Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am
This change (as you've likely seen) is quite tricky.
Linking __ANYTHING__ in rpmio/*.c with ${RPM_LDADD_COMMON}
instead of ${RPMIO_LDADD_COMMON} will have build failures.
If you're not seeing the failures, you are not looking hard enough.
The solution is (or will be) for tools that
need more than -lrpmio is to move the sources to tools/*.
However, its not at all clear whether there is desire or need for these
executables, that is a very different issue than fixing
compiler warnings and such.
73 de Jeff
On Dec 25, 2008, at 1:32 PM, Per Øyvind Karlsen wrote:
RPM Package Manager, CVS Repository
http://rpm5.org/cvs/
Server: rpm5.org Name: Per Øyvind Karlsen
Root: /v/rpm/cvs Email: pkarl...@rpm5.org
Module: rpm Date: 25-Dec-2008 19:32:01
Branch: HEAD Handle: 2008122518320100
Modified files:
rpm CHANGES
rpm/rpmio Makefile.am
Log:
tpw: fix linking against librpm.
Summary:
RevisionChanges Path
1.2711 +1 -0 rpm/CHANGES
1.200 +1 -1 rpm/rpmio/Makefile.am
patch -p0 '@@ .'
Index: rpm/CHANGES
=
=
=
=
=
=
==
$ cvs diff -u -r1.2710 -r1.2711 CHANGES
--- rpm/CHANGES25 Dec 2008 18:18:42 - 1.2710
+++ rpm/CHANGES25 Dec 2008 18:32:01 - 1.2711
@@ -1,5 +1,6 @@
5.2a2 - 5.2a3:
+- proyvind: tpw: fix linking against librpm.
- proyvind: rpmtar: fix build breakage.
- proyvind: rpmcpio: fix build breakage.
- proyvind: tmacro: fix build breakage.
@@ .
patch -p0 '@@ .'
Index: rpm/rpmio/Makefile.am
=
=
=
=
=
=
==
$ cvs diff -u -r1.199 -r1.200 Makefile.am
--- rpm/rpmio/Makefile.am 15 Dec 2008 23:33:58 - 1.199
+++ rpm/rpmio/Makefile.am 25 Dec 2008 18:32:01 - 1.200
@@ -311,7 +311,7 @@
tput_LDADD = $(RPMIO_LDADD_COMMON)
tpw_SOURCES = tpw.c
-tpw_LDFLAGS = $(RPMIO_LDADD_COMMON)
+tpw_LDFLAGS = $(RPM_LDADD_COMMON) $(RPMIO_LDADD_COMMON)
trpmio_SOURCES = trpmio.c
trpmio_LDADD = $(RPMIO_LDADD_COMMON)
@@ .
__
RPM Package Managerhttp://rpm5.org
CVS Sources Repositoryrpm-...@rpm5.org
smime.p7s
Description: S/MIME cryptographic signature
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am
2008/12/5 Jeff Johnson [EMAIL PROTECTED] What needed fixing? I have built using internal lua daily without these changes. Can you try to pin down what changed so that this patch is now needed please? Likely trying to build rpm-5.1.6, without these changes, would be the easiest way to identify if there is a regression. 73 de Jeff I noticed something being odd about this myself, I've built internal lua with cvs checkout from a couple of weeks ago that didn't have this issue, now it showed up, both me and afb has experienced it, but even while I tried, I didn't manage to figure it out.. I'll try again later.. :)
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am
2008/12/5 Jeff Johnson [EMAIL PROTECTED] The flaw will be in -lrpmmisc, not in -lrpmio, is my guess. And its likely my portability brain fart somehow if reproducible. Doing cd misc cvs diff -D 2 months ago will show all my changes. There's not a whole lot different in misc/* afaict. I have seen some issues trying to sort out the SELinux textrel issue by attempting ld -pie ... linkage. But I abandoned those efforts, none of those changes are on HEAD. Note that its *very* easy to get tricked by -lrpmisc because some of the make dependencies are not included. Adding those dependencies through AutoFu would be fairly difficult, what is there is OK for most usage cases even if a developer can get occaisionally tricked. You're of course right, as usual! ;) I tend to always forget about misc/librpmmisc.vers when I haven't touched it for a while, then stabbing wildy around in the dark taking the usual (but wrong) route to investigate it. Fixed now! :)
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am adler32.c digest.c gzdi...
I have mixed feelings about this patch ...
Building rpm --without-zlib (and --without-db and --without-sqlite
and ...) leads
only to a useless /usr/bin/rpm imho.
But its easier to do the mechnical compile engineering than it is to
try and describe why building rpm without -lz and -ldb is unlikely to
be useful.
73 de Jeff
On Sep 28, 2008, at 5:50 PM, Jeff Johnson wrote:
RPM Package Manager, CVS Repository
http://rpm5.org/cvs/
__
__
Server: rpm5.org Name: Jeff Johnson
Root: /v/rpm/cvs Email: [EMAIL PROTECTED]
Module: rpm Date: 28-Sep-2008
23:50:28
Branch: HEAD Handle: 2008092821502700
Added files:
rpm/rpmio adler32.c
Modified files:
rpm CHANGES
rpm/rpmio Makefile.am digest.c gzdio.c rpmio.c
Log:
- jbj: rpmio: permit adler32/crc32/crc64 in spite of --without-
zlib.
- jbj: rpmio: eliminate HAVE_ZLIB_H tests, use WITH_ZLIB tests
instead.
Summary:
RevisionChanges Path
1.2578 +2 -0 rpm/CHANGES
1.194 +1 -1 rpm/rpmio/Makefile.am
2.1 +140 -0 rpm/rpmio/adler32.c
2.41+113 -25rpm/rpmio/digest.c
2.13+2 -2 rpm/rpmio/gzdio.c
1.148 +7 -7 rpm/rpmio/rpmio.c
__
__
patch -p0 '@@ .'
Index: rpm/CHANGES
==
==
$ cvs diff -u -r1.2577 -r1.2578 CHANGES
--- rpm/CHANGES 27 Sep 2008 13:00:13 - 1.2577
+++ rpm/CHANGES 28 Sep 2008 21:50:27 - 1.2578
@@ -1,5 +1,7 @@
5.1.0 - 5.2a0:
+- jbj: rpmio: permit adler32/crc32/crc64 in spite of --
without-zlib.
+- jbj: rpmio: eliminate HAVE_ZLIB_H tests, use WITH_ZLIB
tests instead.
- rse: rpmio: allow newer fnmatch to build on non-Linux
platforms, too.
- jbj: rpmio: upgrade fnmatch to version that implements
FNM_EXTMATCH.
- jbj: rpmpsm: simplify runTriggers() by adding
rpmTriggersLoop().
@@ .
patch -p0 '@@ .'
Index: rpm/rpmio/Makefile.am
==
==
$ cvs diff -u -r1.193 -r1.194 Makefile.am
--- rpm/rpmio/Makefile.am 25 Sep 2008 19:39:28 - 1.193
+++ rpm/rpmio/Makefile.am 28 Sep 2008 21:50:27 - 1.194
@@ -7,7 +7,7 @@
SUBDIRS = # tests
EXTRA_DIST = \
- fnmatch_loop.c getdate.y rpmcpio.c rpmcpio.h \
+ adler32.c fnmatch_loop.c getdate.y rpmcpio.c rpmcpio.h \
rpmgenbasedir.c rpmgenpkglist.c rpmgensrclist.c \
rpmtar.c rpmtar.h \
tdir.c tfts.c tget.c tglob.c thkp.c thtml.c tinv.c tkey.c
tmire.c \
@@ .
patch -p0 '@@ .'
Index: rpm/rpmio/adler32.c
==
==
$ cvs diff -u -r0 -r2.1 adler32.c
--- /dev/null 2008-09-28 23:45:21 +0200
+++ adler32.c 2008-09-28 23:50:28 +0200
@@ -0,0 +1,140 @@
+/* adler32.c -- compute the Adler-32 checksum of a data stream
+ * Copyright (C) 1995-2004 Mark Adler
+ * For conditions of distribution and use, see copyright notice
in zlib.h
+ */
+
+#define BASE 65521UL/* largest prime smaller than 65536 */
+#define NMAX 5552
+/* NMAX is the largest n such that 255n(n+1)/2 + (n+1)(BASE-1)
= 2^32-1 */
+
+#define DO1(buf,i) {adler += (buf)[i]; sum2 += adler;}
+#define DO2(buf,i) DO1(buf,i); DO1(buf,i+1);
+#define DO4(buf,i) DO2(buf,i); DO2(buf,i+2);
+#define DO8(buf,i) DO4(buf,i); DO4(buf,i+4);
+#define DO16(buf) DO8(buf,0); DO8(buf,8);
+
+/* use NO_DIVIDE if your processor does not do division in
hardware */
+#ifdef NO_DIVIDE
+# define MOD(a) \
+do { \
+if (a = (BASE 16)) a -= (BASE 16); \
+if (a = (BASE 15)) a -= (BASE 15); \
+if (a = (BASE 14)) a -= (BASE 14); \
+if (a = (BASE 13)) a -= (BASE 13); \
+if (a = (BASE 12)) a -= (BASE 12); \
+if (a = (BASE 11)) a -= (BASE 11); \
+if (a = (BASE 10)) a -= (BASE 10); \
+if (a = (BASE 9)) a -= (BASE 9); \
+if (a = (BASE 8)) a -= (BASE 8); \
+if (a = (BASE 7)) a -= (BASE 7); \
+if (a = (BASE 6)) a -= (BASE 6); \
+if (a = (BASE 5)) a -= (BASE 5); \
+if (a = (BASE 4)) a -= (BASE 4); \
+if (a = (BASE 3)) a -= (BASE 3); \
+if (a = (BASE 2)) a -= (BASE 2); \
+if (a = (BASE 1)) a -= (BASE 1); \
+if (a = BASE) a -= BASE; \
+} while (0)
+# define MOD4(a) \
+do { \
+if (a = (BASE 4)) a -= (BASE 4); \
+if (a = (BASE 3)) a -= (BASE 3); \
+if (a = (BASE 2)) a -= (BASE 2); \
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am rpmgrep.1
(Dunno why this msg is mot getting through ... 3rd time's a charm) For various rpmio development reasons, I needed PCRE expressions applied to HTML content delivered by plain HTTP (not DAV enabled) transport. In order to achieve that goal, I've rewritten pcregrep (from pcre-7.6) to use -lpopt and -lrpmio. IMHO, the result has uses outside of rpm (and rpmio), so I'm going to install the executable (at least through rpm-5.1 development) in bindir (i.e. /usr/bin). We'll see later about whether /usr/bin/rpmgrep should be included in rpm-5.1 (or not). For now, I need to hear problem reports with rpmgrep, and that simply isn't going to happen unless I install /usr/bin/rpmgrep in PATH. Here's a very brief intro to what rpmgrep adds to pcregrep (note the URL argument, and I hope the HTML in the output spew makes it through mail): $ ./rpmgrep Fedora http://jbj.org/ titleTest Page for the Apache HTTP Server on Fedora/title h1Fedora strongTest Page/strong/h1 pFor information on Fedora, please visit the a href=http://fedoraproject.org/;Fedora Project website/a./p pYou are free to use the images below on Apache and Fedora powered HTTP servers. Thanks for using Apache and Fedora!/p pa href=http://httpd.apache.org/;img src=/icons/apache_pb2.gif alt=[ Powered by Apache ]//a a href=http:// fedoraproject.org/img src=/icons/poweredby.png alt=[ Powered by Fedora ] width=88 height=31 //a/p (aside) Yes, the same functionality could have be done with $ curl http://jbj.org/ | grep Fedora if I were writing a grep program. (aside) I'm not writing a grep program, but rather using rpmgrep as an external executable to stabilize PCRE patterns, hierarchical path traversal, and HTTP transport before enabling the same functionality within rpm itself. The output of rpm grep --help is appended below. Note that everwhere file is mentioned, a URI should be able to be substituted. That's what rpmio is about. I'll get a rpmgrep man page together soonishly ... Enjoy! 73 de Jeff [EMAIL PROTECTED] rpmio]$ ./rpmgrep --help Usage: lt-rpmgrep [OPTION...] -A, --after-context=number set number of following context lines -B, --before-context=number set number of prior context lines --color matched text color option --colourmatched text colour option -C, --context=numberset number of context lines, before after -c, --count print only a count of matching lines per FILE -D, --devices=actionhow to handle devices, FIFOs, and sockets -d, --directories=actionhow to handle directories -e, --regex(p) specify pattern (may be used more than once) -F, --fixed_strings patterns are sets of newline- separated strings -f, --file=path read patterns from file --file-offsets output file offsets, not text -H, --with-filename force the prefixing filename on output -h, --no-filename suppress the prefixing filename on output -i, --ignore-case ignore case distinctions -l, --files-with-matchesprint only FILE names containing matches -L, --files-without-match print only FILE names not containing matches --label=nameset name for standard input --line-offsets output line numbers and offsets, not text --locale=locale use the named locale -M, --multiline run in multiline mode -N, --newline=type set newline type (CR, LF, CRLF, ANYCRLF or ANY) -n, --line-number print line number with output lines -o, --only-matching show only the part of the line that matched -q, --quiet suppress output, just set return code -r, --recursive recursively scan sub-directories --exclude=pattern exclude matching files when recursing --include=pattern include matching files when recursing -s, --no-messages suppress error messages -u, --utf-8 use UTF-8 mode -V, --version print version information and exit -v, --invert-match select non-matching lines -w, --word-regexforce patterns to match only as words -x, --line-regexforce patterns to match only whole lines Common options for all rpmio executables: -D, --define='MACRO EXPR' define MACRO with value EXPR --undefine='MACRO' undefine MACRO -E, --eval='EXPR'
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am testit.sh tfts.c
On Feb 10, 2008, at 1:46 AM, Dan Nicholson wrote: On Feb 9, 2008 5:29 AM, Jeff Johnson [EMAIL PROTECTED] wrote: On Feb 9, 2008, at 8:09 AM, Jeff Johnson wrote: Or fix the linkage. How about the attached patch instead? WORKSFORME. And I'm still shopping ideas for how to add make check to rpmio intelligently. That seems fine. Also, you can put the C test programs in check_PROGRAMS and they'll only be built when `make check' is run. Right now, tdir tfts tget tglob are all built whether you want to run check or not. Thanks. Ralf has already cleaned up my sloppiness. After a bit of noodling, what I think I'm gonna do is swipe a copy of pcregrep.c and rewrite to use Fopen() instead of fopen(), etc, using other rpmio peculier structures and coding style internally to pcregrep.c. That basically means that I can also swipe the make check tests (and framework) frpm pcregrep to bootstrap automated rpmio testing as well. So if you don't like/want similar to pcre-7.6 make check does in rpmio/*, please suggest better alternatives or a different approach. What will still need thought is the client - server aspects of rpmio make check. Sure I can test for network connectivity and disable URI testing if end-point URI's are not accessible. But disabling != testing The better (what I've seen) approach is what neon does. The neon test harness lives on both sides of client - server localhost connection, basically a server is built and started by make check. Doing similar in rpmio is likely more work/time than I have this month, sigh. I'll get there eventually ... 73 de Jeff __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am testit.sh tfts.c
On Feb 9, 2008, at 4:03 AM, Ralf S. Engelschall wrote: On Fri, Feb 08, 2008, Jeff Johnson wrote: [...] +- jbj: tie tget/tdir/tglob/tfts to a rpmio make check. much more to do ... [...] LINT = splint -EXTRA_DIST = gengpg.sh tdir.c tfts.c tget.c thkp.c tput.c tglob.c tinv.c tkey.c trpmio.c lookup3.c tpw.c librpmio.vers +EXTRA_DIST = gengpg.sh thkp.c tput.c tinv.c tkey.c trpmio.c lookup3.c tpw.c librpmio.vers -EXTRA_PROGRAMS = tdir tfts tget tglob thkp tinv tkey tmacro tmagic tmire tput tpw trpmio tsw dumpasn1 lookup3 +EXTRA_PROGRAMS = thkp tinv tkey tmacro tmagic tmire tput tpw trpmio tsw dumpasn1 lookup3 + +noinst_PROGRAMS = tdir tfts tget tglob +noinst_SCRIPTS = testit.sh + +TESTS = testit.sh Sorry, Jeff, this cannot be done. It worked for you only because you still have not built from scratch. Else you would have noticed that this requires build/ and lib/ to be built _BEFORE_ rpmio/ stuff and this breaks the library dependency hierarchy! Then something else is broken. There's nothing in this set of includes #include system.h #include rpmio_internal.h #include rpmmacro.h #include rpmcb.h #include argv.h #include popt.h #include debug.h that should need anything other than -lrpmio and -lpopt and dpendents. Here is what is broken: make[2]: *** No rule to make target `../build/librpmbuild.la', needed by `tdir'. Stop. There's nothing in -lrpmbuild needed by tdir. It is important that the test programs in rpmio/ which need build/ and lib/ are marked as EXTRA_ stuff so they are *not* built automatically during the the regular build. By moving them to noinst_ they are attempted to be build and this breaks the regular build procedure now. I had to back this change out in order to unbreak the regular building. Sure. But there's nothing in those programs that should need anything outside of -lrpmio. If you really wish those test programs to be build automatically I recommend to move all of them into a separate directory (e.g. tools/) and additionally perhaps allow one to disable their building via --without-tests, too. Or fix the linkage. 73 de Jeff __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am testit.sh tfts.c
On Feb 9, 2008, at 4:03 AM, Ralf S. Engelschall wrote: On Fri, Feb 08, 2008, Jeff Johnson wrote: [...] +- jbj: tie tget/tdir/tglob/tfts to a rpmio make check. much more to do ... [...] LINT = splint -EXTRA_DIST = gengpg.sh tdir.c tfts.c tget.c thkp.c tput.c tglob.c tinv.c tkey.c trpmio.c lookup3.c tpw.c librpmio.vers +EXTRA_DIST = gengpg.sh thkp.c tput.c tinv.c tkey.c trpmio.c lookup3.c tpw.c librpmio.vers -EXTRA_PROGRAMS = tdir tfts tget tglob thkp tinv tkey tmacro tmagic tmire tput tpw trpmio tsw dumpasn1 lookup3 +EXTRA_PROGRAMS = thkp tinv tkey tmacro tmagic tmire tput tpw trpmio tsw dumpasn1 lookup3 + +noinst_PROGRAMS = tdir tfts tget tglob +noinst_SCRIPTS = testit.sh + +TESTS = testit.sh Sorry, Jeff, this cannot be done. It worked for you only because you still have not built from scratch. Else you would have noticed that this requires build/ and lib/ to be built _BEFORE_ rpmio/ stuff and this breaks the library dependency hierarchy! It is important that the test programs in rpmio/ which need build/ and lib/ are marked as EXTRA_ stuff so they are *not* built automatically during the the regular build. By moving them to noinst_ they are attempted to be build and this breaks the regular build procedure now. I had to back this change out in order to unbreak the regular building. If you really wish those test programs to be build automatically I recommend to move all of them into a separate directory (e.g. tools/) and additionally perhaps allow one to disable their building via --without-tests, too. Ralf S. Engelschall [EMAIL PROTECTED] www.engelschall.com __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am testit.sh tfts.c
On Feb 9, 2008 5:29 AM, Jeff Johnson [EMAIL PROTECTED] wrote: On Feb 9, 2008, at 8:09 AM, Jeff Johnson wrote: Or fix the linkage. How about the attached patch instead? WORKSFORME. And I'm still shopping ideas for how to add make check to rpmio intelligently. That seems fine. Also, you can put the C test programs in check_PROGRAMS and they'll only be built when `make check' is run. Right now, tdir tfts tget tglob are all built whether you want to run check or not. http://www.gnu.org/software/automake/manual/automake.html#Tests -- Dan __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am testit.sh tfts.c
I'm shopping ideas for designing a reliable make check for rpmio. Below are some examples of what needs testing and how to test. I've written exercisers for the basic functionality: tgetStat a path, and, iff a file, Fopen/Fread/Fclose the contents tdirwalk a directory using Opendir/Readdir tglob use Glob(3) to expand glob patterns tftsuse Fts(3) to walk a directory tree. Paths can be normal file paths as well as file:///, ftp://, http:// or https:// URI's. I'm almost positive that normal paths and file:/// URL's are fully functional. Most of rpmio was built and tested against ftp:// URI's, so I expect that functionality to be mostly (Stat() on a ftp:// directory URI seems broken, and so tfts is broken atm) fixed in the next day or two: [EMAIL PROTECTED] rpmio]$ ./tget ftp://localhost/pub/jbj/time.xml = ftp://localhost/pub/jbj/time.xml 396969608 1543245544 -rw-r--r-- 1 0 0 10018 2005-07-29 00:00:00 ftp://localhost/pub/jbj/time.xml stat:1 0.01 MB 0.087706 secs get:1 0.010018 MB 0.003858 secs [EMAIL PROTECTED] rpmio]$ ./tdir ftp://localhost/pub/jbj = ftp://localhost/pub/jbj/: 162 entries opendir:1 0.000162 MB 0.091381 secs [EMAIL PROTECTED] rpmio]$ ./tglob ftp://localhost/pub/jbj/* = ftp://localhost/pub/jbj/* glob:1 0.00 MB 0.092328 secs [EMAIL PROTECTED] rpmio]$ ./tfts ftp://localhost/pub/jbj = (0/0) dirs/files in: ftp://localhost/pub/jbj/ fts:1 0.00 MB 0.031525 secs Now that http://rpm5.org/files is DAV enabled, I've been using that to develop the exercisers. ATM, Fts(3) won't walk a tree, only a top level directory, but the other exercisers appear functional: [EMAIL PROTECTED] rpmio]$ ./tget http://rpm5.org/files/rpm/rpm-5.1/ readme.txt = http://rpm5.org/files/rpm/rpm-5.1/readme.txt 2332648353 0 -rw-r--r-- 1 0 0 157 2008-01-05 05:01:58 http:// rpm5.org/files/rpm/rpm-5.1/readme.txt stat:1 0.01 MB 0.743715 secs get:1 0.000157 MB 0.426203 secs [EMAIL PROTECTED] rpmio]$ ./tdir http://rpm5.org/files/rpm/rpm-5.1 = http://rpm5.org/files/rpm/rpm-5.1/: 5 entries opendir:1 0.05 MB 0.744295 secs [EMAIL PROTECTED] rpmio]$ ./tglob http://rpm5.org/files/rpm/rpm-5.1/* = http://rpm5.org/files/rpm/rpm-5.1/* glob:1 0.00 MB 0.748786 secs [EMAIL PROTECTED] rpmio]$ ./tfts http://rpm5.org/files/rpm/rpm-5.1 = (1/1) dirs/files in: http://rpm5.org/files/rpm/rpm-5.1/ fts:1 0.01 MB 3.551848 secs I have not yet looked at plain http:// URI's in any detail. I know that tglob will not function at all with non-DAV URL's ... checking ... yes more to do with non-DAV URI's: [EMAIL PROTECTED] rpmio]$ ./tget http://rpm5.org/sources.rss.xml = http://rpm5.org/sources.rss.xml 3586396879 40 -rw-r--r-- 1 0 0 41348 2008-02-08 17:45:07 http:// rpm5.org/sources.rss.xml stat:1 0.01 MB 0.765802 secs get:1 0.041348 MB 1.054655 secs [EMAIL PROTECTED] rpmio]$ ./tdir http://rpm5.org/ Segmentation fault [EMAIL PROTECTED] rpmio]$ ./tglob http://rpm5.org/* = http://rpm5.org/* Segmentation fault [EMAIL PROTECTED] rpmio]$ ./tfts http://rpm5.org/ = (0/1) dirs/files in: http://rpm5.org/ fts:1 0.00 MB 1.004163 secs How should a make check rpmio test harness be designed? As always its the error, not the functional, paths that will need explicit verification through rpmio. 73 de Jeff On Feb 8, 2008, at 5:20 PM, Jeff Johnson wrote: RPM Package Manager, CVS Repository http://rpm5.org/cvs/ __ __ Server: rpm5.org Name: Jeff Johnson Root: /v/rpm/cvs Email: [EMAIL PROTECTED] Module: rpm Date: 08-Feb-2008 23:20:03 Branch: HEAD Handle: 2008020822200201 Added files: rpm/rpmio testit.sh Modified files: rpm CHANGES rpm/rpmio Makefile.am tfts.c Log: - jbj: tie tget/tdir/tglob/tfts to a rpmio make check. much more to do ... Summary: RevisionChanges Path 1.2157 +1 -0 rpm/CHANGES 1.122 +7 -2 rpm/rpmio/Makefile.am 2.1 +16 -0 rpm/rpmio/testit.sh 2.18+1 -1 rpm/rpmio/tfts.c __
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am mire.c
On Mon, Feb 04, 2008, Jeff Johnson wrote: On Feb 4, 2008, at 3:09 PM, Ralf S. Engelschall wrote: On Sat, Feb 02, 2008, Jeff Johnson wrote: - jbj: mire.c: add STANDALONE test exerciser. [...] -EXTRA_PROGRAMS = tdir tfts tget thkp tmacro tput tglob tinv tkey tmagic trpmio tsw dumpasn1 lookup3 tpw +EXTRA_PROGRAMS = tdir tfts tget tglob thkp tinv tkey tmacro tmagic tmire tput tpw trpmio tsw dumpasn1 lookup3 [...] +tmire_SOURCES = tmire.c +tmire_LDADD = $(RPM_LDADD) [...] Jeff, just for your information: there is no rpmio/tmire.c in CVS. It doesn't hurt, but I guess you wanted to commit this file and just forgot to add rpmio/tmire.c... Sorry ... how are you catching these? make distcheck? [...] No, I do a regular make dist and just closely inspect the output. Ralf S. Engelschall [EMAIL PROTECTED] www.engelschall.com __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am mire.c
On Feb 4, 2008, at 3:09 PM, Ralf S. Engelschall wrote: On Sat, Feb 02, 2008, Jeff Johnson wrote: - jbj: mire.c: add STANDALONE test exerciser. [...] -EXTRA_PROGRAMS = tdir tfts tget thkp tmacro tput tglob tinv tkey tmagic trpmio tsw dumpasn1 lookup3 tpw +EXTRA_PROGRAMS = tdir tfts tget tglob thkp tinv tkey tmacro tmagic tmire tput tpw trpmio tsw dumpasn1 lookup3 [...] +tmire_SOURCES = tmire.c +tmire_LDADD = $(RPM_LDADD) [...] Jeff, just for your information: there is no rpmio/tmire.c in CVS. It doesn't hurt, but I guess you wanted to commit this file and just forgot to add rpmio/tmire.c... Sorry ... how are you catching these? make distcheck? If so, I'll explicitly check before checking in on -r rpm-5_0 73 de Jeff __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am mire.c
On Sat, Feb 02, 2008, Jeff Johnson wrote: - jbj: mire.c: add STANDALONE test exerciser. [...] -EXTRA_PROGRAMS = tdir tfts tget thkp tmacro tput tglob tinv tkey tmagic trpmio tsw dumpasn1 lookup3 tpw +EXTRA_PROGRAMS = tdir tfts tget tglob thkp tinv tkey tmacro tmagic tmire tput tpw trpmio tsw dumpasn1 lookup3 [...] +tmire_SOURCES = tmire.c +tmire_LDADD = $(RPM_LDADD) [...] Jeff, just for your information: there is no rpmio/tmire.c in CVS. It doesn't hurt, but I guess you wanted to commit this file and just forgot to add rpmio/tmire.c... Ralf S. Engelschall [EMAIL PROTECTED] www.engelschall.com __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org
Re: [CVS] RPM: rpm/ CHANGES rpm/rpmio/ Makefile.am mire.c
On Feb 4, 2008, at 3:41 PM, Ralf S. Engelschall wrote: Sorry ... how are you catching these? make distcheck? [...] No, I do a regular make dist and just closely inspect the output. OK. Typically I do make dist iff changing po/* files, I'll look harder. The other approach is to s/EXTRA_PROGRAMS/noinst_PROGRAMS/ (if that makes sense: i.e. don't distribute test programs, just leave in cvs). I'm not sure any of the test programs are useful to anyone but me. 73 de Jeff __ RPM Package Managerhttp://rpm5.org Developer Communication Listrpm-devel@rpm5.org
