February 2007 - developers-list

Re: [Monetdb-developers] [Monetdb-checkins] MonetDB/src/common stream.mx, 1.133.2.2, 1.133.2.3
by Ying Zhang 18 Feb '07

18 Feb '07

On Fri, Feb 16, 2007 at 07:07:00PM +0000, Niels Nes wrote: > Update of /cvsroot/monetdb/MonetDB/src/common > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv11260/src/common > > Modified Files: > Tag: MonetDB_1-16 > stream.mx > Log Message: > removed stream_r/wstream_byteorder > replaced with stream_set_byteorder > > the block_stream now always sends the blocksize/lastblock short in little > endian format (ie this is now fully controlled in streams). > The stream_set_byteorder only stores if the other end has little or bigendian > byte orders (this can later be use when we send/receive binary data) > > > > Index: stream.mx > =================================================================== > RCS file: /cvsroot/monetdb/MonetDB/src/common/stream.mx,v > retrieving revision 1.133.2.2 > retrieving revision 1.133.2.3 > diff -u -d -r1.133.2.2 -r1.133.2.3 > --- stream.mx 16 Feb 2007 15:27:20 -0000 1.133.2.2 > +++ stream.mx 16 Feb 2007 19:06:57 -0000 1.133.2.3 > @@ -555,6 +554,22 @@ > return s->byteorder; > } > > +void > +stream_set_byteorder(stream *s, char bigendian) > +{ > +#ifdef STREAM_DEBUG > + printf("stream_set_byteorder %s, %d\n", s->name ? s->name : "<unnamed>"); > +#endif > + assert(s->access == ST_READ); Hi Niels, Won't this assert kill all write streams, since their access is set to ST_WRITE? If I use MapiClient, I get the following: (in shell 1) $ Mserver # Monet Database Server V4.16.1 # Copyright (c) 1993-2007, CWI. All rights reserved. # Compiled for x86_64-redhat-linux-gnu/64bit with 64bit OIDs; # dynamically linked. # Visit http://monetdb.cwi.nl/ for further information. MonetDB>module(mapi); MonetDB> (in shell 2) $ MapiClient MapiClient: ../../../src/common/stream.mx:563: stream_set_byteorder: Assertion `s->access == 0' failed. Aborted In 'connect_to_server', mid->to->access is set to ST_WRITE. Jennie > + s->type = ST_BIN; > +#ifdef WORDS_BIGENDIAN > + s->byteorder = bigendian ? 1234 : 3412; > +#else > + s->byteorder = bigendian ? 3412 : 1234; > +#endif > +} > + > + > void > close_stream(stream *s) > { > @@ -576,22 +591,6 @@ > } > > stream * > -stream_rstream_byteorder(stream *s, char bigendian) > -{ > -#ifdef STREAM_DEBUG > - printf("stream_rstream_byteorder %s, %d\n", s->name ? s->name : "<unnamed>", bigendian); > -#endif > - assert(s->access == ST_READ); > - s->type = ST_BIN; > -#ifdef WORDS_BIGENDIAN > - s->byteorder = bigendian ? 1234 : 3412; > -#else > - s->byteorder = bigendian ? 3412 : 1234; > -#endif > - return s; > -} > - > -stream * > stream_wstream(stream *s) > { > #ifdef STREAM_DEBUG > @@ -604,22 +603,6 @@ > return s; > } > > -stream * > -stream_wstream_byteorder(stream *s, char bigendian) > -{ > -#ifdef STREAM_DEBUG > - printf("stream_wstream_byteorder %s, %d\n", s->name ? s->name : "<unnamed>"); > -#endif > - assert(s->access == ST_WRITE); > - s->type = ST_BIN; > -#ifdef WORDS_BIGENDIAN > - s->byteorder = bigendian ? 1234 : 3412; > -#else > - s->byteorder = bigendian ? 3412 : 1234; > -#endif > - return s; > -} > - > #define EXT_LEN 4 > static const char * > get_extention(const char *file) > @@ -2222,6 +2205,9 @@ > /* the last bit tells whether a flush is in there, it's not > * at this moment, so shift it to the left */ > blksize <<= 1; > +#ifdef WORDS_BIGENDIAN > + blksize = short_int_SWAP(blksize); > +#endif > if (!stream_writeSht(s->s, blksize) || s->s->write(s->s, s->buf, 1, s->nr) != (ssize_t) s->nr) { > ss->errnr = WRITE_ERROR; > return -1; > @@ -2451,6 +2437,10 @@ > return NULL; > if ((b = bs_create(s)) == NULL) > ns->errnr = OPEN_ERROR; > + /* blocksizes have a fixed little endian byteorder */ > +#ifdef WORDS_BIGENDIAN > + s->byteorder = 3412; /* simply != 1234 */ > +#endif > ns->type = s->type; > ns->access = s->access; > ns->read = bs_read; > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Monetdb-checkins mailing list > Monetdb-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-checkins

1 0

Re: [Monetdb-developers] [Monetdb-checkins] MonetDB4/conf MonetDB.conf.in, 1.3.2.1, 1.3.2.2
by Stefan Manegold 16 Feb '07

16 Feb '07

Peter, "datadir" is "datadir", just as individual as prefix, exec_prefix, libdir, etc., and hence not (necessarily) identical to exec_prefix. IMHO, you should set datadir=@Xdatadir@ instead of datadir=@Xexec_prefix@ Stefan On Fri, Feb 16, 2007 at 11:38:45PM +0000, Peter Boncz wrote: > Update of /cvsroot/monetdb/MonetDB4/conf > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv10130/conf > > Modified Files: > Tag: MonetDB_4-16 > MonetDB.conf.in > Log Message: > MonetDB.conf > - export configure 'datadir' so that it is visible in monetdb env ^^^^^^^ [...] > Index: MonetDB.conf.in > =================================================================== > RCS file: /cvsroot/monetdb/MonetDB4/conf/MonetDB.conf.in,v > retrieving revision 1.3.2.1 > retrieving revision 1.3.2.2 > diff -u -d -r1.3.2.1 -r1.3.2.2 > --- MonetDB.conf.in 12 Feb 2007 22:49:23 -0000 1.3.2.1 > +++ MonetDB.conf.in 16 Feb 2007 23:38:42 -0000 1.3.2.2 > @@ -7,6 +7,7 @@ > # Where MonetDB is installed > prefix=@Xprefix@ > exec_prefix=@Xexec_prefix@ > +datadir=@Xexec_prefix@ ^^^^^^^ ^^^^^^^^^^^^ > > # GDK Section > #============ > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Monetdb-checkins mailing list > Monetdb-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-checkins > -- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4312 |

1 0

Re: [Monetdb-developers] ! Compiling 'Current': Results for MonetDB DIFFER from 2007.02.15_04-03-15's !
by Stefan Manegold 16 Feb '07

16 Feb '07

On Fri, Feb 16, 2007 at 11:55:20AM +0100, Jan Rittinger wrote: > Stefan, > > I just got the testing mail, clicked the link and chose ``Shortend > "cross-check-list"'' > (http://monetdb.cwi.nl/testing/projects/monetdb/Current/MonetDB/.Makes3/inde…) > > Now I see that everything fails... (Thus I think the testweb is *not* > completely updated.) That was also why I did not see any problems on > Wednesday... What does the top say? "Results of 2007.02.16_04-43-27 ..." or "Results of 2007.02.15_04-03-15 ..." If the latter, tell your browser to reload the page. Stefan > Jan > > On 02/16/2007 11:26 AM, Project account wrote with possible deletions: > >Compilation results of MonetDB in > >/export/scratch2/monet/Testing/Current DIFFER from those of > >2007.02.15_04-03-15 ! > > > >Maybe YOUR checkins since 15.02.2007, 04:03:15 caused some of the > >problems: > > > >buildtools jan.rittinger(a)in.tum.de sjoerd(a)cwi.nl > >MonetDB sjoerd(a)cwi.nl > > > >Check http://monetdb.cwi.nl/Development/TestWeb/Current/MonetDB/ for > >details ! > > > > > >Old file: Mmake..2007.02.15_04-03-15..out > >New file: Mmake..2007.02.16_04-43-27..out > >========================================= > > > > > >Summary: > >======== > > > > * 13 Systems now produce output > > * 2 Systems still don't produce output > > > > * Results for 13 systems that now produce output > > - No output for 10 tests > > > > > >Details: > >======== > > > >13 Systems now produce output: > >D8.8.0-G32/32d, F4-G64/64d, F4-n64/64d, F6-G32/32d, F6-G64/32d, > >F6-G64/64d, F6-G64/64s, F6-n64/32d, F6-n64/64d, G1.12.6-G32/32d, > >R4-n64/64d, S9.3-G32/32d, S5.10-G32/32d > > > >2 Systems still don't produce output: > >C4.4-G32/32d, D8.8.0-G64/64d > > > > > >Results for 13 systems that now produce output: > >----------------------------------------------- > >Summary: > > - No output for 10 tests > >Details: > >No output for 10 tests: > >D8.8.0 G32/32d dist/MonetDB-1.17.1.tar.gz - > >D8.8.0 G32/32d rpm/rpm/RPMS/MonetDB-1.17.1-1_.rpm - > >D8.8.0 G32/32d rpm/rpm/RPMS/MonetDB-devel-1.17.1-1_.rpm - > >D8.8.0 G32/32d rpm/rpm/SRPMS/MonetDB-1.17.1-1.src.rpm - > >G1.12.6 G32/32d rpm/rpm/RPMS/MonetDB-1.17.1-1_.rpm - > >G1.12.6 G32/32d rpm/rpm/RPMS/MonetDB-devel-1.17.1-1_.rpm - > >G1.12.6 G32/32d rpm/rpm/SRPMS/MonetDB-1.17.1-1.src.rpm - > >S5.10 G32/32d rpm/rpm/RPMS/MonetDB-1.17.1-1_.rpm - > >S5.10 G32/32d rpm/rpm/RPMS/MonetDB-devel-1.17.1-1_.rpm - > >S5.10 G32/32d rpm/rpm/SRPMS/MonetDB-1.17.1-1.src.rpm - > > > > -- > Jan Rittinger > Database Systems > Technische Universität München (Germany) > http://www-db.in.tum.de/~rittinge/ -- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4312 |

3 2

Re: [Monetdb-developers] [Monetdb-pf-checkins] pathfinder EXTENSIONS, 1.7.2.1, 1.7.2.2
by Peter Boncz 14 Feb '07

14 Feb '07

Hi Jens, Thanx. One correction: we *do* support the upd:put() operator. I is actually named fn:put(). For the moment, we only accept file URLs. For security reasons, these must start with a relative path. Peter > Update of /cvsroot/monetdb/pathfinder > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv29416 > > Modified Files: > Tag: XQuery_0-16 > EXTENSIONS > Log Message: > Clean up things that grep cannot cover. > > > Index: EXTENSIONS > =================================================================== > RCS file: /cvsroot/monetdb/pathfinder/EXTENSIONS,v > retrieving revision 1.7.2.1 > retrieving revision 1.7.2.2 > diff -u -d -r1.7.2.1 -r1.7.2.2 > --- EXTENSIONS 13 Feb 2007 22:39:08 -0000 1.7.2.1 > +++ EXTENSIONS 14 Feb 2007 09:17:20 -0000 1.7.2.2 > @@ -17,9 +17,12 @@ > > "do rename" TargetExpr "into" NewNameExpr > > -while the Update Facility uses "as" instead of "with". Sorry > +while the Update Facility uses "as" instead of "into". Sorry > for the inconvenience. > > +(The same issue has been discussed in the W3C XQuery bug > +tracking system: http://www.w3.org/Bugs/Public/show_bug.cgi?id=4176.) > + > We do not currently support the upd:put() operator. > > 2. Transitive Closure Operator > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Monetdb-pf-checkins mailing list > Monetdb-pf-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-pf-checkins >

1 0

[Monetdb-developers] XQUF: "rename . with ." => "rename . into ." ?
by Stefan Manegold 13 Feb '07

13 Feb '07

Dear all, maybe I missed the repective checkin (-message), but has the syntax of do rename . with . recently (read yesterday) been changed into do rename . into . ======== .../pathfinder $ find * -type f | xargs grep --color 'do.*rename.*into' -------- compiler/core/fs.brg: * [[ do rename <TargetExpr> into <NewNameExpr> ]] compiler/parser/parser.y:RenameExpr : "do rename" TargetExpr "into" NewNameExpr ======== ?? If this was intended, we should also update the repective documentation and tests! ======== .../pathfinder $ find * -type f | xargs grep --color 'do.*rename.*with' -------- EXTENSIONS: "do rename" TargetExpr "with" NewNameExpr tests/Update/Tests/update.xq: do rename exactly-one($elem/@attribute) with "Attribute", tests/Update/Tests/update.xq: do rename exactly-one($elem/@dummy) with "foo", tests/Update/Tests/update.xq: return do rename exactly-one($elem) with "Element" tests/Update/Tests/update.xq: do rename exactly-one($testdoc//processing-instruction(pi1000)) with "bar", tests/Update/Tests/update.xq: do rename exactly-one($testdoc//processing-instruction(pi5000)) with "xyzzy", ======== http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103… ======== Stefan ps: "find" & "grep" are you friends --- quite handy to make such changes *consistenstly* ... -- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4312 |

1 1

Re: [Monetdb-developers] [Monetdb-pf-checkins] pathfinder/tests/XRpc/Tests add_dbl.xq.in, 1.2, 1.2.2.1 add_int.xq.in, 1.2, 1.2.2.1 concatStr.xq.in, 1.2, 1.2.2.1 echoAll.xq.in, 1.2, 1.2.2.1 echoVoid.xq.in, 1.2, 1.2.2.1 emptySeq1.xq.in, 1.2, 1.2.2.1 emptySeq2.xq.in, 1.2, 1.2.2.1 emptySeq3.xq.in, 1.2, 1.2.2.1 emptySeq4.xq.in, 1.2, 1.2.2.1 insertNode.xq.in, 1.2, 1.2.2.1 multi_clnt_s00.xq.in, 1.2, 1.2.2.1 multi_clnt_s01.xq.in, 1.2, 1.2.2.1 multi_clnt_s02.xq.in, 1.2, 1.2.2.1 returnAncestor.xq.in, 1.3, 1.3.2.1 ret
by Stefan Manegold 13 Feb '07

13 Feb '07

Forgot to mention: To make this work properly, you also need to cvs-update & make-install MonetDB to get the latest version of Mtest.py . Stefan On Tue, Feb 13, 2007 at 08:03:50PM +0000, Stefan Manegold wrote: > Update of /cvsroot/monetdb/pathfinder/tests/XRpc/Tests > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv9407/tests/XRpc/Tests > > Modified Files: > Tag: XQuery_0-16 > add_dbl.xq.in add_int.xq.in concatStr.xq.in echoAll.xq.in > echoVoid.xq.in emptySeq1.xq.in emptySeq2.xq.in emptySeq3.xq.in > emptySeq4.xq.in insertNode.xq.in multi_clnt_s00.xq.in > multi_clnt_s01.xq.in multi_clnt_s02.xq.in returnAncestor.xq.in > returnAncestorOrSelf.xq.in returnChild.xq.in > returnDescendant.xq.in returnDescendantOrSelf.xq.in > returnFollowing.xq.in returnFollowingSibling.xq.in > returnParent.xq.in returnPreceding.xq.in > returnPrecedingSibling.xq.in returnSelf.xq.in > xrpcConvert.xq.in xrpcConvert2.xq.in > xrpcConvert2.xrpcfun2.xq.in xrpcfun.xq > Removed Files: > Tag: XQuery_0-16 > add_dbl.prologue.src add_int.prologue.src > concatStr.prologue.src echoAll.prologue.src > echoVoid.prologue.src emptySeq1.prologue.src > emptySeq2.prologue.src emptySeq3.prologue.src > emptySeq4.prologue.src insertNode.prologue.src > multi_clnt.prologue.src returnAncestor.prologue.src > returnAncestorOrSelf.prologue.src returnChild.prologue.src > returnDescendant.prologue.src > returnDescendantOrSelf.prologue.src > returnFollowing.prologue.src > returnFollowingSibling.prologue.src returnParent.prologue.src > returnPreceding.prologue.src > returnPrecedingSibling.prologue.src returnSelf.prologue.src > rpcd_start.mil xrpcConvert.prologue.src > xrpcConvert2.prologue.src > Log Message: > > made XRpc tests work, again: > > Not only has "xquery_httpd_port" been renamed into "xrpc_port", > but also the defaults have changed: > While the former used to default to 48080 on the server side (via MonetDB.conf) > and on the client side (hard-wired in the code), > the latter now defaults to "" (not set) in MonetDB.conf, and hence to > mapi_port+1 in the server code; in the client code, the default is still > hard-wired to 48080. > > Hence, to make the XRPC tests work, again, in the Mtest enviroment > (which dynamically sets mapi_port randomly to avoid conflict between > concurrent Mtest runs and/or with other programs that happen to use > the default port(s)), we need to also set xrpc_port dynamically (now > done by Mtest), and use the dynamically set xrpc_port (via $XRPCPORT) > in the respective tests. > > Moreover, when module pathfinder is loaded, it now autimatically starts > an xrpc-server; hence, we must not try to start a second xrpc-server in > the XRPC tests. > -- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4312 |

1 0

Re: [Monetdb-developers] [Monetdb-checkins] MonetDB5/src/tools mserver5.mx, 1.27, 1.28
by Sjoerd Mullender 13 Feb '07

13 Feb '07

On 2007-02-13 10:55, Fabian wrote: > Update of /cvsroot/monetdb/MonetDB5/src/tools > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv30761/src/tools > > Modified Files: > mserver5.mx > Log Message: > Don't set the version as a gdk_XXX variable, instead "hardcode" it by > using the VERSION define. This way a user won't get an "old mserver" > by using an "old config file". > > > Index: mserver5.mx > =================================================================== > RCS file: /cvsroot/monetdb/MonetDB5/src/tools/mserver5.mx,v > retrieving revision 1.27 > retrieving revision 1.28 > diff -u -d -r1.27 -r1.28 > --- mserver5.mx 19 Jan 2007 14:41:24 -0000 1.27 > +++ mserver5.mx 13 Feb 2007 09:55:54 -0000 1.28 > @@ -156,7 +156,7 @@ > char *msg = mo_find_option(set, setlen, "monet_welcome"); > > if (msg && strcmp(msg, "yes") == 0) { > - printf("# MonetDB Server v%s\n", GDKgetenv("gdk_version")); > + printf("# MonetDB Server v" VERSION "\n"); > printf("# Copyright (c) 1993-2007 CWI, all rights reserved\n"); > printf("# Compiled for %s/" SZFMT "bit with " SZFMT "bit OIDs %s linked\n", HOST, (size_t) (sizeof(ptr) * 8), (size_t) (sizeof(oid) * 8), linkinfo); > #ifdef MONET_GLOBAL_DEBUG Using printf("# MonetDB Server v%s\n", VERSION); may be better. It'll even work when somebody somehow puts a % in VERSION. -- Sjoerd Mullender

2 1

Re: [Monetdb-developers] [Monetdb-checkins] MonetDB4 html.mk, 1.2, NONE
by Jan Rittinger 12 Feb '07

12 Feb '07

As there seems to be an effort to clean up the repositories I have an idea to make Martin happy :) As you all know the current setup does not allow Martin to introduce personal files like e.g. 'todo.mk' in the sources (at least not at the top level). The reason is a nice '*' in am.py. With Stefan latest changes only rpm.mk and extra.mk remain in the various sources (whereas extra.mk is only used in pathfinder). My proposal thus is to apply the following small diff. It would replace '*.mk' with rpm.mk and an additional extra.mk can be integrated conditionally only. Any objections? Index: autogen/autogen/am.py =================================================================== RCS file: /cvsroot/monetdb/buildtools/autogen/autogen/am.py,v retrieving revision 1.31 diff -r1.31 am.py 1095c1095,1098 < include $(top_builddir)/*.mk --- > if HAVE_EXTRA_MK > include $(top_builddir)/extra.mk > endif > include $(top_builddir)/rpm.mk Index: configure.ag =================================================================== RCS file: /cvsroot/monetdb/pathfinder/configure.ag,v retrieving revision 1.99 diff -r1.99 configure.ag 37a38,39 > AM_CONDITIONAL(HAVE_EXTRA_MK, true) > > -----Original Message----- > From: monetdb-checkins-bounces(a)lists.sourceforge.net > [mailto:monetdb-checkins-bounces@lists.sourceforge.net] On > Behalf Of Stefan Manegold > Sent: Friday, February 09, 2007 9:05 PM > To: monetdb-checkins(a)lists.sourceforge.net > Subject: [Monetdb-checkins] MonetDB4 html.mk,1.2,NONE > > Update of /cvsroot/monetdb/MonetDB4 > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv22609 > > Removed Files: > Tag: MonetDB_4-16 > html.mk > Log Message: > > documentation from MonetDB4/doc/ > has been moved to monetweb/Docs/MonetDB4/ > > hence, we can remove it here. > > also removed html.mk, since the respective work of generating > documentation from sources is now done by monetweb/Docs/* > > > --- html.mk DELETED --- > > > -------------------------------------------------------------- > ----------- > Using Tomcat but need to do more? Need to support web > services, security? > Get stuff done quickly with pre-integrated technology to make > your job easier. > Download IBM WebSphere Application Server v.1.0.1 based on > Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057& dat=121642 > _______________________________________________ > Monetdb-checkins mailing list > Monetdb-checkins(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/monetdb-checkins > >

2 1

Re: [Monetdb-developers] [Monetdb-pf-checkins] pathfinder/compiler/mil milprint_summer.c, 1.318.2.9, 1.318.2.10
by Stefan Manegold 11 Feb '07

11 Feb '07

On Sun, Feb 11, 2007 at 11:43:49AM +0000, Peter Boncz wrote: > Update of /cvsroot/monetdb/pathfinder/compiler/mil > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv30433/compiler/mil > > Modified Files: > Tag: XQuery_0-16 > milprint_summer.c > Log Message: > - fixed bug in MIL epilogue after error (resulted in another error on "-") > - changed output of pf:collections() and pf:documents() info functions > to return nodes rather thanm strings > - introduced new convenience builtin pf:mil() ^^^^^^^^ Peter (out of pure curiosity) what's the functionality/semantic of this one? (Needless to mention that this appears to be a new *feature* introduced on the *release branch* --- hence, some more information as "justification" would be very much appreciated.) Stefan -- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4312 |

1 0

Re: [Monetdb-developers] trying to use monetdb pathfinder
by p.a.boncz 09 Feb '07

09 Feb '07

Dear Martijn (et al.) I am currently working on the documentation that describes all the new XQuery functionality (updates, xrpc, collections, node-ids) . I expect it to come online next week -- I will announce it here on the mailing lists. We are also working on a 64-bits windows build, and we have found a number of bugs. Therefore, in I expect a new minor (bug-fix) release to appear soon, i.e. in the scope of two weeks. So, bugs that are reported now might well get fixed very soon. So please report them. I do recognize that it is hard to test functionality that is not documented, - but please bear with me for a couple of days more. Thanks, Peter

3 2