*** wendall911 has left #gnuenterprise
*** SachaS has joined #gnuenterprise
*** btami has joined #gnuenterprise
*** johannesV has joined #gnuenterprise
*** sjc has joined #gnuenterprise
*** kilo has joined #gnuenterprise
<kilo> good morning
<johannesV> hi kilo
<kilo> hi johannesV
<btami> hi johannesV
<btami> can you fix the release script, it doesn't links the common-0.5.10
<btami> it thinks 0.5.9 > 0.5.10
<johannesV> wow
<johannesV> which release script ?
<btami> in releases current page
<btami> good question :)
<johannesV> common/utils/release ... ?
<johannesV> i think we better wait for reinhard to do this fix ...
<johannesV> since he had a better overview about the release process
<btami> ok
*** reinhard has joined #gnuenterprise
<johannesV> ah, there he is ... :)
<reinhard> heh
* reinhard hides
*** dimas has quit IRC
<reinhard> btami: you talking about the web site?
<btami> yes
<reinhard> ok
<reinhard> thanks
<btami> np
<reinhard> hmmm
<kilo> hi reinhard
<reinhard> i have an interesting problem then
<johannesV> i like interesting problems :)
<kilo> do you want to talk about it? 8-))
<reinhard> just putting together the details
<kilo> johannesV: does postgresql support altering a column spec from NULL to NOT NULL?
<johannesV> kilo: no
<johannesV> it cannot
<johannesV> because there migth be already NULL values in there
<johannesV> and theirfore a constraint violation would occur
<kilo> ok, just the same here in iterbase and it did look sane
<johannesV> what do you mean with 'did look sane' ?
<johannesV> have you expected it to work ?
<kilo> no, it should not work, it is ok if it does not work
<johannesV> ok
<kilo> btami extended a table (reports, foobulation), added a column 'ID'
<kilo> but the scripted generated code that first created the column, then tried to alter it to NOT NULL
<johannesV> but that's another case: here it cannot work because after adding the new column all rows have a NULL value in there. so if the db would then add the constraint it would violate it in that moment
<kilo> and then it did not work...
*** dimas has joined #gnuenterprise
<reinhard> aaahhh
<reinhard> finally found that script
<kilo> hmm hmm hmmmmmm
<kilo> how can then it be done?
<reinhard> ok
<reinhard> here is my problem
<reinhard> currently i have a list of filenames like
<reinhard> gnue-appserver-0.1.1
<reinhard> gnue-appserver-0.1.2
<reinhard> gnue-appserver-0.2.0
<reinhard> gnue-appserver-0.2.1
<reinhard> gnue-appserver-0.2.9
<reinhard> gnue-appserver-0.2.10
<reinhard> to find out the newest version, i currently do
<reinhard> sort | tail --lines=1
<reinhard> what appearantly doesn't work
* reinhard wonders if he should just release gnue-common 0.6.0
<reinhard> ;)
<reinhard> or maybe
<reinhard> we should switch to hex version numbers
<johannesV> reinhard: sort -n | tail --lines=1
<reinhard> 0.2.A
<johannesV> this should do the trick
<kilo> eh, you were quicker johannesV
<johannesV> ah no
<johannesV> it does not
<johannesV> 10 is still in between
<johannesV> :(
<johannesV> damn
<reinhard> I think sort -n only takes the *first* numeric part
<johannesV> exactly
<johannesV> sort takes only the first arg
<johannesV> but let me try for 5 mins
<btami> johannesV: i have a foobulations table with 3 old field, and the table is empty
<btami> the .gsd contains a new id field
<btami> defaultwith="serial"
<btami> as you can see in reports samples
<btami> the scripter creates this lines:
<btami> CREATE GENERATOR foobulations_id_seq;
<btami> ALTER TABLE foobulations ADD id smallint;
<btami> CREATE TRIGGER trg_id FOR foobulations ACTIVE BEFORE INSERT POSITION 0 AS BEGIN IF (NEW.id IS NULL) THEN NEW.id = GEN_ID (foobulations_id_seq,1); END
<btami> ALTER TABLE foobulations ALTER COLUMN id SET NOT NULL;
<btami> and firebird screams on last line
<btami> (-104, 'isc_dsql_prepare: Dynamic SQL Error. SQL error code = -104. Token unknown - line 1, char 42. SET. ')
<btami> whay scripter not creates
<btami> ALTER TABLE foobulations ADD id smallint NOT NULL;
<btami> instead ?
<johannesV> because this usually won't work
<kilo> reinhard: cat <nameslist> | tail --lines=1
<btami> adding a new NOT NULL column ?
<johannesV> btami, does interbase accept that last alter statement ?
<btami> as the docs writes, yes
<btami> but i will try in a sec...
<johannesV> reinhard, why won't you write a simple python script doing that sorting ? :)
<btami> johannesV: yes, it accepts it
<johannesV> ok, so i can change the creation for interbase
<btami> why just for interbase
<btami> isn't it standard sql ?
<johannesV> will have check it for the others
<btami> ok
<btami> thx
<johannesV> i thought it failed for postgres, for example
<kilo> on an empty table it should work on a table. if there is at least onw row in a table, it should not
<johannesV> as i was thinking it is not sql-standard anyway ... but I'll see
<kilo> s/onw/one
<johannesV> kilo, correct
<johannesV> it cannot if the table contains information (as i've explained above)
<johannesV> btami, try it with a table having one row ... :)
<btami> the problem (alter table .... add ...) is independent from generators/serial/etc. imo
<johannesV> the question is: why does interbase fail on that alter column set not null
<btami> it fails if it has generator+trigger
<btami> but in general
<btami> alter table ... add .. not null have to work
<btami> just adding a new column with an "alter table ..." statement
<btami> can have usual column defs
<johannesV> right, but if there is data in the table this statement can *not* work because this column has NULL for all *preexisting* rows and so violates the constraint NOT NULL
<btami> yes, it's true, but
*** holycow has quit IRC
<btami> it can have default clause too
<btami> and then it's _can_ be valid
<johannesV> reinhard, here it is:sort -t . -k 1,1 -k 2,2n -k 3,3n foo
<johannesV> where foo is the input
<johannesV> -t . uses the "." as field delimiter
<johannesV> -k 1,1 means "sort by the first field (which is the fielname including the -)"
<johannesV> -k 2,2n means: sort by the second field numerically)
<johannesV> and -k 3,3n is the same for the last number
<reinhard> cool
<reinhard> you were just about 30 seconds faster
<reinhard> i was already at
<reinhard>  sort -t . -k 1 -k 2n -k 3n x
<reinhard> but thanks
<johannesV> ah ok
<johannesV> but there is one problem left
<reinhard> it assures me that it's the correct solution :)
<reinhard> you mean
<johannesV> foo-2.3.2 is greater than foo-10.2.3
<reinhard> when the major version number gets > 10?
<johannesV> right
<reinhard> ok
<reinhard> we'll talk about in 20 years then ;-)
<johannesV> you would have to add an additional sorting layer
<kilo> written in C
<kilo> really, why not sort with a little python thingy?
<reinhard> i like writing shell scripts for stuff other people would use python or perl for
<reinhard> it's my hobby
<reinhard> ;-)
<kilo> mazochist...
<johannesV>  sort -t"-" foo | sort -t . -k 2,2n -k 3,3n
<johannesV> this would do the job
<btami> johannesV: all i want to notice: we don't have to make scripter too smart, just follow the sql standard, and
<reinhard> johannesV: cool
<reinhard> thanks
<btami> let the dbms complain if something conflicts
<johannesV> gnu textutils are a great thing !!
<johannesV> btami, but that's what it tries to do
<btami> ok, but why it not uses 1 line to add a new column then ?
<kilo> if you write a line like above with those sort blablabla, will yuo remember after 1 year what it does?
<johannesV> btami, adding new columns to an existing table which introduce a NOT NULL constraint is problematic
<btami> yes, i know
<reinhard> kilo: sure
<reinhard> kilo: you may ask me in a year from now
<btami> but let the db backend report it
<reinhard> :)
<johannesV> kilo, if you put a # first sort by filename and major number, then by minor
<johannesV> i would
<kilo> reinhard: ok, will try to remember to ask if you remember 8-)))
<johannesV> btami, the only right solution would be the following:
<johannesV> create a temporary table, copy all data into it, drop original table (or data) and modify it, copy back all data, and *INSERTING* a default value for the not-null-added column
<johannesV> there's no other way to handle this case ...
<btami> if the table is empty, there is no problem with this sql statement
<johannesV> this would be a nice task if there's only one table affected, but if this table has ref. constraints to other tables one cannot remove data without handling all that other tables ...
<btami> but there is a problem with a 2 line solution
<johannesV> btami, the two-line solution was thought to handle "default" values
<johannesV> where NOT NULL might be a bad 'default' value
<johannesV> but if you add a column f.e. having a "DEFAULT 'foo'" it should work this way
<johannesV> (as i've tested it worked *only* this way)
<btami> yes, it's true
<btami> i think the root of the problem is in firebird side, koz it cant handle generators in default clause
<btami> ok, forget all my complaints above :)
<btami> btw. one issue is remaining with firebird drivers creation
<btami> i'v put it in #TODO in the code
<btami> the before insert trigger code generation have to be 2 versions
<johannesV> one for direct execution and one for file-dump ?
<btami> 1) when putting it in a file, it have to use SET TERM...
<btami> 2) when directed to kinterbasdb
<btami> yes
<btami> i'v just changed the default to 2) in the code
<btami> and havn't implemented 1)
<johannesV> well, i saw that ... and i've thought it's a bit dangerous ... :)
<btami> not more dangerous than having not working 1 :)
<johannesV> yeah
<johannesV> it doesn't fit into the construction of schema-generation anyway ...
<johannesV> *!!!!* interbase
<btami> :))))
* btami is waiting for postgresql 8.0, it will ship native win port as i know
<reinhard> btami: current download page is fixed
<reinhard> thanks again for your report
<btami> my pleasure
<reinhard> please update exe files
<btami> i will, just have the win build infrastructure at home only
<reinhard> sorry
<reinhard> not meaning to say you should do it *now*
<btami> :)
<reinhard> it's just a little bug still existing in the page generation
<reinhard> it tests if the exe file is same or newer version than the .tar.gz
<reinhard> and if exe is older version, it leaves it out completely
<btami> no problem
<reinhard> so that people don't download exe of obsolete version
<reinhard> and this test does not work for 0.5.9 vs. 0.5.10
<reinhard> so while not displaying appserver .exe
<reinhard> it still displays common 0.5.9 exe
<btami> ok
<reinhard> johannesV: SQLite has indexes??
<reinhard> wow
<johannesV> yeah :)
<johannesV> reinhard, have you already looked at navigator ?
<reinhard> you mean
<reinhard> tested the gtk2 ui
<johannesV> right
<reinhard> or created the curses ui?
<reinhard> ok
<reinhard> no
<reinhard> but will do over weekend
*** dimas_ has joined #gnuenterprise
<johannesV> *lol* creating the curses ui for navigator will be a bit more work to do
<reinhard> johannesV: yes
<reinhard> johannesV: could you please update all Info.py files
<reinhard> with notes like "This driver does not implement index introspection"
<reinhard> where appropriate
<johannesV> ok
<reinhard> so the driver info for the web page is up to date and complete
<reinhard> thanks
<reinhard> of course only for drivers where introspection works apart from indexes :)
*** dimas has quit IRC
<johannesV> btami, do you need the sql-output of schema-creator ?
<johannesV> cause adding that divergent output depending on flags given to gnue-schema is really a problem
<btami> no, i used it only for debugging
<SachaS> night
*** SachaS has quit IRC
<johannesV> has anybody use parameters in report-datasources successfully ?
<johannesV> ok, got it
<johannesV> sorry, had a typo in there and there was an error in DataObject.py not reporting the real exception
<kilo> all-time-svn-commits
<kilo> jcater: 1904
<kilo> reinhard: 1006
<kilo> jamest: 672
<kilo> siesel: 441
<kilo> ntiffin: 439
<kilo> treshna: 439
<kilo> johannesV: 403
<kilo> dneighbo: 369
<kilo> btami: 318
<kilo> arturas: 127
<kilo> jmeskill: 118
<kilo> baumannd: 88
<kilo> kilo: 64
<kilo> jbailey: 23
<kilo> cspence: 18
<kilo> dimas_: 10
<kilo> madlocke: 4
<kilo> micah: 2
<kilo> root: 4
<kilo> gnue: 2
<kilo> arno: 1
<kilo> dhill: 1
<kilo> florin: 1
<kilo> psu_gnue: 1
<kilo> some people committed by different names, these were summed up
<kilo> reinhard's commit number is under inspection because of release script usage 8-))
<kilo> johannesV: is it possible that from a form's block's OnNewRecord trigger i cannot access an appserver class' procedure? what is the time relation between Form::OnNewRecord and AppServer_ClassX::OnInit?
<btami> bbl
*** btami has quit IRC
<johannesV> i think form's OnNewRecord is called 'before' the recordset is created
<johannesV> but i'd have to check
<johannesV> kilo, it should work
<johannesV> in GFBlock's newRecord () function the new record get's created (-> appserver-class of the block's datasource is created)
<johannesV> and then the PRE-FOCUSIN, followed by ON-NEWRECORD and POST-FOCUSIN triggers are fired
<kilo> hmmm
<kilo> it tells me there is no such function as 'newNumber', but there is...
<johannesV> what is 'newNumber' ?
<kilo> a procedure in INV_Head class
<johannesV> kilo, if you call procedures from forms-triggers, please make sure to use the block's call statement
<johannesV> like blkPerson.call ('...') not the datasource
<reinhard> what?
<reinhard> derek 369 commits?
<kilo> why?
<reinhard> scary
<johannesV> (this has changed with the oninit-stuff)
<kilo> reinhard: lol
<johannesV> kilo, iirc it was because if the function raises an error the state of the current record is set correctly
<kilo> hmm, ok, will try it soon. thx
<reinhard> kilo: ah, yes, i wanted to remind you to change that in the wiki
<reinhard> off to daughter's birthday party
<reinhard> l8r all
*** reinhard has quit IRC
<kilo> ah, ok. didn't notice it has changed in sample.gfd too...
<kilo> strange. same error message now, though using blkINV_Head.call(...)
<johannesV> kilo, can i reproduce this using gnue-contrib ?
<kilo> hmmm, i'll commit it there in a minute
<johannesV> ok
<kilo> johannesV: gnue-contrib up to date
<kilo> now the form INV_Head.gfd fails.
<kilo> if you comment out blkINV_Head OnNewrecord trigger, it is OK
<johannesV> ok, have the form up
<johannesV> what do i have to do ?
<kilo> you have the form up and running???
<johannesV> ok, i've forgotton to uncomment the call-part
<kilo> no need to uncomment
<kilo> it is in blkINV_Head
<johannesV> ok, got it to work
<johannesV> :)
<johannesV> just change "newNumber" to "INV_newNumber"
<johannesV> :)
<kilo> ooooooooooooooooooooooooooooo
* kilo steps down...
* kilo retire
<kilo> i was looking at it for at least 1 hour...
<johannesV> :)
<johannesV> oh
<johannesV> my little daughter woke up ...
<johannesV> :)
<johannesV> bbl
<kilo> bbl, thx johannesV
*** kilo has quit IRC
*** sjc has quit IRC
*** jamest has joined #gnuenterprise
*** Amorphous has quit IRC
*** johannesV_ has joined #gnuenterprise
*** johannesV has quit IRC
*** Amorphous has joined #gnuenterprise
*** sjc has joined #gnuenterprise
*** bluesbaron_ has joined #gnuenterprise
*** bluesbaron_ has left #gnuenterprise
*** SachaS has joined #gnuenterprise
*** nickr has joined #gnuenterprise
*** jcater has joined #gnuenterprise
*** cilkay has quit IRC
*** wendall911 has joined #gnuenterprise
*** cilkay has joined #gnuenterprise
*** holycow has joined #gnuenterprise
*** johannesV_ has quit IRC
<jcater> ----------
<jcater> (for the logs)
<jcater> reinhard: Bogus commit to update website.
<jcater> you can also just do:
<jcater>  $ ssh reinhard@ash.gnuenterprise.org signal-website-update
<jcater> ----------
*** cilkay has quit IRC
<chillywilly> hey you guys what do you think of this: http://www.libertyetech.com/logo_raw.png
<jcater> spiffy
<chillywilly> now I just need a website
<chillywilly> :P
*** kilo has joined #gnuenterprise
*** mapangojoe has joined #gnuenterprise
*** sjc has quit IRC
*** sjc has joined #gnuenterprise
*** jemfinch has joined #gnuenterprise
<jemfinch> howdy, folx0rs.
<jemfinch> bigbrother`: user list --capability owner
<bigbrother`> jemfinch: jcater
<jemfinch> jcater: you in?
<jcater> maybe
<jemfinch> jcater: sweet.
<jemfinch> I may need your help :)
<jemfinch> bigbrother`: version
<bigbrother`> jemfinch: The current (running) version of this Supybot is 0.79.999.  The newest version available online is 0.80.0pre2.
<jcater> ok?
<jemfinch> (I'm doing the testing that doesn't require your help right now :))
<chillywilly> who the hell are you jemfinch ?
<jemfinch> chillywilly: I'm the author of Supybot.
<chillywilly> ok
<jcater> lol
<jcater> jemfinch: please excuse our door greeter
<jemfinch> hehe
* jemfinch is used to all kinds of IRCers.
<chillywilly> I'll be good
<chillywilly> supybot is pretty rad
<jemfinch> chillywilly: thanks.
<jcater> my favorite
<jcater> @weather 38654
<bigbrother`> jcater: The current temperature in Olive Branch, Mississippi is 81°F (3:40 PM CDT on October 22, 2004). Conditions: Scattered Clouds. Humidity: 65%. Dew Point: 68°F. Pressure: 29.92 in (1013 hPa). Visibility: 3.0 miles (4.8 kilometers). UV: 2 out of 12
<jcater> :)
<jcater> that rocks my socks
<chillywilly> I think you guys need to find your own sayings
<chillywilly> that one is mine :)
<jcater> beg your pardon?
<jcater> we said that as kids
<chillywilly> sure you did ;)
<chillywilly> is it time to go home yet?
<jcater> yes
<chillywilly> I hate this fucking place
<chillywilly> lalala
<chillywilly> ooops, this thing is logged
<chillywilly> damn you bigbrother` !!!
<chillywilly> you're going down
<chillywilly> bigbrother`: die
<chillywilly> darn
<chillywilly> no die commmand
*** jemfinch has left #gnuenterprise
<kilo> @weather LHUD
<bigbrother`> kilo: The current temperature in Szeged, Hungary is 58°F. Conditions: Mist. Humidity: 94%. Dew Point: 57°F. Wind: NNE at 2 mph (4 km/h). Pressure: 30.18 in (1022 hPa). Visibility: 2.0 miles (3.0 kilometers). UV: 0 out of 12
<kilo> better stay home and drink a dram o' whisky...
*** moo has joined #gnuenterprise
*** moo has left #gnuenterprise
*** holycow has quit IRC
*** SachaS has quit IRC
*** jamest has quit IRC
*** jcater has quit IRC
*** kilo has quit IRC
<wendall911> @weather 83864
<bigbrother`> wendall911: The current temperature in Sandpoint, Idaho is 37°F (5:10 PM MDT on October 22, 2004). Conditions: Overcast. Humidity: 81%. Dew Point: 32°F. Wind: WSW at 5 mph (7 km/h). Pressure: 29.72 in (1006 hPa). Visibility: 5.0 miles (8.0 kilometers). UV: 0 out of 12
*** sjc has quit IRC
*** mapangojoe has quit IRC
*** jcater has joined #gnuenterprise
*** jamest has joined #gnuenterprise
*** wendall911 has quit IRC
*** jcater has quit IRC
*** jamest has quit IRC