Testing for the v8.17 release

Testing for the v8.17 release
(using the v8.16.900 release candidate build)

Search for your name on the checklist page to find relevant items, either
reply when you finish an item (please indicate which item/s is/are done),
or check it off yourself on the checklist page. Also, if you have any
commits that should have been picked, make sure that the changes are in.

Be sure to finish your testing by the 27th. Otherwise the release
will move on without your input.

The checklist page is at:
     https://github.com/racket/racket/wiki/Release-Checklist-8.17

Release candidates are at:
     https://pre-release.racket-lang.org

Please use these installers (or source bundles) -- don't test from
your own git clone (don't test the `master' branch by mistake!).
To get the tests, you can do this:

  cd ...racket-root...
  ./bin/raco pkg install -i --auto main-distribution-test

@mike When I run 'racket-tests' in 'htdp-test/tests/test-engine' in drracket, I get three test failures. Can you check these out please? Thanks.

@jbclements I don't think this is a show-stopper but we should wait for a reply from @mike

  1. When I close all windows and use cmd-N to open a new one, I don’t get a new window. (I see this in the drracket @ HEAD too.)
    — I do get a new window via a mouse gesture and the drop-down menu.

2a. Using F1 for docs fails. It does bring up a page with a search window using a file URL.

2b. Searching the docs fails. When I enter big-bang in the search window of my browser, it seems to do something for perhsp a second or two and then switches from the “home page” to the “results page” but shows no results.

2c. Some of the links on the right menu work (e.g. the Release Notes link (local) and Report Bug link (global) ) and some don’t (e.g. Acknowledgment and License).

  1. I separately reported a problem with one test-engine file to @mike via Discourse,

(I don’t think either one of these is a show-stopper but all of them are annoying.)

This has been a bug for a while now. I see that in v8.8 it behaves that way (beeping when you type cmd-n after closing the last drracket window) and in v8.3 it does something worse, not letting you close the last window at all.

I opened an issue (on racket/racket) for the inability to create a window using cmd-N. I also checked to see whether this issue was already present in racket/draw, didn't find anything.

I'm not able to reproduce this. Specifically, when I open a new file in the Racket language, enter

#lang racket

+

... place the cursor on the plus, and hit F1, I get a browser window that does show results for +.

I'm running macOS Sonoma 14.6.1, in case that's relevant.

Oh... actually, the search term is part of the file URL. I ran it again with a more ascii-friendly name (string-append), and it took me to this url:

file:///Users/clements/Library/Racket/8.16.900/doc/search/index.html?q=string-append

Is yours different? This could easily be messiness at the URL/browser/OS layer?

... and, unsurprisingly, @EmEf I'm not seeing 2b or 2c either; definitely sounds like it's all the same problem somehow. What browser/OS are you using? I'm using Firefox 137.0.2 + macOS Sonoma 14.6.1.

Are you referring to a different thread, or the mention you made earlier in this thread? :

Yes, we know that this is an intermittent bug. But this behavior has shown up for a year now on laptops of people I was trying to help, and it now affects mine. (It seems browser-independent, and by creating a new account on my machine I was able to replicate it.)

On May 3, 2025, at 2:26 PM, John Clements via Racket Discourse notifications@racket.discoursemail.com wrote "Are you referring to a different thread, or the mention you made earlier in this thread? “

Yes, I pinged him directly.

Perhaps I missed it: did you say what OS this was? Also, have you tried manually entering a URL such as file:///Users/clements/Library/Racket/8.16.900/doc/search/index.html?q=string-append (mutatis mutandis) to see whether that works? My guess is that the OS is blocking URLs that are file pointers into the user's Library folder, I recall a problem like this before.

John, I re-built my workday Racket from source .. and these issues disappeared, including for the release candidate.
This is clearly a strange issue, which I knew and which is why I didn’t think it was a show-stopper. Having said that,
I have encountered other people (students) who had the exact problem and got pretty frustrated because of it.

Did you hear back from @mike about the test-engine failures? I will write to him, I'd love to start the final build.

No, I have not.