Hi Tom,
Exchanged interpretation of XON and XOFF.
It might help if you wrote the same explanation about their meanings being reversed in your email that you did in the patch: Alexandre's not likely to go check up on bugzilla. --Juan
On 25/10/2007, Juan Lang <juan.lang(a)gmail.com> wrote:
Hi Tom,
Exchanged interpretation of XON and XOFF.
It might help if you wrote the same explanation about their meanings being reversed in your email that you did in the patch: Alexandre's not likely to go check up on bugzilla.
Tests would also be useful here as well, so that there is not a regression. - Reece
On 25/10/2007, Juan Lang <juan.lang(a)gmail.com> wrote:
Tests would also be useful here as well, so that there is not a regression.
That's good general advice, but hard to implement without a serial device to test against, which our test systems won't have.
You could always skip the tests when a serial connection is not present. That way, you get coverage on systems that do have them. It should be possible to configure Virtual Machines that have serial ports suitable for use in these tests. If there are other factors that make this complicated to test (such as getting repeatable results from the serial devices on machines that have serial ports configured) then it is not worth the effort to create tests for this case. Are there any existing tests for the serial port code? If there are, that would suggest that this is possible. - Reece
On 10/26/07, Reece Dunn <msclrhd(a)googlemail.com> wrote:
On 25/10/2007, Juan Lang <juan.lang(a)gmail.com> wrote:
Tests would also be useful here as well, so that there is not a regression.
That's good general advice, but hard to implement without a serial device to test against, which our test systems won't have.
You could always skip the tests when a serial connection is not present. That way, you get coverage on systems that do have them. It should be possible to configure Virtual Machines that have serial ports suitable for use in these tests.
If there are other factors that make this complicated to test (such as getting repeatable results from the serial devices on machines that have serial ports configured) then it is not worth the effort to create tests for this case.
Are there any existing tests for the serial port code? If there are, that would suggest that this is possible.
If we have serial port tests, then there aren't nearly enough. Check on bugzilla, virtually every serial port patch introduces at least 1 new regression, and these regressions generally don't get fixed, resulting in constantly deteriorating serial port support. Testing serial port hardware is very difficult because you need the tests to work on Windows, and you need everybody to have the hardware. You can emulate a serial port, but that requires kernel-mode drivers in at least Windows, and emulating hardware that plugs into the emulated port isn't much easier. Out of interest, Java and Mono both have serial port classes, do they do any serial port regression testing?
- Reece
Damjan
* On Fri, 26 Oct 2007, Damjan Jovanovic wrote:
* On 10/26/07, Reece Dunn wrote:
* On 25/10/2007, Juan Lang wrote:
Tests would also be useful here as well, so that there is not a regression.
That's good general advice, but hard to implement without a serial device to test against, which our test systems won't have.
You could always skip the tests when a serial connection is not present. That way, you get coverage on systems that do have them. It should be possible to configure Virtual Machines that have serial ports suitable for use in these tests. ... You can emulate a serial port, but that requires kernel-mode drivers in at least Windows, and emulating hardware that plugs into the emulated port isn't much easier.
Yes, but this is possible. I believe, the ideal state would be when MinGW support for compiling device drivers + Wine support for running KMD is good enough. It probably should be a separate software project, though, as it will require very specific driver coding and testing-on-Windows manpower. No?
On Thu, 25 Oct 2007, Juan Lang wrote:
Tests would also be useful here as well, so that there is not a regression.
That's good general advice, but hard to implement without a serial device to test against, which our test systems won't have.
The tests could be set to only run in interactive mode. Then a Yess/No dialog could be displayed to the user to ask him if he has a connected serial device. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ "Utilisateur" (nom commun) : Mot utilisé par les informaticiens en lieu et place d'"idiot".
participants (5)
-
Damjan Jovanovic -
Francois Gouget -
Juan Lang -
Reece Dunn -
Saulius Krasuckas