Help formatting numbers without printing to screen

[xra7en]

I am curious what is the difference with what is wanted here and this:
https://www.qb64.org/forum/index.php?topic=840.msg100462#msg100462

almost nothing, except that I did rewrite that a little cleaner, and now it seems to work, but if you run the code I just posted in a loop, using an increment of - say - 2.4 and deleay it about 1 second or less (there is a reason for that chaos), starting var 10000000.23 (some crazy decimal.. just for use) what I see when I run it, is that around the 1million spot the comma disappears for a second then reappears.

here is a sample you can try

Code: QB64: [Select]

1. counter = 1
2. somenumber = 1000000000.23
3. DO
4.     LOCATE 1, 1, 0
5.  
6.     PRINT somenumber
7.     somenumber = somenumber + 2.4
8.  
9.  
10.     PRINT num2str$(somenumber, 3) 'use the code function i posted earlier
11.     _DELAY (1)
12. LOOP UNTIL counter = 1000
13.  

the million comma seems to come and go for some reason, and I think it is how I am doing the modulous.(old fashion way vs the "%" method)
Your question is very valid. The only reason I ask is that sometimes my OCD kicks in, and I think, in this case, that I am not seeing the forest for the trees with the PRINT USING format. since it does most the work. And wondering if I am on the right track writing a custom funciton that does almost the same thing.

oooorrrrrr

is there a flag I am missing in the PRINT USING to rid the space padding ?

[SMcNeill]

That may be a glitch with the display flickering.

Try adding _DISPLAY to your loop to stop flicker.

If that doesn’t work, put a check to count commas (a DO LOOP with INSTR can do that easily enough), and then pause execution when a value appears without it.  Print that value to the screen, and we can use it to figure out why the comma would appear sometimes, but not others.

Once the fringe case is identified (if it exists), then it can be debugged/accounted for.  ;)

[bplus]

Code: QB64: [Select]

1. DIM somenumber AS DOUBLE '<<<<<<<<<<< added
2. counter = 1
3. somenumber = 1000000000.23
4. DO
5.     CLS '<<<<<<<<<<<<< added to make sure seeing fresh number clearly
6.     LOCATE 1, 1, 0 ' <<<< 0 to turn off cursor???
7.  
8.  
9.     'somenumber = somenumber + 2.4
10.     somenumber = somenumber + RND * 1000 - 500 '<<< to make more interesting
11.     PRINT somenumber '<<<<<<<<<<<<<<<<< moved to compare equal values
12.     PRINT xnum2str$(somenumber, 3) 'use the code function i posted earlier  '<<<< added the x in front
13.     _DELAY (1)
14. LOOP UNTIL counter = 1000
15.  
16.  
17.  
18. FUNCTION xnum2str$ (num AS DOUBLE, dec AS INTEGER)
19.  
20.     '// convert to number delimted string
21.  
22.     DIM number AS STRING
23.     DIM precision AS STRING
24.     DIM convert AS STRING
25.     DIM tempStr AS STRING
26.     DIM I AS INTEGER
27.  
28.     convert = STR$(num) '                                               convert it all to a strig
29.     number = LEFT$(convert, INSTR(convert, ".") - 1) '                  store the integer
30.     precision = MID$(convert, INSTR(convert, "."), LEN(convert)) '      store the decimal
31.  
32.     tempStr = ""
33.     FOR I = LEN(number) TO 1 STEP -1 '                                  create comma delimted string
34.         tempStr = MID$(convert, I, 1) + tempStr
35.         IF I / 3 = INT(I / 3) THEN tempStr = "," + tempStr
36.     NEXT
37.     number = tempStr
38.  
39.  
40.     tempStr = "" '                                                      adjust len of precisio to match
41.     FOR I = 1 TO LEN(precision)
42.         tempStr = tempStr + MID$(precision, I, 1)
43.         IF I >= dec THEN EXIT FOR
44.     NEXT
45.     precision = tempStr
46.     xnum2str$ = number + precision
47. END SUB
48.  
49.  

This looks solid as a rock without _display using QB64 X64. I will complain precision 3, gives only 2 decimal places.

Testing with X32 and negative numbers (dec set at 2), I find error in code:


I also see that my commatose code is not handling neg numbers quite right either. dang!

[freetrav]

Just as a random side comment on this subject, I've seen other BASICs with either an SPRINT/SPRINT USING statement (SPRINT -> PRINT to a string) or a USING$ function, which is probably implemented by "riding" the code of PRINT USING. Perhaps it might be worth thinking about implementing _USING$(pattern$,value)?

[SMcNeill]

Code: QB64: [Select]

1. DIM somenumber AS DOUBLE '<<<<<<<<<<< added
2. counter = 1
3. somenumber = 1000000000.23
4. DO
5.     CLS '<<<<<<<<<<<<< added to make sure seeing fresh number clearly
6.     LOCATE 1, 1, 0 ' <<<< 0 to turn off cursor???
7.  
8.  
9.     'somenumber = somenumber + 2.4
10.     somenumber = somenumber + RND * 1000 - 500 '<<< to make more interesting
11.     PRINT somenumber '<<<<<<<<<<<<<<<<< moved to compare equal values
12.     PRINT xnum2str$(somenumber, 3) 'use the code function i posted earlier  '<<<< added the x in front
13.     _DELAY (1)
14. LOOP UNTIL counter = 1000
15.  
16.  
17.  
18. FUNCTION xnum2str$ (num AS DOUBLE, dec AS INTEGER)
19.  
20.     '// convert to number delimted string
21.  
22.     DIM number AS STRING
23.     DIM precision AS STRING
24.     DIM convert AS STRING
25.     DIM tempStr AS STRING
26.     DIM I AS INTEGER
27.  
28.     convert = STR$(num) '                                               convert it all to a strig
29.     number = LEFT$(convert, INSTR(convert, ".") - 1) '                  store the integer
30.     precision = MID$(convert, INSTR(convert, "."), LEN(convert)) '      store the decimal
31.  
32.     tempStr = ""
33.     FOR I = LEN(number) TO 1 STEP -1 '                                  create comma delimted string
34.         tempStr = MID$(convert, I, 1) + tempStr
35.         IF I / 3 = INT(I / 3) THEN tempStr = "," + tempStr
36.     NEXT
37.     number = tempStr
38.  
39.  
40.     tempStr = "" '                                                      adjust len of precisio to match
41.     FOR I = 1 TO LEN(precision)
42.         tempStr = tempStr + MID$(precision, I, 1)
43.         IF I >= dec THEN EXIT FOR
44.     NEXT
45.     precision = tempStr
46.     xnum2str$ = number + precision
47. END SUB
48.  
49.  

This looks solid as a rock without _display using QB64 X64. I will complain precision 3, gives only 2 decimal places.

Testing with X32 and negative numbers (dec set at 2), I find error in code:


I also see that my commatose code is not handling neg numbers quite right either. dang!

Just check for negative values.  If found, use ABS() on the number, generate the result, and then add a “-“ to the end result.  ;)

[bplus]

Yeah I have it fixed in commatose:

Code: QB64: [Select]

1. RANDOMIZE TIMER
2. DIM somenumber AS DOUBLE '<<<<<<<<<<< added
3. counter = 1
4. somenumber = 0
5. DO
6.     CLS '<<<<<<<<<<<<< added to make sure seeing fresh number clearly
7.     LOCATE 1, 1, 0 ' <<<< 0 to turn off cursor???
8.  
9.     'somenumber = somenumber + 2.4
10.     somenumber = somenumber - RND * 10 + 4 '<<< to make more interesting  drift down
11.     PRINT "               PRINT somenumber: "; somenumber '<<<<<<<<<<<<<<<<< moved to compare equal values
12.     PRINT " PRINT xnum2str$(somenumber, 3): "; xnum2str$(somenumber, 3) 'use the code function i posted earlier  '<<<< added the x in front
13.     PRINT "PRINT Commatose$(somenumber, 3): "; Commatose$(somenumber, 3)
14.     _LIMIT 60
15. LOOP UNTIL counter = 1000
16.  
17.  
18.  
19. FUNCTION xnum2str$ (num AS DOUBLE, dec AS INTEGER)
20.  
21.     '// convert to number delimted string
22.  
23.     DIM number AS STRING
24.     DIM precision AS STRING
25.     DIM convert AS STRING
26.     DIM tempStr AS STRING
27.     DIM I AS INTEGER
28.  
29.     convert = STR$(num) '                                               convert it all to a strig
30.     number = LEFT$(convert, INSTR(convert, ".") - 1) '                  store the integer
31.     precision = MID$(convert, INSTR(convert, "."), LEN(convert)) '      store the decimal
32.  
33.     tempStr = ""
34.     FOR I = LEN(number) TO 1 STEP -1 '                                  create comma delimted string
35.         tempStr = MID$(convert, I, 1) + tempStr
36.         IF I / 3 = INT(I / 3) THEN tempStr = "," + tempStr
37.     NEXT
38.     number = tempStr
39.  
40.  
41.     tempStr = "" '                                                      adjust len of precisio to match
42.     FOR I = 1 TO LEN(precision)
43.         tempStr = tempStr + MID$(precision, I, 1)
44.         IF I >= dec THEN EXIT FOR
45.     NEXT
46.     precision = tempStr
47.     xnum2str$ = number + precision
48. END SUB
49.  
50.  
51. FUNCTION Commatose$ (value AS DOUBLE, precision AS INTEGER)
52.     DIM v AS DOUBLE
53.     v = value 'copy so it doesn't get changed
54.     IF v < 0 THEN s$ = "-": v = v * -1 ELSE s$ = ""
55.  
56.     sv$ = LTRIM$(STR$(v)) '<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< ltrim$ remove leading space!
57.     dot = INSTR(sv$, ".")
58.     IF dot THEN
59.         i$ = MID$(sv$, 1, dot - 1)
60.         d$ = LEFT$(MID$(sv$, dot + 1) + STRING$(precision, "0"), precision)
61.     ELSE
62.         i$ = sv$
63.         d$ = STRING$(precision, "0")
64.     END IF
65.     FOR i = LEN(i$) TO 1 STEP -1
66.         c = c + 1
67.         IF c = 4 THEN c = 1: b$ = "," + b$
68.         b$ = MID$(i$, i, 1) + b$
69.     NEXT
70.     Commatose$ = s$ + b$ + "." + d$
71. END FUNCTION
72.  
73.  

[xra7en]

That may be a glitch with the display flickering.

Try adding _DISPLAY to your loop to stop flicker.

If that doesn’t work, put a check to count commas (a DO LOOP with INSTR can do that easily enough), and then pause execution when a value appears without it.  Print that value to the screen, and we can use it to figure out why the comma would appear sometimes, but not others.

Once the fringe case is identified (if it exists), then it can be debugged/accounted for.  ;)

you might be on to something...

[xra7en]

btw, after reading this and my otherone - they are extremely similar, it is the same issue, with a sleight variation - any chance a mod can move this to that thread. Makes more sense.

I think I just solved it too, it is something to do with the way I am parsing the string

using anything larger than 100k or 1mil the comma drifts. while I will look here in a sec, but for public curiosity, is qb64 strings zero based or 1?

abcdef
does "a" position = zero or 1?

[SMcNeill]

btw, after reading this and my otherone - they are extremely similar, it is the same issue, with a sleight variation - any chance a mod can move this to that thread. Makes more sense.

I think I just solved it too, it is something to do with the way I am parsing the string

using anything larger than 100k or 1mil the comma drifts. while I will look here in a sec, but for public curiosity, is qb64 strings zero based or 1?

abcdef
does "a" position = zero or 1?

1st  position

[xra7en]

OK. works!! Got it. Woot

Code: QB64: [Select]

1. FUNCTION num2str$ (num AS DOUBLE, dec AS INTEGER)
2.  
3.     '// convert to number delimted string
4.  
5.     DIM number AS STRING '                                              stores whole number(w/o dec) minus "."
6.     DIM precision AS STRING '                                           stores decimals (minus ".")
7.     DIM convert AS STRING '                                             stores the orginal number in string format
8.     DIM tempStr AS STRING '                                             used to create the new formated string
9.     DIM I, j AS INTEGER '                                               i=loop counter, j= delimeter counter
10.  
11.     convert = STR$(num) '                                               convert it all to a strig
12.     number = LEFT$(convert, INSTR(convert, ".") - 1) '                  store the integer
13.     precision = MID$(convert, INSTR(convert, ".") + 1, LEN(convert)) '  store the decimal
14.     number = RTRIM$(LTRIM$(number)) '                                   trim it up
15.    
16.        IF num < 1000 THEN '                                                if its less than 1000, then no comma needed
17.         num2str$ = number + "." + LEFT$(precision, dec) '               assign function for dec (precision)
18.         EXIT SUB '                                                      leave this sub now
19.     END IF
20.  
21.  
22.  
23.  
24.     j = 1 '                                                             set delimeter counter
25.     tempStr = ""
26.     FOR I = LEN(number) TO 1 STEP -1 '                                  count from the right to left of number
27.  
28.         IF j > 3 THEN '                                                 if delimeter counter is > 3 then
29.             tempStr = "," + tempStr '                                   add a comma
30.             j = 1 '                                                     reset the delimeter counter
31.         END IF
32.  
33.  
34.         tempStr = MID$(number, I, 1) + tempStr '                        add the next digit from "number" to tempstr
35.         j = j + 1 '                                                     increase the delimeter counter
36.  
37.     NEXT '                                                              keep looping
38.     number = tempStr '                                                  reassign number to the actual string created
39.  
40.  
41.     tempStr = "" '                                                      clear tempStr so we can make the percision
42.     FOR I = 1 TO LEN(precision) '                                       start looping through dec
43.         tempStr = tempStr + MID$(precision, I, 1) '                     create it
44.         IF I >= dec THEN EXIT FOR '                                     is it the number of places(dec) we want? exit this loop
45.     NEXT
46.     precision = tempStr '                                               assign the precision var to the newly created decimal length
47.     num2str$ = number + "." + precision '                               return it back as a fll string with "."
48. END SUB
49.  
50.  
51.  

the only way I could get this to work was to use the "J" loop.
I tried the MOD, and I tried an old method of

Code: QB64: [Select]

1. if i/3 = int(i/3) then add a comma

none of them worked.
This does.
Returns a comma delimited string variable from an integer or double

Opinions? Suggestions?

I had to stop looking at my Steam Que so I could focus on this LOL...

[Fifi]

Just as a random side comment on this subject, I've seen other BASICs with either an SPRINT/SPRINT USING statement (SPRINT -> PRINT to a string) or a USING$ function, which is probably implemented by "riding" the code of PRINT USING. Perhaps it might be worth thinking about implementing _USING$(pattern$,value)?

As another side comment: would it be so difficult to interface (implement) two new reserved words with the C/C++ printf() and sprintf() stdio.h functions using their syntax and parameters that would avoid to "translate" the current PRINT USING while being more powerfull?

Just my two cents.

[xra7en]

Just as a random side comment on this subject, I've seen other BASICs with either an SPRINT/SPRINT USING statement (SPRINT -> PRINT to a string) or a USING$ function, which is probably implemented by "riding" the code of PRINT USING. Perhaps it might be worth thinking about implementing _USING$(pattern$,value)?

As another side comment: would it be so difficult to interface (implement) two new reserved words with the C/C++ printf() and sprintf() stdio.h functions using their syntax and parameters that would avoid to "translate" the current PRINT USING while being more powerfull?

Just my two cents.

hey, every penny counts :-)
I used to work with C++/# in my old BBS days. Wish I would have stuck with it, because I went from Telegard->Invision-Infinity (all pascal) to Proboard(C++), very weird change. Not only that, with all the pascal making a hit, I thought that would be the language of choice, and BOOM here comes "C" LOL

OK I digress.....

sprint, print etc.. all output to the screen/console

I needed something that returned a var that does the same thing.
or at minimum a PHP version of "number_format" which does not require a template. Actually, I should check that script source out and see how they did that (slow thinker LOL)

I do like your idea, however Im wanting to keep the langs separate (at least for now).

Nice Idea though

EDIT:
================================

Here is afunctional version in PHP... The original is in C, So will look at that. I personally think this might be a useful function to add in QB64 IMHO

Code: QB64: [Select]

1. function number_format($num){
2.   $num=(string)$num;
3.   $len=strlen($num);
4.   $newnum='';
5.   for ($i=0; $i<$len; ++$i) {
6.     if (($i%3==0) && $i) {
7.       $newnum=','.$newnum;
8.     }
9.     $newnum=$num[$len-$i-1].$newnum;
10.   }
11.   return $newnum;
12. }

[TempodiBasic]

Hi
fine to add meself to this thread again

I have wrote a my routine... :-)
yes! Why? As someone knows the way to write numbers in italian is different from that used here, (also date and some few other things...) so I try to make my algorithm in QB64 working on string format and letting to show international format (, for thousand and . for decimal or that italian that is the inverse system  i.e.  international 1,000,009.999 = italian 1.000.009,999 ). So here my work.

I take your code and add my function.
PS: Just modify the number maker to have greater numbers... and both negative and positive numbers...on program's controls you can exit by keyboard and stop and go by space/anyOtherKey.

Attached are the  images of results...


as you can see in image3
1.  I am not able to manage Commatose$'s output, I have got a flickering passing from negative to positive numbers...
2. xnum2str$ have a bug...

Thanks to read and to reply

PS I have no knowledge in the matter please tell me, but after the decimal point do the digits need no formatting with separator ?


PSS sorry I have forgotten to post the code....

Code: QB64: [Select]

1. RANDOMIZE TIMER
2. DIM somenumber AS DOUBLE '<<<<<<<<<<< added
3. DIM counter AS INTEGER, signs AS INTEGER
4.  
5. counter = 1
6. somenumber = 0
7. DO
8.     CLS '<<<<<<<<<<<<< added to make sure seeing fresh number clearly
9.     LOCATE 1, 1, 0 ' <<<< 0 to turn off cursor???
10.     IF (RND * 5 + 1) > 3 THEN signs = -1 ELSE signs = 1
11.  
12.     somenumber = (somenumber - (RND * 10) + 40000000) * signs
13.     PRINT "                    PRINT somenumber: "; somenumber
14.     PRINT " PRINT xnum2str$(somenumber, 3):      "; xnum2str$(somenumber, 3)
15.     PRINT " PRINT num2str$(somenumber, 3):       "; num2str$(somenumber, 3)
16.     PRINT "PRINT Commatose$(somenumber, 3):       "; Commatose$(somenumber, 3)
17.     PRINT "PRINT MyNumToString$(somenumber, 3, 1)"; MyNumToString$(somenumber, 3, 1)
18.     PRINT "PRINT MyNumToString$(somenumber, 3, 2)"; MyNumToString$(somenumber, 3, 2)
19.     _LIMIT 60
20.     IF _KEYDOWN(32) THEN SLEEP
21. LOOP UNTIL _KEYDOWN(27)
22.  
23.  
24.  
25. FUNCTION MyNumToString$ (num AS DOUBLE, dec AS INTEGER, mode AS INTEGER)
26.     DIM Thousand AS STRING * 1, Decimal AS STRING * 1, NumberStringed AS STRING
27.     DIM DecimalString AS STRING, NumberString AS STRING, Minus AS STRING * 1
28.     DIM DecimalPoint AS INTEGER, LenghtDecimal AS INTEGER, LenghtInteger AS INTEGER
29.     DIM Starter AS INTEGER, Counter AS INTEGER, Steps AS INTEGER
30.     IF mode = 1 THEN
31.         ' english puntaction for number
32.         Thousand = ","
33.         Decimal = "."
34.     ELSEIF mode = 2 THEN
35.         ' italian puntaction for number
36.         Thousand = "."
37.         Decimal = ","
38.     ELSE
39.         PRINT "invalid Mode setting"
40.         EXIT FUNCTION
41.     END IF
42.     Minus = ""
43.     NumberStringed = LTRIM$(STR$(num))
44.     IF (num < 0) AND INSTR(NumberStringed, "-") = 1 THEN
45.         NumberStringed = MID$(NumberStringed, 2)
46.         Minus = "-"
47.     END IF
48.     DecimalPoint = INSTR(NumberStringed, ".")
49.     DecimalString = RIGHT$(NumberStringed, LEN(NumberStringed) - DecimalPoint)
50.     LenghtDecimal = LEN(RIGHT$(NumberStringed, LEN(NumberStringed) - DecimalPoint))
51.     NumberString = LEFT$(NumberStringed, DecimalPoint - 1)
52.     LenghtInteger = LEN(NumberString)
53.  
54.     IF (LenghtInteger < 4) THEN
55.         MyNumToString$ = Minus + NumberString + Decimal + LEFT$(DecimalString, dec)
56.     ELSE
57.         NumberStringed = ""
58.         Starter = (LenghtInteger MOD 3)
59.         Steps = INT(LenghtInteger / 3)
60.         IF Starter > 0 THEN NumberStringed = LEFT$(NumberString, Starter) + Thousand
61.         FOR Counter = 0 TO Steps - 1
62.             NumberStringed = NumberStringed + MID$(NumberString, Starter + (Counter * 3) + 1, 3) + Thousand
63.         NEXT
64.         NumberStringed = LEFT$(NumberStringed, LEN(NumberStringed) - 1)
65.         MyNumToString$ = Minus + NumberStringed + Decimal + LEFT$(DecimalString, dec)
66.     END IF
67. END FUNCTION
68.  
69.  
70. FUNCTION num2str$ (num AS DOUBLE, dec AS INTEGER)
71.  
72.     '// convert to number delimted string
73.  
74.     DIM number AS STRING '                                              stores whole number(w/o dec) minus "."
75.     DIM precision AS STRING '                                           stores decimals (minus ".")
76.     DIM convert AS STRING '                                             stores the orginal number in string format
77.     DIM tempStr AS STRING '                                             used to create the new formated string
78.     DIM I, j AS INTEGER '                                               i=loop counter, j= delimeter counter
79.  
80.     convert = STR$(num) '                                               convert it all to a strig
81.     number = LEFT$(convert, INSTR(convert, ".") - 1) '                  store the integer
82.     precision = MID$(convert, INSTR(convert, ".") + 1, LEN(convert)) '  store the decimal
83.     number = RTRIM$(LTRIM$(number)) '                                   trim it up
84.  
85.     IF num < 1000 THEN '                                                if its less than 1000, then no comma needed
86.         num2str$ = number + "." + LEFT$(precision, dec) '               assign function for dec (precision)
87.         EXIT SUB '                                                      leave this sub now
88.     END IF
89.  
90.  
91.  
92.  
93.     j = 1 '                                                             set delimeter counter
94.     tempStr = ""
95.     FOR I = LEN(number) TO 1 STEP -1 '                                  count from the right to left of number
96.  
97.         IF j > 3 THEN '                                                 if delimeter counter is > 3 then
98.             tempStr = "," + tempStr '                                   add a comma
99.             j = 1 '                                                     reset the delimeter counter
100.         END IF
101.  
102.  
103.         tempStr = MID$(number, I, 1) + tempStr '                        add the next digit from "number" to tempstr
104.         j = j + 1 '                                                     increase the delimeter counter
105.  
106.     NEXT '                                                              keep looping
107.     number = tempStr '                                                  reassign number to the actual string created
108.  
109.  
110.     tempStr = "" '                                                      clear tempStr so we can make the percision
111.     FOR I = 1 TO LEN(precision) '                                       start looping through dec
112.         tempStr = tempStr + MID$(precision, I, 1) '                     create it
113.         IF I >= dec THEN EXIT FOR '                                     is it the number of places(dec) we want? exit this loop
114.     NEXT
115.     precision = tempStr '                                               assign the precision var to the newly created decimal length
116.     num2str$ = number + "." + precision '                               return it back as a fll string with "."
117. END SUB
118.  
119.  
120. FUNCTION xnum2str$ (num AS DOUBLE, dec AS INTEGER)
121.  
122.     '// convert to number delimted string
123.  
124.     DIM number AS STRING
125.     DIM precision AS STRING
126.     DIM convert AS STRING
127.     DIM tempStr AS STRING
128.     DIM I AS INTEGER
129.  
130.     convert = STR$(num) '                                               convert it all to a strig
131.     number = LEFT$(convert, INSTR(convert, ".") - 1) '                  store the integer
132.     precision = MID$(convert, INSTR(convert, "."), LEN(convert)) '      store the decimal
133.  
134.     tempStr = ""
135.     FOR I = LEN(number) TO 1 STEP -1 '                                  create comma delimted string
136.         tempStr = MID$(convert, I, 1) + tempStr
137.         IF I / 3 = INT(I / 3) THEN tempStr = "," + tempStr
138.     NEXT
139.     number = tempStr
140.  
141.  
142.     tempStr = "" '                                                      adjust len of precisio to match
143.     FOR I = 1 TO LEN(precision)
144.         tempStr = tempStr + MID$(precision, I, 1)
145.         IF I >= dec THEN EXIT FOR
146.     NEXT
147.     precision = tempStr
148.     xnum2str$ = number + precision
149. END SUB
150.  
151.  
152. FUNCTION Commatose$ (value AS DOUBLE, precision AS INTEGER)
153.     DIM v AS DOUBLE
154.     v = value 'copy so it doesn't get changed
155.     IF v < 0 THEN s$ = "-": v = v * -1 ELSE s$ = ""
156.  
157.     sv$ = LTRIM$(STR$(v)) '<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< ltrim$ remove leading space!
158.     dot = INSTR(sv$, ".")
159.     IF dot THEN
160.         i$ = MID$(sv$, 1, dot - 1)
161.         d$ = LEFT$(MID$(sv$, dot + 1) + STRING$(precision, "0"), precision)
162.     ELSE
163.         i$ = sv$
164.         d$ = STRING$(precision, "0")
165.     END IF
166.     FOR i = LEN(i$) TO 1 STEP -1
167.         c = c + 1
168.         IF c = 4 THEN c = 1: b$ = "," + b$
169.         b$ = MID$(i$, i, 1) + b$
170.     NEXT
171.     Commatose$ = s$ + b$ + "." + d$
172. END FUNCTION
173.  

[xra7en]

Looks very nice - AND you seem to have a dilemma that I have. those darn leading spaces for digits that do not fill it.
Thus why I tried my hand at making a custom function,

According to the definition - if I read it right:

  #    Denotes a numerical digit. An appropriate number of digits should be used for values received.

so you kinda have to guess the amount of digits  on a dynamic application. 

HOWEVER FreeTrav gave me an Idea (sorta my original question about using the "PRINT USING" to do the brunt of the work, and I came up with this:
It seems to solve the goofy leading spaces for digits that are not there. It adds a hash for each digit dynamically THEN IT USES THE PRINT USING

Code: QB64: [Select]

1. SUB number_format (num AS DOUBLE, flt AS INTEGER)
2.     DIM number AS STRING '                                              stores whole number(w/o dec) minus "."
3.     DIM convert AS STRING '                                             stores the orginal number in string format
4.     DIM I AS INTEGER '                                                  i=loop counter delimeter counter
5.     DIM tpl AS STRING '                                                 .. using template
6.  
7.     convert = STR$(num) '                                               convert it all to a strig
8.     IF RIGHT$(convert, 1) <> "." THEN convert = convert + "." '         weird fix for lower numbers
9.     number = LEFT$(convert, INSTR(convert, ".") - 1) '                  store the integer
10.     number = RTRIM$(LTRIM$(number)) '                                   trim it up
11.  
12.     tpl = "" '                                                          create a template string for USING
13.  
14.     FOR I = 1 TO LEN(number): tpl = tpl + "#": NEXT '                   add a hash per digit per integer digit
15.     tpl = tpl + "," '                                                   set the comma tag
16.     IF flt > 0 THEN '                                                   is the decimal count set?
17.         tpl = tpl + "." '                                               add the decimal tag
18.         FOR I = 1 TO flt: tpl = tpl + "#": NEXT '                       how many places (as per dec)
19.     END IF
20.     '    PRINT "tpl for "; num
21.     '    PRINT tpl
22.     PRINT USING tpl; num '                                              now PRINT USING should look cleaner
23.  
24. END SUB

So now instead of "re-inventing the wheel" I just changed it a little. Now there are no spaces, the decimal and digits can grow dynamically

whatcha think? ya?

[xra7en]

Hi
fine to add meself to this thread again

I have wrote a my routine... :-)
yes! Why? As someone knows the way to write numbers in italian is different from that used here, (also date and some few other things...) so I try to make my algorithm in QB64 working on string format and letting to show international format (, for thousand and . for decimal or that italian that is the inverse system  i.e.  international 1,000,009.999 = italian 1.000.009,999 ). So here my work.

That is very interesting, If I understand...
Italian uses a ","(comma) for a decimal notation and a "."(dot) for thousands separator?
If so we could change the function to ask for a delimiter.
 
thanks for that info.