You can subscribe to this list here.
2000 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(13) |
Oct
(12) |
Nov
(26) |
Dec
(2) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2001 |
Jan
(8) |
Feb
|
Mar
|
Apr
(20) |
May
(31) |
Jun
(7) |
Jul
(6) |
Aug
(56) |
Sep
(2) |
Oct
|
Nov
(3) |
Dec
(1) |
2002 |
Jan
(4) |
Feb
(2) |
Mar
(2) |
Apr
(4) |
May
(2) |
Jun
(20) |
Jul
(31) |
Aug
(14) |
Sep
(30) |
Oct
(14) |
Nov
(13) |
Dec
(32) |
2003 |
Jan
(29) |
Feb
(46) |
Mar
(1) |
Apr
(3) |
May
(9) |
Jun
(3) |
Jul
(7) |
Aug
(6) |
Sep
(5) |
Oct
(4) |
Nov
(7) |
Dec
(5) |
2004 |
Jan
(6) |
Feb
|
Mar
(5) |
Apr
(2) |
May
(3) |
Jun
|
Jul
(3) |
Aug
(3) |
Sep
(4) |
Oct
(4) |
Nov
(5) |
Dec
(3) |
2005 |
Jan
|
Feb
(2) |
Mar
(23) |
Apr
(1) |
May
(5) |
Jun
|
Jul
(5) |
Aug
(1) |
Sep
(1) |
Oct
(4) |
Nov
(4) |
Dec
|
2006 |
Jan
(1) |
Feb
(3) |
Mar
(1) |
Apr
(2) |
May
(3) |
Jun
|
Jul
(1) |
Aug
(10) |
Sep
(3) |
Oct
(2) |
Nov
(3) |
Dec
|
2007 |
Jan
(3) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(2) |
Oct
(1) |
Nov
|
Dec
|
2008 |
Jan
|
Feb
(1) |
Mar
(28) |
Apr
(18) |
May
(1) |
Jun
|
Jul
(4) |
Aug
(1) |
Sep
(1) |
Oct
|
Nov
|
Dec
(20) |
2009 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(2) |
Nov
|
Dec
|
2010 |
Jan
(1) |
Feb
(1) |
Mar
(2) |
Apr
(2) |
May
|
Jun
(5) |
Jul
(1) |
Aug
(2) |
Sep
(10) |
Oct
(3) |
Nov
(4) |
Dec
(2) |
2011 |
Jan
(2) |
Feb
(3) |
Mar
(2) |
Apr
(3) |
May
(1) |
Jun
|
Jul
|
Aug
(5) |
Sep
(2) |
Oct
|
Nov
|
Dec
(1) |
2012 |
Jan
(1) |
Feb
(7) |
Mar
(1) |
Apr
|
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2013 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
(2) |
Jun
(1) |
Jul
|
Aug
(1) |
Sep
|
Oct
(7) |
Nov
(3) |
Dec
|
2014 |
Jan
|
Feb
(3) |
Mar
|
Apr
(3) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(2) |
Nov
|
Dec
|
2015 |
Jan
|
Feb
|
Mar
(2) |
Apr
|
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2016 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Pavel C. <pc...@ib...> - 2008-12-13 10:39:34
|
Hi Mark, Mark O'Donohue napsal(a): > Hi Pavel > > Thanks for the tip, I have got further, but now I am getting groups of > failed tests. > > I am running against the standard ubuntu fb2.0.3 package install, > > $qmtest run -C context.txt functional > > gives: > --- STATISTICS -------------------------------- > > 358 tests total > 88 ( 25%) tests FAIL > 270 ( 75%) tests PASS > > > 88 failures seems a little high. Looking at the first failure, which is > > > test case: functional.arno.derived_tables.derived_tables_06 > > That test is: > > SELECT > dt.* > FROM > Table_10 t10 > JOIN (SELECT * FROM Table_10 t2 WHERE t2.ID = t10.ID) > dt ON (1 = 1); > > When I ran that test using isql it gives the "Column Unknown: T10.ID" as > expected. > > > The QMTest result complained it found an extra "=": > > -Token unknown - line 5, column 2 > -= > > So QMTest has an extra "=" inserted somewhere. > > So any idea of what is going wrong (I have included more details below) ? Now I realized that you may use wrong (outdated) version of qmtest, if you downloaded qmtest from original site. It doesn't contain recent patches and might have trouble with Python 2.5. Please download this version and try again http://www.firebirdsql.org/download/prerelease/python/qmtest-2.4.1070.tar.gz There is also a Windows installer if you want to run it under Windows too. best regards Pavel Cisar IBPhoenix |
From: Mark O'D. <mar...@gm...> - 2008-12-13 10:37:45
|
I found the problem, and Helen will give me an i-told-you-so for it I am sure! It seemed that any test with lines > 64 characters in length fails. even a: /usr/bin/isql-fb < test.sql will ignore the first 64 characters for lines that are longer than 64 characters - this is a problem with the libeditline provided in Unbuntu (and Debian). Mind you it took a bit of digging though and looking at the internals of qmtest and fbqa.py before I figured that out. The bug doesnt occur in our older localized version of libeditline. There isnt any reported bug for Debian or Ubuntu but someone fixed something similar on minix-editline. So I created a bug report for Unbuntu. https://bugs.launchpad.net/ubuntu/+source/editline/+bug/307652 Then having gone that far, it wasn't too hard then to then give them a patch to fix is as well. Philippe Makowski wrote: > Mark, >> And the actual test ran fine, when I did it by hand. >> > and what result if you run only one test, for example > $qmtest run -C context.txt functional.arno.derived_tables.derived_tables_06 > yep, I was doing that, and comparing out put to one that worked, then I added some code within FirebirdTest to debug what was actually sent through to isql, since it does several substitutions to the input. Finally when dumping and comparing stdin/stderr/stdout for a working and non-working cases, I tried the line length. Anyway, a side effect is I've added some debugging code that will with a switch print the SQL code, used for setup and execution when you run the tests. Now where was I... Cheers - Mark |
From: Philippe M. <mak...@fi...> - 2008-12-13 09:06:53
|
Mark, > And the actual test ran fine, when I did it by hand. > and what result if you run only one test, for example $qmtest run -C context.txt functional.arno.derived_tables.derived_tables_06 ? may be you have a qmtest problem which version are you using ? |
From: Mark O'D. <mar...@gm...> - 2008-12-13 03:55:49
|
Hi Helen Helen Borrie wrote: > Mark, > >> >> I am running against the standard ubuntu fb2.0.3 package install, > > Wouldn't you do better installing and testing with the official > project distro, so that you actually know what you're working with? > Careful, Helen they will hear you :-). The they in debian in terms of who maintains the packages there is generally us; at least thats my take from what I read in pkg-firebird-general; and packaging within constrains imposed on you ain't easy. Besides what is wrong with a package name of firebird2_1.5.3 :-). However, my general aim is to test "my" build, but to do that I wanted to run a base test set for reference. The Ubuntu 2.0.3 firebird looks legit it reports: $ isql-fb -z ISQL Version: LI-V2.0.3.12981 Firebird 2.0 Use CONNECT or CREATE DATABASE to specify a database SQL> And the actual test ran fine, when I did it by hand. So I actually suspect some trickery in qmtest. It would be nice to be able to get in and trace it a little better - bring back DECTest thats what I say :-). Cheers - Mark . |
From: Helen B. <he...@tp...> - 2008-12-13 03:09:53
|
Mark, At 01:36 PM 13/12/2008, you wrote: >Hi Pavel > >Thanks for the tip, I have got further, but now I am getting groups of >failed tests. > >I am running against the standard ubuntu fb2.0.3 package install, Wouldn't you do better installing and testing with the official project distro, so that you actually know what you're working with? Ubuntu's stuff comes from Debian and Debian was calling Firebird 1.5 "Firebird 2" for a long time, possibly still is... > $qmtest run -C context.txt functional > >gives: > --- STATISTICS -------------------------------- > > 358 tests total > 88 ( 25%) tests FAIL > 270 ( 75%) tests PASS > > >88 failures seems a little high. Firebird 1.5 (posing as Firebird 2) wouldn't pass any tests set up to test new language features in (the real) Firebird 2.0. Just a thought...on the opposite side of the coin, at least you might be able to demonstrate how much up the creek the Debian/Firebird situation has been at defined points in its evolution. ;-) Helen |
From: Mark O'D. <mar...@gm...> - 2008-12-13 02:36:27
|
Hi Pavel Thanks for the tip, I have got further, but now I am getting groups of failed tests. I am running against the standard ubuntu fb2.0.3 package install, $qmtest run -C context.txt functional gives: --- STATISTICS -------------------------------- 358 tests total 88 ( 25%) tests FAIL 270 ( 75%) tests PASS 88 failures seems a little high. Looking at the first failure, which is test case: functional.arno.derived_tables.derived_tables_06 That test is: SELECT dt.* FROM Table_10 t10 JOIN (SELECT * FROM Table_10 t2 WHERE t2.ID = t10.ID) dt ON (1 = 1); When I ran that test using isql it gives the "Column Unknown: T10.ID" as expected. The QMTest result complained it found an extra "=": -Token unknown - line 5, column 2 -= So QMTest has an extra "=" inserted somewhere. So any idea of what is going wrong (I have included more details below) ? Cheers - Mark Setup is : ./qadbm.py QADBM>platform Linux QADBM>target 2.0.3 QADBM>create fbtest2.0.3 QADBM>copy fbtest2.0.3 QADBM>exit I create a context.txt file for ubuntu/2.0.3 as : temp_directory=/var/tmp/ server_location=localhost: database_location=/var/firebird/test/ suite_database_location=/home/odonohue/work/firebird/testing/qa/qa/trunk/testsuite/fdb/ backup_location=/home/odonohue/work/firebird/testing/qa/qa/trunk/testsuite/fdk/ files_location=/home/odonohue/work/firebird/testing/qa/qa/trunk/testsuite/files/ isc4_path=/var/lib/firebird/2.0/system/security2.fdb user_name=SYSDBA user_password=<a secret noone will ever guess> isql_path=/usr/bin/isql-fb gsec_path=/usr/bin/gsec gstat_path=/usr/bin/gstat gbak_path=/usr/bin/gbak nbackup_path=/usr/bin/nbackup gfix_path=/usr/bin/gfix gpre_path=/usr/bin/gpre Then I run : cd fbtest2.0.3 qmtest run -C context.txt functional It produced the results file: $ ls -lh results.qmr -rw-r--r-- 1 odonohue odonohue 165K 2008-12-13 12:03 results.qmr I was hoping for a basic report and tried most of the options for: $qmtest report -f results.qmr but it just kept replying with : usage: qmtest report [ OPTIONS ] [ result [-e expected] ]+ and although -e is optional, I wasnt sure where the was. So I did the gui thing. $qmtest gui And fired up the browser. I then chose File/Load Results, selected the results.qmr file, and that seemed to work. navagating to the first failue: functional.arno.derived_tables.derived_tables_06 The difference is: Actual result: Statement failed, SQLCODE = -104 Dynamic SQL Error -SQL error code = -104 -Token unknown - line 5, column 2 -= Expected result: Statement failed, SQLCODE = -206 Dynamic SQL Error -SQL error code = -206 -Column unknown -T10.ID -At line 5, column 53 > > The report command is not so important. The run command should produce > the results.qmr file that you can rename (to have platform, fb arch. a > run number, for example wcs1.qmr fro Windows Classic run 1) and send to > me for further processing. > What further processing do you do? |
From: Pavel C. <pc...@ib...> - 2008-12-12 11:58:40
|
Mark O'Donohue napsal(a): > > Then you get to "Configuration" : > > It lists a group of key value pairs. They look nice, but what does > one do with them? They really look like they want to go in a > config file or be environment variables. There is a > /home/fbtest/QMtest/configuration file but that is full of XML so > I dont think they below there - or do they? You need to create a text file with these variables with values valid for your system. Then you will specify this file when running qmtest using the -C option. > The final step is to run qmtest, and list the cmd options. > > Does someone mind filling me in on what to do, in terms of completing > the above howto example. I assume it then goes something like: > > $cd /home/fbtests > $qmtest ls > $qmtest run > > $qmtest report The report command is not so important. The run command should produce the results.qmr file that you can rename (to have platform, fb arch. a run number, for example wcs1.qmr fro Windows Classic run 1) and send to me for further processing. If you want to look at results, I would advise to use the web gui (qmtest qui command). > And I guess it would be handy to run a subset of tests or an > invidiual test, presumaby that is by: > > $qmtest ls functional.arno > $qmtest run functional.arno Yes. best regards Pavel Cisar IBPhoenix |
From: Mark O'D. <mar...@gm...> - 2008-12-12 06:25:23
|
Hi It has been a while since I last ran qmtest for firebird, so I was following the instructions at: http://www.firebirdsql.org/index.php?op=devel&sub=qa&id=qmtest_howto All is dandy, $python qadbm.py > repository /home/fbqa > platform Windows > target 1.5.2 > create /home/fbtests > copy /home/fbtests > exit Then you get to "Configuration" : It lists a group of key value pairs. They look nice, but what does one do with them? They really look like they want to go in a config file or be environment variables. There is a /home/fbtest/QMtest/configuration file but that is full of XML so I dont think they below there - or do they? The final step is to run qmtest, and list the cmd options. Does someone mind filling me in on what to do, in terms of completing the above howto example. I assume it then goes something like: $cd /home/fbtests $qmtest ls $qmtest run $qmtest report And I guess it would be handy to run a subset of tests or an invidiual test, presumaby that is by: $qmtest ls functional.arno $qmtest run functional.arno But at the moment I am stuck on configuration :-). Cheers - Mark |
From: Philippe M. <mak...@fi...> - 2008-09-30 13:55:07
|
Björn Reimer [08-09-30 15.23] : > The current employee.fdb is not up to date at all. So I'd set up an > own examples DB which is not perfect at all but may be a first step. > > So I'd like to ask if the topic "examples database" is part of the doc > subproject or if it does not fit in here. > > Opinions? Are there any more supporters for these two projects? > good news Last year this idea come but nobody started the work Yes we need a new example database that can be used for both documentation and Q/A so please, yes start, and we'll see after if some other pieces have to be added. -- Philippe Makowski http://www.ibphoenix.com Supporting users of Firebird Tel +33 (0) 561058813 |
From: Frank P. <fep...@dr...> - 2008-08-15 17:00:30
|
Just checking... |
From: Philippe M. <mak...@fi...> - 2008-07-10 17:08:18
|
Hello, > > I'm posting this message instead of Vyacheslav. His original message was > banned by SPAM-bot of sourceforge :) (What he must do in this case?) > sorry I don't know, I don't manage the list >> Roman Simakov [08-07-01 11.23] : >> > Done. But I think in this case we will have a set of separate test cases > for 2.x > engine and we can't see full image about FK functionality. Why are you telling that ? I don't understand tests are cumulative, so test tagged for 2.1 apply to 2.1 and 2.5 > And what about full test suite for reference integrity in firebird? This > features are not checked full in QM test now. why not > I think we can develop tests for full checking this features. But this > problem is very important and I think it would be very good to discuss > all possible test cases. Can anybody help as with it? May be use SQL > standard to gather test cases? Are there other ideas? SQL standard are a good starting point yes It lead to another idea that was on air, to have a new example database that could be use for doc and for tests Maybe we could start by this |
From: Roman S. <rom...@re...> - 2008-07-07 12:10:23
|
Hello, Philippe! Many thanks for your comments about this tests. I'm posting this message instead of Vyacheslav. His original message was banned by SPAM-bot of sourceforge :) (What he must do in this case?) > Roman Simakov [08-07-01 11.23] : > >> > I can send you them before commit but I think it's not critical if >> you > review them after commit. >> > I see that they are committed, nice > > Just two point : > > can you use : > #Create second connection - update detail table > con_detail = kdb.connect( > dsn=dsn.encode(), > user=user_name.encode(), > password=user_password.encode() > ) > > instead of : > #Create second connection - update detail table > con_detail = kdb.connect( > dsn=dsn, > user='SYSDBA', > password='masterkey' > ) > > Done. > And seems that a lot of tests are ok for target before 2.5 > so can you change target for all tests that are also ok for 2.x ? > Done. But I think in this case we will have a set of separate test cases for 2.x engine and we can't see full image about FK functionality. And what about full test suite for reference integrity in firebird? This features are not checked full in QM test now. I think we can develop tests for full checking this features. But this problem is very important and I think it would be very good to discuss all possible test cases. Can anybody help as with it? May be use SQL standard to gather test cases? Are there other ideas? > Thanks > -- Philippe Makowski Regards Vyacheslav Ozherelyev Red Soft Corporation |
From: Philippe M. <mak...@fi...> - 2008-07-03 09:36:18
|
Roman Simakov [08-07-01 11.23] : > I can send you them before commit but I think it's not critical if you > review them after commit. I see that they are committed, nice Just two point : can you use : #Create second connection - update detail table con_detail = kdb.connect( dsn=dsn.encode(), user=user_name.encode(), password=user_password.encode() ) instead of : #Create second connection - update detail table con_detail = kdb.connect( dsn=dsn, user='SYSDBA', password='masterkey' ) And seems that a lot of tests are ok for target before 2.5 so can you change target for all tests that are also ok for 2.x ? Thanks -- Philippe Makowski http://www.ibphoenix.com Supporting users of Firebird Tel +33 (0) 561058813 |
From: Roman S. <rom...@re...> - 2008-07-01 09:23:34
|
Hi, We developed a few QM tests for FK testing. They include test cases when master record is active and detail records reference to it. Yesterday, 2008-06-30, we committed the patch improving this handling and fixing the error discussed 2008-04-12 in firebird-devel by report Sean Leyne. We tested them on new Firebird HEAD. Test cases are a new directory in functional.qms and not changed other tests. Only new files. We think these tests will be useful and we'd like to commit them into QA repository. I could commit them using my developer account. I can send you them before commit but I think it's not critical if you review them after commit. Any opinions? Best Regards, Roman Simakov |
From: Philippe M. <mak...@fi...> - 2008-05-02 12:17:36
|
Greg Kay [08-04-29 09.06] : > Hi, > > We've just installed QMTest running against Firebird 2.1 Classic on > linux but out of 551 tests, 22 have errors and 24 fail. > > We installed > > python-2.5.2.msi > pywin32-210.win32-py2.5.exe > kinterbasdb-3.2.win32-FB-1.0-py2.5.exe > qmtest-2.4.win32.exe > > hum you ran test from Windows platform against a Linux Firebird ? Our Qmtests are made to be run on the same box that Firebird is installed > Also, is the difference between "ERROR" and "FAIL" that "ERROR" is some > unexpected exception in the test? yes |
From: Greg K. <gk...@ob...> - 2008-04-29 07:06:07
|
Hi, We've just installed QMTest running against Firebird 2.1 Classic on linux but out of 551 tests, 22 have errors and 24 fail. We installed python-2.5.2.msi pywin32-210.win32-py2.5.exe kinterbasdb-3.2.win32-FB-1.0-py2.5.exe qmtest-2.4.win32.exe Also, is the difference between "ERROR" and "FAIL" that "ERROR" is some unexpected exception in the test? Thanks. Greg |
From: <us...@do...> - 2008-04-29 06:58:43
|
Hi, We've just installed QMTest running against Firebird 2.1 Classic on linux but out of 551 tests, 22 have errors and 24 fail. Should they all pass? We installed: python-2.5.2.msi pywin32-210.win32-py2.5.exe kinterbasdb-3.2.win32-FB-1.0-py2.5.exe qmtest-2.4.win32.exe Also, is the difference between "ERROR" and "FAIL" that "ERROR" is an unexpected exception in the test? Thanks. Greg |
From: Philippe M. <mak...@fi...> - 2008-04-17 15:30:41
|
Jeff Jones - SHERIFFX [08-04-17 17.11] : > Can you give me an example of how to use the -ch switch? I cannot find > any documentation on it. > isql 127.0.0.1:/Volumes/tempo/firebird_db/rdisk2s3 -ch ISO8859_1 -user SYSDBA -password masterkey Database: 127.0.0.1:/Volumes/tempo/firebird_db/rdisk2s3, User: SYSDBA SQL> set; Print statistics: OFF Echo commands: OFF List format: OFF Row Count: OFF Autocommit DDL: ON Access Plan: OFF Access Plan only: OFF Display BLOB type: 1 Set names: ISO8859_1 Column headings: ON Terminator: ; Time: OFF Warnings: ON Bail on error: OFF |
From: Jeff J. - S. <J_...@MC...> - 2008-04-17 15:11:58
|
Can you give me an example of how to use the -ch switch? I cannot find any documentation on it. Thanks! Jeff -----Original Message----- From: fir...@li... [mailto:fir...@li...] On Behalf Of Pavel Cisar Sent: Thursday, April 17, 2008 07:40 AM To: fir...@li... Subject: Re: [Firebird-test] Test will not connect to database (rather long) Jeff Jones - SHERIFFX napsal(a): > I cannot find any documentation of how to use the -ch parameter in a > CONNECT statement. I have found where you use a SET statement before > the CONNECT statement to set the character set, and that seems to work > fine. I have been unable to see how fbqa.py handles the connection. The -ch option is cmdline switch equivalent to SET NAMES command used in isql console before CONNECT/CREATE DATABASE statement. When you pass the database name (and user name and password) as command line parameters to isql, you need to use -ch when you need specific character set for connection, as using SET NAMES is already too late. fbqa.py passes these parameters as cmd switches because initialization scripts and actual tests scripts feeded to isql stdin can't contain CONNECT/CREATE DATABASE commands (for many reasons, for example they don't know the real paths), hence it uses the -ch option. best regards Pavel Cisar IBPhoenix ------------------------------------------------------------------------ - This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/j avaone _______________________________________________ Firebird-test mailing list Fir...@li... https://lists.sourceforge.net/lists/listinfo/firebird-test |
From: Pavel C. <pc...@ib...> - 2008-04-17 14:39:53
|
Jeff Jones - SHERIFFX napsal(a): > I cannot find any documentation of how to use the -ch parameter in a > CONNECT statement. I have found where you use a SET statement before > the CONNECT statement to set the character set, and that seems to work > fine. I have been unable to see how fbqa.py handles the connection. The -ch option is cmdline switch equivalent to SET NAMES command used in isql console before CONNECT/CREATE DATABASE statement. When you pass the database name (and user name and password) as command line parameters to isql, you need to use -ch when you need specific character set for connection, as using SET NAMES is already too late. fbqa.py passes these parameters as cmd switches because initialization scripts and actual tests scripts feeded to isql stdin can't contain CONNECT/CREATE DATABASE commands (for many reasons, for example they don't know the real paths), hence it uses the -ch option. best regards Pavel Cisar IBPhoenix |
From: Jeff J. - S. <J_...@MC...> - 2008-04-17 14:23:25
|
I cannot find any documentation of how to use the -ch parameter in a CONNECT statement. I have found where you use a SET statement before the CONNECT statement to set the character set, and that seems to work fine. I have been unable to see how fbqa.py handles the connection. Jeff -----Original Message----- From: fir...@li... [mailto:fir...@li...] On Behalf Of Pavel Cisar Sent: Wednesday, April 16, 2008 12:44 AM To: fir...@li... Subject: Re: [Firebird-test] Test will not connect to database (rather long) Jeff Jones - SHERIFFX napsal(a): > > When I commented out the "-ch" parameter in the second call, my test > worked. I don't quite understand why, but I am not going to question it > just yet. > > Can you tell my why this would work? Interesting. There is no reason why -ch option (which is character set specification to be used for connection) should make any trouble. Could you try to run isql from system shell with the same parameters that qmtest passes to it when it's run from it? best regards Pavel Cisar IBPhoenix ------------------------------------------------------------------------ - This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/j avaone _______________________________________________ Firebird-test mailing list Fir...@li... https://lists.sourceforge.net/lists/listinfo/firebird-test |
From: Pavel C. <pc...@ib...> - 2008-04-16 07:43:30
|
Jeff Jones - SHERIFFX napsal(a): > > When I commented out the "-ch" parameter in the second call, my test > worked. I don't quite understand why, but I am not going to question it > just yet. > > Can you tell my why this would work? Interesting. There is no reason why -ch option (which is character set specification to be used for connection) should make any trouble. Could you try to run isql from system shell with the same parameters that qmtest passes to it when it's run from it? best regards Pavel Cisar IBPhoenix |
From: Jeff J. - S. <J_...@MC...> - 2008-04-15 22:31:36
|
I am trying to connect to Firebird 1.5.4. And, with the alteration I made, the connection is working. Jeff -----Original Message----- From: fir...@li... [mailto:fir...@li...] On Behalf Of Leyne, Sean Sent: Tuesday, April 15, 2008 03:09 PM To: fir...@li... Subject: Re: [Firebird-test] Test will not connect to database (rather long) > And here is the call in __ExecISQLCommands: > > stdout, stderr= > self.__RunProgram(self.source_code.encode('utf-8'),[self.__con > text["isql > _path"], > self.__dsn, > "-ch", self.character_set, > "-user", self.user_name, > "-password", self.user_password]) > > When I commented out the "-ch" parameter in the second call, > my test worked. I don't quite understand why, but I am not > going to question it just yet. > > Can you tell my why this would work? What version of FB are you trying to connect to/test with? Sean ------------------------------------------------------------------------ - This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/j avaone _______________________________________________ Firebird-test mailing list Fir...@li... https://lists.sourceforge.net/lists/listinfo/firebird-test |
From: Leyne, S. <Se...@br...> - 2008-04-15 22:05:02
|
> And here is the call in __ExecISQLCommands: > > stdout, stderr= > self.__RunProgram(self.source_code.encode('utf-8'),[self.__con > text["isql > _path"], > self.__dsn, > "-ch", self.character_set, > "-user", self.user_name, > "-password", self.user_password]) > > When I commented out the "-ch" parameter in the second call, > my test worked. I don't quite understand why, but I am not > going to question it just yet. > > Can you tell my why this would work? What version of FB are you trying to connect to/test with? Sean |
From: Jeff J. - S. <J_...@MC...> - 2008-04-15 18:07:48
|
Hey! I found the problem! In fbqa.py, here is the call to __RunProgram call in __ExecISQLCommandsBlind: stdout, stderr= self.__RunProgram(self.isql_script.encode('utf-8'),[self.__context["isql _path"], self.__dsn, "-user", self.user_name, "-password", self.user_password]) And here is the call in __ExecISQLCommands: stdout, stderr= self.__RunProgram(self.source_code.encode('utf-8'),[self.__context["isql _path"], self.__dsn, "-ch", self.character_set, "-user", self.user_name, "-password", self.user_password]) When I commented out the "-ch" parameter in the second call, my test worked. I don't quite understand why, but I am not going to question it just yet. Can you tell my why this would work? Jeff -----Original Message----- From: fir...@li... [mailto:fir...@li...] On Behalf Of Pavel Cisar Sent: Monday, April 14, 2008 08:56 AM To: fir...@li... Subject: Re: [Firebird-test] Test will not connect to database (rather long) Hi Jeff, I'm sorry for late answer, but I haven't enough time to dig deep into your problem till now. 1. I can't reproduce your problem, although I tried very hard. 2. I think that I have solved it anyway :-) It seems that certain Windows XP boxes have trouble with unicode. I have failed to pinpoint it down to exact configuration, hence the vague description, but it's for real. Anything older than WinXP can fail miserably with unicode, but XP and newer should work just fine (and mostly do) except when they don't. It seems that your Windows have problem passing unicode strings as parameters for spawned process. The solution should be simple, just call encode method on all parameter values passed to __RunProgram in __ExecISQLCommandsBlind and __ExecISQLCommands. Example: def __ExecISQLCommandsBlind(self): try: stdout, stderr= self.__RunProgram(self.isql_script.encode('utf-8'),[self.__context["isql _path"], self.__dsn, "-user", self.user_name, "-password", self.user_password]) change to: def __ExecISQLCommandsBlind(self): try: stdout, stderr= self.__RunProgram(self.isql_script.encode('utf-8'),[self.__context["isql _path"], self.__dsn, "-user", self.user_name.encode(), "-password", self.user_password.encode()]) This should encode parameters to ASCII by default. You can pass the charset name to the encode if you want something else. best regards Pavel Cisar IBPhoenix ------------------------------------------------------------------------ - This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/j avaone _______________________________________________ Firebird-test mailing list Fir...@li... https://lists.sourceforge.net/lists/listinfo/firebird-test |