World: r3wp
[Core] Discuss core issues
older newer | first last |
Fork 30-Mar-2008 [9792] | (Quick question, I'm new to altme... how do I type in a multiple lines without submitting the message?) |
BrianH 30-Mar-2008 [9793x2] | OK, I guess you're right (just tested). Lit-words seem to be "word-active" in R2. I'll check R3 as well. |
(click the pencil button, 5th from the left. then do ctrl-s to send) | |
Fork 30-Mar-2008 [9795] | (Ah, thank you.) |
BrianH 30-Mar-2008 [9796] | Same in R3. I'll have to do some code review on the mezzanines to check for code that expects the opposite. |
Fork 30-Mar-2008 [9797] | I wonder if this is a bug or a feature? e.g. is there a fundamental part of REBOL depending on this behavior in order to make certain evaluations work... |
BrianH 30-Mar-2008 [9798x3] | I already did a review for other word-active values in a lot of the mezzanines. I just have to check for lit-word! values too. Fortunately lit-word! values assigned to variables are _really_ rare. Most people use word! values, just using lit-words literally. |
That feature you mention is the word-active feature. It's also what causes functions to be evaluated. It's sort of like putting a function reference in the first position of the list in Scheme rather than the other positions. | |
later | |
Fork 30-Mar-2008 [9801] | Later, thank you... |
Gabriele 31-Mar-2008 [9802x4] | Fork: always using "quoting" is actually the source of your problem, as it does not really bring symmetry in. the reason is that ' is not an operator, rather, we have word! and lit-word! as two separate types. |
it is true that you have to know where evaluation happens and where it does not. but this is the key of rebol: since data is code and code is data, there is no explicit sign of what can be evaluated and what cannot. anything can be evaluated, if you make the interpreter evaluate it. | |
so you have to decide whether you are getting something to be evaluated or not. | |
about "word-active" values: i'm not sure lit-word! being word-active is useful, but i'm sure Carl has a good reason for that. it's a good thing to always use :x instead of x when you want to get the value as opposed to evaluate the word (they are the same in most cases except a few types, especially any-function! types and, as you have seen, lit-word!) | |
Fork 31-Mar-2008 [9806x4] | Thanks Gabriele, I think I understand, and knowing the actual answer is helpful in understanding the fundamental springs and pulleys that make the REBOL machine work. |
One of the first errors I encountered while trying to run an installer that was built in REBOL was that a state machine driven by a variable was failing to match a case and falling through to a default, due to a spurious linefeed on one of the cases that was read in. It was attempting to match foo, but had foo^/ ... and so it fell through to the wrong case. So I have been tinkering with a checked enum for REBOL. | |
If anyone has input on to the approach, I'd appreciate feedback: http://pastebin.com/d698d3c16 | |
Oops, deleted some spurious code and that made a typo, will make a better version that prints section headers for the tests... | |
Anton 1-Apr-2008 [9810x5] | I guess you figured it out already, but you can use TRIM to remove leading and trailing whitespace from your input. |
I'm not sure why you are taking this approach, but it seems to be something which you might implement to make another language more useful :) I don't think it's needed in rebol. Perhaps you could explain how you're using this enum/switch in more detail... | |
A small tip: for index 1 (length? cases) 1 [ is usually better as: repeat index length? cases [ | |
(I almost never use FOR. I use REPEAT very often). | |
Fork, perhaps you can show us the original (approximate) example from your installer state machine, with a few example inputs and expected results. Then we can see what started all this :) | |
Fork 1-Apr-2008 [9815x5] | Here is a better documented version: |
http://pastebin.com/m2e06f2b7 | |
I explain the cases in the header | |
Here is the output with performance data of the testing: | |
http://pastebin.com/m20abc2ca | |
Anton 1-Apr-2008 [9820] | Ok, that explains the enum object well. However, it looks like a waste of time to me ! (sorry) but I haven't felt I needed this in rebol for many years ! |
Fork 1-Apr-2008 [9821x2] | I suppose it depends on what kind of code you write |
This specific case came up when I tried to install a large, professional grade REBOL system written by people who are as I understand it very good at REBOL | |
Anton 1-Apr-2008 [9823] | That may be true. |
Fork 1-Apr-2008 [9824] | There is no integrated debugger, and so tracking this kind of stuff down isn't as easy as it might otherwise be |
Anton 1-Apr-2008 [9825] | Can we see what the specifics of the original case are with respect to the enum ? |
Fork 1-Apr-2008 [9826x5] | nextstep: cgi 'nextstep |
So it was getting a CGI variable in an HTTP post | |
Then it did switch/default nextstep [ (various cases, pages long) ] [ the default case ] | |
A bug in the cgi system left a trailing newline on the nextstep variable | |
SO instead of being "apache1" it was "apache1^" | |
Anton 1-Apr-2008 [9831x3] | Ok, so you saw the enum object as an extra level of protection against this kind of bug. |
To more strictly validate input. | |
Or was it just to catch where the bug was ? | |
Fork 1-Apr-2008 [9834x2] | It validates on all sides, it protects against programmer error in the usage as well as things like this where the data source or function has a problem. |
I do like REBOL from what I've seen so far! But I saw somewhere that it was being compared to modeling clay... it's a paradigm shift for people who are used to systems that do a lot of global analysis of your program before it runs the first line of code to catch basic errors. | |
Anton 1-Apr-2008 [9836x2] | Yes, definitely. |
make-enum-type should define 'ret local. | |
Fork 1-Apr-2008 [9838] | I do remember that, though I don't exactly remember what happens if you don't do it |
Anton 1-Apr-2008 [9839] | It sets 'ret global. Check this by running the script then probe ret in the console. |
Fork 1-Apr-2008 [9840x2] | Ah, ok |
So same for test-block | |
older newer | first last |