR]_1&lookInField=SKU&returnField=SAP_ >>>>> SA >>>>> | LES¬Found=TextIfNotFound][/TEXT] | | [TEXT | >>>>> show=F]SYS_VAR_1=[math][SAP_SALES_1]-[interpret][[START_YEAR]_SALES_1][/in >>>>> te >>>>> | rpret][/math][/TEXT] | | [TEXT show=F]BUDGET_SALES_1=[LOOKUP | >>>>> db=data/SAP_sales.db&value=[START_YEAR]_1&lookInField=SKU&returnField=BUDG >>>>> ET >>>>> | _SALES¬Found=TextIfNotFound][/TEXT] | | [TEXT >>>>> show=F]BUDGET_VAR_1=[math][SAP_SALES_1]-[BUDGET_SALES_1][/math][/TEXT] | | >>>>> [TEXT | >>>>> show=F]VAR_PERCENT_1=[math][BUDGET_VAR_1]/[BUDGET_SALES_1]*100[/math][/TEX >>>>> T] >>>>> | | | [DATE_RANGE1b] | >>>> align=right>[Format thousands | >>>>> .2f][interpret][[START_YEAR]_SALES_1][/interpret][/format] | >>>> align=right>[Format thousands][SAP_SALES_1][/format] | >>>> align=right>[Format thousands][SYS_VAR_1][/format] | >>>> align=right>[Format thousands][BUDGET_SALES_1][/format] | >>>> align=right>[Format thousands][BUDGET_VAR_1][/format] | >>>> align=right>[Format .2f][VAR_PERCENT_1][/format]% | | | | >>>>> Regards, | Tim | -- | Tim Robinson | IDFK Web Developments | >>>>> tim@idfk.com.au >>>>> | 114a/40 Yeo Street | Neutral Bay 2089 | Australia | Phone +612 9908 2134 >>>>> | >>>>> Fax >>>>> +612 9908 4837 | | | | >>>>> ------------------------------------------------------------- | This >>>>> message >>>>> is sent to you because you are subscribed to | the mailing list >>>>> . | To unsubscribe, E-mail to: >>>>> | To switch to the DIGEST mode, >>>>> E-mail >>>>> to | Web Archive of this list is >>>>> at: http://webdna.smithmicro.com/ >>>>> >>>>> >>>>> ------------------------------------------------------------- >>>>> This message is sent to you because you are subscribed to >>>>> the mailing list . >>>>> To unsubscribe, E-mail to: >>>>> To switch to the DIGEST mode, E-mail to >>>>> >>>>> Web Archive of this list is at: http://webdna.smithmicro.com/ >>>>> >>>> >>>> >>>> >>>> ------------------------------------------------------------- >>>> This message is sent to you because you are subscribed to >>>> the mailing list . >>>> To unsubscribe, E-mail to: >>>> To switch to the DIGEST mode, E-mail to >>>> >>>> Web Archive of this list is at: http://webdna.smithmicro.com/ >>>> >>> >>> >>> >>> ------------------------------------------------------------- >>> This message is sent to you because you are subscribed to >>> the mailing list . >>> To unsubscribe, E-mail to: >>> To switch to the DIGEST mode, E-mail to >>> >>> Web Archive of this list is at: http://webdna.smithmicro.com/ >>> >> >> >> >> ------------------------------------------------------------- >> This message is sent to you because you are subscribed to >> the mailing list . >> To unsubscribe, E-mail to: >> To switch to the DIGEST mode, E-mail to >> >> Web Archive of this list is at: http://webdna.smithmicro.com/ >> > > > > ------------------------------------------------------------- > This message is sent to you because you are subscribed to > the mailing list . > To unsubscribe, E-mail to: > To switch to the DIGEST mode, E-mail to > > Web Archive of this list is at: http://webdna.smithmicro.com/ > ------------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://webdna.smithmicro.com/ Tim Robinson

DOWNLOAD WEBDNA NOW!

Top Articles:

Talk List

The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...

Related Readings:

WebCat 4.0 & WebMerchant (2000) WebDNA Solutions ... sorry! (1997) search form problem.. (1997) Bug Report, maybe (1997) Encryption of E-mail (1998) calculating ShipCost depending on weight (1997) WCS Newbie question (1997) Falsifying [BROWSERNAME] (1998) WebCatalog Features (1997) WC2.0 Memory Requirements (1997) Configuration Q (1998) emailer w/F2 (1997) Examples Link? (2000) Bug Report, maybe (1997) Searching using numbers (1998) WebDNA to Apache RewriteMap (2005) Grouping search fields, etc. (1997) WebCatalog NT beta 18 now available (1997) List archive status? (1999) Security Question (1997)