I was hoping to release this past week, but was unable to get further testing done.
Before releasing, I try to test a number of large systems that we use here at SIFT with the latest ASDF. This provides a real world add-on to the test suite. Unfortunately, I have not had a chance to make this test. I hope to do so in the next couple of days.
That should either reveal bugs, or lead to an immediate release of 3.0.3.
It would be great if the quicklisp issues could be resolved satisfactorily in time for 3.0.3.
Best, r
On Mon, Oct 7, 2013 at 8:54 AM, Robert P. Goldman rpgoldman@sift.info wrote:
I was hoping to release this past week, but was unable to get further testing done.
Before releasing, I try to test a number of large systems that we use here at SIFT with the latest ASDF. This provides a real world add-on to the test suite. Unfortunately, I have not had a chance to make this test. I hope to do so in the next couple of days.
That should either reveal bugs, or lead to an immediate release of 3.0.3.
It would be great if the quicklisp issues could be resolved satisfactorily in time for 3.0.3.
I took the liberty of committing as 3.0.2.12 a patch that will hopefully fix the issue raised by Zach and blocking adoption by Quicklisp: before to load the asd for foo/bar, asdf will try to load the asd for foo, and so extensions to *system-definition-search-functions* don't have to explicitly know about that for the magic to happen.
Zach, can you test at your leisure whether this satisfies you?
I also added a utility ensure-gethash, some docstrings, and tweaked the webpage and manual.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org There is only one basic human right, the right to do as you damn well please. And with it comes the only basic human duty, the duty to take the consequences. — P. J. O'Rourke
On Mon, 2013-10-07 at 22:49 -0400, Faré wrote:
On Mon, Oct 7, 2013 at 8:54 AM, Robert P. Goldman rpgoldman@sift.info wrote:
I was hoping to release this past week, but was unable to get further testing done.
Before releasing, I try to test a number of large systems that we use here at SIFT with the latest ASDF. This provides a real world add-on to the test suite. Unfortunately, I have not had a chance to make this test. I hope to do so in the next couple of days.
That should either reveal bugs, or lead to an immediate release of 3.0.3.
It would be great if the quicklisp issues could be resolved satisfactorily in time for 3.0.3.
I took the liberty of committing as 3.0.2.12 a patch that will hopefully fix the issue raised by Zach and blocking adoption by Quicklisp: before to load the asd for foo/bar, asdf will try to load the asd for foo, and so extensions to *system-definition-search-functions* don't have to explicitly know about that for the magic to happen.
Zach, can you test at your leisure whether this satisfies you?
I also added a utility ensure-gethash, some docstrings, and tweaked the webpage and manual.
What if foo.asd doesn't define system "foo" ? The current system allows it.
On Tue, Oct 8, 2013 at 3:29 AM, Stelian Ionescu sionescu@cddr.org wrote:
What if foo.asd doesn't define system "foo" ? The current system allows it.
Then trying to (asdf:load-system :foo) will probably fail, and the maintainer will get gripes if he has users. Of maybe he's trying to play clever games and congratulates himself for being more clever than dumb software.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org There are two types of people in the world: People who think there are two kinds of people, and people who don't.
-
Faré -
Robert P. Goldman -
Stelian Ionescu