Re: [Automake-NG] [FYI] make flags analysis: take advantage of GNU make features

Le 13 mai 2013 à 20:08, Stefano Lattarini a écrit : > Hi Akim, glad to read again from you :-) Hi Stefano! I'm happy both to read from you, and to see Automake NG is still alive. Sleeping beauty, awaiting for a kiss from her Stefano. > BTW, Akim, would you like me to update your e-mail addr

Re: [Automake-NG] [FYI] make flags analysis: take advantage of GNU make features

Le 8 mai 2013 à 01:12, Stefano Lattarini a écrit : > +# Shell code that determines whether the current make instance is No longer shell code. > +# running with a given letter option (e.g., -k, -n) that takes > +# no argument. It is either 'true' or 'false', so that it can be > +# easily used

Re: [Automake-NG] Removal of INCLUDES in favour of AM_CPPFLAGS

Le 1 févr. 2013 à 10:35, Stefano Lattarini a écrit : > So, is anyone using or playing with Automake-NG, or at least > contemplating the idea to do so in the short term? Or should > we just let the project die? I subscribe to all the good opinions about NG that have been made here. I would de

Re: [Automake-NG] bug#13351: [IMPORTANT] Dropping support for split '.info' files in mainline Automake

Le 7 janv. 2013 à 20:30, Stefano Lattarini a écrit : > Reference: > > > On 01/04/2013 05:43 PM, Stefano Lattarini wrote: >> Hi Thien-Thi, thanks for the feedback. >> >> On 01/04/2013 03:07 PM, Thien-Thi Nguyen wrote: >>> () Stefano Lattarin

Re: [Automake-NG] [FYI 2/2] built sources: avoid fork bomb in $(BUILT_SOURCES) handling

Le 19 sept. 2012 à 21:49, Stefano Lattarini a écrit : > Well, if the user calls "make all" in the recipe for a $(BUILT_SOURCES), > he will still experience a fork bomb, even after this patch -- but that > is the case also for mainline Automake, so no regression here. On the > other hand, before

Re: [Automake-NG] [FYI 2/2] built sources: avoid fork bomb in $(BUILT_SOURCES) handling

Le 16 sept. 2012 à 20:21, Stefano Lattarini a écrit : > Hi Akim. > > On 09/16/2012 03:48 PM, Akim Demaille wrote: >> >> Le 11 sept. 2012 à 16:49, Stefano Lattarini a écrit : >> >>> Due to how the handling of $(BUILT_SOURCES) was implemented in Automake-

Re: [Automake-NG] [FYI 2/2] built sources: avoid fork bomb in $(BUILT_SOURCES) handling

Le 11 sept. 2012 à 16:49, Stefano Lattarini a écrit : > Due to how the handling of $(BUILT_SOURCES) was implemented in Automake-NG, > a recursive make call in the recipe of any $(BUILT_SOURCES) (or of any of > its prerequisites) would have caused an infinite recursion (complete with > fork bomb,

Re: [Automake-NG] [FYI] [ng] subdirs: merge $(RECURSIVE_CLEAN_TARGETS) into $(RECURSIVE_TARGETS)

Le 4 sept. 2012 à 16:20, Bob Friesenhahn a écrit : > On Mon, 3 Sep 2012, Stefano Lattarini wrote: > >> On 08/31/2012 09:08 PM, Bob Friesenhahn wrote: >>> On Fri, 31 Aug 2012, Akim Demaille wrote: >>>>> + distclean-recursive \ >>>>> + mainta

Re: [Automake-NG] [PATCH 1/7] [ng] dist: refactor: reduce duplication in the 'dist-*' recipes

Le 12 août 2012 à 23:20, Stefano Lattarini a écrit : > * lib/am/distcheck.mk: Here, with heavy refactorings, in particular > introducing ... > (am.dist.all-formats, > am.dist.all-targets, > am.dist.compress-cmd.gzip, > am.dist.uncompress-cmd.gzip, > am.dist.compress-cmd.bzip2, > am.dist.uncompres

Re: [Automake-NG] [FYI 0/4] Some more Texinfo-related patches

Le 12 août 2012 à 14:49, Stefano Lattarini a écrit : > Stefano Lattarini (4): > [ng] rename: am__create_installdir -> am.texi.create-installdir > [ng] rename: am__can_run_installinfo -> am.texi.can-run-installinfo > [ng] refactor: new internal make variable 'am.conf.no-installinfo' > [ng] tex

Re: [Automake-NG] [FYI 00/10] Minor changes to Texinfo support (mostly cosmetic)

Le 12 août 2012 à 13:19, Stefano Lattarini a écrit : > Stefano Lattarini (10): > [ng] rename: am__texibuild_dvi_or_pdf -> am.texi.build.dvi-or-pdf > [ng] rename: am__texibuild_html -> am.texi.build.html > [ng] rename: am__texibuild_info -> am.texi.build.info > [ng] cosmetics: remove a stray e

Re: [Automake-NG] [PATCH 08/10] [ng] texi: be more aware of possible failures in recipes

Le 12 août 2012 à 13:19, Stefano Lattarini a écrit : > @@ -90,12 +90,10 @@ define am.texi.build.html > -I $(@D) -I $(srcdir)/$(@D) \ > -o $(@:.html=.htp) $<; \ Why bother we such names instead of $@.tmp or $@-t? We no longer have to deal with

Re: [Automake-NG] [FYI] [ng] tests: avoid spurious error with ctags from older Emacs and XEmacs

Le 12 août 2012 à 00:06, Stefano Lattarini a écrit : > diff --git a/t/ctags.sh b/t/ctags.sh > index f3f8a6a..22d 100755 > --- a/t/ctags.sh > +++ b/t/ctags.sh > @@ -44,7 +44,7 @@ test-ctags: ctags > grep 'main' sub1/tags > grep 'choke_me' sub1/tags && exit 1; : > grep 'subsub

Re: [Automake-NG] [FYI] [ng] contrib: rewrite and fix 'check-html' and 'recheck-html'

Le 11 août 2012 à 21:52, Stefano Lattarini a écrit : > That has actually been broken for some time, due to the changes > done in Automake-NG proper in the meantime. Thanks for this. Do you have tests? How do you check this?

Re: [Automake-NG] [FYI] [ng] refactor: remove all uses of '$(am__strip_dir)'

Le 31 août 2012 à 13:25, Akim Demaille a écrit : >> - @$(if $(%DIR%_PYTHON),$(MKDIR_P) "$(DESTDIR)$(%NDIR%dir)") >> + @$(if $(and $(%NDIR%dir),$(%DIR%_PYTHON)),$(strip \ >> + )$(MKDIR_P) "$(DESTDIR)$(%NDIR%dir)") > > I don'

Re: [Automake-NG] What is the purpose of split info files?

Le 12 août 2012 à 12:03, Stefano Lattarini a écrit : > Hello everybody. Hi all! > In Automake, the need to support split '.info' output files adds some > annoying complexity in several places; and for backward-compatibility, > that is sadly bound not to change, surely not in the near future, an

Re: [Automake-NG] [FYI] [ng] refactor: in declaration of Texinfo-related files to be cleaned

Le 11 août 2012 à 20:57, Stefano Lattarini a écrit : > + $clean_files{"$dpfx.dvi"} = CLEAN; > + $clean_files{"$dpfx.pdf"} = CLEAN; > + $clean_files{"$dpfx.ps"} = CLEAN; I should probably read at the whole code, but really, this has always seemed reversed to me: why $clean_files{

Re: [Automake-NG] [FYI] [ng] texi: less use of transforms in 'texinfos.am'

Le 11 août 2012 à 20:17, Stefano Lattarini a écrit : > + # Append to dirs, not files, because the files in '$*clean' can also > + # contain any directory created by "makeinfo --html", as well as the > + # '*.t2d' and '*.t2p' directories used by texi2dvi and texi2pdf. Can't we use only t2d? c

Re: [Automake-NG] [FYI] [ng] subdirs: merge $(RECURSIVE_CLEAN_TARGETS) into $(RECURSIVE_TARGETS)

Le 11 août 2012 à 19:09, Stefano Lattarini a écrit : > diff --git a/lib/am/subdirs.mk b/lib/am/subdirs.mk > index 1d39596..9e7 100644 > --- a/lib/am/subdirs.mk > +++ b/lib/am/subdirs.mk > @@ -14,13 +14,17 @@ > ## You should have received a copy of the GNU General Public License > ## along wit

Re: [Automake-NG] [FYI] [ng] dist: fixup: run slower compressors first

Le 11 août 2012 à 11:51, Stefano Lattarini a écrit : > For better parallelism in "make dist". This was already the case > for mainline Automake, but our recent changes to the dist-related > code in Automake-NG had broken that invariant. > > * automake.in (handle_dist): Here, use 'unshift' rathe

Re: [Automake-NG] [FYI] [ng] dist: remove obsolete distribution formats

Le 11 août 2012 à 11:21, Stefano Lattarini a écrit : > Those formats being the 'shar' archive and the tar archive compressed > with 'compress'. Support for them is kept in mainline Automake for > the sake of backward compatibility only, and modern software releases > should not use those formats

Re: [Automake-NG] [FYI] [ng] disthook: use a variable, not a transform, to decide whether to run it

Le 10 août 2012 à 16:13, Stefano Lattarini a écrit : > This is mostly a preparatory refactoring in view of future changes. > > * automake.in (handle_dist): Here, define the internal variable > 'am.conf.handle-distcheck-hook' to "yes" if a user-defined > 'distcheck-hook' rule, to the empty string

Re: [Automake-NG] [FYI] [ng] dejagnu: include verbatim

Le 10 août 2012 à 13:10, Stefano Lattarini a écrit : > -## --- ## > -## Building site.exp. ## > -## --- ## > - > -## Note that in the rule we don't directly generate site.exp to avoid > -## the possibility of a corrupted site.exp if make is interrupted. > -## Jim

Re: [Automake-NG] [FYI] [ng] refactor: drop use of ?HOST?, ?BUILD? and ?TARGET? transforms

Le 10 août 2012 à 11:58, Stefano Lattarini a écrit : > Instead, use the new make variables $(am.conf.host-triplet), > $(am.conf.build-triplet) and $(am.conf.host-triplet) are defined. I guess the last one is 'target', and I parse error on this sentence. > * automake.in (preprocess_file): Drop '

Re: [Automake-NG] [FYI] [ng] tags: move yet more processing to GNU make

Le 10 août 2012 à 00:05, Stefano Lattarini a écrit : > This will also simplify future changes aimed at making the current > 'tags.am' makefile fragment included more "verbatim". > > * lib/am/tags.am (am.tags.include-option): New internal variable, the > include-like option used by the $(EATGS) p

Re: [Automake-NG] [PATCH 2/5] [ng] refactor: remove all uses of $(am__nobase_strip{, _setup})

Le 9 août 2012 à 12:56, Stefano Lattarini a écrit : > Prefer using GNU make built-ins $(patsubst) and $(notdir) instead. > This entails a partial or complete rewrites of several install and/or > uninstall rules. > > This change doesn't offer any serious simplification, being just a > step in the

Re: [Automake-NG] [FYI] [ng] refactor: remove all uses of '$(am__strip_dir)'

Hi! Le 8 août 2012 à 12:40, Stefano Lattarini a écrit : > On 08/08/2012 12:36 PM, Stefano Lattarini wrote: >> Prefer using GNU make built-in '$(notdir)' instead. This change doesn't >> offer any serious simplification, but is just a step in the general >> direction of moving more non-trivial pro

Re: [Automake-NG] [FYI] general: include verbatim makefile fragments in output Makefiles

Le 2 août 2012 à 19:35, Stefano Lattarini a écrit : > Here it is (see attachment). OK? It is. Yet I realize that this goes against what we talked about a while ago. I personally structure my build-aux directories. I have build-aux/m4, build-aux/make (which contains *.mk that I include in va

Re: [Automake-NG] [FYI] general: include verbatim makefile fragments in output Makefiles

Le 2 août 2012 à 16:18, Stefano Lattarini a écrit : >> Why hide it? It will show anyway. IMHO, this should be >> build-aux/automake/ or build-aux/mk. >> > I guess you use "build-aux" to mean "the directory specified by > AC_CONFIG_AUX_DIR", right? Yes :) > If yes, I agree this would offer mo

Re: [Automake-NG] [FYI] [ng] automake: remove support for threaded execution

Le 30 juil. 2012 à 21:27, Stefano Lattarini a écrit : > Why are we removing such a nice feature, especially in a world where > multicore systems are becoming the norm? Sorry I took so long to answer. IMHO there are two major features provided by multiple makefiles. One is that you can "scope"

Re: [Automake-NG] [FYI] general: include verbatim makefile fragments in output Makefiles

Le 2 août 2012 à 11:47, Stefano Lattarini a écrit : > Instead of copying their contents in each output Makefile. > > * automake.in (verbatim): Instead of copying the given Makefile fragment > in the output makefile, copy it in the '.mk' subdirectory of the top-level > source directory (i.e., the

Re: [Automake-NG] [FYI] cosmetics: remove an obsolete comment, fix a typo in another one

Le 1 août 2012 à 12:49, Stefano Lattarini a écrit : > +# Test to make sure subdirs in EXTRA_DIST work. > +# Also tests to make sure *srcdir is properly handled. > +# Also test the situation when the list of distributed files contains where? Or "Also test when the list…" > # a directory and a f

Re: [Automake-NG] [FYI] [ng] cleanup: unused function '&almost_verbatim' removed

Le 31 juil. 2012 à 18:24, Stefano Lattarini a écrit : > Superseded in all its current uses by '&verbatim'. > > * automake.in (almost_verbatim): Delete. This closes a whole set of refactoring changes. Congrats!

Re: [Automake-NG] [FYI] [ng] rename: verbatim makefile fragments: *.am -> *.mk

Le 31 juil. 2012 à 18:05, Stefano Lattarini a écrit : > This make it clear and manifest the fact that they are not parsed makes and I'm not sure if manifest is a name or a verb here :) Adjust according to your grammar. Or maybe "Make clear and manifest that they…"? > like the other '*.am' file

Re: [Automake-NG] [FYI] [ng] refactor: check-typos.am: include verbatim

Le 31 juil. 2012 à 17:48, Stefano Lattarini a écrit : > It contains not more automake-time %TRANSFORM%, so that's the right It no longer contains? > thing to do. I would s/,.*/./.

Re: [Automake-NG] [FYI] [ng] rename: am.rm-f -> am.hack.rm-f, am.rm-rf -> am.hack.rm-rf

Le 31 juil. 2012 à 19:58, Stefano Lattarini a écrit : > Signed-off-by: Stefano Lattarini > --- > > So that they won't be mistaken for some kind of "general utility functions". > > lib/am/header-vars.mk | 8 > 1 file changed, 4 insertions(+), 4 deletions(-) > > diff --git a/lib/am/head

Re: [Automake-NG] [FYI] [ng] rename: am__ensure_target_dir_exists -> am.cmd.ensure-target-dir-exists

Le 31 juil. 2012 à 19:30, Stefano Lattarini a écrit : > Signed-off-by: Stefano Lattarini > --- > lib/am/common-targets.mk | 2 +- > lib/am/depend2.am| 4 ++-- > lib/am/header-vars.mk| 2 +- > lib/am/lex.am| 2 +- > lib/am/library.am| 2 +- > lib/am/ltlibrary.am |

Re: [Automake-NG] [FYI] [ng] rename: am__ensure_dir_exists -> am.cmd.ensure-dir-exists

Le 31 juil. 2012 à 19:29, Stefano Lattarini a écrit : > Signed-off-by: Stefano Lattarini > --- > lib/am/depend2.am | 2 +- > lib/am/header-vars.mk | 4 ++-- > t/ensure-dir-exists.sh | 10 +- > 3 files changed, 8 insertions(+), 8 deletions(-) I'm fine with this commit message, very

Re: [Automake-NG] [FYI] [ng] header vars: remove function $(am__mkdir)

Le 31 juil. 2012 à 19:09, Stefano Lattarini a écrit : > * lib/am/header-vars.am (am__mkdir): Delete. > (am__ensure_dir_exists): Simply use $(MKDIR) directly. Great!

automake-ng@gnu.org

Le 31 juil. 2012 à 00:56, Stefano Lattarini a écrit : > * automake.in (handle_clean): Here, the argument '$makefile'. And remove > the now-redundant description of the function. > (generate_makefile): Adjust. Ok.

Re: [Automake-NG] [FYI] [ng] cleanup: remove two almost-unused global vars: {am, in}_file_name

Le 30 juil. 2012 à 20:52, Stefano Lattarini a écrit : > diff --git a/automake.in b/automake.in > index f8e5c9b..b746b8a 100644 > --- a/automake.in > +++ b/automake.in > @@ -497,11 +497,8 @@ my $am_file = 'BUG'; > ## -- ## > > # Basename and relative dir of

automake-ng@gnu.org

Le 31 juil. 2012 à 00:56, Stefano Lattarini a écrit : > * automake.in (handle_clean): Here. > * lib/am/clean.am: Related changes ans simplifications. and OK.

Re: [Automake-NG] [PATCH 2/4] [ng] clean: prefer internal variables over automake-time %TRANSFORMS%

Le 31 juil. 2012 à 00:56, Stefano Lattarini a écrit : > This will be especially useful in the next change. > > * automake.in (handle_clean): Define several new private make variables > 'am.clean.*.f.auto' and 'am.clean.*.d.auto'. Drop the %TRANSFORMS% when > processing ... > (lib/am/clean.am):

Re: [Automake-NG] [PATCH 1/4] [ng] rename: am__*clean_* -> am.clean.*.*

Le 31 juil. 2012 à 00:56, Stefano Lattarini a écrit : > More precisely, do these renames: > > am__mostlyclean_files -> am.clean.mostly.f > am__clean_files-> am.clean.normal.f > am__distclean_files-> am.clean.dist.f > am__maintclean_files -> am.clean.maint.f > > am__mostl

Re: [Automake-NG] [FYI] [ng] clean: simplify code for removal of '.am' dir

Le 30 juil. 2012 à 22:10, Stefano Lattarini a écrit : > * lib/am/am-dir.am (am--distclean-amdir): Delete this phony rule and > the 'distclean-am' dependency on it ... > (am__distclean_dirs): ... simply append $(am.dir) to this instead. OK.

Re: [Automake-NG] [FYI] [ng] cleanup: remove on almost-unused global vars: am_relative_dir

Le 30 juil. 2012 à 21:04, Stefano Lattarini a écrit : > * automake.in ($am_relative_dir): Delete, it was only used once ... > (generate_makefile): ... in here, so it's simpler to inline its > expansion. > (initialize_per_input): Don't reset the deleted variable. ok.

Re: [Automake-NG] [FYI] [ng] cleanup: remove on almost-unused global vars: topsrcdir

Le 30 juil. 2012 à 20:54, Stefano Lattarini a écrit : > * automake.in ($topsrcdir): Delete, it was only used once ... > (handle_LIBOBJS_or_ALLOCA): ... in here, so it's simpler to inline > its expansion. Improve formatting of immediately surrounding code > a little while we are at it. > (initial

Re: [Automake-NG] [FYI] [ng] general: new internal vars to hold path of current Makefile{, .in, .am}

Le 30 juil. 2012 à 20:20, Stefano Lattarini a écrit : > This is preferable to having to use the '%MAKEFILE%', '%MAKEFILE-IN%' > and '%MAKEFILE-AM%' transforms in several places, and will enable us > to do more sweeping refactorings in the future. > > * automake.in (generate_makefile): Define new

Re: [Automake-NG] [FYI] [ng] clean: prefer '#' comments over '##' ones

Le 30 juil. 2012 à 19:20, Stefano Lattarini a écrit : > * lib/am/clean.am: Here. So that they will be visible also in > the generated Makefiles. ok.

Re: [Automake-NG] [FYI 1/2] [ng] refactor: make '&verbatim' polymorphic in its return value

Le 30 juil. 2012 à 19:03, Stefano Lattarini a écrit : > * automake.in (verbatim): If used in empty context, continue to act > as before, and append the read makefile fragment to '$output_verbatim'. > But if used in scalar or list context, return it instead. A similar comment in the code might be

Re: [Automake-NG] [FYI 2/2] [ng] automake: can copy makefile fragments really verbatim

Le 30 juil. 2012 à 18:45, Stefano Lattarini a écrit : > This will allow us to actually put such fragments in distributed > files, and then include them *at make runtime*, instead of copying > their contents in every single makefile (as is done now, a legacy > from mainline Automake). > > * autom

Re: [Automake-NG] [FYI] [ng] refactor: handle 'all', 'check' and 'install' target in on '.am' file

Le 30 juil. 2012 à 17:56, Stefano Lattarini a écrit : > * lib/am/common-targets.am: New file, superseding and encompassing ... > * lib/am/check-target.am, all-target.am, lib/am/install.am: ... these > ones. > * Makefile.am (dist_am_DATA): Adjust. > * automake.in (handle_all_and_check): Remove. >

Re: [Automake-NG] [FYI] [ng] check: deps for 'all' target listed in '$(am.all.targets)'

Le 30 juil. 2012 à 17:34, Stefano Lattarini a écrit : > On 07/30/2012 05:31 PM, Akim Demaille wrote: >> >> Le 30 juil. 2012 à 16:28, Stefano Lattarini a écrit : >> >>> * automake.in (handle_all_and_check): Drop transform 'ALL-DEPS' when

Re: [Automake-NG] [FYI] [ng] check: avoid extra recursive make invocations

Le 30 juil. 2012 à 17:34, Stefano Lattarini a écrit : > Ah, but I'm not sure which these corner cases might actually be. I just > suspect they might exist. If somebody will ever stumble in one of them > and report it to us, then we will document it in NEWS (maybe with a > suggested workaround).

Re: [Automake-NG] [FYI] [ng] check: deps for 'all' target listed in '$(am.all.targets)'

Le 30 juil. 2012 à 16:28, Stefano Lattarini a écrit : > * automake.in (handle_all_and_check): Drop transform 'ALL-DEPS' when > processing the 'all-target.am' file. > (generate_makefile): Define '$(am.all.targets)' to the list of > dependencies for the 'all' target. > * lib/am/all-target.am: Use i

Re: [Automake-NG] [FYI] [ng] check: avoid extra recursive make invocations

Le 30 juil. 2012 à 14:04, Stefano Lattarini a écrit : > After this change, "make check" will not cause any recursive make > invocation by default (unless there are $(SUBDIRS) to descend into, of > course). One little price to pay is that the targets 'check-TESTS', > 'check-DEJAGNU' and 'check-lo

Re: [Automake-NG] [FYI] [ng] refactor: move hack for libtool installs from automake to install.am

Le 30 juil. 2012 à 17:06, Stefano Lattarini a écrit : > * automake.in (generate_makefile): Move the hack necessary to make the > installation of libtool libraries and binaries dependent on them work > "on degenerate systems even with make -j" (according to the comments of > the original authors :

Re: [Automake-NG] [FYI] [ng] check: deps for 'check' target listed in '$(am.test-suite.deps)'

Le 30 juil. 2012 à 17:06, Stefano Lattarini a écrit : > On 07/30/2012 04:57 PM, Akim Demaille wrote: >> >> Le 30 juil. 2012 à 14:02, Stefano Lattarini a écrit : >> >>> (generate_makefile): Define '$(am.test-suite.deps)' to the list of >>> depe

Re: [Automake-NG] [FYI] [ng] check: deps for 'check' target listed in '$(am.test-suite.deps)'

Le 30 juil. 2012 à 14:02, Stefano Lattarini a écrit : > (generate_makefile): Define '$(am.test-suite.deps)' to the list of > dependencies for the 'check' target. Define to? or as? > diff --git a/lib/am/check-target.am b/lib/am/check-target.am > index 225c566..852941b 100644 > --- a/lib/am/check

automake-ng@gnu.org

Le 30 juil. 2012 à 16:29, Stefano Lattarini a écrit : > * automake.in (handle_tests_dejagnu): This is just a trivial > one-liner no, so merge it … now > (handle_tests): ... in here. And remove a redundant heading > comment. Ack.

Re: [Automake-NG] [PATCH 5/5] [ng] cleanup: drop redundant use of a transform

Le 30 juil. 2012 à 15:28, Stefano Lattarini a écrit : > * lib/am/all-target.am: Use '$(am.config-hdr.local)' directly, > instead of the '%LOCAL-HEADERS%' transform. > * automake.in (handle_all_and_check): Drop that transform. OK.

Re: [Automake-NG] [PATCH 4/5] [ng] automake: remove useless code duplication

Le 30 juil. 2012 à 15:28, Stefano Lattarini a écrit : > This is a pure refactoring with no semantic changes intended. > > * automake.in (handle_all_and_check, handle_tags): Do not re-calculate > the list of "local" input or output header files (specified by > AC_CONFIG_HEADERS in configure.ac);

Re: [Automake-NG] [PATCH 3/5] [ng] vars: list input config headers in an internal make variable as well

Le 30 juil. 2012 à 15:28, Stefano Lattarini a écrit : > This is only a preparatory change in view of future refactorings. > > * automake.in (handle_config_headers): Define a new make variable > 'am.config-hdr.local.in'. > * t/confh-internals.sh: New test. OK.

Re: [Automake-NG] [PATCH 2/5] [ng] vars: separate config headers in two vars (local and non-local ones)

Le 30 juil. 2012 à 15:28, Stefano Lattarini a écrit : > Where the "local" are those which are in the same subdirectory of the > Makefile.am being currently processed, and the "non-local" ones are > not (and thus need to be prepended with $(top_builddir)). > > * automake.in (handle_config_headers

Re: [Automake-NG] [PATCH 1/5] [ng] automake: move processing of config-header rules in its own function

Le 30 juil. 2012 à 15:28, Stefano Lattarini a écrit : > This is a pure refactoring with no semantic changes intended. It will > only be useful for future changes. > > * automake.in (handle_configure): Move definition of $(AM_CONF_HEADERS) > from here ... > (handle_config_headers): ... to this n

Re: [Automake-NG] [FYI] [ng] parallel-tests: define $(TEST_SUITE_LOG) in makefile fragment

Le 30 juil. 2012 à 10:23, Stefano Lattarini a écrit : > diff --git a/lib/am/parallel-tests.am b/lib/am/parallel-tests.am > index 3ad7de6..116d178 100644 > --- a/lib/am/parallel-tests.am > +++ b/lib/am/parallel-tests.am > @@ -309,6 +309,10 @@ define am.setup-test-harness-workdir > && sed '

Re: [Automake-NG] [PATCH] [ng] refactor: consider $(EXEEXT) used if non-empty at make runtime

Le 30 juil. 2012 à 10:02, Stefano Lattarini a écrit : > To check whether we should handle $(EXEEXT), simply check for > "ifdef EXEEXT" in our Makefile fragments, rather than also checking > whether EXEEXT has been set as a configure-time AC_SUBST. This > simplifies the logic, and move it complet

Re: [Automake-NG] [FYI] [ng] vars: names of iterating vars in $(foreach) loops shorter & simpler

Le 24 juil. 2012 à 18:48, Stefano Lattarini a écrit : >>> The first two arguments, VAR and LIST, are expanded before anything >>> else is done; note that the last argument, TEXT, is *not* expanded at >>> the same time. >> >> So, indeed, in the following example, every use of the >> outer $(v) is

Re: [Automake-NG] [FYI] [ng] check: use few more GNU make capabilities

Le 27 juil. 2012 à 17:19, Stefano Lattarini a écrit : > We can do so now that the 'parallel-tests.am' file is now > processed "almost verbatim" by Automake. > > * lib/am/parallel-tests.am (check-TESTS): Here, use "ifeq" to check > whether $(AM_LAZY_CHECK) is "yes". This removes the need for ...

Re: [Automake-NG] [FYI] [ng] check: refactor 'am.test-suite.handle-suffix'

Le 28 juil. 2012 à 02:22, Stefano Lattarini a écrit : > -define am.test-suite.handle-suffix > - > -ifeq ($$(call am.vars.is-undef,$$(call am__tpfx,$1)LOG_DRIVER),yes) > -$$(call am__tpfx,$1)LOG_DRIVER = $(SHELL) $(am.conf.aux-dir)/test-driver > +define am.test-suite.handle-suffix.helper > +ifeq (

Re: [Automake-NG] [FYI] [ng] refactor: use new function 'am.vars.is-undef' ...

Le 28 juil. 2012 à 01:29, Stefano Lattarini a écrit : > * lib/am/header-vars.am (SUBDIRS): ... in the definition of this (instead > of using hand-rolled almost-equivalent) ... > ($(1)LOG_DRIVER, TEST_EXTENSIONS): ... and of these (instead of resorting > to weaker and unsafer '?=' assignment). Ac

Re: [Automake-NG] [FYI] [ng] tests: do not run very expensive tests by default

Le 28 juil. 2012 à 13:50, Stefano Lattarini a écrit : > Some tests in the Automake testsuite (especially those testing for bugs > w.r.t. command-line length limits) require a *long* time to run. For > example, the test 'parallel-tests-many.sh' (the most extreme example of > this issue) takes ~ 1

Re: [Automake-NG] [PATCH] [ng] serial-tests: remove one level of indentation in a recipe

Le 28 juil. 2012 à 13:10, Stefano Lattarini a écrit : > test -z "$$skipped" || echo "$${col}$$skipped$${std}"; \ > test -z "$$report" || echo "$${col}$$report$${std}"; \ > echo "$${col}$$dashes$${std}"; \ > - test "$$failed" -eq 0; \ > - else :; fi > + test "$$fail

Re: [Automake-NG] [FYI] [ng] rename: am__dir -> am.dir, am__abs_dir -> am.abs-dir, etc.

Le 28 juil. 2012 à 10:55, Stefano Lattarini a écrit : > Complete list of renames: > > am__dir -> am.dir > am__abs_dir -> am.abs-dir > am__top_dir -> am.top-dir > am__abs_top_dir -> am.abs-top-dir Fine. FWIW, since "abs" qualifies am.dir, I would rather have use "am.

Re: [Automake-NG] [FYI] [ng] refactor: read 'parallel-tests.am' more verbatim

Le 27 juil. 2012 à 15:43, Stefano Lattarini a écrit : > * automake.in (handle_tests): Process 'parallel-tests.am' with > '&almost_verbatim', not with '&file_contents'. As a related > change, we now *must* include 'color-tests.am' explicitly. > * lib/am/parallel-tests.am: Related adjustments. In

Re: [Automake-NG] [FYI 00/19] [ng] Several renaming patches

Le 27 juil. 2012 à 15:10, Stefano Lattarini a écrit : > And few minor related ones. Already pushed as obvious (the testsuite > still passes on Fedora and Solaris 11). (Ack){19}.

Re: [Automake-NG] [FYI] [ng] header vars: fix comment on why DESTDIR is not explicitly initialized

Le 27 juil. 2012 à 10:39, Stefano Lattarini a écrit : > * lib/am/header-vars.am (DESTDIR): It's not because we want to allow I guess s/not//. > it to be defined from the environment, for compatibility with mainline > Automake. For better clarity, define an (empty) default with the line > "DEST

Re: [Automake-NG] [FYI] [ng] vars: names of iterating vars in $(foreach) loops shorter & simpler

Le 24 juil. 2012 à 16:03, Stefano Lattarini a écrit : >> I might be stretching things too far, but imagine that some >> one defines v in her Makefile.am, at the top level, because >> for instance she can factor longish definitions in EXTRA_DIST. >> >> v = headers/include/pkg >> EXTRA_DIST = $(v)

Re: [Automake-NG] [FYI] [ng] rename: am__newline -> am.chars.newline

Le 24 juil. 2012 à 16:03, Stefano Lattarini a écrit : > On 07/24/2012 03:54 PM, Akim Demaille wrote: >> >> Le 24 juil. 2012 à 15:40, Stefano Lattarini a écrit : >> >>> diff --git a/lib/am/header-vars.am b/lib/am/header-vars.am >> >> Maybe these vars s

Re: [Automake-NG] [FYI] [ng] vars: names of iterating vars in $(foreach) loops shorter & simpler

Le 24 juil. 2012 à 15:45, Stefano Lattarini a écrit : > In GNU make, after a "$(foreach x,...)" loop, the make variable $(x): > > - is restored to its original variable it had before the 'foreach' >loop, if it was already set; > > - re-unset otherwise. > > This behaviour is checked by th

Re: [Automake-NG] [FYI] [ng] rename: am__newline -> am.chars.newline

Le 24 juil. 2012 à 15:40, Stefano Lattarini a écrit : > diff --git a/lib/am/header-vars.am b/lib/am/header-vars.am Maybe these vars should be defined elsewhere. > index 6efad03..bf2ad73 100644 > --- a/lib/am/header-vars.am > +++ b/lib/am/header-vars.am > @@ -52,7 +52,7 @@ am.chars.squote := ' >

Re: [Automake-NG] [PATCH 3/4] [ng] rename: am__configure_deps -> am.remake.configure-deps

All these are fine. Le 22 juil. 2012 à 19:08, Stefano Lattarini a écrit : > - define_variable ('am__configure_deps', INTERNAL, @configuredeps); > + define_variable ('am.remake.configure-deps', INTERNAL, @configuredeps); Wouldn't some "am.deps.*" make sense?

Re: [Automake-NG] [FYI] [ng] tests: fix test driver botch-up on Solaris

Le 24 juil. 2012 à 10:13, Stefano Lattarini a écrit : > This issue is very similar to the one fixed by commit v1.12.2-31-g587e0c6. > > The test 't/memoize.sh' was producing a '.log' file with few overly-long > lines (more than 12k characters long) and, when Solaris XPG4 awk was in > use, that wa

Re: [Automake-NG] [FYI] [ng] tests: relax even more a grepping check on configure output

Le 24 juil. 2012 à 10:12, Stefano Lattarini a écrit : > This is a follow-up to today's commit v1.12.2-594-geee3aff. > > * t/subpkg.sh: Here: do bee too picky about the verb declension do bee? don't be I guess, but there might be a reference to Harry Potter that I don't grasp :) > used in a "c

Re: [Automake-NG] [PATCH 6/6] [ng] recheck: don't exceed command line limits, even with many failed tests

Le 21 juil. 2012 à 10:50, Stefano Lattarini a écrit : > +## Re-run the relevant tests, without hitting command-line length limits. > + echo am__test_bases="$$bases" | \ > + $(MAKE) -f- -f$(firstword $(MAKEFILE_LIST)) \ > + $(TEST_SUITE_LOG) .am/doing-recheck=yes Nice and

Re: [Automake-NG] [PATCH 5/6] [ng] coverage: recheck with many failed tests

Le 21 juil. 2012 à 10:50, Stefano Lattarini a écrit : > * t/parallel-tests-many.sh: Extend to check that the 'recheck' target > works even when a huge number of tests (~ 30 thousands) have failed 30k? > in the previous testsuite run. Currently this doesn't work, and cause causes > the test t

Re: [Automake-NG] [PATCH 4/6] [ng] check: refactor for less duplication and better performances

Le 21 juil. 2012 à 10:50, Stefano Lattarini a écrit : > * lib/am/parallel-tests.am (am__count_test_results): Adjust this awk > program to emit a shell snippet to be executed by the calling recipe ... > ($(TEST_SUITE_LOG)): ... here. This avoid the need to call the program > in $(am__count_test_r

Re: [Automake-NG] [PATCH 3/6] [ng] check: use awk rather than grep+xargs to count test results

Le 21 juil. 2012 à 10:50, Stefano Lattarini a écrit : > * lib/am/parallel-tests.am ($(TEST_SUITE_LOG)): Here, with the help > of ... > (am__count_test_results): ... this new internal variable. Ack.

Re: [Automake-NG] [FYI] [ng] tests: enhance spy tests on patter rules

Le 22 juil. 2012 à 09:50, Stefano Lattarini a écrit : > On 07/22/2012 09:18 AM, Akim Demaille wrote: >> >> Le 22 juil. 2012 à 08:57, Stefano Lattarini a écrit : >> >>> +# Two check are done here: >> >> checks >> > Fixed. > >>>

Re: [Automake-NG] [PATCH 1/6] [ng] coverage: testing with lots of test scripts

Le 22 juil. 2012 à 09:58, Stefano Lattarini a écrit : >> I personally use it extensively in my test suites, as I find this >> much more legible: >> >> ! list_logs | grep . >> > Ah, but this doesn't do what you expect: > > $ bash -c '! echo x | grep .'; echo st = $? > x > st = 0 > $ bash -c

Re: [Automake-NG] [PATCH 2/6] [ng] parallel-tests: do not exceed command line length limits

Le 21 juil. 2012 à 10:50, Stefano Lattarini a écrit : > Fixes automake bug#7868. > > Two things worth noting: > > + a "make recheck" issued after a huge number of tests have failed > can still hit command-line length issues; > > + the check-recipes now contain (first among the Automake-gener

Re: [Automake-NG] [PATCH 1/6] [ng] coverage: testing with lots of test scripts

Le 21 juil. 2012 à 10:50, Stefano Lattarini a écrit : > See long-standing automake bug#7868. > > * t/parallel-tests-many.sh: Simplify and enhance. Among the other things, I would s/the //. > this test now tries running ~ 30 thousands tests. Currently fails on I would also 30,000, I don't li

Re: [Automake-NG] [FYI] [ng] tests: enhance spy tests on patter rules

Le 22 juil. 2012 à 08:57, Stefano Lattarini a écrit : > +# Two check are done here: checks > +# - if we have two pattern rules from which the same file (or set of s/we have // (th shrtr, th btr :). > +# files) can be built, both are applicable, and both results in the > +# same target st

Re: [Automake-NG] [PATCH 0/2] Two simple refactorings

Le 21 juil. 2012 à 16:14, Stefano Lattarini a écrit : > Will push these shortly. > > Stefano Lattarini (2): > automake: new function &almost_verbatim (small refactoring) > automake: inline '&handle_install' Sure.

Re: [Automake-NG] [PATCH 00/33] Several renamings of private make variables

Le 21 juil. 2012 à 14:55, Stefano Lattarini a écrit : > The first bach of "am__foo -> am.foo" renamings. See also this recent > patch that announced the upcoming renamings: > > > Stefano Lattarini (33): > [ng] rename: am__di

Re: [Automake-NG] [PATCH] vars: names for internal make variables: 'am.foo' and 'am.foo.bar-baz'

Le 18 juil. 2012 à 20:35, Stefano Lattarini a écrit : > That is the new preferred naming scheme: 'am.foo' where we would > have previously used something like 'am__foo', and 'am.foo.bar-baz' > where we would have previously used something like 'am__foo_bar_baz' > or 'am__foo__bar_baz'. Sane idea

Re: [Automake-NG] [PATCH 3/3] [ng] dist: do not exceed command line length limits, even with many files

Le 14 juil. 2012 à 23:49, Stefano Lattarini a écrit : >> That is, it's hard-coded into the POSIX shell grammar that a >> simple_command can start with redirections in the cmd_prefix, but a >> compound_command can ONLY have redirections after the end of the >> compound command, if you are being po

Re: [Automake-NG] [PATCH 3/3] [ng] dist: do not exceed command line length limits, even with many files

Le 13 juil. 2012 à 12:19, Stefano Lattarini a écrit : > + @while read file; do \ > ## Always look for the file or directory to distribute in the build > ## directory first, in VPATH spirit. > if test -f $$file || test -d $$file; then d=.; else d=$(srcdir); fi; \ > @@ -191,7 +200,7 @@

Re: [Automake-NG] [PATCH 1/3] [ng] dist: memoize some internal variables

Le 13 juil. 2012 à 12:19, Stefano Lattarini a écrit : > * lib/am/distdir.am: Here, so that we will be able to use them several > times with no performance impact. Fine.

Re: [Automake-NG] [PATCH 2/3] [ng] coverage: distributing lots of files

Le 13 juil. 2012 à 12:19, Stefano Lattarini a écrit : > * t/dist-many.sh: New test, try distributing ~ 30 thousands files. > Currently fails on several systems (e.g., Linux 2.6.30 on i686, > Solaris 10 on i86pc). > * t/dist-many2.sh: New test, check that our distribution rules do not > hit errors

  1   2   >