Hi All, As mentioned by Tom, a third opinion is sought on the aforementioned minor test change. @Tom - I'd prefer to use the full version of the error messages to keep things clear, though I would not object to using your abbreviated version if it means getting the change in. So long as it passes when it should pass, and fails only when it should fail, clarity is a "would be nice", but ultimately a secondary priority. Thanks for your feedback. :) Best Regards Adam Farley IBM Runtimes "Thomas Stüfe" <thomas.stuefe@gmail.com> wrote on 03/03/2020 10:52:10:

From: "Thomas Stüfe" <thomas.stuefe@gmail.com> To: Adam Farley8 <adam.farley@uk.ibm.com> Cc: core-libs-dev <core-libs-dev@openjdk.java.net> Date: 03/03/2020 10:52 Subject: [EXTERNAL] Re: RFR: 8239365: ProcessBuilder/Basic.java test modifications for AIX execution

This is why I always was against handing up the result of strerror to the user :) The same problem we would have when running with different locales. We should have a platform agnostic string table in the java lib for that purpose...

As for the test, looks good, but I personally would shorten the AIX patterns a bit or maybe try to find a short form fitting all platforms (e.g. "[Pp]ermission").

But thats just idle bikeshedding, lets see what others think.

Cheers Thomas

On Tue, Mar 3, 2020 at 11:43 AM Adam Farley8 <adam.farley@uk.ibm.com> wrote: Hi All,

Reviews and sponsor requested for a small test change.

Short version: When an AIX machine has the file set "bos.msg.en_US.rte",

the error messages are not in a form that the test expects, causing failure.

The simplest option appears to be adding the second potential form of the message into the regex (see webrev).

http://cr.openjdk.java.net/~afarley/8239365.1/webrev/

Bug: https://bugs.openjdk.java.net/browse/JDK-8239365

Best Regards

Adam Farley IBM Runtimes Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number

741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

Hi Tom, Roger, No, I didn't see that. I'm not sure why. Sorry about that Roger. :) Roger's proposal seems a fair compromise. Here's a webrev for it: http://cr.openjdk.java.net/~afarley/8239365.2/webrev If this meets both of your approvals, could I ask for a volunteer to sponsor and merge, please? Best Regards Adam Farley IBM Runtimes "Thomas Stüfe" <thomas.stuefe@gmail.com> wrote on 21/04/2020 18:29:23:

From: "Thomas Stüfe" <thomas.stuefe@gmail.com> To: Adam Farley8 <adam.farley@uk.ibm.com> Cc: core-libs-dev <core-libs-dev@openjdk.java.net>, Roger Riggs <roger.riggs@oracle.com> Date: 21/04/2020 18:29 Subject: [EXTERNAL] Re: RFR: 8239365: ProcessBuilder/Basic.java test modifications for AIX execution

Hi Adam,

I do not have strong emotions, leave it up to you. But did you read Rogers mail from 21.feb?

http://mail.openjdk.java.net/pipermail/core-libs-dev/2020-February/064878.ht...

Cheers, Thomas

On Thu, Mar 5, 2020 at 3:32 PM Adam Farley8 <adam.farley@uk.ibm.com>

wrote:

Hi All,

As mentioned by Tom, a third opinion is sought on the aforementioned minor test change.

@Tom - I'd prefer to use the full version of the error messages to keep things clear, though I would not object to using your abbreviated version if it means getting the change in.

So long as it passes when it should pass, and fails only when it should fail, clarity is a "would be nice", but ultimately a secondary priority.

Thanks for your feedback. :)

Best Regards

Adam Farley IBM Runtimes

"Thomas Stüfe" <thomas.stuefe@gmail.com> wrote on 03/03/2020 10:52:10:

...

From: "Thomas Stüfe" <thomas.stuefe@gmail.com> To: Adam Farley8 <adam.farley@uk.ibm.com> Cc: core-libs-dev <core-libs-dev@openjdk.java.net> Date: 03/03/2020 10:52 Subject: [EXTERNAL] Re: RFR: 8239365: ProcessBuilder/Basic.java test modifications for AIX execution

This is why I always was against handing up the result of strerror to the user :) The same problem we would have when running with different locales. We should have a platform agnostic string table in the java lib for that purpose...

As for the test, looks good, but I personally would shorten the AIX patterns a bit or maybe try to find a short form fitting all platforms (e.g. "[Pp]ermission").

But thats just idle bikeshedding, lets see what others think.

Cheers Thomas

On Tue, Mar 3, 2020 at 11:43 AM Adam Farley8 <adam.farley@uk.ibm.com>

...

Hi All,

Reviews and sponsor requested for a small test change.

Short version: When an AIX machine has the file set "bos.msg.en_US.rte", the error messages are not in a form that the test expects, causing failure.

The simplest option appears to be adding the second potential form of the message into the regex (see webrev).

http://cr.openjdk.java.net/~afarley/8239365.1/webrev/

Bug: https://bugs.openjdk.java.net/browse/JDK-8239365

Best Regards

Adam Farley IBM Runtimes Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number

wrote: 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

Hi Adam, ok, looks good. I will sponsor it.  (Seeing Thomas's email). Thanks, Roger On 4/22/20 12:38 PM, Adam Farley8 wrote:

Hi Tom, Roger,

No, I didn't see that. I'm not sure why.

Sorry about that Roger. :)

Roger's proposal seems a fair compromise.

Here's a webrev for it: http://cr.openjdk.java.net/~afarley/8239365.2/webrev

If this meets both of your approvals, could I ask for a volunteer to sponsor and merge, please?

Best Regards

Adam Farley IBM Runtimes

"Thomas Stüfe" <thomas.stuefe@gmail.com> wrote on 21/04/2020 18:29:23:

...

From: "Thomas Stüfe" <thomas.stuefe@gmail.com> To: Adam Farley8 <adam.farley@uk.ibm.com> Cc: core-libs-dev <core-libs-dev@openjdk.java.net>, Roger Riggs <roger.riggs@oracle.com> Date: 21/04/2020 18:29 Subject: [EXTERNAL] Re: RFR: 8239365: ProcessBuilder/Basic.java test modifications for AIX execution

Hi Adam,

I do not have strong emotions, leave it up to you. But did you read Rogers mail from 21.feb?

http://mail.openjdk.java.net/pipermail/core-libs-dev/2020-February/064878.ht...

...

Cheers, Thomas

On Thu, Mar 5, 2020 at 3:32 PM Adam Farley8 <adam.farley@uk.ibm.com>

wrote:

...

Hi All,

As mentioned by Tom, a third opinion is sought on the aforementioned minor test change.

@Tom - I'd prefer to use the full version of the error messages to keep things clear, though I would not object to using your abbreviated version if it means getting the change in.

So long as it passes when it should pass, and fails only when it should fail, clarity is a "would be nice", but ultimately a secondary priority.

Thanks for your feedback. :)

Best Regards

Adam Farley IBM Runtimes

"Thomas Stüfe" <thomas.stuefe@gmail.com> wrote on 03/03/2020 10:52:10:

...

From: "Thomas Stüfe" <thomas.stuefe@gmail.com> To: Adam Farley8 <adam.farley@uk.ibm.com> Cc: core-libs-dev <core-libs-dev@openjdk.java.net> Date: 03/03/2020 10:52 Subject: [EXTERNAL] Re: RFR: 8239365: ProcessBuilder/Basic.java test modifications for AIX execution

This is why I always was against handing up the result of strerror to the user :) The same problem we would have when running with different locales. We should have a platform agnostic string table in the java lib for that purpose...

As for the test, looks good, but I personally would shorten the AIX patterns a bit or maybe try to find a short form fitting all platforms (e.g. "[Pp]ermission").

But thats just idle bikeshedding, lets see what others think.

Cheers Thomas

On Tue, Mar 3, 2020 at 11:43 AM Adam Farley8 <adam.farley@uk.ibm.com> wrote: Hi All,

Reviews and sponsor requested for a small test change.

Short version: When an AIX machine has the file set "bos.msg.en_US.rte", the error messages are not in a form that the test expects, causing failure.

The simplest option appears to be adding the second potential form of the message into the regex (see webrev).

http://cr.openjdk.java.net/~afarley/8239365.1/webrev/

Bug: https://bugs.openjdk.java.net/browse/JDK-8239365

Best Regards

Adam Farley IBM Runtimes Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU