Locally we don't use any tabs in lisp files, and my emacs enforces that. Seems like that added a bunch of whitespace diffs to the patch. Sorry about that.
Best, r
The patch was applied. I should probably use the same emacs setting to avoid tabs (what do you use?).
Also, I bumped the version for good measure.
I intend to release 2.27 next week. Can people test on their systems and their implementations?
Xach, are things all clear with Quicklisp now?
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org No man would listen to you talk if he didn't know it was his turn next. — Edgar Watson Howe
Faré wrote:
The patch was applied. I should probably use the same emacs setting to avoid tabs (what do you use?).
Let me see....
In my slime-config.el:
(add-hook 'lisp-mode-hook '(lambda () ;; (setq show-trailing-whitespace t) (make-local-variable 'write-contents-hooks) (add-hook 'write-contents-hooks 'kill-all-tabs)))
and in my init.el:
;;;--------------------------------------------------------------------------- ;;; Kill tabs ;;;--------------------------------------------------------------------------- (defun kill-all-tabs () "Code lifted from jwz for untabifying before save. Note that this was originally java-mode-untabify." (interactive) (save-excursion ;; stop killing trailing whitespace... ;; (goto-char (point-min)) ;; (while (re-search-forward "[ \t]+$" nil t) ;; (delete-region (match-beginning 0) (match-end 0))) (goto-char (point-min)) (if (search-forward "\t" nil t) (untabify (1- (point)) (point-max)))) nil)
;; tab characters make shared editing a total mess. I'm killing them for all ;; edits I make. [2008/12/23:rpg] (setq-default indent-tabs-mode nil)
(setq whitespace-style '(indentation::space empty space-before-tab::space trailing space-after-tab::space))
Key insight is due to jwz -- and the time I worked with someone who used some windows editor that insisted that tabs were 8 characters wide, and whose editor turned all our indents into tab characters that broke the layout.
Our local svn repositories are configured to refuse to accept tab characters in commits.
This *mostly* works, but can be annoying when doing vendor branch installs of upstream code with tabs.
cheers, r
Faré fahree@gmail.com writes:
The patch was applied. I should probably use the same emacs setting to avoid tabs (what do you use?).
Also, I bumped the version for good measure.
I intend to release 2.27 next week. Can people test on their systems and their implementations?
Xach, are things all clear with Quicklisp now?
There is an unusual issue with the "periods" library that I can't quite puzzle out. It does not have any trouble with ASDF 2.26, but it fails to load from FASL, under some circumstances, with the latest commit to master. It seems like something in my build environment is triggering it, but I can't reproduce it reliably enough to figure it out.
I'll keep you posted.
Zach
I intend to release 2.27 next week. Can people test on their systems and their implementations?
Xach, are things all clear with Quicklisp now?
There is an unusual issue with the "periods" library that I can't quite puzzle out. It does not have any trouble with ASDF 2.26, but it fails to load from FASL, under some circumstances, with the latest commit to master. It seems like something in my build environment is triggering it, but I can't reproduce it reliably enough to figure it out.
I'll keep you posted.
Do you logs of some of these failures? Maybe I can reproduce at home, or, looking at the .asd and knowing what I changed, figure a scenario that explains the failure.
Faré fahree@gmail.com writes:
I intend to release 2.27 next week. Can people test on their systems and their implementations?
Xach, are things all clear with Quicklisp now?
There is an unusual issue with the "periods" library that I can't quite puzzle out. It does not have any trouble with ASDF 2.26, but it fails to load from FASL, under some circumstances, with the latest commit to master. It seems like something in my build environment is triggering it, but I can't reproduce it reliably enough to figure it out.
I'll keep you posted.
Do you logs of some of these failures? Maybe I can reproduce at home, or, looking at the .asd and knowing what I changed, figure a scenario that explains the failure.
http://report.quicklisp.org/periods/2012-12-12/failtail.txt is the typical failure log. The use of the SERIES package is conditionalized on the :PERIODS-USE-SERIES feature, and I can't find any place where that is in effect.
Zach
Also, I'd like to know if any package in Quicklisp relies on some ASDF internals I'd like to rename. This is especially harder when users define methods on a generic function — you can't rename the function, then. Is there an easy way to download EVERYTHING (excluding old versions) so I may grep the results?
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org Democracy is a pathetic belief in the collective wisdom of individual ignorance. — H.L. Mencken
On Thu, Dec 13, 2012 at 9:27 AM, Zach Beane xach@xach.com wrote:
Faré fahree@gmail.com writes:
I intend to release 2.27 next week. Can people test on their systems and their implementations?
Xach, are things all clear with Quicklisp now?
There is an unusual issue with the "periods" library that I can't quite puzzle out. It does not have any trouble with ASDF 2.26, but it fails to load from FASL, under some circumstances, with the latest commit to master. It seems like something in my build environment is triggering it, but I can't reproduce it reliably enough to figure it out.
I'll keep you posted.
Do you logs of some of these failures? Maybe I can reproduce at home, or, looking at the .asd and knowing what I changed, figure a scenario that explains the failure.
http://report.quicklisp.org/periods/2012-12-12/failtail.txt is the typical failure log. The use of the SERIES package is conditionalized on the :PERIODS-USE-SERIES feature, and I can't find any place where that is in effect.
Zach
Faré fahree@gmail.com writes:
Also, I'd like to know if any package in Quicklisp relies on some ASDF internals I'd like to rename. This is especially harder when users define methods on a generic function — you can't rename the function, then. Is there an easy way to download EVERYTHING (excluding old versions) so I may grep the results?
Sure. The easiest way is this:
(map nil 'ql-dist:ensure-installed (ql-dist:provided-releases t))
Then all sources currently in Quicklisp will be present in ~/quicklisp/dists/quicklisp/software/.
Someone(tm) should write a search engine that can answer questions like yours instantly via offline indexing of all public CL code!
Zach
Faré fahree@gmail.com writes:
Also, I'd like to know if any package in Quicklisp relies on some ASDF internals I'd like to rename. This is especially harder when users define methods on a generic function — you can't rename the function, then. Is there an easy way to download EVERYTHING (excluding old versions) so I may grep the results?
Also, if you want to make such changes on a branch, I can pull from that branch and test building the Quicklisp universe and let you know what fails.
Zach
-
Faré -
Robert Goldman -
Zach Beane