Dealing with complex strings

[Horizon]

Hello, 

I'm working on a program that extracts values from a serial device, which Just Basic is great for . The device has fixed labels within its output strings and for simplicity I've populated a couple of example strings. As we can see below the results associated with these 'labels' can vary so, utilising mid$ would probably not be the solution here.



first$ = "*START VARONE=10.7 VARTWO=56.789 VARTHREE=Quick fox VARFOUR=ABC-D VARFIVE=12 VARSIX=1 VARSEVEN=NO VAREIGHT=22.5"

second$ = "*START VARONE=12.34 VARTWO=75.23 VARTHREE=Lazy dog days VARFOUR=ABC-D2 VARFIVE=8 VARSIX=19 VARSEVEN=YES VAREIGHT=27.88"



So by using word$ we can extract string up to the next space. In this example print word$(first$, 2) would result in "VARONE=10.7". We can then create a new variable varone$ and for this we can use right$ to extract the wanted$ of value "10.7" but, this is only OK if we know its length in symbols.

'Strings

first$ =     "*START VARONE=10.7 VARTWO=56.789 VARTHREE=Quick fox VARFOUR=DVB-S2 VARFIVE=12 VARSIX=1 VARSEVEN=NO VAREIGHT=22.5"

second$ = "*START VARONE=12.34 VARTWO=75.23 VARTHREE=Lazy dog days VARFOUR=DVB-S2 VARFIVE=8 VARSIX=19 VARSEVEN=YES VAREIGHT=27.88"



print word$(first$, 2)

varone$ = word$(first$, 2)

wanted$ = right$(varone$, 4)

print wanted$




If we replace first$ with second$ we would get an incorrect wanted$ value of "2.34"

When looking at VARTHREE this can be a short phrase with additional spaces so word$ may not be the best tool here resulting in:


VARTHREE=Quick           from first$


or


VARTHREE=Lazy            from second$



Where the next space prevents the capture of the whole phrase.


Thoughts on solutions:


Part 1 Numeric values

For the numeric values we can utilise right$ and somehow the "=" chr$(61) equals symbol to use this as a end of wanted string marker (reading from right to left), but this is where I'm struggling with the code to execute this kind of filtering.



Part 2 Phrase values (VARTHREE)

For the text values this is complicated by the spaces when they occur so word$ may not be appropriate here, perhaps the next variable label (which is a fixed value of VARFOUR=) would be useful here again reading backwards until the "=" chr$(61) symbol is encountered so that spaces are included in your wanted$. So, perhaps a method of grabbing the string between word$(first$, 4) and word$(first$, 5) would capture the whole phrase and enable easy filtering into the wanted$. Capturing the whole phrase is baking my noodle somewhat.


Any guidance on the above would be most welcome.

[tsh73]

As they said there is more then one way to skin a cat
One of ways

1             VARONE        10.7
2             VARTWO        56.789
3             VARTHREE      Quick fox
4             VARFOUR       ABC-D
5             VARFIVE       12
6             VARSIX        1
7             VARSEVEN      NO
8             VAREIGHT      22.5


first$ = "*START VARONE=10.7 VARTWO=56.789 VARTHREE=Quick fox VARFOUR=ABC-D VARFIVE=12 VARSIX=1 VARSEVEN=NO VAREIGHT=22.5"
second$ = "*START VARONE=12.34 VARTWO=75.23 VARTHREE=Lazy dog days VARFOUR=ABC-D2 VARFIVE=8 VARSIX=19 VARSEVEN=YES VAREIGHT=27.88"

varNames$="VARONE VARTWO VARTHREE VARFOUR VARFIVE VARSIX VARSEVEN VAREIGHT"

s$=first$
's$=second$
N = 8
for i = 1 to N
   var$=word$(varNames$, i)
   p1=instr(s$, var$)
   p2=len(s$)+1
   if i<N then p2=instr(s$,word$(varNames$, i+1))
   sub1$=mid$(s$,p1,p2-p1)
   p3=instr(sub1$, "=")
   print i, var$,  mid$(sub1$, p3+1)
next


[Horizon]

Thanks tsh73

I think my solution would have been very complex to overcome these varying output conditions. This is very elegant, and I'm already learning a great deal from it.

I appreciate the help

[Rod]

I would describe what tsh73 is showing you as utilising tags. So the first tag is VARONE= finding this tag with instr let’s you find the data immediately following the = sign. Usually the device will be sending specific length or format numeric, devices usually send specific length text as well. So knowing which tag you are dealing with can let you fetch back specific length data. However your next tag might define the end of the data for the previous tag, so variable length data can be easily found.

So find where instr VARONE= is and where VARTWO= is and the data between is the data for VARONE=