Re: [gentoo-dev] USE_EXPAND and IUSE

Re: [gentoo-dev] USE_EXPAND and IUSE

[Alin Nastac]

[-- Attachment #1: Type: text/plain, Size: 558 bytes --]

Jason Stubbs wrote:

>So there have been many complaints about how USE_EXPANDed flags don't belong 
>in IUSE. There haven't actually been any reasons given though. :P
>
>  
>
net-dialup/pppconfig-2.3.11-r1 depends on LINGUAS, but the list of
supported languages is created in pkg_unpack, based on what tarball
contains.

The IUSE thing would raise 2 problems:
 - lower maintainability - not really a major problem
 - if user wants all languages, it will be forced to select them one by
one - now if LINGUAS is empty, all available languages gets installed.


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 256 bytes --]

Re: [gentoo-dev] USE_EXPAND and IUSE

[Sven Wegener]

[-- Attachment #1: Type: text/plain, Size: 1671 bytes --]

On Sun, Jun 19, 2005 at 07:48:27PM +0300, Alin Nastac wrote:
> Jason Stubbs wrote:
> 
> >So there have been many complaints about how USE_EXPANDed flags don't belong 
> >in IUSE. There haven't actually been any reasons given though. :P
> net-dialup/pppconfig-2.3.11-r1 depends on LINGUAS, but the list of
> supported languages is created in pkg_unpack, based on what tarball
> contains.
> 
> The IUSE thing would raise 2 problems:
>  - lower maintainability - not really a major problem
>  - if user wants all languages, it will be forced to select them one by
> one - now if LINGUAS is empty, all available languages gets installed.

We're talking about USE_EXPANDed USE flags. The LINGUAS variable will
still exist and the way you use LINGUAS in net-dialup/pppconfig doesn't
involve USE flags. So you don't need to add them to IUSE. We currently
don't have a policy that enforces the use of the expanded linguas_* USE
flags instead of directly using the LINGUAS variable. Maybe we should
first decide, if it's OK to use LINGUAS directly. If not, we need to
modify the packages, which is tricky for packages like pppconfig that
take the list of available languages from the tarball.

IMHO using the linguas_* USE flags makes better binary packages because
the USE flags are handled by portage and shown on emerge -pv output.
Portage will merge the binary package using the compiled USE flags and
will not react on current USE settings. Same is for LINGUAS, but using
the USE flags the user knows on emerge -pv with which language support
the package was compiled.

Sven

-- 
Sven Wegener
Gentoo Linux Developer
http://www.gentoo.org/

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

[gentoo-dev] USE_EXPAND and IUSE

[Jason Stubbs]

[-- Attachment #1: Type: text/plain, Size: 1289 bytes --]

Hi all,

So there have been many complaints about how USE_EXPANDed flags don't belong 
in IUSE. There haven't actually been any reasons given though. :P

I've assumed that the reasons they haven't been added thus far are due to what 
emerge's output would look like if they were. So I've taken the liberty of 
fixing up the output a bit. Now IUSE flags that are expansions of USE_EXPAND 
will be outputted like this:

[ebuild  N    ] kde-base/kde-i18n-3.4.1  +arts -debug +kdeenablefinal 
-xinerama LINGUAS="-ar -bg -bn -br -bs -ca -cs -cy -da -de -el -en_GB -eo -es 
-et -eu -fi -fr -fy -ga -he -hi -hsb -hu -is -it +ja -lt -mk -nb -nds -nl -nn 
-pa -pl -pt -pt_BR -ro -ru -se -sk -sl -sr -sr@Latn -sv -ta -tg -tr -uk 
-zh_CN" 0 kB

The format follows the same as standard USE flags but are separated out into 
the individual variables that enable them. Of course, the ebuilds themselves 
need to be updated to add the appropriate flags to IUSE.

Although I believe they should be, I'll leave it up to others to decide 
whether said flags should be documented in use.desc. However, if it is 
decided to not document them in use.desc, they won't be dropped from 
repoman's checks until there is some other way that they can be documented.

Regards,
Jason Stubbs

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] USE_EXPAND and IUSE

[Jason Stubbs]

[-- Attachment #1: Type: text/plain, Size: 1113 bytes --]

On Monday 20 June 2005 01:48, Alin Nastac wrote:
> Jason Stubbs wrote:
> > So there have been many complaints about how USE_EXPANDed flags don't
> > belong in IUSE. There haven't actually been any reasons given though. :P
>
> net-dialup/pppconfig-2.3.11-r1 depends on LINGUAS, but the list of
> supported languages is created in pkg_unpack, based on what tarball
> contains.

What happened to determinism and predictability? From the user's standpoint, 
there is no way to know what is going to be installed until it is actually 
installed.

> The IUSE thing would raise 2 problems:
>  - lower maintainability - not really a major problem
>  - if user wants all languages, it will be forced to select them one by
> one - now if LINGUAS is empty, all available languages gets installed.

FRITZCAPI_CARDS FCDSL_CARDS VIDEO_CARDS INPUT_DEVICES LINGUAS

Do any of these not follow the same pattern of empty var == all installed? Any 
objections to making that a requirement for adding a new USE_EXPAND variable? 
If so, the emerge *display* issue would be no problem.

Regards,
Jason Stubbs

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] USE_EXPAND and IUSE

[Donnie Berkholz]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jason Stubbs wrote:
> FRITZCAPI_CARDS FCDSL_CARDS VIDEO_CARDS INPUT_DEVICES LINGUAS
> 
> Do any of these not follow the same pattern of empty var == all installed? Any 
> objections to making that a requirement for adding a new USE_EXPAND variable? 
> If so, the emerge *display* issue would be no problem.

x11-drm dies if you don't set VIDEO_CARDS. At the time, I thought it was
better to make the user understand the available video cards and set
them manually, since settings of multiple cards are allowed. Perhaps
that's worth rethinking.

Thanks,
Donnie
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFCt6GPXVaO67S1rtsRAhtgAJ97XgL92Tj/3aW4xYQMcx5pMzYTWwCgj++F
kdOxDd1RJNOJCMOmizr2zGk=
=2Fyn
-----END PGP SIGNATURE-----
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] USE_EXPAND and IUSE

[Alin Nastac]

[-- Attachment #1: Type: text/plain, Size: 1222 bytes --]

Jason Stubbs wrote:

>On Monday 20 June 2005 01:48, Alin Nastac wrote:
>  
>
>>net-dialup/pppconfig-2.3.11-r1 depends on LINGUAS, but the list of
>>supported languages is created in pkg_unpack, based on what tarball
>>contains.
>>    
>>
>
>What happened to determinism and predictability? From the user's standpoint, 
>there is no way to know what is going to be installed until it is actually 
>installed.
>
>  
>
I don't really understand why you consider pppconfig unpredictable.
the ebuild displays what languages will be installed.

>>The IUSE thing would raise 2 problems:
>> - lower maintainability - not really a major problem
>> - if user wants all languages, it will be forced to select them one by
>>one - now if LINGUAS is empty, all available languages gets installed.
>>    
>>
>
>FRITZCAPI_CARDS FCDSL_CARDS VIDEO_CARDS INPUT_DEVICES LINGUAS
>
>Do any of these not follow the same pattern of empty var == all installed? Any 
>objections to making that a requirement for adding a new USE_EXPAND variable? 
>If so, the emerge *display* issue would be no problem.
>  
>
This would be fine as long as LINGUAS do not appear in IUSE. When
LINGUAS var is empty, the "equery uses" report would be an abomination .

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 256 bytes --]

Re: [gentoo-dev] USE_EXPAND and IUSE

[Donnie Berkholz]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jason Stubbs wrote:
> Any reason why INPUT_DEVICES is needed in USE_EXPAND rather than just as local 
> USE flags? It's kind of the odd one out among the above.

The reasoning for that is that hardware support doesn't make sense as
USE flags, so it should be something else. In this case, that was
INPUT_DEVICES. We haven't been able to take much advantage of it yet,
but it may work out better after X's upstream modularization.

Thanks,
Donnie
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFCuEBpXVaO67S1rtsRApqXAKCssJLG1lWtsoNXWGz05I8vGoxw1wCgvPGF
SaEd0RXRIJqok+oOjgUg39Y=
=SZxT
-----END PGP SIGNATURE-----
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] USE_EXPAND and IUSE

[Alec Warner]

Donnie Berkholz wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Jason Stubbs wrote:
> 
>>Any reason why INPUT_DEVICES is needed in USE_EXPAND rather than just as local 
>>USE flags? It's kind of the odd one out among the above.
> 
> 
> The reasoning for that is that hardware support doesn't make sense as
> USE flags, so it should be something else. In this case, that was
> INPUT_DEVICES. We haven't been able to take much advantage of it yet,
> but it may work out better after X's upstream modularization.
   So those make no sense but 3dnow,sse,sse2,makemyprocessorgrow are 
fine as USE flags?  Those also enable support for special hardware.

> 
> Thanks,
> Donnie
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.1 (GNU/Linux)
> 
> iD8DBQFCuEBpXVaO67S1rtsRApqXAKCssJLG1lWtsoNXWGz05I8vGoxw1wCgvPGF
> SaEd0RXRIJqok+oOjgUg39Y=
> =SZxT
> -----END PGP SIGNATURE-----

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] USE_EXPAND and IUSE

[Donnie Berkholz]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Alec Warner wrote:
> Donnie Berkholz wrote:
> The reasoning for that is that hardware support doesn't make sense as
> USE flags, so it should be something else. In this case, that was
> INPUT_DEVICES. We haven't been able to take much advantage of it yet,
> but it may work out better after X's upstream modularization.
> 
>>   So those make no sense but 3dnow,sse,sse2,makemyprocessorgrow are fine
>> as USE flags?  Those also enable support for special hardware.

I think we had that argument already. Try searching the archives and see
whether you can find it.

Thanks,
Donnie
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFCuElEXVaO67S1rtsRAhDpAKDrizHHPj7QVEYmfsuaekGagBsdxACeNDbO
0MEGWzTPSh5IDbvZwDtjnZs=
=61aD
-----END PGP SIGNATURE-----
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] USE_EXPAND and IUSE

[Jason Stubbs]

[-- Attachment #1: Type: text/plain, Size: 1187 bytes --]

On Tuesday 21 June 2005 14:35, Alin Nastac wrote:
> I don't really understand why you consider pppconfig unpredictable.
> the ebuild displays what languages will be installed.

Because there is no way to foretell what will be installed without actually 
initiating an install.

> > >The IUSE thing would raise 2 problems:
> > > - lower maintainability - not really a major problem
> > > - if user wants all languages, it will be forced to select them one by
> > >one - now if LINGUAS is empty, all available languages gets installed.
> >
> > FRITZCAPI_CARDS FCDSL_CARDS VIDEO_CARDS INPUT_DEVICES LINGUAS
> >
> > Do any of these not follow the same pattern of empty var == all installed?
> > Any objections to making that a requirement for adding a new USE_EXPAND
> > variable? If so, the emerge *display* issue would be no problem.
>
> This would be fine as long as LINGUAS do not appear in IUSE. When
> LINGUAS var is empty, the "equery uses" report would be an abomination .

The whole point is to add this stuff to IUSE and improve the output 
accordingly. Using current tools as a reason to not improve things isn't 
really logical.

Regards,
Jason Stubbs

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] USE_EXPAND and IUSE

[Jason Stubbs]

[-- Attachment #1.1: Type: text/plain, Size: 1166 bytes --]

On Wednesday 22 June 2005 05:28, Jason Stubbs wrote:
> On Tuesday 21 June 2005 14:35, Alin Nastac wrote:
> > This would be fine as long as LINGUAS do not appear in IUSE. When
> > LINGUAS var is empty, the "equery uses" report would be an abomination .
>
> The whole point is to add this stuff to IUSE and improve the output
> accordingly. Using current tools as a reason to not improve things isn't
> really logical.

And here's a messy patch that took me ten minutes (after looking at equery 
code for the first time) that does just that. Output as follows:

$ ./equery uses portage
...
[ Found these USE variables for sys-apps/portage-2.0.51.22-r1 ]
 U I
 - - build   : !!internal use only!! DO NOT SET THIS FLAG YOURSELF!, used for 
creating build images and the first half of bootstrapping.
 - - selinux : !!internal use only!! Security Enhanced Linux support, this 
must be set by the selinux profile or breakage will occur

$ ./equery vars portage
...
[ Found these environment variables for sys-apps/portage-2.0.51.22-r1 ]
 U I
----- LINGUAS
 + + en : <unknown>
 - - fr : <unknown>
 + + ja : <unknown>


Regards,
Jason Stubbs

[-- Attachment #1.2: equery-use_expand.patch --]
[-- Type: text/x-diff, Size: 7741 bytes --]

--- /usr/bin/equery	2005-06-22 05:29:36.000000000 +0900
+++ equery	2005-06-22 06:14:45.000000000 +0900
@@ -451,6 +451,8 @@
 		except IOError:
 			print_warn(5, "Could not load USE flag descriptions from " + ppath(gentoolkit.settings["PORTDIR"] + "/profiles/use.desc"))
 
+		use_expand = gentoolkit.settings["USE_EXPAND"].lower().split()
+
 		# Load local USE flag descriptions
 		try:
 			fd = open(gentoolkit.settings["PORTDIR"]+"/profiles/use.local.desc")
@@ -483,12 +485,19 @@
 			matches_found += 1
 			
 			bestver = p.get_cpv()
-			iuse = p.get_env_var("IUSE")
-		  
-			if iuse:
-				usevar = iuse.split()
-			else:
-				usevar = []
+			iuse = p.get_env_var("IUSE").split()
+
+			for usevar in iuse[:]:
+				found = False
+				for usehead in use_expand:
+					if usevar.startswith(usehead+"_"):
+						found = True
+						break
+				if found:
+					iuse.remove(usevar)
+
+			usevar = iuse
+			usevar.sort()
 
 			inuse = []
 			if p.is_installed():
@@ -570,6 +579,206 @@
 			   "  " + pp.localoption("-a, --all") + "	 - include non-installed packages\n"
 
 
+class CmdDisplayVars(Command):
+	"""Advanced report of a package's USE flags"""
+	def __init__(self):
+		self.default_opts = {
+			"installedOnly" : True
+			}
+	def parseArgs(self, args):
+
+		query = ""
+		need_help = 0
+		opts = self.default_opts
+		skip = 0
+		
+		for i in xrange(len(args)):
+
+			if skip:
+				skip -= 1
+				continue
+			x = args[i]
+			
+			if x in ["-h","--help"]:
+				need_help = 1
+				break
+			elif x in ["-a", "--all"]:
+				opts["installedOnly"] = False
+			else:
+				query = x
+
+		if need_help or query == "":
+			print_info(0, self.longHelp())
+			sys.exit(-1)
+			
+		return (query, opts)
+
+	def perform(self, args):
+
+		(query, opts) = self.parseArgs(args)
+
+		if not Config["piping"]:
+			print_info(3, "[ Searching for packages matching " + pp.pkgquery(query) + "... ]")
+		
+		matches = gentoolkit.find_packages(query, True)
+
+		if not matches:
+			die(3, "No matching packages found for \"" + pp.pkgquery(query) + "\"")
+
+
+		useflags = gentoolkit.settings["USE"].split()
+		usedesc = {}
+		uselocaldesc = {}
+
+		# Load global USE flag descriptions
+		try:
+			fd = open(gentoolkit.settings["PORTDIR"]+"/profiles/use.desc")
+			usedesc = {}
+			for line in fd.readlines():
+				if line[0] == "#":
+					continue
+				fields = line.split(" - ")
+				if len(fields) == 2:
+					usedesc[fields[0].strip()] = fields[1].strip()
+		except IOError:
+			print_warn(5, "Could not load USE flag descriptions from " + ppath(gentoolkit.settings["PORTDIR"] + "/profiles/use.desc"))
+
+		# Load local USE flag descriptions
+		try:
+			fd = open(gentoolkit.settings["PORTDIR"]+"/profiles/use.local.desc")
+			for line in fd.readlines():
+				if line[0] == "#":
+					continue
+				fields = line.split(" - ")
+				if len(fields) == 2:
+					catpkguse = re.search("([a-z]+-[a-z]+/.*):(.*)", fields[0])
+					if catpkguse:
+						if not uselocaldesc.has_key(catpkguse.group(1).strip()):
+							uselocaldesc[catpkguse.group(1).strip()] = {catpkguse.group(2).strip() : fields[1].strip()}
+						else:
+							uselocaldesc[catpkguse.group(1).strip()][catpkguse.group(2).strip()] = fields[1].strip()
+		except IOError:
+				print_warn(5, "Could not load USE flag descriptions from " + path(gentoolkit.settings["PORTDIR"] + "/profiles/use.desc"))
+
+		if not Config["piping"]: 
+			print_info(3, "[ Colour Code : " + pp.useflagon("set") + " " + pp.useflagoff("unset") + " ]")
+			print_info(3, "[ Legend	: Left column  (U) - USE flags from make.conf			  ]")
+			print_info(3, "[		   : Right column (I) - USE flags packages was installed with ]")
+
+		use_expand = gentoolkit.settings["USE_EXPAND"].lower().split()
+
+		# Iterate through matches, printing a report for each package
+		matches_found = 0
+		for p in matches:
+
+			if not p.is_installed() and opts["installedOnly"]:
+				continue
+	
+			matches_found += 1
+			
+			bestver = p.get_cpv()
+			iuse = p.get_env_var("IUSE").split()
+
+			use_expand_map = {}
+
+			for usevar in iuse[:]:
+				found = False
+				for usehead in use_expand:
+					if usevar.startswith(usehead+"_"):
+						found = True
+						use_expand_map[usevar] = usehead
+						break
+				if not found:
+					iuse.remove(usevar)
+
+			usevar = use_expand_map.keys()
+			usevar.sort()
+
+			inuse = []
+			if p.is_installed():
+				used = p.get_use_flags().split()
+			else:
+				# cosmetic issue here as noninstalled packages don't have "used" flags
+				used = useflags
+
+			# store (inuse, inused, flag, desc)
+			output = []
+
+			for u in usevar:
+				inuse = 0
+				inused = 0
+				try:
+					desc = usedesc[u]
+				except KeyError:
+					try:
+						desc = uselocaldesc[p.get_category() + "/" + p.get_name()][u]
+					except KeyError:
+						desc = ""
+
+				if u in p.get_settings("USE"):
+					inuse = 1
+				if u in used:
+					inused = 1
+
+				output.append((inuse, inused, u, desc))
+
+			# pretty print
+			if output:
+				if not Config["piping"]:
+					print_info(0, "[ Found these environment variables for " + pp.cpv(bestver) + " ]")
+					print_info(3, pp.emph(" U I"))
+
+				maxflag_len = 0
+				for inuse, inused, u, desc in output:
+					if len(u) - len(use_expand_map[u]) - 1 > maxflag_len:
+						maxflag_len = len(u)-len(use_expand_map[u])-1
+
+				prev_usehead = ""
+				for in_makeconf, in_installed, flag, desc in output:
+					if use_expand_map[flag] != prev_usehead:
+						prev_usehead = use_expand_map[flag]
+						print "-----",prev_usehead.upper()
+					markers = ["-","+"]
+					colour = [pp.useflagoff, pp.useflagon]
+					if Config["piping"]:
+						print_info(0, markers[in_makeconf] + flag[len(prev_usehead)+1:])
+					else:
+						if in_makeconf != in_installed:
+							print_info(0, pp.emph(" %s %s" % (markers[in_makeconf], markers[in_installed])), False)
+						else:
+							print_info(0, " %s %s" % (markers[in_makeconf], markers[in_installed]), False)
+	
+						print_info(0, " " + colour[in_makeconf](flag[len(prev_usehead)+1:].ljust(maxflag_len)), False)
+
+						# print description
+						if desc:
+							print_info(0, " : " + desc)
+						else:
+							print_info(0, " : <unknown>")
+			else:
+				if not Config["piping"]:
+					print_info(1, "[ No environment variables found for " + pp.cpv(p.get_cpv()) + "]")
+
+		if Config["verbosityLevel"] >= 2:
+			if matches_found == 0:
+				s = ""
+				if opts["installedOnly"]:
+					s = "installed "
+				die(3, "No " + s + "packages found for " + pp.pkgquery(query))
+				
+					
+	def shortHelp(self):
+		return pp.localoption("<local-opts> ") + pp.pkgquery("pkgspec") + " - display environment variables for " + pp.pkgquery("pkgspec")
+	def longHelp(self):
+		return "Display environment for a given package\n" + \
+			   "\n" + \
+			   "Syntax:\n" + \
+			   "  " + pp.command("vars") + pp.localoption(" <local-opts> ") + pp.pkgquery("pkgspec") + \
+			   "\n" + \
+			   pp.localoption("<local-opts>") + " is either of: \n" + \
+			   "  " + pp.localoption("-a, --all") + "	 - include non-installed packages\n"
+
+
 class CmdDisplayDepGraph(Command):
 	"""Display tree graph of dependencies for a query"""
 
@@ -1403,6 +1612,7 @@
 	"depends" : CmdListDepends(),
 	"hasuse" : CmdFindUSEs(),
 	"uses" : CmdDisplayUSEs(),
+	"vars" : CmdDisplayVars(),
 	"depgraph" : CmdDisplayDepGraph(),
 	"changes" : CmdDisplayChanges(),
 	"size" : CmdDisplaySize(),
@@ -1427,6 +1637,7 @@
 	"s" : "size",
 	"t" : "stats",
 	"u" : "uses",
+	"v" : "vars",
 	"w" : "which"
 }
 

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] USE_EXPAND and IUSE

[Jason Stubbs]

[-- Attachment #1: Type: text/plain, Size: 1213 bytes --]

On Tuesday 21 June 2005 14:11, Donnie Berkholz wrote:
> Jason Stubbs wrote:
> > FRITZCAPI_CARDS FCDSL_CARDS VIDEO_CARDS INPUT_DEVICES LINGUAS
> >
> > Do any of these not follow the same pattern of empty var == all
> > installed? Any objections to making that a requirement for adding a new
> > USE_EXPAND variable? If so, the emerge *display* issue would be no
> > problem.
>
> x11-drm dies if you don't set VIDEO_CARDS. At the time, I thought it was
> better to make the user understand the available video cards and set
> them manually, since settings of multiple cards are allowed. Perhaps
> that's worth rethinking.

It seems we have the following behaviour when unset:

FRITZCAPI_CARDS => Selective based on "pcmcia" and "usb" USE flags
FCDSL_CARDS => All
VIDEO_CARDS => Dies
INPUT_DEVICES => None
LINGUAS => All

It would be nice to unify the behaviour among these. I like either the "All" 
or the "Dies" option though. While the FRITZCAPI_CARDS behaviour is 'probably 
what the user wants', it is not predictable.

Any reason why INPUT_DEVICES is needed in USE_EXPAND rather than just as local 
USE flags? It's kind of the odd one out among the above.

Regards,
Jason Stubbs

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] USE_EXPAND and IUSE

[Jason Stubbs]

[-- Attachment #1: Type: text/plain, Size: 843 bytes --]

On Wednesday 22 June 2005 02:07, Donnie Berkholz wrote:
> Alec Warner wrote:
> > Donnie Berkholz wrote:
> > The reasoning for that is that hardware support doesn't make sense as
> > USE flags, so it should be something else. In this case, that was
> > INPUT_DEVICES. We haven't been able to take much advantage of it yet,
> > but it may work out better after X's upstream modularization.
> >
> >>   So those make no sense but 3dnow,sse,sse2,makemyprocessorgrow are
> >> fine as USE flags?  Those also enable support for special hardware.
>
> I think we had that argument already. Try searching the archives and see
> whether you can find it.

I tried searching but no luck. Care to elaborate on how "building optional 
support for drivers" doesn't fall under the umbrella of "building optional 
support"?

-- 
Jason Stubbs

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]