Menu
▾
▴
genex-dev — GeneX developers
You can subscribe to this list here.
| 2001 |
Jan
(135) |
Feb
(57) |
Mar
(84) |
Apr
(43) |
May
(77) |
Jun
(51) |
Jul
(21) |
Aug
(55) |
Sep
(37) |
Oct
(56) |
Nov
(75) |
Dec
(23) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2002 |
Jan
(32) |
Feb
(174) |
Mar
(121) |
Apr
(70) |
May
(55) |
Jun
(20) |
Jul
(23) |
Aug
(15) |
Sep
(12) |
Oct
(58) |
Nov
(203) |
Dec
(90) |
| 2003 |
Jan
(37) |
Feb
(15) |
Mar
(14) |
Apr
(57) |
May
(7) |
Jun
(40) |
Jul
(36) |
Aug
(1) |
Sep
(56) |
Oct
(38) |
Nov
(105) |
Dec
(2) |
| 2004 |
Jan
|
Feb
(117) |
Mar
(69) |
Apr
(160) |
May
(165) |
Jun
(35) |
Jul
(7) |
Aug
(80) |
Sep
(47) |
Oct
(23) |
Nov
(8) |
Dec
(42) |
| 2005 |
Jan
(19) |
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(2) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2010 |
Jan
|
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: <ja...@op...> - 2001-04-11 15:38:46
|
"Harry J. Mangalam" <hj...@nc...> writes: > So to finish out your thought, the implication of the XMLization of > the table defs is that ports to other RDBMSs are now a matter of a > platform-specific script, which will largely be the same for all. > > So MySQL, Oracle, and Sybase ports are now just a click away (so to > speak..) Correct! Thanks for finishing my thought for me (it's amazing how slowly one's brain can work after midnight ;-) So now to add a new RDBMS follow these steps: * Create DBMS specific xml2sql() function in XMLUtils.pm * Create DBMS specific DB creation script (presumably by modifying create_genex_db.pl) * Port DB/xml/functions.xml from PL/pgsql to DBMS specific procedure language. * Added DBMS case to conditionals in DBUtils:: - create_select_sql() - create_insert_sql() - create_update_sql() - last_id() Since most DBMS systems are SQL92 compliant these days, the changes are small. jas. |
|
From: Harry J. M. <hj...@nc...> - 2001-04-11 15:21:03
|
So to finish out your thought, the implication of the XMLization of the table defs is that ports to other RDBMSs are now a matter of a platform-specific script, which will largely be the same for all. So MySQL, Oracle, and Sybase ports are now just a click away (so to speak..) hjm On 10 Apr 2001, Jason E. Stewart wrote: > Hey All, > > The DB schema and the API are both finally calming down to the point > that it's worth making a 1.1.0 beta release. Here are the changes I've > just committed: > > DB creation: > * Like functions, triggers are adding on a one-by-one basis, and if > they already exist in the DB, they are dropped first. This means > that like functions, you can debug triggers conveniently. > * The Postgres table definition files are gone. All table definitions > are stored in XML, and converted to SQL at DB creation time. This > means that for each table there is a single definition file to > define the DB storage and the Perl API. > > API: > * update_db() now is more cautious. It firsts checks if the object has > any differences from what's in the DB, and if so, it only updates > those fields that are different. > * all Classes are created from the central XML Table Definition > files. > > jas. > > _______________________________________________ > Genex-dev mailing list > Gen...@li... > http://lists.sourceforge.net/lists/listinfo/genex-dev > |
|
From: <ja...@op...> - 2001-04-11 06:20:33
|
Hey, In a success for the new API, the WWW interfaces for users to enter and update their Contact information are now working!!! Due to the bifurcation of the old Contact table into three tables, a fair bit of the internal logic needed to be modified. And it works!! Along the way I also had to fix the script that entered Controlled Vocab info into the DB (an, ahem, minor oversight ;-) I'll clean it up a bit and check it in tomorrow. jas. |
|
From: <ja...@op...> - 2001-04-11 01:49:45
|
Hey All, The DB schema and the API are both finally calming down to the point that it's worth making a 1.1.0 beta release. Here are the changes I've just committed: DB creation: * Like functions, triggers are adding on a one-by-one basis, and if they already exist in the DB, they are dropped first. This means that like functions, you can debug triggers conveniently. * The Postgres table definition files are gone. All table definitions are stored in XML, and converted to SQL at DB creation time. This means that for each table there is a single definition file to define the DB storage and the Perl API. API: * update_db() now is more cautious. It firsts checks if the object has any differences from what's in the DB, and if so, it only updates those fields that are different. * all Classes are created from the central XML Table Definition files. jas. |
|
From: Harry M. <man...@ho...> - 2001-04-10 23:41:57
|
Smack it and tell me what breaks. http://genex.ncgr.org/genex/download/genex-server/GeneX-Server-1.0.1_RC1.tar.gz Release Notes for Release 1.0.1 ------------------------------- This is second public release, mainly a more stable bugfix edition and more features for developers. We have had a number of developer releases - thanks to our testers. Changes for 1.0.1 ----------------- * many patches and fixes, mostly having to do with broken links and installation glitches (see Changelog for the full list). * CyberT changes: - hdarray R lib updated to reflect increased confidence in Bayesian stats - form & analytical scripts merged to make further changes easier (now 1 form and 1 analytical script; Paired & C+E analyses now coherent in terms of features. - output format changed to present all results sorted on raw p value and moved data 1 tab/column right so it doesn't conflict with header info New Features ------------ * No really new features for users, hopefully the last time we have to say this. * CVS repository now available from genex.ncgr.org (see: http://genex.ncgr.org/CVS.note for info) -- Cheers, Harry Harry J Mangalam -- (949) 856 2847 (v&f) -- hj...@nc... || man...@ho... |
|
From: Harry M. <man...@ho...> - 2001-04-09 22:24:03
|
Anyone having changes/patches to the 1.0 tree should check them in. (I think I'm the only person working on this tree, but just to be sure...) This should be the last release before a large re-working of the CVS tree to merge re-root the new dev branch. The new release structure has Jason's server-side data-loader code incorporated and the new DB structure to handle it. -- Cheers, Harry Harry J Mangalam -- (949) 856 2847 (v&f) -- hj...@nc... || man...@ho... |
|
From: <ja...@op...> - 2001-04-09 16:23:11
|
"Harry J. Mangalam" <hj...@nc...> writes: > Jason, that's great! As soon as it's in CVS, give me a call or email and > maybe i can start adding both species info (or more usefully) arrays to it. OK. So all my changes are in. You will be able to modify DB/curated_data/species.xml and add any species you would like. There is quite a collection already, but only ecoli and saccharomyces have chromosome data. Create a DB: * DB/tdscripts/create_genex_db.pl --dbname=??? --dir=DB/tdscripts --name=?? * Genex/scripts/create-db-accounts.pl --dbname=??? --file=DB/curated_data/contact-db.xml --name=?? Add the species * Genex/scripts/species-insert.pl --dbname=??? --file=DB/curated_data/species.xml --name=?? Now, the problem starts when you've alread got the species in the DB, and just want to modify one of the pieces. species-insert.pl has a --update option that worked fine before I moved over to the new XML API, but I didn't implement it for the XML API. I just discovered I need it, and am implementing it as we speak, but it requires modifying the underlying code in each Bio::Genex:: class's update_db() method. Here's the issue: 1) check that something is different, and only then call UPDATE 2) only update the fields that are different The old code assumes you know what you're doing, and just calls UPDATE with the value for every attribute. I'll let you know when it's in. jas. |
|
From: Harry J. M. <hj...@nc...> - 2001-04-09 15:48:52
|
Jason, that's great! As soon as it's in CVS, give me a call or email and maybe i can start adding both species info (or more usefully) arrays to it. hjm |
|
From: <ja...@op...> - 2001-04-09 03:19:59
|
Celebrate Good times, Come ON! (for all you disco fans out there ;-) In an eight hour marathon coding session, I added XML API support for the Species and Chromosome tables, and enter all species info into my copy of the genex2 DB. species-insert.pl is now 50% of it's original size, and the logic is dramatically simpler, because all the work is now done by the XML API. The importance of this is that the DataLoader is now ready to begin importing Arrays into the DB. The species info was the last piece required before Arrays could be entered. Tables conquered (data in DB and working API): GroupLink ContactType UserSec GroupSec Security Contact ContactOwner Chromosome Species Changes in CVS shortly, jas. |
|
From: <ja...@op...> - 2001-04-06 16:50:49
|
"Michael Pear" <mic...@ho...> writes:
> One thing I found in that process is that the "get_all_objects" and
> "lookup" database helper routines die when they find that there are
> no entries in the database that match the criteria given. This means
> that I can't use them to check whether something already exists or
> should be created new. I think it would be good to change this
> behavior for version 2 so that an "Undef" is returned if nothing
> matches, but so that processing doesn't stop.
Hmmm.... if get_all_objects() is behaving that way, it is a definate
bug. Here is the code that checks for that exact condition:
# if there were no objects, return. decide whether to return an
# empty list or an empty arrayref using wantarray
unless ($sth->rows()) {
return () if wantarray;
return []; # if not wantarray
}
If you can give me an example of where it breaks I'd be grateful.
So you should be getting back an empty list.... The lookup_* routines
are ancient and crusty, and probably want to be completely overhauled
(if not just deleted). They existed prior to the class-based API when
Peter Hraber and I needed to get data into the DB. CLASS->new(),
CLASS->get_objects(), and CLASS->get_all_objects() should be used
CLASS->instead.
Hope you straighten out your RAID problems, soon. Could use some more
of your very useful feedback on genex2...
jas.
|
|
From: <ja...@op...> - 2001-04-06 16:43:28
|
"Todd Peterson" <tf...@nc...> writes: > Thank you for your comments. Postgres works pretty well. As far as I know, > nobody is planning on implementing MySQL. Oracle was being considered, but > for $$$. I'd bite the bullet and go with Postgres. Hi Damian, There are some important differences between MySQL and Postgres. A fair amount of my code assumes that it's safe to run things in a transaction and fail without modifying the DB. As I understand MySQL doesn't have transactions and never will. It is also likely that MySQL implements less of SQL92, and some of the SQL the backend generates might not work. That said, the amount of change that needs to happen to test MySQL is pretty small, and I'd be happy to work together with you to get MySQL up and running. I have built a DBMS abstraction layer into the API, so all that should need to be done is adding the MySQL pieces for: * create_select_sql() * create_update_sql() * create_insert_sql() * last_id() Probably 100 lines of new code all together, all of which (except for last_id) can be copied from the existing Pg implementation. If, however, you want it to happen quickly, your best bet is to buckle down and use Pg. It will take me a week a least to get around to looking at MySQL. > > my name is Damian Silvera. I am contacting you from the University of > > Miami School of Medicine. We are quite interested in implementing your > > product (good job by the way) in our environment. I have a ticky-tacky > > request/question to ask of you before I embark on this project, however. > > My first venture into the land of open source database engines was mSQL. > > When MySQL got up and running, I quickly made the switch and have not > > looked in any other direction. I have heard wonderful things about > > PostgreSQL over the years, but have never implemented it. Are there > > plans to incorporate support for MySQL or should I RTFM on PostgreSQL? |
|
From: Todd P. <tf...@nc...> - 2001-04-06 15:43:41
|
Thank you for your comments. Postgres works pretty well. As far as I know, nobody is planning on implementing MySQL. Oracle was being considered, but for $$$. I'd bite the bullet and go with Postgres. Todd Peterson Software Developer ----- Original Message ----- From: "Damian Silvera" <dsi...@ne...> To: <gen...@li...> Sent: Thursday, April 05, 2001 6:40 PM Subject: [GeneX-dev] genx db engine > Good evening, > > my name is Damian Silvera. I am contacting you from the University of > Miami School of Medicine. We are quite interested in implementing your > product (good job by the way) in our environment. I have a ticky-tacky > request/question to ask of you before I embark on this project, however. > My first venture into the land of open source database engines was mSQL. > When MySQL got up and running, I quickly made the switch and have not > looked in any other direction. I have heard wonderful things about > PostgreSQL over the years, but have never implemented it. Are there > plans to incorporate support for MySQL or should I RTFM on PostgreSQL? > > Regards > > > _______________________________________________ > Genex-dev mailing list > Gen...@li... > http://lists.sourceforge.net/lists/listinfo/genex-dev > |
|
From: Todd P. <tf...@nc...> - 2001-04-06 15:40:41
|
Improved draft mode added. Experiment set file name passed as an argument to the tool causes tool to open the experiment set. http://genex.ncgr.org/genex/download/curation_tool Todd Peterson NCGR mailto:tf...@nc... (505) 995-4417 |
|
From: Damian S. <dsi...@ne...> - 2001-04-06 00:41:11
|
Good evening, my name is Damian Silvera. I am contacting you from the University of Miami School of Medicine. We are quite interested in implementing your product (good job by the way) in our environment. I have a ticky-tacky request/question to ask of you before I embark on this project, however. My first venture into the land of open source database engines was mSQL. When MySQL got up and running, I quickly made the switch and have not looked in any other direction. I have heard wonderful things about PostgreSQL over the years, but have never implemented it. Are there plans to incorporate support for MySQL or should I RTFM on PostgreSQL? Regards |
|
From: <ja...@op...> - 2001-04-05 19:58:14
|
Hey, I'm finally getting around to commiting all the changes necessary to get data into the new genex2 DB. It should now be possible for anyone to create a sample genex2 DB by running to following commands for the toplevel genex-server directory: # create the DB DB/tdscripts/create_genex_db.pl --dbname=genex2 --dir=DB/tdscripts # populate the basic tables Genex/scripts/create-db-accounts.pl --name=USER@HOSTNAME --dbname=genex2 --file=DB/curated_data/contact-db.xml Some of the highlights of the changes: GeneXML DTD changes, to support adding of security info: * DTD/usf.dtd (Repository): removed <canonical_sequence_feature_list> <USF>:species_id changed to #IMPLIED * DTD/genexml.dtd (Repository): added <user_list> and <security_list> to <header>, and <user> and <security>. <contact> now uses attributes for all information instead of sub-elements <group>:name renamed to :group_name to match DB. Added <group>:description, and <group>:owner_id renamed <member>:contact_id to :user_id new basic DB account file: * DB/curated_data/contact-db.xml (Repository): Genex.pm API changes: 2001-04-05 Jason E. Stewart <ja...@op...> * XMLUtils/XMLUtils.pm.in (Repository): now uses XML::Xerces added Bio::Genex::GeneXML::DOMParser subclass Bio::Genex::GeneXML no longer parses in test XML file just to get &VERSION; added DESTROY() to remove any circular refs to parser and document stored in Bio::Genex::GeneXML instances Added xml2db() methods for: Contact ContactType ContactOwner Security GroupSec GroupLink UserSec * t/XMLUtils.t (Repository): * t/UserSec.t (Repository): * t/TestDB.pm (Repository): * t/GroupSec.t (Repository): * t/GroupLink.t (Repository): * t/Genex.t (Repository): * t/ContactType.t (Repository): * t/ContactOwner.t (Repository): * t/Contact.t (Repository): updated to use new result() idiom. updated to test new tables and classes * scripts/make_classes.pl (Repository): Now handles new tables: ContactOwner ContactType TableAdmin Security removed pkey line from ControlledVocab, since it is now specified as a validation table without a primary key * scripts/create-db-accounts.pl.in (Repository): Now handles inserting data into ALL account related tables: Contact ContactType ContactOwner Security GroupSec GroupLink UserSec * Genex.pm.in (Repository): Added is_object() helper method. This falls between ref() and isa(). |
|
From: Harry M. <man...@ho...> - 2001-04-04 18:25:44
|
Hi Reinhard, Thanks for taking a look and actually reading thru the docs! And yes, you're right - currently the only significance testing is as you described below - actually it's the same application with a Bayesian option. Of the approaches that you mention, the first is certainly a good idea - we have several collaborators requesting (or screaming, depending on the collaborator), for such an analysis. Karen Schlauch at VBI (ka...@vb...) is working on such an approach as am I (altho she's probably doing it de novo the right way, and I'm trying to find an approximately appropriate R library that I can mangle). The second approach is also a good idea, as it's pretty powerful and fast - rank order is easy to calculate and the ranking decreases noise a bit. Both of these are pretty easy to implement given a bit of time and attention; the harder part is how to partition the data (and how to allow the user to do so easily) so that the test makes sense in a biological domain. The KW test applied to *complete sets* of normalized data even over dramatically different conditions will not find a difference. One way is to allow partitioning on the basis of predetermined gene sets such as the MIPS or KEGG or [insert your favorite here] classification, or allow lists of IDs to be uploaded (GeneX already supports this). Do you have suggestions as to how to do this? thanks again for your input! Harry > Hi there! > I have not yet installed GeneX, but will probably be using it later this > year. However, during a brief overview of the GeneX features, it appears > as though there are only two ways to detect (statitically valid) > differentially expressed genes: via t-test with bonferroni correction > and a bayesian modeller. > I think two additional features would be extremely helpful: > 1. Implementation of an ANOVA-like routine that identifies > differentially expressed genes in, for example, developmental studies > where no baseline condition can be defined, and that contains >3 > conditions with a number of replicates; > 2. Inplementation of non-parametric tests like Kruskal-Wallis, also in > conjunction with 1. > Please let me know what you think of these points. > Greetings > Dr. Reinhard Hoffmann > Basel Institute for Immunology > Grenzacherstr. 487 > 4005 Basel Switzerland > Fon +4161-6051425 > Fax +4161-6051364 > email rho...@in... -- Cheers, Harry Harry J Mangalam -- (949) 856 2847 (v&f) -- hj...@nc... || man...@ho... |
|
From: Harry J. M. <hj...@nc...> - 2001-04-04 18:02:11
|
Great! I'm planning to do a bit more cleanup and testing over this week and cut a new one on the weekend - We don't need to release in sync, but if that's convenient for you, lets try to make it available by Monday...? hjm On Wed, 4 Apr 2001, Todd Peterson wrote: > > > Todd, it sounds like the Curation Tool is coming up to another minor > release > > > as well. > > > > > Yep, could do that tomorrow or Fri. The new feature is most awesome, it > really streamlines the operation of the tool. Makes it actually usable. > > Todd > > > _______________________________________________ > Genex-dev mailing list > Gen...@li... > http://lists.sourceforge.net/lists/listinfo/genex-dev > |
|
From: Todd P. <tf...@nc...> - 2001-04-04 17:39:00
|
> > Todd, it sounds like the Curation Tool is coming up to another minor release > > as well. > > Yep, could do that tomorrow or Fri. The new feature is most awesome, it really streamlines the operation of the tool. Makes it actually usable. Todd |
|
From: Reinhard H. <rho...@in...> - 2001-04-04 15:10:56
|
Hi there! I have not yet installed GeneX, but will probably be using it later this year. However, during a brief overview of the GeneX features, it appears as though there are only two ways to detect (statitically valid) differentially expressed genes: via t-test with bonferroni correction and a bayesian modeller. I think two additional features would be extremely helpful: 1. Implementation of an ANOVA-like routine that identifies differentially expressed genes in, for example, developmental studies where no baseline condition can be defined, and that contains >3 conditions with a number of replicates; 2. Inplementation of non-parametric tests like Kruskal-Wallis, also in conjunction with 1. Please let me know what you think of these points. Greetings Dr. Reinhard Hoffmann Basel Institute for Immunology Grenzacherstr. 487 4005 Basel Switzerland Fon +4161-6051425 Fax +4161-6051364 email rho...@in... |
|
From: Bill B. <wd...@nc...> - 2001-04-04 03:33:07
|
on 4/3/01 1:20 PM, Harry Mangalam at man...@ho... wrote: > Because of all the problems with sourceforge, I'm redirecting requests to > Sourceforge to NCGR. Seems to work OK and will leave us with a single Dev > page to update. Should SF get its collective act > together, we can move back at some point. > > Anne is changing the institutional NCGR web page and in coordinating the GeneX > info page at .../research/genex with the genex page, she also caught a bunch > of bad links that I didn't. The ones she > caught are now patched on genex and in CVS. They are: > > 2001-04-03 Harry Mangalam <hj...@nc... || man...@ho...> > * Anne Rafferty caught a number of bad links in the GeneX tree - corrected > them in CVS and also on genex.ncgr.org by direct editing. > -download link > http://genex.ncgr.org/download/index.shtml -> > http://genex.ncgr.org/genex/download/ > -bugtracking link > http://genex.ncgr.org:8080 -> > http://genex.ncgr.org/bugzilla > -SourceForge GeneX Development Site remaned to GeneX Development Site > and linked to http://genex.ncgr.org > -number of download-link-related changes in the rcluster branch > -removed link to Old GeneX index (why was THAT there?) > -updated contacts to reflect reality > > NB: Once I've tested the current CVS tree, I'll cut another release of GeneX > to reflect all the changes from version 1 and place it in the download > section. > > Todd, it sounds like the Curation Tool is coming up to another minor release > as well. > > Don't forget to get me your additions to the Biweekly thing. Glad to hear that you're taking the step and voting with our feet. Would you send a note to SF letting them (whoever they are) know that we are no longer associating with their site and the reason why. Also, can you give me an assessment of any possible damage to NCGR that being on a site that does not work may have caused us. -- Bill Beavis wd...@nc... |
|
From: <ja...@op...> - 2001-04-03 21:54:04
|
"Harry J. Mangalam" <hj...@nc...> writes: > OK - post it and then let's chat to make sure I understand the fine > details > - I thought you just marked the cut with a label and ran an export from the > CVS tree. That's the beef, pretty much. 1) Make a ReleaseNotes doc, that indicates what's changed. This I create mainly from the ChangeLog files (which is why I'm meticulous about maintaining the ChangeLogs). 2) cvs tag Rel-X_Y_Z 3) cvs export -r Rel-X_Y_Z -d GeneX-Server-X.Y.Z genex-server 4) run the installer 5) run through a subset of the functionality to convince myself it works. If all is well proceed to 6), otherwise: 5a) make changes to source 5b) cvs commit 5c) cvs tag Rel-X_Y_Z-tryN 5d) go back to 3) with -r Rel-X_Y_Z-tryN upload the files: 6) tar zxf GeneX-Server-X.Y.Z.tar.gz GeneX-Server-X.Y.Z 7) scp GeneX-Server-X.Y.Z.tar.gz dev...@ge...:/home/httpd/html/genex/download/genex-server 8) scp GeneX-Server-X.Y.Z/README dev...@ge...:/home/httpd/html/genex/download/genex-server/00_README 8) scp GeneX-Server-X.Y.Z/RELEASE dev...@ge...:/home/httpd/html/genex/download/genex-server/00_RELEASE 9) ssh genebox.ncgr.org 10) cd /home/httpd/html/genex/download/genex-server 11) mv OLD-GENEX-SERVER archive 12) cvs checkout gxtop 13) modify content/main.ssi to add the new release event with a URL to the RELEASE, README, and tarball jas. |
|
From: Harry J. M. <hj...@nc...> - 2001-04-03 21:11:59
|
OK - post it and then let's chat to make sure I understand the fine details - I thought you just marked the cut with a label and ran an export from the CVS tree. hjm On 3 Apr 2001, Jason E. Stewart wrote: > "Harry Mangalam" <man...@ho...> writes: > > > NB: Once I've tested the current CVS tree, I'll cut another release > > of GeneX to reflect all the changes from version 1 and place it in > > the download section. > > Harry, I've never actually written up my 'How to make a GeneX-Server > Release' document. I can talk to you in person, or I can post it > here. That way it can get committed to CVS, and future generations > will have a guideline. > > jas. > > _______________________________________________ > Genex-dev mailing list > Gen...@li... > http://lists.sourceforge.net/lists/listinfo/genex-dev > |
|
From: <ja...@op...> - 2001-04-03 21:04:41
|
"Harry Mangalam" <man...@ho...> writes: > NB: Once I've tested the current CVS tree, I'll cut another release > of GeneX to reflect all the changes from version 1 and place it in > the download section. Harry, I've never actually written up my 'How to make a GeneX-Server Release' document. I can talk to you in person, or I can post it here. That way it can get committed to CVS, and future generations will have a guideline. jas. |
|
From: Harry M. <man...@ho...> - 2001-04-03 20:22:19
|
Because of all the problems with sourceforge, I'm redirecting requests to Sourceforge to NCGR. Seems to work OK and will leave us with a single Dev page to update. Should SF get its collective act together, we can move back at some point. Anne is changing the institutional NCGR web page and in coordinating the GeneX info page at .../research/genex with the genex page, she also caught a bunch of bad links that I didn't. The ones she caught are now patched on genex and in CVS. They are: 2001-04-03 Harry Mangalam <hj...@nc... || man...@ho...> * Anne Rafferty caught a number of bad links in the GeneX tree - corrected them in CVS and also on genex.ncgr.org by direct editing. -download link http://genex.ncgr.org/download/index.shtml -> http://genex.ncgr.org/genex/download/ -bugtracking link http://genex.ncgr.org:8080 -> http://genex.ncgr.org/bugzilla -SourceForge GeneX Development Site remaned to GeneX Development Site and linked to http://genex.ncgr.org -number of download-link-related changes in the rcluster branch -removed link to Old GeneX index (why was THAT there?) -updated contacts to reflect reality NB: Once I've tested the current CVS tree, I'll cut another release of GeneX to reflect all the changes from version 1 and place it in the download section. Todd, it sounds like the Curation Tool is coming up to another minor release as well. Don't forget to get me your additions to the Biweekly thing. -- Cheers, Harry Harry J Mangalam -- (949) 856 2847 (v&f) -- hj...@nc... || man...@ho... |
|
From: Michael P. <mic...@ho...> - 2001-03-27 15:03:15
|
In the database definition for 1.0,
the private key als_pk in al_spots is missing the default
generation of a value from a sequence and a definition
as a unique index. This doesn't seem to cause any operational
problems within the tool set of Genex V1.0...I ran into it as an issue for
some experimentation I am doing outside the current tool set, and thought
it best to correct it early. Jason's approach to an updated schema
for V1.1 will correct the issue in that version.
The following assumes you have a database installed and have added some data.
1) From "psql genex", change any null values for newly added data to appropriate
key values using:
UPDATE al_spots set als_pk = nextval ('al_spots_als_pk_seq') where als_pk = null;
2) pg_dump genex >/tmp/genex.dump
3) Edit genex.dump and modify the line
CREATE TABLE "al_spots" (
"als_pk" int4,
to
CREATE TABLE "al_spots" (
"als_pk" int4 DEFAULT nextval('al_spots_als_pk_seq'::text) NOT NULL,
and adding a line along with the other INDEX defnitions:
CREATE UNIQUE INDEX "al_spots_pkey" on "al_spots" using btree ( "als_pk" "int4_ops" );
4) dropdb genex
createdb genex
5) Then reload the database with "psql genex </tmp/genex.dump"
Regards,
Michael Pear
|
54 messages has been excluded from this view by a project administrator.
