Rebol/RED parsing. Save ASCII codes from parsed string to variable?

Question

I stumbled into RED language the other day and spend (more or less literally) the last 24h "learning" it. I exhausted my googling skills trying to find solution for a simple problem that just evades my skills and logic, so hopefully somebody here can lead me to path to righteousness.

As the title suggest, I tried to parse a simple string (any string of random text, really), get the individual char(acter)s and then tried to save them into a variable. (Another string/array/any type really)

The best I could do was with code using:

alpha: charset [#"a" - #"z"]
testString: "this is just random rambling to test parsing!"

as prerequisites and something like this when (trying to) parse:

probe parse teststring [copy text to alpha (append text2 to-integer(to-char text)) to end]

Saves the (first letter) ascii code to text2 and running the script several times properly adds the (same first letter) ASCII code several times in a row:

CONSOLE OUTPUT

>> probe parse teststring [copy text to alpha (append text2 to-integer(to-char text)) to end]
true
== true
>> probe text2
"34"
== "34"
>> probe parse teststring [copy text to alpha (append text2 to-integer(to-char text)) to end]
true
== true
>> probe text2
"3434"
== "3434"

Obviously my parsing is not really "looping" the individual characters of the string or not saving them properly while doing it. Maybe my parsing really takes the whole string from TO till END and I try to then convert that to ASCII code or something else is happening here?

ANY help will be greatly appreciated, as I cannot possibly advance with my RED learning before solving this dilemma and understanding how the parsing really works in RED.

I think you're missing some stuff in your prerequisites test code to reproduce, like `text2: ""` — Geeky I, Oct 01 '17 at 14:53
Very true, text2 was not declared indeed. Was a sloppy paste it seems, sorry for that. — Stacking For Heap, Oct 01 '17 at 14:55
Welcome to Red language! I faced similar problem a few days back: https://stackoverflow.com/questions/46260181/variable-value-in-foreach-of-red-language — rnso, Oct 02 '17 at 07:14

sqlab · Answer 1 · 2017-10-01T13:52:36.017

I am not sure, if I understand your question and what you want to achieve, but if you are looking for all the ascii representation of the chars you can get that with

asciis: []
parse teststring [some [set a  alpha (append asciis to-integer a )| skip]]
== true
>> asciis
== [116 104 105 115 105 115 106 117 115 116 114 97 110 100 111 109 114 97 109 98 108 105 110 103 116 111 116 101 115 116 112 97 114 115 105 110 103]

some is one of the available words responsible for the looping

There are some issues in your trial. It would give some errors before doing any conversion. text2 is probably declared before. to alpha would give an empty string "" as you try to copy up to the first alpha character. Remember to goes up to and not including the target. You can not convert an empty string to a character. If we assume until now no error occurred, then you are still in front of your string and you go straight to the end of your string.

Some documentation about Red parse. see Iteration about looping

You can debug your parsing either with parse-trace or just put a simple (probe text) after the part of your rule you want to investigate:

Thank you, that was EXACTLY what I was trying to achieve. Can you explain what happens in my code and why it only gave the first character ('s ASCII code)? Did it return the whole string between my parse conditions and then "to-char and to-integer" converted it to single character and its ascii code? — Stacking For Heap, Oct 01 '17 at 12:34
No love for COLLECT? `collect [parse [... (keep to integer! a) ...]]` (or if Red-only `parse [collect [some [... keep (to integer! a) ...]]]`) — rgchris, Oct 01 '17 at 15:28

Rebol/RED parsing. Save ASCII codes from parsed string to variable?

1 Answers1