I'm currently using Racket 7.6 (non-CS) on Windows, and it feels like `raco
test` is much slower these days. Specifically I am seeing
`raco test tests.rkt` takes 10 seconds. This is fine the first time, but
repeating the test immediately also takes 10 seconds. It never creates any
"compiled" di
I have a scribble document with a string representation of some code. The
string is "(+ 1 2) \r\n (* 2 3)" which eventually gets sent through my
reader.
When I run "Scribble HTML" in DrRacket it works. When I run "raco setup" it
works. But when I run "raco test" my reader fails with
(exn:fail:
I'd expect it to work for a single file but fail for multiple files,
because the default `--process` mode for `raco test` won't pass along
the `-S` directory.
Probably `--process` mode should pass along the right collection paths.
Meanwhile, if that's the problem, then using `--places` mode or
`--
Hi,
I'd like to invoke 'raco test' with additional collection paths passed in via
the command line. I'm trying this: racket.exe -S -l- raco test ,
but that -S parameter doesn't seem to take effect.
Thanks,
Dan
--
You received this message because you are subscribed to the Google Groups
"
Hi Alex,
It is a bit confusing. There's a lot of history to the message.
It used to be that rackunit only had the test-suites, test-cases, and
the checks. It tries very hard to NOT count the checks as "tests".
Later, the checks were exposed so you didn't have to but them in
test-suites and test-c
On Friday, March 3, 2017 at 7:18:27 AM UTC+8, schuster wrote:
> The problem is that a test-case expression runs a test immediately; it does
> not return a test object to be run later. In your case, the test runs while
> my-test-case is being defined, then no test at all actually runs
> (my-test-
The problem is that a test-case expression runs a test immediately; it does
not return a test object to be run later. In your case, the test runs while
my-test-case is being defined, then no test at all actually runs
(my-test-case is just #).
A test-suite expression, however, *does* construct a te
I'm a bit confused about how raco test is reporting the number of tests run, if
a test case is defined outside a test suite. If I run the program below:
#lang racket
(require rackunit)
(require rackunit/text-ui)
(define my-test-case
(test-case
"My Test case"
On Mar 30, 2015, at 10:16 PM, Robby Findler wrote:
> On Monday, March 30, 2015, Alexander D. Knauth wrote:
> Well the point of using the configure-runtime was that it would run when it
> was the “main" program so that exact-decimals would work in the repl, but if
> this file was required as a
On Monday, March 30, 2015, Alexander D. Knauth wrote:
> Well the point of using the configure-runtime was that it would run when
> it was the “main" program so that exact-decimals would work in the repl,
> but if this file was required as a library, it wouldn’t set the parameter
> for the other p
Well the point of using the configure-runtime was that it would run when it was
the “main" program so that exact-decimals would work in the repl, but if this
file was required as a library, it wouldn’t set the parameter for the other
program.
But should the tests be thought of as the “main” pro
I believe the intention of configure-runtime is that it runs only when
your file is run as the main program and not when your program is run
as a library (via require). So I don't think you can use that facility
to implement the exact-decimal language.
Robby
On Mon, Mar 30, 2015 at 8:56 PM, Alexa
For my exact-decimal meta-language, I had some tests that depend on the
read-decimal-as-inexact parameter being set to #f by configure-runtime.
It seems like raco test isn’t running that before running the tests. Is that
intentional?
(by the way the tests are not within a test submodule)
--
13 matches
Mail list logo
