Return-Path: Received: from smtp1.axis.com (smtp1.axis.com [195.60.68.17]) by mail.openembedded.org (Postfix) with ESMTP id 18CE2619B9 for ; Tue, 3 Mar 2020 13:15:44 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=axis.com; l=2664; q=dns/txt; s=axis-central1; t=1583241346; x=1614777346; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=fPs8blzKeeeNEArAxq4rdelHSIAxakOfN+CZ+NY8gzI=; b=bWdpxAu4rA/aM+HknxTK45pTS8ZOIKScqH+x3ujsjDXx1dJrqvnZ7/VQ x8VIloIGp6OxIHLadktOy07e2BGGIBBBXeB0QR21DWsGII8nusq7cXZbA ixsqSsmnWnkycyJVvHct1n7PABl0Di4wT2Xc57fOOnEz66Rb6dgQ3nh+X RRyyIp/07ug5fqJs18SY7OP/5exFfmthkLCfHgmxr1LscdWVsVk5bN7l1 4HPrOxjCD76BQ86IROD2VXJ1KrbdzrrZnpYNoMB/bZVZPk3us4fZ/wBgU NzwXfCNlh+kW5luToJudgQBj5wDIzDwIZ8ptaF+fn30xgzUxHc+1xqSwK w==; IronPort-SDR: bwUkQ4+pjZ8iO9eEh2s6njdARcHKfNJKJiVG9qNXwoZ8xI3ll3Vukojm7GxrEu3Ugf2fGrK9lm jp8KBNkBhmVu7+KBoPiFLQ218gPPNuUs+FeXX3JnU19t3qWKmuzB4+7IS25es34VDBIbwhXfYa SpJePqvftx8uYEdKsah5mfkJ5oZ5y5ipJVFqfEl5AECMVlXg0p0OXhCJNnmw/jEE4MmxiudJUR cd+QwO9VhMwmuhdifNwB2/oW+9RXXdguV/38mu69uM3k9Kl0I3LpIBWgprc86PJSYaJQn9mpVX SsE= X-IronPort-AV: E=Sophos;i="5.70,511,1574118000"; d="scan'208";a="6042478" From: Peter Kjellerstedt To: "Robert P. J. Day" , OE Core mailing list Thread-Topic: [OE-core] is there ever a compelling reason for "FILESEXTRAPATHS_append := ..."? Thread-Index: AQHV774hmsJz5zIW+0uS6YrxyoT826gzjxKAgANI5DA= Date: Tue, 3 Mar 2020 13:15:45 +0000 Message-ID: <673fe5c88b2842ea8a023562a2043be3@XBOX03.axis.com> References: In-Reply-To: Accept-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.0.5.60] MIME-Version: 1.0 Subject: Re: is there ever a compelling reason for "FILESEXTRAPATHS_append := ..."? X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 03 Mar 2020 13:15:45 -0000 Content-Language: en-US Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable > -----Original Message----- > From: openembedded-core-bounces@lists.openembedded.org bounces@lists.openembedded.org> On Behalf Of Robert P. J. Day > Sent: den 1 mars 2020 12:51 > To: OE Core mailing list > Subject: Re: [OE-core] is there ever a compelling reason for > "FILESEXTRAPATHS_append :=3D ..."? >=20 > On Sun, 1 Mar 2020, Robert P. J. Day wrote: >=20 > > > > occasionally, i run across an existing bbappend file which contains > > > > FILESEXTRAPATHS_append :=3D ... > > > > and that makes me nervous as i don't see the rationale in *appending* > > to that variable in a bbappend file -- seems to defeat the purpose of > > potentially overriding what's in the original .bb recipe. > > > > now, sometimes it's obvious that it's broken as i'm currently > > looking at an example: > > > > FILESEXTRAPATHS_append :=3D "${THISDIR}/${PN}:" > > > > which makes no sense at all with the colon at the end. but other than > > something so clearly broken, are there situations to append to > > FILESEXTRAPATHS? that just seems counter-productive, but maybe i'm > > missing something. >=20 > figured i'd scan for examples of that in some of the many layers i > have checked out and found the following: >=20 > meta-boundary/recipes-qt/qt4/qt4-embedded_%.bbappend:FILESEXTRAPATHS_appe= nd :=3D "${THISDIR}/files:" >=20 > meta-intel/recipes-core/zlib/zlib-intel_1.2.11.1.jtkv6.3.bb:FILESEXTRAPAT= HS_append =3D ":${COREBASE}/meta/recipes-core/zlib/zlib" >=20 > meta-qt5/recipes-qt/qt5/qt5-ptest.inc:FILESEXTRAPATHS_append :=3D ":${THI= SDIR}/ptest" >=20 > meta-virtualization/recipes-networking/openvswitch/openvswitch_git.bb:FIL= ESEXTRAPATHS_append :=3D "${THISDIR}/${PN}-git:" >=20 > note that a couple seem clearly(?) wrong (like that last one), but Although contradictory to logic, it is apparently correct to use=20 ":" both with _prepend and _append when adding to=20 FILESEXTRAPATHS (see the default value set by bitbake.conf, which is=20 "__default:"). However, problems _will_ arise if different bbappend=20 files for the same recipe do it differently... It can also be noted=20 that the QA test in insane.bbclass suggests to use ":" with=20 _append, further adding to the confusion. > i'm open to clarification as to when this is useful. (i think i see > the application in that meta-intel layer example, although should that > use "OEROOT" rather than "COREBASE"?) No, ${COREBASE} is correct. (OEROOT is not a bitbake variable, it is=20 only available in the oe-init-build-env script.) >=20 > rday //Peter