From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.org!not-for-mail From: Jason Rumney Newsgroups: gmane.emacs.devel Subject: Re: Emacs choosing a font with no latin characters when passed a bogus -fn Date: Sat, 21 Jun 2008 23:48:01 +0100 Message-ID: <485D8521.4050003@gnu.org> References: NNTP-Posting-Host: lo.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Trace: ger.gmane.org 1214088529 19969 80.91.229.12 (21 Jun 2008 22:48:49 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Sat, 21 Jun 2008 22:48:49 +0000 (UTC) Cc: emacs-devel Mailinglist To: Juanma Barranquero Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Sun Jun 22 00:49:34 2008 Return-path: Envelope-to: ged-emacs-devel@m.gmane.org Original-Received: from lists.gnu.org ([199.232.76.165]) by lo.gmane.org with esmtp (Exim 4.50) id 1KABtj-0004im-TJ for ged-emacs-devel@m.gmane.org; Sun, 22 Jun 2008 00:49:32 +0200 Original-Received: from localhost ([127.0.0.1]:37252 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1KABsv-0005jc-0k for ged-emacs-devel@m.gmane.org; Sat, 21 Jun 2008 18:48:41 -0400 Original-Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1KABsp-0005hm-3j for emacs-devel@gnu.org; Sat, 21 Jun 2008 18:48:35 -0400 Original-Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1KABsn-0005fg-Cn for emacs-devel@gnu.org; Sat, 21 Jun 2008 18:48:34 -0400 Original-Received: from [199.232.76.173] (port=53488 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1KABsn-0005fX-2X for emacs-devel@gnu.org; Sat, 21 Jun 2008 18:48:33 -0400 Original-Received: from mk-outboundfilter-4.mail.uk.tiscali.com ([212.74.114.32]:45533) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1KABsm-0004EZ-Tx for emacs-devel@gnu.org; Sat, 21 Jun 2008 18:48:33 -0400 Original-X-Trace: 101010094/mk-outboundfilter-2.mail.uk.tiscali.com/F2S/$ACCEPTED/freedom2Surf-customers/83.67.23.108 X-SBRS: None X-RemoteIP: 83.67.23.108 X-IP-MAIL-FROM: jasonr@gnu.org X-IP-BHB: Once X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ArMEAHciXUhTQxds/2dsb2JhbACBW6xP X-IronPort-AV: E=Sophos;i="4.27,684,1204502400"; d="scan'208";a="101010094" X-IP-Direction: IN Original-Received: from i-83-67-23-108.freedom2surf.net (HELO wanchan.jasonrumney.net) ([83.67.23.108]) by smtp.f2s.tiscali.co.uk with ESMTP; 21 Jun 2008 23:48:16 +0100 Original-Received: from [192.168.249.27] (chiko.jasonrumney.net [192.168.249.27]) by wanchan.jasonrumney.net (Postfix) with ESMTP id AAEFA2EC; Sat, 21 Jun 2008 23:49:11 +0100 (BST) User-Agent: Thunderbird 2.0.0.14 (Windows/20080421) In-Reply-To: X-Enigmail-Version: 0.95.6 OpenPGP: id=8086879D X-detected-kernel: by monty-python.gnu.org: Genre and OS details not recognized. X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Original-Sender: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Errors-To: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Xref: news.gmane.org gmane.emacs.devel:99644 Archived-At: Juanma Barranquero wrote: > Until a few weeks ago, Emacs crashed when passed a non-existent > font/fontset, like > > emacs -Q -fn "-*-nonexistent" > > Now it does not crash, but I'm still getting a weird effect (as > witnessed in the font log below, and the attached image). > > AFAICS, Emacs is selecting fonts (Estrangelo Edessa, Gautami, Latha, > Mangal, etc.) that do not have Latin characters. In fact, it has filtered the list of fonts down to those that do not contain support for any recognized Windows "charset", since the registry you supplied in the xfld pattern was not mapped by w32-charset-info-alist. Fonts for languages with poor OS support tend to be deliberately broken so they can be used with various kludge solutions on older versions of Windows, so it is not surprising that they map ASCII codepoints to one or more characters in other scripts rather than returning the .notdef glyph or including a basic ASCII subset. I added a script condition to the entry for iso10646-1 fonts for latin characters in fontset.el, which should result in a random latin font being selected instead of using these fonts. > sort-by: -*-unspecified-unspecified-unspecified-*-428567-*-nonexistent Where did the 428567 come from? This looks like a real bug, though it does later get reverted to 16, so probably never noticeable.