bug DEC(something,1) (SOLVED)

[rikky]

Hello

Sorry for reporting a (probable) bug in the wrong section, but the bug section is currently occupied by a question from Alexfish.

I'm using the latest Fossil from today.

From the documentation we have two examples:

PRINT DEC("AB1E")
PRINT DEC("00010101", 1)
The response on the first command is 43806

The response on the second command however is:

ERROR: signal for SEGMENTATION FAULT received - memory invalid or array out of bounds? Try to compile the program with TRAP LOCAL to find the cause.
It becomes stranger even.

string$ = "00010101"
flag$ = "1"

PRINT "DEC(" & CHR$(34) & string$ & CHR$(34) & "," &  flag$ & ")"
PRINT DEC(string$,VAL(flag$))
Haha, now it works.
Except it doesn't give the right answer.

DEC("00010101",1)
65793
should be 1 + 4 + 16  = 21, or something.

 Rik.

[Pjot]

Hi rikky,

Thanks for your post. The cause of this issue is the documentation

In fact, the last argument should lie between 2 and 36. So if you want to convert from a binary string, then the value should be '2':

PRINT DEC("00010101", 2)


This will print the correct result '21' in decimal.

I will update the documentation so it reflects the correct functionality. Also, I will add a check to make sure that the '1' cannot be used.

Best regards
Peter

[rikky]

Nice.
So now we can also recount octal values.
I wasn't aware.

PRINT DEC("112",8)
Gives the correct value: 74

However the whole thing with variables still doesn't work. 

PRINT

string$ = "00010101"
flag$ = "2"

PRINT "DEC(" & CHR$(34) & string$ & CHR$(34) & "," &  flag$ & ")"
PRINT DEC(string$,VAL(flag$))
PRINT "should be: " & STR$(DEC("00010101",2))

PRINT

string$ = "112"
flag$ = "8"

PRINT "DEC(" & CHR$(34) & string$ & CHR$(34) & "," &  flag$ & ")"
PRINT DEC(string$,VAL(flag$))
PRINT "should be: " & STR$(DEC("112",8))

PRINT
response:

DEC("00010101",2)
65793
should be: 21

DEC("112",8)
274
should be: 74
 Rik.

PS: Am I missing OCT$(x) as a BaCon command?   

[Pjot]

Hi rikky,

You code sample works fine for me:

DEC("00010101",2)
21
should be: 21

DEC("112",8)
74
should be: 74


Can you try the latest BaCon from fossil and let me know your result?

Also I am curious on which platform you are testing?

BR
Peter

[rikky]

Oh, I was already using the latest version from this morning, but I'm fosseling a new one at this moment.
I don't think it will make a difference.
Is there a way to know what fossil version (creation date?) I'm using?
(just to be able to cross check.)

I'm using OS$ = Linux armv7l

uname -r = 4.19.49v6v7-aufs (kernel)

cat /etc/os-release = 
PRETTY_NAME="Raspbian GNU/Linux 9 (stretch)"
NAME="Raspbian GNU/Linux"
VERSION_ID="9"
VERSION="9 (stretch)"
ID=raspbian
ID_LIKE=debian
HOME_URL="http://www.raspbian.org/"
SUPPORT_URL="http://www.raspbian.org/RaspbianForums"
BUG_REPORT_URL="http://www.raspbian.org/RaspbianBugs"

 Rik.

[rikky]

Oh....
solved.

The new fossil gives the right answers.

I've absolutely no idea how that can be.
If nothing changed in the source code around the DEC(x) routine, this should not be possible.

But it is.

I think it is impossible now to find the cause, so  ..
I supposed this one is solved.

Thanks..
 Rik.
 

[Pjot]

Thanks Rik,

FYI: I have mapped the value '1' to '2' to maintain backwards compatibility.

Best regards,
Peter