[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Chicken-janitors] #1548: Extra modules leak into environment when u
From: |
Chicken Trac |
Subject: |
Re: [Chicken-janitors] #1548: Extra modules leak into environment when using -j |
Date: |
Wed, 10 Oct 2018 15:14:45 -0000 |
#1548: Extra modules leak into environment when using -j
---------------------------------+-------------------
Reporter: megane | Owner:
Type: defect | Status: new
Priority: major | Milestone: 5.1
Component: compiler | Version: 5.0
Resolution: | Keywords: macro
Estimated difficulty: |
---------------------------------+-------------------
Comment (by megane):
Replying to [comment:3 megane]:
> Replying to [comment:2 sjamaan]:
> > I'm not sure this is an entirely correct comparison, given that
`(import)` is now equivalent to `(use)` in CHICKEN 4. If you do `csc -s
m.scm -j m && csi -qbn -e '(use m) (print (string-null? "") (string-null?
"1"))'`, you'll notice that it prints the same as C5. So, IMO it's not a
true regression and we should probably postpone this to 5.1, especially
given that fixing this would probably involve changing the import
machinery, which might lead to new bugs and cause other unexpected changes
this late in the process.
>
> On CHICKEN 4 `csc -s m.scm -j m && csi -qbn -e '(use m) (print (string-
null? "") (string-null? "1"))` gives me `Error: unbound variable: string-
null?`. That is the expected behavior.
There was a PEBKAC issue here. I was compiling the `(import srfi-13)`
version, when I should have used the `(use srfi-13)` version. No wonder
`string-null?` wasn't found. So my regression judgement is wrong.
--
Ticket URL: <https://bugs.call-cc.org/ticket/1548#comment:4>
CHICKEN Scheme <https://www.call-cc.org/>
CHICKEN Scheme is a compiler for the Scheme programming language.