From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.io!.POSTED.ciao.gmane.io!not-for-mail From: Eli Zaretskii Newsgroups: gmane.emacs.devel Subject: Re: Avoid duplicate emacs.exe / emacs-$version.exe Date: Sun, 29 Mar 2020 05:31:32 +0300 Message-ID: <83o8sf6ge3.fsf@gnu.org> References: <865zeoctg3.fsf@csic.es> <83v9mo5om8.fsf@gnu.org> <86y2rkb4a8.fsf@csic.es> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Injection-Info: ciao.gmane.io; posting-host="ciao.gmane.io:159.69.161.202"; logging-data="59127"; mail-complaints-to="usenet@ciao.gmane.io" Cc: emacs-devel@gnu.org To: Juan =?utf-8?Q?Jos=C3=A9_Garc=C3=ADa-Ripoll?= Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane-mx.org@gnu.org Sun Mar 29 04:32:04 2020 Return-path: Envelope-to: ged-emacs-devel@m.gmane-mx.org Original-Received: from lists.gnu.org ([209.51.188.17]) by ciao.gmane.io with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1jINk3-000FJ1-Hg for ged-emacs-devel@m.gmane-mx.org; Sun, 29 Mar 2020 04:32:03 +0200 Original-Received: from localhost ([::1]:33744 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jINk2-0001ks-JQ for ged-emacs-devel@m.gmane-mx.org; Sat, 28 Mar 2020 22:32:02 -0400 Original-Received: from eggs.gnu.org ([2001:470:142:3::10]:58853) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jINjT-0001DX-Kb for emacs-devel@gnu.org; Sat, 28 Mar 2020 22:31:28 -0400 Original-Received: from fencepost.gnu.org ([2001:470:142:3::e]:44313) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1jINjT-0008S1-GG; Sat, 28 Mar 2020 22:31:27 -0400 Original-Received: from [176.228.60.248] (port=1160 helo=home-c4e4a596f7) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1jINjR-0007tj-Uq; Sat, 28 Mar 2020 22:31:26 -0400 In-Reply-To: <86y2rkb4a8.fsf@csic.es> (juanjose.garciaripoll@gmail.com) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-devel-bounces+ged-emacs-devel=m.gmane-mx.org@gnu.org Original-Sender: "Emacs-devel" Xref: news.gmane.io gmane.emacs.devel:245921 Archived-At: > From: Juan José García-Ripoll > > Date: Sat, 28 Mar 2020 21:41:51 +0100 > > > No, they don't waste space, at least not by default. When you install > > Emacs, the installation procedure produces a hard link with another > > name to the same file data. These two names are just 2 different > > names that refer to the same disk space. [...] > > This means your installation procedure is modified, or maybe you > > installed a binary someone else produced, in which case the archive > > used to package the binaries didn't support hard links. You can > > restore the hard link by removing onhe of the copies and making a hard > > link to the remaining copy under the other name. > > I am reporting figures that come from either (a) the official release > from ftp.gnu.org, (b) the prerelease *.zip files from alpha.gnu.org and > (c) the official build process as reported in emacs-27/nt. AFAIK, the ZIP archive indeed doesn't support hard links, at least not on MS-Windows. But you can recreate the hard link locally, if you want, after unzipping the archive. > - Building from emacs-27 branch (Savannah's git), following the > instructions from nt/ (i.e. configure + make install), creates two > identical files, emacs.exe and emacs-27.0.90.exe. Space usage is as in > 26.3, about 123Mb per executable. They aren't 2 files, they are 2 _names_. Look at the link count of each file, and you will see 2, not 1. > So I am using stock files, never my own copies. I am using also standard > procedures. I do not understand why the executable sizes differ so much > between release, prerelease and built from sources. However, there are > no symbolic links happening at all. Indeed, MSYS's "ln" as used in the > autoconf build process does not seem to work: it just copies the file. I didn't say symbolic links, I said hard links. > $ ln -sf foo faa Not "ln -s", just "ln". > So maybe you are discussing what happens in Linux or Mac systems? It happens on all supported systems.