popsearch-help Mailing List for POPsearch
Status: Beta
Brought to you by:
mrider
You can subscribe to this list here.
| 2004 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(5) |
Oct
|
Nov
(25) |
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2005 |
Jan
|
Feb
(2) |
Mar
(1) |
Apr
(4) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Matthew T. R. <mr...@po...> - 2005-04-27 18:33:15
|
This major release includes: * RSS and email feeds can now be created for each directory. You can download the latest release from https://sourceforge.net/project/showfiles.php?group_id=107274&package_id=115847&release_id=323636 |
|
From: Matthew T. R. <mr...@po...> - 2005-04-24 15:22:48
|
Did you install the prerequisites package. http://prdownloads.sourceforge.net/popsearch/prerequisites-1.3.tar.bz2?download > -------- Original Message -------- > Subject: [Popsearch-help] Error compile popsearch > From: Adriano Frare <al...@e-...> > Date: Fri, April 22, 2005 1:11 pm > To: pop...@li... > > Dear Friends, > > > I am trying compile popsearch and I have received follow messages: > > ================== BEGIN ================================= > In file included from IndexedCollection_factory.h:142, > from BerkeleyDB_bridge.h:276, > from BerkeleyDB_bridge.cc:36: > File.h:255:19: magic.h: No such file or directory > BerkeleyDB_bridge.cc: In constructor > `BerkeleyDB_bridge::BerkeleyDB_bridge(const char*, const char*, const > databaseMode&, const databaseTypes&, const int&, const int&, const int&, > const int&, const int&, const int&, const char*, const char*, const int&, > const int&, const long int&, const int&, const int&)': > BerkeleyDB_bridge.cc:1298: warning: passing negative value `-1' for argument > passing 2 of `int chown(const char*, unsigned int, unsigned int)' > BerkeleyDB_bridge.cc:1298: warning: argument of negative value `-1' to ` > unsigned int' > BerkeleyDB_bridge.cc: In constructor > `BerkeleyDB_bridge::BerkeleyDB_bridge(const char*, const databaseMode&, > const databaseTypes&, const int&, const int&, const int&, const int&, > const > int&, const int&, const char*, const char*, const int&, const int&, const > long int&, const int&, const int&)': > BerkeleyDB_bridge.cc:2253: warning: passing negative value `-1' for argument > passing 2 of `int chown(const char*, unsigned int, unsigned int)' > BerkeleyDB_bridge.cc:2253: warning: argument of negative value `-1' to ` > unsigned int' > BerkeleyDB_bridge.cc: In member function `long int > BerkeleyDB_bridge::purge()': > BerkeleyDB_bridge.cc:2647: warning: assignment to `long int' from `double' > BerkeleyDB_bridge.cc:2647: warning: argument to `long int' from `double' > BerkeleyDB_bridge.cc: In member function `long int > BerkeleyDB_bridge::trickleWrite()': > BerkeleyDB_bridge.cc:2772: warning: assignment to `long int' from `double' > BerkeleyDB_bridge.cc:2772: warning: argument to `long int' from `double' > BerkeleyDB_bridge.cc:2932: warning: assignment to `long int' from `double' > BerkeleyDB_bridge.cc:2932: warning: argument to `long int' from `double' > BerkeleyDB_bridge.cc: In member function `CollectionData_factory* > BerkeleyDB_bridge::cachelessLookup(const std::string&, Status&, const > readFlag&, DB_TXN*, int)': > BerkeleyDB_bridge.cc:4: Internal compiler error in cp_expand_stmt, at > cp/semantics.c:2168 > Please submit a full bug report, > with preprocessed source if appropriate. > See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions. > make[3]: ** [BerkeleyDB_bridge.lo] Erro 1 > make[3]: Leaving directory `/data/appl/popsearch-0.4.1/src/common' > make[2]: ** [all] Erro 2 > make[2]: Leaving directory `/data/appl/popsearch-0.4.1/src/common' > make[1]: ** [all-recursive] Erro 1 > make[1]: Leaving directory `/data/appl/popsearch-0.4.1/src' > make: ** [all-recursive] Erro 1 > > ================== END =================================== > > > Thanks for help. > > > Adriano > > > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click > _______________________________________________ > Popsearch-help mailing list > Pop...@li... > https://lists.sourceforge.net/lists/listinfo/popsearch-help |
|
From: Adriano F. <al...@e-...> - 2005-04-22 17:12:00
|
Dear Friends,
I am trying compile popsearch and I have received follow messages:
================== BEGIN =================================
In file included from IndexedCollection_factory.h:142,
from BerkeleyDB_bridge.h:276,
from BerkeleyDB_bridge.cc:36:
File.h:255:19: magic.h: No such file or directory
BerkeleyDB_bridge.cc: In constructor
`BerkeleyDB_bridge::BerkeleyDB_bridge(const char*, const char*, const
databaseMode&, const databaseTypes&, const int&, const int&, const int&,
const int&, const int&, const int&, const char*, const char*, const int&,
const int&, const long int&, const int&, const int&)':
BerkeleyDB_bridge.cc:1298: warning: passing negative value `-1' for argument
passing 2 of `int chown(const char*, unsigned int, unsigned int)'
BerkeleyDB_bridge.cc:1298: warning: argument of negative value `-1' to `
unsigned int'
BerkeleyDB_bridge.cc: In constructor
`BerkeleyDB_bridge::BerkeleyDB_bridge(const char*, const databaseMode&,
const databaseTypes&, const int&, const int&, const int&, const int&,
const
int&, const int&, const char*, const char*, const int&, const int&, const
long int&, const int&, const int&)':
BerkeleyDB_bridge.cc:2253: warning: passing negative value `-1' for argument
passing 2 of `int chown(const char*, unsigned int, unsigned int)'
BerkeleyDB_bridge.cc:2253: warning: argument of negative value `-1' to `
unsigned int'
BerkeleyDB_bridge.cc: In member function `long int
BerkeleyDB_bridge::purge()':
BerkeleyDB_bridge.cc:2647: warning: assignment to `long int' from `double'
BerkeleyDB_bridge.cc:2647: warning: argument to `long int' from `double'
BerkeleyDB_bridge.cc: In member function `long int
BerkeleyDB_bridge::trickleWrite()':
BerkeleyDB_bridge.cc:2772: warning: assignment to `long int' from `double'
BerkeleyDB_bridge.cc:2772: warning: argument to `long int' from `double'
BerkeleyDB_bridge.cc:2932: warning: assignment to `long int' from `double'
BerkeleyDB_bridge.cc:2932: warning: argument to `long int' from `double'
BerkeleyDB_bridge.cc: In member function `CollectionData_factory*
BerkeleyDB_bridge::cachelessLookup(const std::string&, Status&, const
readFlag&, DB_TXN*, int)':
BerkeleyDB_bridge.cc:4: Internal compiler error in cp_expand_stmt, at
cp/semantics.c:2168
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
make[3]: ** [BerkeleyDB_bridge.lo] Erro 1
make[3]: Leaving directory `/data/appl/popsearch-0.4.1/src/common'
make[2]: ** [all] Erro 2
make[2]: Leaving directory `/data/appl/popsearch-0.4.1/src/common'
make[1]: ** [all-recursive] Erro 1
make[1]: Leaving directory `/data/appl/popsearch-0.4.1/src'
make: ** [all-recursive] Erro 1
================== END ===================================
Thanks for help.
Adriano
|
|
From: Matthew T. R. <mr...@po...> - 2005-04-11 19:24:28
|
This maintenance release includes: * Performance tuning of the indexer. * Changed the layout of the search results. * Added logic to recover from database deadlocks. |
|
From: Matthew T. R. <mr...@po...> - 2005-03-29 15:53:26
|
It's been over four months since the last POPsearch release. During this time I took a step back and tried to determine the future direction of POPsearch. Some of the things that were completed: POPsearch should be easy to use. One command starts the user interface and performs the indexing, not the two step process that exists today, where you have to index everything before it can be searched. With the new POPsearch, you can search as the data is being indexed. I also wanted to work on the scalability and reliability of POPsearch. After load testing POPsearch on large mailboxes that contain thousands of email messages, and directories that contain thousands of files, it was apparent that POPsearch suffered from scalability issues. As a result, I've redesigned almost a dozen choke points, and now POPsearch indexes at a much faster pace. Since indexing is a CPU intensive process, the indexing should only occur when the machine is idle. POPsearch reliability has also been a priority issue. Past releases have suffered from cores at various times, and the removal of these cores has been a high priority over the past few months. While it may appear that POPsearch development had stopped, I'm happy to say that it hasn't. I just didn't feel comfortable with the major flaws that I saw with POPsearch, and I didn't want to apply a band-aid to problems as they were uncovered by the user community. So I felt that I needed to take a long look at the system and redesign several areas for the benefit of everyone. I hope that it will be worth the wait. You can download the latest release from https://sourceforge.net/project/showfiles.php?group_id=107274&package_id=115847&release_id=316232 |
|
From: Matthew T. R. <mr...@po...> - 2005-02-14 17:41:47
|
It looks like you may need to update to the latest version of the Berkeley database. http://www.sleepycat.com/webforms/postdownload.php?fname=db-4.3.27.NC.tar.gz&prod=core After you install the latest version of Berkeley, my recommendation is to reconfigure POPsearch (with ./configure) then: make clean make make install But you may want to hold off on this version of POPsearch until version 0.4 is released. It's been almost three months since the last POPsearch release. During this time I took a step back and tried to determine the future direction of POPsearch. Some of the things that I've been working on: 1) POPsearch should be easy to use. One command starts the user interface and performs the indexing, not the two step process that exists today where you have to index everything before it can be searched. With the new POPsearch, you'll be able to search as the data is being indexed. 2) I also wanted to work on the scalability and reliability of POPsearch. After load testing POPsearch on large mailboxes that contain thousands of email messages, and directories that contain thousands of files, it was apparent that POPsearch suffered from scalability issues. As a result, I've redesigned almost a dozen choke points, and now POPsearch indexes at a much faster pace. 3) The POPsearch compiles take a long time. So I've been working to reduce the size of large source files. 4) Since indexing is a time consuming CPU intensive process, the indexing should only occur when the machine is idle. 5) POPsearch reliability has also been a priority issue. Past releases have suffered from cores at various times, and the removal of these cores has been a high priority over the past few months. While it may appear that POPsearch development had stopped, I'm happy to say that it hasn't. I just didn't feel comfortable with the major flaws that I saw with POPsearch, and I didn't want to apply a band-aid to problems as they were uncovered by the user community. So I felt that I needed to take a long look at the system and redesign several areas for the benefit of everyone. I hope that it will be worth the wait. > -------- Original Message -------- > Subject: [Popsearch-help] Popsearch 0.3.3 Compile Issues > From: "Brad Bowling" <bbo...@gm...> > Date: Fri, February 11, 2005 4:37 pm > To: pop...@li... > > Good afternoon, > > I am trying to build popsearch 0.3.3 on a gentoo box, gcc version > 3.3.5, all dependencies are satisfied. Everything compiles just fine, > but the linking fails with the following error: > > /bin/sh ../../libtool --mode=link g++ -g -O2 -L. -lpthread -o > build main.o -L../common -L. -L/usr/local/lib -lcommon -lbuild > -lMagick -lMagick++ -ldb -lbz2 -lfl -laspell > g++ -g -O2 -o .libs/build main.o -L/root/popsearch-0.3.3/src/build > -L/root/popsearch-0.3.3/src/common -L/usr/local/lib > /root/popsearch-0.3.3/src/common/.libs/libcommon.so > /root/popsearch-0.3.3/src/build/.libs/libbuild.so -L/usr/lib > -L/usr/i686-pc-linux-gnu/bin -L/usr/i686-pc-linux-gnu/lib -lm > -lpthread /usr/lib/libMagick++.so /usr/lib/libWand.so > /usr/lib/libMagick.so /usr/lib/libtiff.so /usr/lib/libjpeg.so -lz > /usr/lib/libltdl.so -ldl > -L/usr/lib/gcc-lib/i686-pc-linux-gnu/../../../i686-pc-linux-gnu/lib > -ldb -lbz2 -lfl /usr/lib/libaspell.so > //usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5/libstdc++.so -Wl,--rpath > -Wl,/usr/local/lib -Wl,--rpath > -Wl,//usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5 > /root/popsearch-0.3.3/src/common/.libs/libcommon.so: undefined > reference to `txn_abort' > /root/popsearch-0.3.3/src/common/.libs/libcommon.so: undefined > reference to `txn_begin' > /root/popsearch-0.3.3/src/common/.libs/libcommon.so: undefined > reference to `txn_commit' > collect2: ld returned 1 exit status > > > Has anyone seen this error before, or have any idea how I can get around it? > > Thank you for your help, > > Brad > > > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click > _______________________________________________ > Popsearch-help mailing list > Pop...@li... > https://lists.sourceforge.net/lists/listinfo/popsearch-help |
|
From: Brad B. <bbo...@gm...> - 2005-02-11 21:37:30
|
Good afternoon, I am trying to build popsearch 0.3.3 on a gentoo box, gcc version 3.3.5, all dependencies are satisfied. Everything compiles just fine, but the linking fails with the following error: /bin/sh ../../libtool --mode=link g++ -g -O2 -L. -lpthread -o build main.o -L../common -L. -L/usr/local/lib -lcommon -lbuild -lMagick -lMagick++ -ldb -lbz2 -lfl -laspell g++ -g -O2 -o .libs/build main.o -L/root/popsearch-0.3.3/src/build -L/root/popsearch-0.3.3/src/common -L/usr/local/lib /root/popsearch-0.3.3/src/common/.libs/libcommon.so /root/popsearch-0.3.3/src/build/.libs/libbuild.so -L/usr/lib -L/usr/i686-pc-linux-gnu/bin -L/usr/i686-pc-linux-gnu/lib -lm -lpthread /usr/lib/libMagick++.so /usr/lib/libWand.so /usr/lib/libMagick.so /usr/lib/libtiff.so /usr/lib/libjpeg.so -lz /usr/lib/libltdl.so -ldl -L/usr/lib/gcc-lib/i686-pc-linux-gnu/../../../i686-pc-linux-gnu/lib -ldb -lbz2 -lfl /usr/lib/libaspell.so //usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5/libstdc++.so -Wl,--rpath -Wl,/usr/local/lib -Wl,--rpath -Wl,//usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.5 /root/popsearch-0.3.3/src/common/.libs/libcommon.so: undefined reference to `txn_abort' /root/popsearch-0.3.3/src/common/.libs/libcommon.so: undefined reference to `txn_begin' /root/popsearch-0.3.3/src/common/.libs/libcommon.so: undefined reference to `txn_commit' collect2: ld returned 1 exit status Has anyone seen this error before, or have any idea how I can get around it? Thank you for your help, Brad |
|
From: Matthew T. R. <mr...@po...> - 2004-11-19 17:07:53
|
This maintenance release includes: * Documentation changes. * Now filtering Emacs backup files from the indexer. * Now ignoring comments in the file exclusions. * Added a logout button. * Changed the way that low space warnings are handled. * Added statistics to the web browser. * Performance tuning of the popbuild process. * Minor change to the directory display. * Display warning for systems with weak thread support. * Removed post processing from the PDF conversion. * Made some changes to the build configuration. |
|
From: Matthew T. R. <mr...@po...> - 2004-11-12 15:57:58
|
Is it still running? If you type: ps axww | grep build What do you see? If you look at the last messages in file .POPsearch/logs/systemLog Does it say that popbuild finished? Also, are you using version 0.3.2 that I released yesterday? > -------- Original Message -------- > Subject: [Popsearch-help] popbuild > From: "Jeffrey Mutonho" <ejb...@gm...> > Date: Fri, November 12, 2004 9:13 am > To: "Matthew T. Rider" <mr...@po...> > Cc: pop...@li... > > Why does popbuild die gracefully?I ran "stop_popbuild" 6 hours ago , > and now trying to run another build i get the message "popbuild is > already running > " > How can I stop popbuild ? > > jeff > > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Sybase ASE Linux Express Edition - download now for FREE > LinuxWorld Reader's Choice Award Winner for best database on Linux. > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click > _______________________________________________ > Popsearch-help mailing list > Pop...@li... > https://lists.sourceforge.net/lists/listinfo/popsearch-help |
|
From: Jeffrey M. <ejb...@gm...> - 2004-11-12 14:13:42
|
Why does popbuild die gracefully?I ran "stop_popbuild" 6 hours ago , and now trying to run another build i get the message "popbuild is already running " How can I stop popbuild ? jeff |
|
From: Matthew T. R. <mr...@po...> - 2004-11-11 22:21:53
|
This maintenance release includes: * Documentation changes. * Made some changes to the build configuration. * Removed obsolete restart logic. * Changed the help messages. * Fixed a core in the email indexing phase. * Performance tuning of the popbuild process. * Fixed a bug with file names that have shell redirects. * Fixed a bug with PDF conversion failures. * Fixed a bug with RTF conversion failures. * Fixed a bug with WordPerfect conversion failures. * Conversion scripts now shut down when they are told to do so. * Minor change to the create account web page. * Now filtering Emacs backup files from the indexer. |
|
From: Matthew T. R. <mr...@po...> - 2004-11-08 16:46:49
|
Glad to hear that you got things running on debian. It sounds like the external PDF conversion is causing the problem, and POPsearch won't shutdown until it regains control from the external PDF conversion that it started. You can look at .POPsearch/logs/systemLog but I doubt that it will show anything about the PDF that you're having problems with. One thing to try is running the conversion yourself by typing: pdftohtml -c -noframes /path/to/your/pdf HTML When you run this, do you get a file called HTML.html? If you don't and it keeps chewing up your CPU, then perhaps I'll have to switch to another PDF converter (fortunately there are more of them out there). To get around the problems that you're having, I would exclude the PDF from the indexing process. You can do this by moving it from your home account to another location, or by editing .POPsearch/data/fileExclusions If you type delete_databases and then start popbuild again, let it index a handful of files, and then shut it down by running stop_popbuild. Then see if your data looks correct by typing popsearch -noIndexing Note that this stops the index process, but you can restart it at a later time from the browser or by running popbuild from the command line. > -------- Original Message -------- > Subject: [Popsearch-help] Tribulations trying to use popsearch > From: "Don Rozenberg" <ro...@mc...> > Date: Sun, November 07, 2004 8:41 pm > To: pop...@li... > > Hi, > > I really applaud what you are trying to do. Getting a better handle > on the data that I have on ones machine seems like a real gain for > everybody. > > However, I seem to be having many problems. I am running a Debian > variant, Knoppix, on a 1.3 GHz Pentium III machine with 512 MB of > storage. My home directory occupies 16 GB of disk space. I downloaded > and built popsearch-0.3.1. I also downloaded all the things that the > documentation said might be causing me a problem: BerkeleyDB, > IMageMagic, aspell, WordNet, pdftohtml, libxml2, and libxslt. > > When I executed 'popobuild', the system tookoff at 100% cpu load and a > list of files started coming out. However, the list of files stopped > with a pdf and the cpu load stayed at 100% for hours. When I tried > executing 'stop_popbuild' I got the following message: > > ** shutdown received, waiting for queues to empty ** > > A half hour later, I entered 'stop_popbuild -now' and got the message > > ** immediate shutdown received, waiting for indexing to finish ** > > Twenty minutes later, I did a ps and a 'kill -9' on the popbuild > process and the build subprocesses. It also seemed as though another > pdfConvert.sh process was the process running at 100%. I also killed > that one. And got my machine back to 3% cpu usage. > > > I also tried 'popsearch' which opened up Konquerer pointing to > 'http://127.0.0.1:3001' but nothing seemed to be happening there. So I > killed the command with control-C. At that point, Konquerer displayed > the page for creating the user account. > > At this point I don't really know what is going on. The command > 'du -sh .POPsearch' returns 16M, so there is a lot of stuff was generated. > > A final point is that when I entered emacs to type this message, I > got the following emacs problem: > > (error "aspell exited with code 127") > > Goofing around a little more with aspell, I discovered that kde sets > my language to 'us' and so the command 'aspell -c popsearch.txt' > fails. I responded by setting my environmental variable LANGUAGE to > 'en' and now emacs and 'aspell -c popsearch.txt' works. > > So I removed the flag files from .POPsearch entered 'popbuild -verbose' > and got > > recovering from database problem > > directory /home/rozen > > and a bunch of dots. > > I have now spent 10 minutes in a single 'build' process. So I killed > that process, removed my ~/.POPsearch directory and started 'popbuild > -verbose'. It ran for several minutes and appeared to hang at the same > file as earlier today. But I also got the message 'recovering from > database problem' and 'directory /home/rozen' plus the same fist full > of dots referred to above. And, of course, my cpu is pegged at 100%. > > -- > Don Rozenberg > 707-882-3601 > ro...@mc... > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Sybase ASE Linux Express Edition - download now for FREE > LinuxWorld Reader's Choice Award Winner for best database on Linux. > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click > _______________________________________________ > Popsearch-help mailing list > Pop...@li... > https://lists.sourceforge.net/lists/listinfo/popsearch-help |
|
From: Don R. <ro...@mc...> - 2004-11-08 01:42:00
|
Hi, I really applaud what you are trying to do. Getting a better handle on the data that I have on ones machine seems like a real gain for everybody. However, I seem to be having many problems. I am running a Debian variant, Knoppix, on a 1.3 GHz Pentium III machine with 512 MB of storage. My home directory occupies 16 GB of disk space. I downloaded and built popsearch-0.3.1. I also downloaded all the things that the documentation said might be causing me a problem: BerkeleyDB, IMageMagic, aspell, WordNet, pdftohtml, libxml2, and libxslt. When I executed 'popobuild', the system tookoff at 100% cpu load and a list of files started coming out. However, the list of files stopped with a pdf and the cpu load stayed at 100% for hours. When I tried executing 'stop_popbuild' I got the following message: ** shutdown received, waiting for queues to empty ** A half hour later, I entered 'stop_popbuild -now' and got the message ** immediate shutdown received, waiting for indexing to finish ** Twenty minutes later, I did a ps and a 'kill -9' on the popbuild process and the build subprocesses. It also seemed as though another pdfConvert.sh process was the process running at 100%. I also killed that one. And got my machine back to 3% cpu usage. I also tried 'popsearch' which opened up Konquerer pointing to 'http://127.0.0.1:3001' but nothing seemed to be happening there. So I killed the command with control-C. At that point, Konquerer displayed the page for creating the user account. At this point I don't really know what is going on. The command 'du -sh .POPsearch' returns 16M, so there is a lot of stuff was generated. A final point is that when I entered emacs to type this message, I got the following emacs problem: (error "aspell exited with code 127") Goofing around a little more with aspell, I discovered that kde sets my language to 'us' and so the command 'aspell -c popsearch.txt' fails. I responded by setting my environmental variable LANGUAGE to 'en' and now emacs and 'aspell -c popsearch.txt' works. So I removed the flag files from .POPsearch entered 'popbuild -verbose' and got recovering from database problem directory /home/rozen and a bunch of dots. I have now spent 10 minutes in a single 'build' process. So I killed that process, removed my ~/.POPsearch directory and started 'popbuild -verbose'. It ran for several minutes and appeared to hang at the same file as earlier today. But I also got the message 'recovering from database problem' and 'directory /home/rozen' plus the same fist full of dots referred to above. And, of course, my cpu is pegged at 100%. -- Don Rozenberg 707-882-3601 ro...@mc... |
|
From: Matthew T. R. <mr...@po...> - 2004-11-05 18:19:54
|
This maintenance release includes: * Fixed a bug with the display of images in local html pages. * Documentation changes. * Worked on RPM support. * Now indexing image metadata. * Adding unistd.h were necessary. * Remote access support for FreeBSD. * Made some changes to the build configuration. * Made a change to the flex parser. * Fixed a bug in the indexer. * Postscript is now being converted by ps2ascii. * Fixed a socket problem. |
|
From: Matthew T. R. <mr...@po...> - 2004-11-05 15:44:55
|
You know Reuben it almost sounds like another application has a library named libwn because the link says it's finding the library, but it can't find the symbols. If you su to root and type: find /. -name libwn.\* -ls I'm wondering if you'll come up with more than one library. > -------- Original Message -------- > Subject: RE: [Popsearch-help] flex: new problems in 0.3.0 > From: "Reuben Thomas" <rr...@sc...> > Date: Thu, November 04, 2004 5:58 pm > To: "Matthew T. Rider" <mr...@po...> > Cc: pop...@li... > > > On your WordNet problem (which is really my problem), the build is > > looking for WordNet in > > > > /usr/WordNet-2.0/lib > > > > or > > > > /usr/local/WordNet-2.0/lib > > > > I'm putting in a fix for this (which I hope to release tomorrow), but in > > the meantime, you can edit the Makefile in src/irs, look for something > > like -L/usr/WordNet-2.0/lib and insert -L/usr/lib > > This doesn't help. It's fairly obvious why it doesn't help: /usr/lib is > already in the library path, and several other libraries in /usr/lib are > already linked correctly. > > -- > http://rrt.sc3d.org/ | resident, a. unable to leave (Bierce) |
|
From: Matthew T. R. <mr...@po...> - 2004-11-05 15:33:23
|
Unfortunately I just found this problem on the suite of browsers that I test on. I'm going to fix this in a new release on SourceForge today. I'll send out an announcement as soon as I can. > -------- Original Message -------- > Subject: [Popsearch-help] Account creation > From: "Jeffrey Mutonho" <ejb...@gm...> > Date: Fri, November 05, 2004 3:04 am > To: pop...@li... > Cc: mr...@po... > > I've launched Popsearch for the first time and I got the account > creation window ( at http://127.0.0.1:2552/).I entered all the > necessary details , hit the Create Account button and then nothing > happens.At the bottom of the browser (Mozilla), I do see the "Sending > request to 127.0.0.1..." message.I waited and waited and waited...made > some coffee ...came back...still "Sending request to 127.0.0.1..." > :(( > Whats going on here? > > > jeff mutonho > > Registered Linux user number 366042 > > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Sybase ASE Linux Express Edition - download now for FREE > LinuxWorld Reader's Choice Award Winner for best database on Linux. > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click > _______________________________________________ > Popsearch-help mailing list > Pop...@li... > https://lists.sourceforge.net/lists/listinfo/popsearch-help |
|
From: Jeffrey M. <ejb...@gm...> - 2004-11-05 08:04:49
|
I've launched Popsearch for the first time and I got the account creation window ( at http://127.0.0.1:2552/).I entered all the necessary details , hit the Create Account button and then nothing happens.At the bottom of the browser (Mozilla), I do see the "Sending request to 127.0.0.1..." message.I waited and waited and waited...made some coffee ...came back...still "Sending request to 127.0.0.1..." :(( Whats going on here? jeff mutonho Registered Linux user number 366042 |
|
From: Reuben T. <rr...@sc...> - 2004-11-04 22:59:02
|
> On your WordNet problem (which is really my problem), the build is > looking for WordNet in > > /usr/WordNet-2.0/lib > > or > > /usr/local/WordNet-2.0/lib > > I'm putting in a fix for this (which I hope to release tomorrow), but in > the meantime, you can edit the Makefile in src/irs, look for something > like -L/usr/WordNet-2.0/lib and insert -L/usr/lib This doesn't help. It's fairly obvious why it doesn't help: /usr/lib is already in the library path, and several other libraries in /usr/lib are already linked correctly. -- http://rrt.sc3d.org/ | resident, a. unable to leave (Bierce) |
|
From: Matthew T. R. <mr...@po...> - 2004-11-04 22:46:27
|
It's not a debian bug. Berkeley changed some of their interfaces in version 4.1 and POPsearch wasn't configured to handle it properly. Now it checks the Berkeley version, and picks the correct interface. On your WordNet problem (which is really my problem), the build is looking for WordNet in /usr/WordNet-2.0/lib or /usr/local/WordNet-2.0/lib I'm putting in a fix for this (which I hope to release tomorrow), but in the meantime, you can edit the Makefile in src/irs, look for something like -L/usr/WordNet-2.0/lib and insert -L/usr/lib I hope this gets you up and running. > -------- Original Message -------- > Subject: RE: [Popsearch-help] flex: new problems in 0.3.0 > From: "Reuben Thomas" <rr...@sc...> > Date: Thu, November 04, 2004 5:22 pm > To: "Matthew T. Rider" <mr...@po...> > Cc: pop...@li... > > > On the Berkeley problem, I'm wondering if you had multiple db.h and > > libdb.so files on your system. Because it appears that the version of > > db.h that you picked up, didn't match the version of libdb.so > > A little explanation is in order. I had the dev files for db4.1 > installed (because 4.2 didn't work with earlier versions of popsearch). > I had the runtime libraries for both 4.1 and 4.2. > > As I showed in the error messages, most of the symbols were versioned > (with suffix 4001) so seem to have been picked up from the right > library. The db.h header was for 4.1. I wonder if the txn_* functions > weren't properly versioned, hence the problems. I couldn't find any bugs > filed in the Debian bug-tracker, and couldn't be bothered to do more > investigation (e.g. remove the run-times for 4.2), so instead just tried > installing the dev files for 4.2 instead, and since you'd fixed > popsearch so that now works, it built and I looked no further. > > In short, it might be a Debian bug, but in any case I wouldn't worry > about it; it's best just to build with 4.2. > > > Before you started to configure POPsearch for the first time, did you > > download a version of Berkeley before you started, or did you try and > > build with the version that was bundled with the operating system? > > I used the bundled versions. > > > On the WordNet problem, is the following file installed on your system? > > > > /usr/local/WordNet-2.0/lib/libwn.a > > Yes, libwn.a is installed, though as /usr/lib/libwn.a, and as I said: > > >> ... (all the symbols above seem to be defined in libwn). > > Here there won't be a versioning problem, since there's only one libwn. > > -- > http://rrt.sc3d.org/ | plagiarism, n. the mind burgles |
|
From: Reuben T. <rr...@sc...> - 2004-11-04 22:22:24
|
> On the Berkeley problem, I'm wondering if you had multiple db.h and > libdb.so files on your system. Because it appears that the version of > db.h that you picked up, didn't match the version of libdb.so A little explanation is in order. I had the dev files for db4.1 installed (because 4.2 didn't work with earlier versions of popsearch). I had the runtime libraries for both 4.1 and 4.2. As I showed in the error messages, most of the symbols were versioned (with suffix 4001) so seem to have been picked up from the right library. The db.h header was for 4.1. I wonder if the txn_* functions weren't properly versioned, hence the problems. I couldn't find any bugs filed in the Debian bug-tracker, and couldn't be bothered to do more investigation (e.g. remove the run-times for 4.2), so instead just tried installing the dev files for 4.2 instead, and since you'd fixed popsearch so that now works, it built and I looked no further. In short, it might be a Debian bug, but in any case I wouldn't worry about it; it's best just to build with 4.2. > Before you started to configure POPsearch for the first time, did you > download a version of Berkeley before you started, or did you try and > build with the version that was bundled with the operating system? I used the bundled versions. > On the WordNet problem, is the following file installed on your system? > > /usr/local/WordNet-2.0/lib/libwn.a Yes, libwn.a is installed, though as /usr/lib/libwn.a, and as I said: >> ... (all the symbols above seem to be defined in libwn). Here there won't be a versioning problem, since there's only one libwn. -- http://rrt.sc3d.org/ | plagiarism, n. the mind burgles |
|
From: Matthew T. R. <mr...@po...> - 2004-11-04 19:04:23
|
Thanks Reuben for finding these problems. I fixed the flex problem, and it'll go out in the next release. On the Berkeley problem, I'm wondering if you had multiple db.h and libdb.so files on your system. Because it appears that the version of db.h that you picked up, didn't match the version of libdb.so Before you started to configure POPsearch for the first time, did you download a version of Berkeley before you started, or did you try and build with the version that was bundled with the operating system? On the WordNet problem, is the following file installed on your system? /usr/local/WordNet-2.0/lib/libwn.a > -------- Original Message -------- > Subject: [Popsearch-help] flex: new problems in 0.3.0 > From: "Reuben Thomas" <rr...@sc...> > Date: Thu, November 04, 2004 5:48 am > To: pop...@li... > > 1. parser.cc doesn't compile from 0.3.0, because there is no declaration > of > yywrap. Adding "%option noyywrap" to the start of parser.l appears to > fix it. > > > 2. Unfortunately I get a link error I can't fix: > > /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `txn_abort' > /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `txn_begin' > /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `txn_commit' > > I don't understand this because the symbols are defined in libdb, and in > the header file, and -ldb is included in the link line. My only guess is > that this is some C/C++ problem, but the header file does declare the > functions above as extern "C". > > However, if I leave out -ldb, I get the following additional errors: > > /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `db_env_create_4001' > /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `db_create_4001' > /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `db_strerror_4001' > > which suggests that it might just be a problem with symbol versioning. > > If I build against libdb4.2 instead of 4.1, it works fine, though... > > > 3. The one I haven't so far managed to work out, however, is another > link error: > > /home/rrt/download/popsearch-0.3.0/src/irs/.libs/libirs.so: undefined reference to `free_syns' > /home/rrt/download/popsearch-0.3.0/src/irs/.libs/libirs.so: undefined reference to `in_wn' > /home/rrt/download/popsearch-0.3.0/src/irs/.libs/libirs.so: undefined reference to `wninit' > /home/rrt/download/popsearch-0.3.0/src/irs/.libs/libirs.so: undefined reference to `findtheinfo_ds' > > Again, -lwn is included on the link line (all the symbols above seem to > be defined in libwn). > > -- > http://rrt.sc3d.org/ | traddutore, v.t. traditore (Anon) > > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Sybase ASE Linux Express Edition - download now for FREE > LinuxWorld Reader's Choice Award Winner for best database on Linux. > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click > _______________________________________________ > Popsearch-help mailing list > Pop...@li... > https://lists.sourceforge.net/lists/listinfo/popsearch-help |
|
From: Reuben T. <rr...@sc...> - 2004-11-04 10:48:34
|
1. parser.cc doesn't compile from 0.3.0, because there is no declaration of yywrap. Adding "%option noyywrap" to the start of parser.l appears to fix it. 2. Unfortunately I get a link error I can't fix: /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `txn_abort' /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `txn_begin' /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `txn_commit' I don't understand this because the symbols are defined in libdb, and in the header file, and -ldb is included in the link line. My only guess is that this is some C/C++ problem, but the header file does declare the functions above as extern "C". However, if I leave out -ldb, I get the following additional errors: /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `db_env_create_4001' /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `db_create_4001' /home/rrt/download/popsearch-0.3.0/src/common/.libs/libcommon.so: undefined reference to `db_strerror_4001' which suggests that it might just be a problem with symbol versioning. If I build against libdb4.2 instead of 4.1, it works fine, though... 3. The one I haven't so far managed to work out, however, is another link error: /home/rrt/download/popsearch-0.3.0/src/irs/.libs/libirs.so: undefined reference to `free_syns' /home/rrt/download/popsearch-0.3.0/src/irs/.libs/libirs.so: undefined reference to `in_wn' /home/rrt/download/popsearch-0.3.0/src/irs/.libs/libirs.so: undefined reference to `wninit' /home/rrt/download/popsearch-0.3.0/src/irs/.libs/libirs.so: undefined reference to `findtheinfo_ds' Again, -lwn is included on the link line (all the symbols above seem to be defined in libwn). -- http://rrt.sc3d.org/ | traddutore, v.t. traditore (Anon) |
|
From: Matthew T. R. <mr...@po...> - 2004-11-02 16:52:29
|
Matt, I have a FreeBSD machine that I test things on, and I'm able to able to get things working. I'll check the version, but I think I'm running FreeBSD 5.2.1. It looks like I'm running gcc3.3.3 with success. Unfortunately you may need to upgrade you compiler at http://gcc.gnu.org The gcc upgrade can be somewhat tricky, so feel free to respond to the list if you still need help. Thanks. From: Matt Swinbourne <m@it...> Configure Fails on FreeBSD 4.10-RELEASE 2004-11-01 18:57 Hi All, I"ve been trying to compile popsearch-0.3.0 on FreeBSD 4.10-RELEASE. Output from ./configure: ./configure checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking for gawk... no checking for mawk... no checking for nawk... nawk checking whether make sets $(MAKE)... yes checking whether to enable maintainer-specific portions of Makefiles... no ./configure: line 1859: arch: command not found checking for db.h checking for libdb.so checking for db_recover checking for gcc... gcc checking for C compiler default output... configure: error: C compiler cannot create executables See `config.log" for more details. Config.log follows: This file contains any messages produced by compilers while running configure, to aid debugging if configure makes a mistake. |
|
From: Matthew T. R. <mr...@po...> - 2004-11-02 16:39:48
|
Jeff,
You need to install aspell along with the english dictionary.
From: Jeffrey Mutonho <ejbengine@gm...>
Installation error
2004-11-01 07:58
I"m trying to install PopSearch on RedHat 7.2 machine.I"ve installed
Berkeley DB , WordNet and ImageMagick.
After running "./configure" under my popsearch-0.3.0 directory and
then executing "make" I"m getting the error message :
cc1plus: warning: changing search order for system directory
"/usr/local/include"
cc1plus: warning: as it has already been specified as a non-system
directory
In file included from aspell_bridge.cc:21:
aspell_bridge.h:46:20: aspell.h: No such file or directory
In file included from aspell_bridge.cc:21:
aspell_bridge.h:66: syntax error before `*" token
aspell_bridge.h:67: syntax error before `*" token
aspell_bridge.cc: In constructor `aspell_bridge::aspell_bridge()":
aspell_bridge.cc:32: class `aspell_bridge" does not have any field
named `
_speller"
aspell_bridge.cc:33: `new_aspell_config" undeclared (first use this
function)
aspell_bridge.cc:33: (Each undeclared identifier is reported only once
for each
function it appears in.)
aspell_bridge.cc:33: class `aspell_bridge" does not have any field
named `
_config"
aspell_bridge.cc:35: `AspellCanHaveError" undeclared (first use this
function)
aspell_bridge.cc:35: `aspellError" undeclared (first use this function)
aspell_bridge.cc:47: `_config" undeclared (first use this function)
aspell_bridge.cc:47: `aspell_config_replace" undeclared (first use this
function)
aspell_bridge.cc:50: `new_aspell_speller" undeclared (first use this
function)
aspell_bridge.cc:51: `aspell_error_number" undeclared (first use this
function)
aspell_bridge.cc:55: `aspell_error_message" undeclared (first use this
function)
aspell_bridge.cc:65: `_speller" undeclared (first use this function)
aspell_bridge.cc:65: `to_aspell_speller" undeclared (first use this
function)
aspell_bridge.cc: In destructor `aspell_bridge::~aspell_bridge()":
aspell_bridge.cc:93: `delete_aspell_config" undeclared (first use this
function)
aspell_bridge.cc:96: `delete_aspell_speller" undeclared (first use this
function)
aspell_bridge.cc: In member function `std::queue<std::string,
std::deque<std::string, std::allocator<std::string> > >*
aspell_bridge::check(const std::string&)":
aspell_bridge.cc:111: ISO C++ forbids declaration of `AspellWordList"
with no
type
aspell_bridge.cc:111: uninitialized const `AspellWordList"
aspell_bridge.cc:111: parse error before `*" token
aspell_bridge.cc:112: `AspellStringEnumeration" undeclared (first use
this
function)
aspell_bridge.cc:112: `replacements" undeclared (first use this
function)
aspell_bridge.cc:127: `aspell_speller_check" undeclared (first use this
function)
aspell_bridge.cc:130: `suggestions" undeclared (first use this
function)
aspell_bridge.cc:130: `aspell_speller_suggest" undeclared (first use
this
function)
aspell_bridge.cc:131: `aspell_word_list_elements" undeclared (first use
this
function)
aspell_bridge.cc:133: `aspell_string_enumeration_next" undeclared
(first use
this function)
aspell_bridge.cc:136: `delete_aspell_string_enumeration" undeclared
(first use
this function)
make[3]: *** [aspell_bridge.lo] Error 1
make[3]: Leaving directory
`/home/maven/popsearch/popsearch-0.3.0/src/common"
make[2]: *** [all] Error 2
make[2]: Leaving directory
`/home/maven/popsearch/popsearch-0.3.0/src/common"
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/maven/popsearch/popsearch-0.3.0/src"
make: *** [all-recursive] Error 1
I can see that there"s a message about a missing header file :
In file included from aspell_bridge.cc:21:
aspell_bridge.h:46:20: aspell.h: No such file or directory
Where is that file suppose to come from?
jeff mutonho
|
|
From: Matthew T. R. <mr...@po...> - 2004-11-02 16:22:34
|
Reuben, A couple of the modules are too big, and I should have known better. Unfortunately, some of the modules grew at a rapid rate, as I was trying to get things written and out the door. During development, I was using gcc3.2 on a system that didn't have a large amount of memory, and I was able to get things to compile. Since then, I upgraded to gcc3.3.4 and I'm now running out of memory as well. It seems like they've changed the optimizer in gcc, and now it's using a lot more memory than it did before. I started to decompose some of the larger modules into smaller ones, and I'm now able to get things to compile in the latest release. Although some of the modules are still too big, so there's still more work that needs to be done in this area. Thanks. From: Reuben Thomas <rrt@sc...> Build problem: g++ out of memory 2004-09-11 17:21 During make I got the following error: /bin/sh ../../libtool --mode=compile g++ -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\" -DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"popsearch\" -DVERSION=\"0.1.3\" -DYYTEXT_POINTER=1 -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -I. -I. -D_REENTRANT -DTHREADSAFE -pthread -D__need_timeval -Wno-deprecated -DLinux -march=i686 -mcpu=i686 -I. -I../../src/common -I/usr/local/include -I/usr/local -g -O2 -c -o TopicManager.lo `test -f "TopicManager.cc" || echo "./"`TopicManager.cc g++ -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\" -DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"popsearch\" -DVERSION=\"0.1.3\" -DYYTEXT_POINTER=1 -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -I. -I. -D_REENTRANT -DTHREADSAFE -pthread -D__need_timeval -Wno-deprecated -DLinux -march=i686 -mcpu=i686 -I. -I../../src/common -I/usr/local/include -I/usr/local -g -O2 -c TopicManager.cc -MT TopicManager.lo -MD -MP -MF .deps/TopicManager.TPlo -fPIC -DPIC cc1plus: out of memory allocating 1582874352 bytes after a total of 20807680 bytes i.e. out of memory error (using a total of about 1.5Gb of physical RAM plus swap space) for compiling TopicManager.cc. I have g++ 3.3.4 running on Debian sarge. Looking at the code, this error does not come as a total surprise, as there seems to be a monstrous multi-thousand line method parseToDictionary. Indeed, this one method seems to occupy most of the 14,000+ lines of the file. It"s bigger, I think, than the largest program I have ever written (admittedly, not that big). Since it"s so big, it would seem reasonable either to refactor the code or have some manual twiddling in the makefiles to use a low level of optimisation when compiling this file. Any chance of that? Of course, it may also indicate a problem with g++; I"m not competent to say. -- http://rrt.sc3d.org/ | Si hoc legere scis nimium eruditionis habes (Anon) |