SqueakDBX August 2010

squeakdbx@lists.squeakfoundation.org

7 participants
8 discussions

Re: [SqueakDBX] Character encoding and pooling in SqueakDBX and Glorp
by Mariano Martinez Peck 22 Oct '10

22 Oct '10

On Fri, Aug 27, 2010 at 8:10 AM, Panu Suominen <panu.j.m.suominen(a)gmail.com>wrote: > Thank you for excelent feedback. > thanks for your changes :) > > 2010/8/26 Mariano Martinez Peck <marianopeck(a)gmail.com>: > > - I guess a lot of people doesn't need UTF (or other) encoding. And I > guess > > that the transformation brings some overhead in the query result > mosty...(I > > should run the benchmarks and see if there is difference). But I was > > thinking if we could make this optional. > > I was thinking that too. Apparently no encoding means that only 7bit > characters should be used? > > I am totally newbie with encoding. I have no idea about how many bits hahah. The only thing I would like is that with "no encoding" I do what we were doing until this change. > > or something like that....but the idea of making it optional not to have > > overhead if not required, and being able to be enabled from the app code. > > This probably need some thinking. Because if one uses database through > GLORP the encoding parameter should be passed somehow to lower levels. > Yes, I saw this after sending all the emails :( > Would the ConnectionSettings in SqueakDBX level be right place? ahhhh yes, I forgot about ConnectionSettings :) That sounds more reasonable than Connection. FOr all your changes and comments it seems you understood SqueakDBX quite well and fast heheh > And in > Glorp I think the platform level seems to be good. Then in Glorp one > could express > (Login new) > database: (PostgreSQLPlatform withEncoding: #utf8); > username: 'yyyyy'; > password: 'xxxx'; > connectString: 'host' , '_' , 'database' > > (PostgreSQLPlatform withEncoding: #utf8) would be short cut for > (PostgreSQLPlatform new encoding: #utf8). > > > > maybe we can just set again the NullEncoder in that case? or raise an > error ? what do you think? > > Yep, error checking code was not up to the task in the first try. > Probably raise an error. Espesially if user has decided to use > encoding (and data needs it). If > code would continue silently with NullEncoder the data read would be > corrupted. > > I agree. Raise an error with a really clear message. I would like to add tests for this encoding stuff, if it makes sense. I try to keep tests updated and trying to cover most of the features. > > > Ok.... I run the benchmarks and I was rigth > > If I understood correctly berformance dropped over 10%. It is quite > much. Have to see if that could be improved. > If you want, you can try it by yourself: http://www.squeakdbx.org/Benchmarks Easy, evaluate:* DBXBenchmark* facility: *DBXPostgreFacility* facilityForBenchmark. *DBXTinyBenchmarks* new runAll. > > > In addition, if someone has its own database using a particular encoding, > and doesn't implement openConnection (that does the query, like you did in > postgres) because he doesn't knwo, this > > will be a problem beasue the database will have an encoding X and you > will be encoding always with UTF8 ;) > > This is a problem. However I do feel that encoding should be checked > when database is connected (at least when user has selected that > encoding should be used). > yes, but only when the user has selected encoded > Also endocoding should be property of the connection. If it would be > class variable it would be impractical to connect to different > databases with different encoding. > > you are right, sorry about that. > I think one possibility to takle these problems would be > EncoderFactory user passes to the code. This factory would set encoder > in connection but weren't we going to put it in connectionSettings? > based either on the database query > or by preselected value. The query option would ask the connection > about encoding (using message encodingName or something like that). If > connection cant > handle the message then exception is raised (user chosed to > automatically select encoding in setting where code is not up to the > task). > > The preselected option would just set right encoder to the connection. > I think that validation of this choice should be skipped because it > would render this option to be basically the same than the automatic > option. Preselected encoding could be NullEncoder. > > Uhh sorry, you lost me in this last paragraph. Maybe if I see the code I understand better : > If this sound sensible I will try to implement the changes in the > weekend but cant promise I have the time. :) > > That would be cool, but don't worry, no stress. BTW...just by curious. where/why are you trying to use SqueakDBX? cheers mariano

3 24

Character encoding and pooling in SqueakDBX and Glorp
by Panu Suominen 27 Aug '10

27 Aug '10

I meant to reply to everyone but shaky hands pressed the wrong button... ---------- Forwarded message ---------- From: Panu Suominen <panu.j.m.suominen(a)gmail.com> Date: 2010/8/27 Subject: Re: [SqueakDBX] Character encoding and pooling in SqueakDBX and Glorp To: Mariano Martinez Peck <marianopeck(a)gmail.com> Thank you for excelent feedback. 2010/8/26 Mariano Martinez Peck <marianopeck(a)gmail.com>: > - I guess a lot of people doesn't need UTF (or other) encoding. And I guess > that the transformation brings some overhead in the query result mosty...(I > should run the benchmarks and see if there is difference). But I was > thinking if we could make this optional. I was thinking that too. Apparently no encoding means that only 7bit characters should be used? > or something like that....but the idea of making it optional not to have > overhead if not required, and being able to be enabled from the app code. This probably need some thinking. Because if one uses database through GLORP the encoding parameter should be passed somehow to lower levels. Would the ConnectionSettings in SqueakDBX level be right place? And in Glorp I think the platform level seems to be good. Then in Glorp one could express (Login new) database: (PostgreSQLPlatform withEncoding: #utf8); username: 'yyyyy'; password: 'xxxx'; connectString: 'host' , '_' , 'database' (PostgreSQLPlatform withEncoding: #utf8) would be short cut for (PostgreSQLPlatform new encoding: #utf8). > maybe we can just set again the NullEncoder in that case? or raise an error ? what do you think? Yep, error checking code was not up to the task in the first try. Probably raise an error. Espesially if user has decided to use encoding (and data needs it). If code would continue silently with NullEncoder the data read would be corrupted. > Ok.... I run the benchmarks and I was rigth If I understood correctly berformance dropped over 10%. It is quite much. Have to see if that could be improved. > In addition, if someone has its own database using a particular encoding, and doesn't implement openConnection (that does the query, like you did in postgres) because he doesn't knwo, this > will be a problem beasue the database will have an encoding X and you will be encoding always with UTF8 ;) This is a problem. However I do feel that encoding should be checked when database is connected (at least when user has selected that encoding should be used). Also endocoding should be property of the connection. If it would be class variable it would be impractical to connect to different databases with different encoding. I think one possibility to takle these problems would be EncoderFactory user passes to the code. This factory would set encoder in connection based either on the database query or by preselected value. The query option would ask the connection about encoding (using message encodingName or something like that). If connection cant handle the message then exception is raised (user chosed to automatically select encoding in setting where code is not up to the task). The preselected option would just set right encoder to the connection. I think that validation of this choice should be skipped because it would render this option to be basically the same than the automatic option. Preselected encoding could be NullEncoder. If this sound sensible I will try to implement the changes in the weekend but cant promise I have the time. :) -- Panu -- Panu

1 0

Character encoding and pooling in SqueakDBX and Glorp
by Panu Suominen 26 Aug '10

26 Aug '10

We have postgresql database using UTF-8 charset and I had problems doing queries against the database using umlaut characters. It looked like there was no character conversion done to the statements going to opendbx api. Or is it that I just did not find it? However I implemented a little fix that encodes strings going and coming from OpenDBX and mechanism that queries the encoding database uses and attaches this information to the connection. Unfortunately the charset query is currently only implemented for postgresql (show client_encoding). Where should I post these changes? Directly to squeaksource? Also is there a connection pooling facility for SqueakDBX or Glorp? -- Panu

2 6

SqueakDBX and MSSQL server
by Herbert König 21 Aug '10

21 Aug '10

Hi, i read through the list archives to find out how to connect to a MSSQL server under Win XP. Which dlls do I need? From my MySQL experiments I know that I need the backend dll which I took out of a mail from Mariano (Hi Mariano, how are you?) and a client dll from MS. Some mails suggest I also need the freeTDS library but the website says freeTDS is for Linux. So where to get the TDS library (if necessary) and what's the name of MS client library? As our ERP system uses ODBC to access MS SQL server and SqueakDBX also has a class DBXOdbcMSSQLFacility maybe it's easier to go through ODBC? Speed is not an issue, I expect no more than a few 1000 elements in any result set. Also which port to use to connect to MSSQL? I welcome every suggestion. Cheers, Herbert

2 1

how to use SqueakDBX with Oracle in Mac OS [WAS] Fwd: Problem with Oracle in Mac OS
by Mariano Martinez Peck 19 Aug '10

19 Aug '10

For those who needs to run SqueakDBX or OpenDBX with Oracle in Mac OS. Here is what Colin had to do in order to make it work. Cheers Mariano ---------- Forwarded message ---------- From: Colin Doherty <colin.doherty(a)gmail.com> Date: Thu, Aug 19, 2010 at 12:10 PM Subject: Re: Problem with Oracle in Mac OS To: Mariano Martinez Peck <marianopeck(a)gmail.com> Hi Mariano, This is what I did if you want to add it to the documentation: Platform: OSX 10.6.4 with Squeak 4.1, Oracle Instant Client for OSX 10.2 Installed SqueakDBX using Monticello. Installed Oracle Instant Client (downloaded base, sqlplus and sdk packages) to here: /usr/local/instantclient10_2 Created a subdirectory in /usr/local/instantclient10_2 called tns, containing tnsnames.ora file. Added the following environment settings to .bash_profile for compilation of libopendbx-1.5.0: DYLD_LIBRARY_PATH=/usr/local/instantclient10_2 ORACLE_HOME=/usr/local/instantclient10_2 TNS_ADMIN=/usr/local/instantclient10_2/tns Compiled libopendbx-1.5.0 using these flags: CFLAGS="-m32" CPPFLAGS="-m32 -I/usr/local/instantclient10_2/sdk/include" LDFLAGS="-arch i386 -L/usr/local/instantclient10_2" ./configure --disable-utils --with-backends=oracle make make install To get SqueakDBX/Oracle to work when Squeak is launched as an OSX application need to create the following launch configuration file /etc/launchd.conf, containing: setenv DYLD_LIBRARY_PATH /usr/local/instantclient10_2 setenv ORACLE_HOME /usr/local/instantclient10_2 setenv TNS_ADMIN /usr/local/instantclient10_2/tns Need to reboot for launchd.conf to take effect. You also need to add these environment settings to your .bash_profile in order to run squeak with SqueakDBX/Oracle from command line: DYLD_LIBRARY_PATH=/usr/local/lib:/usr/local/instantclient10_2 ORACLE_HOME=/usr/local/instantclient10_2 TNS_ADMIN=/usr/local/instantclient10_2/tns Cheers Colin On Thu, Aug 19, 2010 at 8:36 AM, Mariano Martinez Peck < marianopeck(a)gmail.com> wrote: > > > On Thu, Aug 19, 2010 at 2:48 AM, Colin Doherty <colin.doherty(a)gmail.com>wrote: > >> Hi, >> >> I've got it working. The problem turned out to be that environment that >> launched applications in OSX was very minimal and wasn't including all the >> required paths. You have to add the DYLD_LIBRARY_PATH to the Oracle >> libraries for the systems /etc/launchd.conf file and it then works, no >> memory errors etc. >> >> > Great news! I had a similar problem with DYLD_LIBRARY_PATH and oracle but > in Linux. See this link > http://www.squeakdbx.org/Oracle > > In http://www.squeakdbx.org/Different%20backends%20under%20Mac > you see there is no entry for Oracle ;) > So....if you want to give us a hand and help with the documentation, please > let me know. I can create you a user or to send me text. > > thanks > > mariano > > > >> Thanks for all your help previously. >> >> Cheers >> Colin >> >> >> On Wed, Aug 18, 2010 at 6:25 PM, Mariano Martinez Peck < >> marianopeck(a)gmail.com> wrote: >> >>> >>> >>> On Wed, Aug 18, 2010 at 7:21 PM, Colin Doherty <colin.doherty(a)gmail.com>wrote: >>> >>>> That's quite possible given my limited C skills. This is what I did, >>>> just added switch statement and declared error variable: >>>> >>>> >>> Don't worry. I am C limited also...Norbert is the expert here ;) >>> >>> >>>> sb4 error; >>>> >>>> if( ( conn->errcode = OCIEnvCreate( &env, OCI_THREADED, NULL, NULL, >>>> NULL, NULL, 0, NULL ) ) != OCI_SUCCESS ) >>>> { >>>> switch( conn->errcode ) >>>> { >>>> case OCI_SUCCESS: >>>> snprintf( conn->errmsg, OCI_ERROR_MAXMSG_SIZE, >>>> "Success" ); >>>> break; >>>> case OCI_SUCCESS_WITH_INFO: >>>> snprintf( conn->errmsg, OCI_ERROR_MAXMSG_SIZE, >>>> "Success with info" ); >>>> break; >>>> case OCI_NEED_DATA: >>>> snprintf( conn->errmsg, OCI_ERROR_MAXMSG_SIZE, >>>> "Need data" ); >>>> break; >>>> case OCI_NO_DATA: >>>> snprintf( conn->errmsg, OCI_ERROR_MAXMSG_SIZE, >>>> "No data" ); >>>> break; >>>> case OCI_ERROR: >>>> OCIErrorGet( (dvoid*) conn->err, 1, NULL, >>>> &error, (text*) conn->errmsg, OCI_ERROR_MAXMSG_SIZE, OCI_HTYPE_ERROR >>>> ); >>>> break; >>>> case OCI_INVALID_HANDLE: >>>> snprintf( conn->errmsg, OCI_ERROR_MAXMSG_SIZE, >>>> "Invalid handle" ); >>>> break; >>>> case OCI_STILL_EXECUTING: >>>> snprintf( conn->errmsg, OCI_ERROR_MAXMSG_SIZE, >>>> "Still executing" ); >>>> break; >>>> case OCI_CONTINUE: >>>> snprintf( conn->errmsg, OCI_ERROR_MAXMSG_SIZE, >>>> "Continue" ); >>>> break; >>>> default: >>>> snprintf( conn->errmsg, OCI_ERROR_MAXMSG_SIZE, >>>> "Unknown error" ); >>>> break; >>>> } >>>> >>>> printf("2-hndl-cleanup"); >>>> printf("ODBX_ERR_NOMEM1 error: %d \n", conn->errcode); >>>> oracle_odbx_error(handle); >>>> oracle_priv_handle_cleanup( handle ); >>>> return -ODBX_ERR_NOMEM; >>>> } >>>> >>> >>> And all the output of these printf where are they? did you check in the >>> console? If you run Squeak from the console, and squeak uses OpenDBX >>> binaries that you modified and added this log, they should appear in the >>> console output....But....be SURE you are using your own compiled and >>> modified opendbx binary :) >>> >>> do you know in which of the case it is entering? >>> >>> Thanks for all the testing you are doing...hope we can get it ;) >>> >>> cheers >>> >>> Mariano >>> >>> >>> >>>> >>>> However, the memory error was occurring in Squeak/Pharo with SqueakDBX >>>> before I did this, so even if there is a memory bug above it doesn't explain >>>> that error. >>>> >>> >>> maybe you are right. >>> >>> >>>> >>>> Cheers >>>> Colin >>>> >>>> >>>> On Wed, Aug 18, 2010 at 2:44 PM, Mariano Martinez Peck < >>>> marianopeck(a)gmail.com> wrote: >>>> >>>>> >>>>> >>>>> On Wed, Aug 18, 2010 at 3:17 PM, Colin Doherty < >>>>> colin.doherty(a)gmail.com> wrote: >>>>> >>>>>> Did you do what Norbert said about using the oracle functions to get >>>>>> more info about the error ? >>>>>> >>>>>> - Yes, I believe so, that's what produced the additional log entries >>>>>> from what I can see: >>>>>> >>>>>> >>>>> Can you paste please the code you did in OpenDBX for this? Maybe there >>>>> is something wrong there and that's what is causing the malloc error, and >>>>> not the real cause of the error ;) >>>>> >>>>> CC Norbert also. >>>>> >>>>> >>>>> >>>>>> Aug 15 21:54:09 Colin-Dohertys-MacBook >>>>>> [0x0-0x29029].org.squeak.SqueakAllInOne41[9243]: Squeak VM >>>>>> Opt(9243,0xa035f500) malloc: *** error for object 0xf1f2f3f4: pointer being >>>>>> freed was not allocated >>>>>> Aug 15 21:54:09 Colin-Dohertys-MacBook >>>>>> [0x0-0x29029].org.squeak.SqueakAllInOne41[9243]: *** set a breakpoint in >>>>>> malloc_error_break to debug >>>>>> Aug 15 21:54:09 Colin-Dohertys-MacBook Squeak VM Opt[9243]: Squeak VM >>>>>> Opt(9243,0xa035f500) malloc: *** error for object 0xf1f2f3f4: pointer being >>>>>> freed was not allocated\n*** set a breakpoint in malloc_error_break to debug >>>>>> >>>>>> From what I've read malloc_error_break is an OSX memory error, >>>>>> presumably it's trying to free memory on an object that doesn't exist ? >>>>>> It looks like you can use Xcode to set a malloc_error_break but I >>>>>> don't know how you would hook SqueakDBX to this, if it's possible ? >>>>>> >>>>>> >>>>> This is "complicated" since you have to compile the SqueakVM with XCode >>>>> and maybe put a breakpoint from there.... >>>>> >>>>> >>>>> >>>>>> Cheers >>>>>> Colin >>>>>> >>>>>> >>>>>> On Wed, Aug 18, 2010 at 9:39 AM, Mariano Martinez Peck < >>>>>> marianopeck(a)gmail.com> wrote: >>>>>> >>>>>>> >>>>>>> >>>>>>> On Sun, Aug 15, 2010 at 10:58 PM, Colin Doherty < >>>>>>> colin.doherty(a)gmail.com> wrote: >>>>>>> >>>>>>>> Works with odbxtest too: >>>>>>>> >>>>>>>> >>>>>>> ok....So..opendbx is working correctly. The problem is then with >>>>>>> SqueakDBX. >>>>>>> >>>>>>> >>>>>>>> Seem to have got some further debug info by adding the additional >>>>>>>> error code suggested for the OutOfMemory error occurring in Squeak, not sure >>>>>>>> if it helps: >>>>>>>> >>>>>>>> >>>>>>> I am not sure if this helps. Did you do what Norbert said about using >>>>>>> the oracle functions to get more info about the error ? >>>>>>> >>>>>>> >>>>>>> " use the OCI error functions and pass the return value to them (like >>>>>>> in oracle_odbx_error()). Maybe you get a descriptive error message and a >>>>>>> hint for further investigations." >>>>>>> >>>>>>> Probably with that we can know which is the problem... >>>>>>> >>>>>>> thanks for the testing >>>>>>> >>>>>>> Mariano >>>>>>> >>>>>>> >>>>>>> >>>>>>>> Aug 15 21:54:09 Colin-Dohertys-MacBook >>>>>>>> [0x0-0x29029].org.squeak.SqueakAllInOne41[9243]: Squeak VM >>>>>>>> Opt(9243,0xa035f500) malloc: *** error for object 0xf1f2f3f4: pointer being >>>>>>>> freed was not allocated >>>>>>>> Aug 15 21:54:09 Colin-Dohertys-MacBook >>>>>>>> [0x0-0x29029].org.squeak.SqueakAllInOne41[9243]: *** set a breakpoint in >>>>>>>> malloc_error_break to debug >>>>>>>> Aug 15 21:54:09 Colin-Dohertys-MacBook Squeak VM Opt[9243]: Squeak >>>>>>>> VM Opt(9243,0xa035f500) malloc: *** error for object 0xf1f2f3f4: pointer >>>>>>>> being freed was not allocated\n*** set a breakpoint in malloc_error_break to >>>>>>>> debug >>>>>>>> >>>>>>>> Thanks >>>>>>>> Colin >>>>>>>> >>>>>>>> On Sun, Aug 15, 2010 at 9:16 PM, Mariano Martinez Peck < >>>>>>>> marianopeck(a)gmail.com> wrote: >>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> On Sun, Aug 15, 2010 at 9:18 PM, Colin Doherty < >>>>>>>>> colin.doherty(a)gmail.com> wrote: >>>>>>>>> >>>>>>>>>> Hi Mariano, Norbert, >>>>>>>>>> >>>>>>>>>> I've got the odbxplustest to work with Oracle, it needed the fully >>>>>>>>>> qualified service name defined on the server rather than that defined >>>>>>>>>> locally in the TNS settings to function correctly: >>>>>>>>>> >>>>>>>>>> >>>>>>>>> and what about odbxtest ?? not plus....as that's uses c++ >>>>>>>>> >>>>>>>>> is it working too? >>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> However, in order to get around the problem of Squeak not finding >>>>>>>>>> the Oracle client library referenced by the odbx library I had to create the >>>>>>>>>> /b/32_216/rdbms/lib/ directory referenced from within the Oracle library >>>>>>>>>> libclntsh.dylib.10.1 distributed with Oracle Instant Client. When I try to >>>>>>>>>> then use this in Squeak I get a DBXFatalError: FATAL OpenDBX: Out of memory >>>>>>>>>> - note no errors are logged to the squeak debug log when this occurs. >>>>>>>>>> >>>>>>>>>> >>>>>>>>> Could you try what Norbert say to use Oracle funcions to see more >>>>>>>>> details about the problem? >>>>>>>>> >>>>>>>>> thanks a lot in advance >>>>>>>>> >>>>>>>>> mariano >>>>>>>>> >>>>>>>>> >>>>>>>>>> So it appears odbx is working but SqueakDBX seems to have an issue >>>>>>>>>> ? >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>> Thanks >>>>>>>>>> Colin >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> On Mon, Aug 2, 2010 at 8:27 PM, Mariano Martinez Peck < >>>>>>>>>> marianopeck(a)gmail.com> wrote: >>>>>>>>>> >>>>>>>>>>> Hi Norbert. Colin (in this mail, so please reply to all), has a >>>>>>>>>>> problem with Oracle in Mac OS. >>>>>>>>>>> >>>>>>>>>>> He can configure and compiles without any problem. But when >>>>>>>>>>> running the opendbx tests, he has the following error: >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> 1. Run: >>>>>>>>>>> Conn::Conn() >>>>>>>>>>> Caught exception: Out of memory >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> That problem is in odbx_init_oracle, as it is answering >>>>>>>>>>> ODBX_ERR_NOMEM in oracle_odbx_init() where it does: >>>>>>>>>>> >>>>>>>>>>> if( ( conn->errcode = OCIEnvCreate( &env, OCI_THREADED, >>>>>>>>>>> NULL, NULL, NULL, NULL, 0, NULL ) ) != OCI_SUCCESS ) >>>>>>>>>>> { >>>>>>>>>>> printf("2-hndl-cleanup"); >>>>>>>>>>> oracle_priv_handle_cleanup( handle ); >>>>>>>>>>> return -ODBX_ERR_NOMEM; >>>>>>>>>>> } >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> The error code was: ODBX_ERR_NOMEM1 error: -1 >>>>>>>>>>> >>>>>>>>>>> So OCIEnvCreate is answering -1.... >>>>>>>>>>> >>>>>>>>>>> I had the same problem in Windows, and my problem is that I was >>>>>>>>>>> using ANOTHER oci.dll. In his case, he checked and only has ONE library. >>>>>>>>>>> >>>>>>>>>>> So, we have no idea what to do. >>>>>>>>>>> >>>>>>>>>>> Any quotes? >>>>>>>>>>> >>>>>>>>>>> Thanks in advance, >>>>>>>>>>> >>>>>>>>>>> Mariano >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> On Mon, Aug 2, 2010 at 7:43 PM, Colin Doherty < >>>>>>>>>>> colin.doherty(a)gmail.com> wrote: >>>>>>>>>>> >>>>>>>>>>>> Hi, >>>>>>>>>>>> >>>>>>>>>>>> I've added the printf() to oracle_basic.c and it's failing here: >>>>>>>>>>>> >>>>>>>>>>>> if( ( conn->errcode = OCIEnvCreate( &env, OCI_THREADED, >>>>>>>>>>>> NULL, NULL, NULL, NULL, 0, NULL ) ) != OCI_SUCCESS ) >>>>>>>>>>>> { >>>>>>>>>>>> printf("2-hndl-cleanup"); >>>>>>>>>>>> oracle_priv_handle_cleanup( handle ); >>>>>>>>>>>> return -ODBX_ERR_NOMEM; >>>>>>>>>>>> } >>>>>>>>>>>> >>>>>>>>>>>> Cheers >>>>>>>>>>>> Colin >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>> On Sun, Aug 1, 2010 at 8:43 PM, Mariano Martinez Peck < >>>>>>>>>>>> marianopeck(a)gmail.com> wrote: >>>>>>>>>>>> >>>>>>>>>>>>> maybe you can edit oracle_basic.c and put some flags in the >>>>>>>>>>>>> function oracle_odbx_init() >>>>>>>>>>>>> There, you will see several times it is returned >>>>>>>>>>>>> -ODBX_ERR_NOMEM >>>>>>>>>>>>> (the error you have) >>>>>>>>>>>>> >>>>>>>>>>>>> So maybe we should detect WHERE it is failing, and then, why. >>>>>>>>>>>>> >>>>>>>>>>>>> You can just add some printf() and then just recompile. >>>>>>>>>>>>> >>>>>>>>>>>>> I know this is crap....but it is the only thing I can do. >>>>>>>>>>>>> >>>>>>>>>>>>> cheers >>>>>>>>>>>>> >>>>>>>>>>>>> mariano >>>>>>>>>>>>> >>>>>>>>>>>>> >>>>>>>>>>>>> On Sun, Aug 1, 2010 at 9:39 PM, Mariano Martinez Peck < >>>>>>>>>>>>> marianopeck(a)gmail.com> wrote: >>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> ---------- Forwarded message ---------- >>>>>>>>>>>>>> From: OpenDBX devel list < >>>>>>>>>>>>>> libopendbx-devel(a)lists.sourceforge.net> >>>>>>>>>>>>>> Date: Tue, Jun 9, 2009 at 9:34 PM >>>>>>>>>>>>>> Subject: Re: [opendbx] OpenDBX binaries for Windows >>>>>>>>>>>>>> To: libopendbx-devel(a)lists.sourceforge.net >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> Hi Mariano >>>>>>>>>>>>>> >>>>>>>>>>>>>> > I tried again with your dll for oracle and I get the SAME >>>>>>>>>>>>>> error I got when >>>>>>>>>>>>>> > I compiled by myself. I have the OUT_OF_MEMORY problem. See >>>>>>>>>>>>>> mail with >>>>>>>>>>>>>> > subject: 'Oracle in Windows' (i couldnt find it in gmane). >>>>>>>>>>>>>> Perhaps during >>>>>>>>>>>>>> > this week I can put some flags and see where the problem is. >>>>>>>>>>>>>> >>>>>>>>>>>>>> Did you found something new? >>>>>>>>>>>>>> >>>>>>>>>>>>>> ODBX_ERR_NOMEM is probably returned by oracle_odbx_init() when >>>>>>>>>>>>>> a handle >>>>>>>>>>>>>> couldn't be allocated. >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> Norbert >>>>>>>>>>>>>> -- >>>>>>>>>>>>>> OpenPGP public key >>>>>>>>>>>>>> http://www.linuxnetworks.de/norbert.pubkey.asc >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> ------------------------------------------------------------------------------ >>>>>>>>>>>>>> Crystal Reports - New Free Runtime and 30 Day Trial >>>>>>>>>>>>>> Check out the new simplified licensing option that enables >>>>>>>>>>>>>> unlimited >>>>>>>>>>>>>> royalty-free distribution of the report engine for externally >>>>>>>>>>>>>> facing >>>>>>>>>>>>>> server and web deployment. >>>>>>>>>>>>>> http://p.sf.net/sfu/businessobjects >>>>>>>>>>>>>> >>>>>>>>>>>>>> _______________________________________________ >>>>>>>>>>>>>> libopendbx-devel mailing list >>>>>>>>>>>>>> libopendbx-devel(a)lists.sourceforge.net >>>>>>>>>>>>>> https://lists.sourceforge.net/lists/listinfo/libopendbx-devel >>>>>>>>>>>>>> http://www.linuxnetworks.de/doc/index.php/OpenDBX >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>> >>>>>> >>>>> >>>> >>> >> >

1 0

Re: Problems installing glorp
by Mariano Martinez Peck 13 Aug '10

13 Aug '10

On Fri, Aug 13, 2010 at 10:18 PM, Douglas Naka <douglaseiji(a)yahoo.com.br>wrote: > Hi, Mariano, > > Now all the openDBX tests are all passing, but I'm having problems to > configure the environment to run the GLORP tests. Could you help me please? > I read this tutorial ( > http://www.assembla.com/wiki/show/SqueakDBX/GlorpDBX_Tutorial) that says > that I have to modify this class: > DBXGlorpBackendFacility<http://www.assembla.com/wiki/show/SqueakDBX/DBXGlorpBackendFacility>. But I dont have this class at my image. Do you know what I need to do to > set up the environment to run Glorp tests? > > Hi. In order to run sucessfully Glorp tests you should need to change the database connection settings used by them. To do this, change the method #glorpBackendFacility of the class that you want to use (depending on which database). Once this is done, it comes the tricky part: Open a test runner, and run the test DBXGlorpMainBackendTestPostgresql >> testSetToBaseTestAFacility that is not actually a test, it just sets en enviorment. Now...you should be able to run all the Glorp tests under the packages: GlorpTest and GlorpTest-DBTests. Let me know if it works. BTW you can post in the squeakdbx mailing list. Cheers Mariano > Thanks. > Douglas > > ------------------------------ > *De:* Mariano Martinez Peck <marianopeck(a)gmail.com> > *Para:* Douglas Naka <douglaseiji(a)yahoo.com.br> > *Enviadas:* Terça-feira, 3 de Agosto de 2010 14:46:48 > *Assunto:* Re: Problems installing glorp > > HI. I don't know if you are aware, but we (me and others) modified Glorp so > that it can support different database drivers and not only postgresql. We > called this new Glorp, GlorpDBX as it is part of the SqueakDBX project. > > http://www.squeakdbx.org/GLORP%20integration > > The recommened way to install GlorpDBX is using Metacello. There is a > configuration for GlorpDBX, but I didn't do for Glorp. Sorry, did't have too > much time. > > So...if you want to install GlorpDBX + SqueakDBX driver, you have to follow > this instructions: > > http://www.squeakdbx.org/Installation > > (at the end, where it speaks about metacello configs). > > I suggest you read about Metacello. > > If you want to install the original Glorp, you should do it by yourself. > What I suggest is that if you want the postgres native driver (instead of > squeakdbx) you create a new config based on ConfigurationOfGlorpDBX but > that doesn't load SqueakDBX, but the postgres driver. > > Just see the difference with GlorpPharoLoader >> > loadGlorpWithNativePostgreSQLDriver > > Finally, the installer is just a shortcut to do it by code. You can do what > Installer do, by hand, using the Monticello browser. > > cheers > > mariano > > > On Tue, Aug 3, 2010 at 7:34 PM, Douglas Naka <douglaseiji(a)yahoo.com.br>wrote: > >> I forgot to say, I'm using Pharo 1.0. >> >> >> ------------------------------ >> *De:* Douglas Naka <douglaseiji(a)yahoo.com.br> >> *Para:* marianopeck(a)gmail.com >> *Enviadas:* Terça-feira, 3 de Agosto de 2010 14:16:11 >> *Assunto:* Problems installing glorp >> >> Hi Mariano, >> >> I was trying to install Glorp and I'm having problems. Could you help me? >> >> I evaluated this: >> >> |*instClass| >> *instClass := Smalltalk at: #Installer ifAbsent: [ScriptLoader >> loadLatestPackage: 'Installer-Core' fromSqueaksource: 'Installer'. Smalltalk >> >> at: #Installer]. >> instClass ss project: 'Glorp'; install: 'GlorpPharoLoader'. >> (Smalltalk at: #GlorpPharoLoader) loadGlorp. >> >> but I got the error: "doesNotUnderstand: #squeakSource" >> >> >> Do you know what is missing? >> >> Thanks >> Douglas >> >> >> >> >> >> >> > > > >

1 0

Problem In Ubuntu 8.02
by Diogenes Moreira 12 Aug '10

12 Aug '10

Hi folks.. i'm trying use GlorpDBX in Pharo 1.1 on a Ubuntu 8.02 I install opendbx without problems (from source) but im get "unable to find function address" here es my context varibles.. TERM=xterm SHELL=/bin/bash SSH_CLIENT=24.232.62.194 60419 1874 PG_HOME=/usr/local/pgsql SSH_TTY=/dev/pts/0 PGPORT=5499 LD_LIBRARY_PATH=:/usr/local/pgsql/lib:/usr/local/lib:/usr/lib MAIL=/var/mail/diogenes PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/pgsql/bin:/usr/local/lib:/usr/lib PWD=/home/diogenes/www EDITOR=vim PGHOST=localhost HISTCONTROL=ignoreboth SHLVL=1 HOME=/home/diogenes LOGNAME=diogenes LESSOPEN=| /usr/bin/lesspipe %s LESSCLOSE=/usr/bin/lesspipe %s %s PGDATESTYLE=European, SQL OLDPWD=/home/diogenes _=/usr/bin/env I forgot something? Thanks in Advance

2 3

MAC OSX 10.6 + squeakDBX + Oracle 10.2 issues
by Colin Doherty 01 Aug '10

01 Aug '10

Hi, I've tried installing squeakDBX and got it installed using the 32 bit Oracle 10.2.0.4 instant client using this config: CFLAGS="-m32" CPPFLAGS="-m32 -I/usr/local/instantclient10_2/sdk/include" LDFLAGS="-arch i386 -L/usr/local/instantclient10_2" ./configure --disable-utils --with-backends="oracle" However, I'm getting the following error when I try to use it in either squeak 4.1 or pharo 1.1: Out of memory error This is from the log: Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading /Users/DevelopmentSmalltalk/Pharo/Plugins/opendbx.so Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading /Users/Development/Smalltalk/Pharo/Plugins/opendbx.dylib Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading /Users/Development/Smalltalk/Pharo/Plugins/libopendbx Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading /Users/Development/Smalltalk/Pharo/Plugins/libopendbx.so Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading /Users/Development/Smalltalk/Pharo/Plugins/libopendbx.dylib Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading ./opendbx.bundle/Contents/MacOS/opendbx Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading ./opendbx Jul 31 20:57:47: --- last message repeated 1 time --- Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading ./opendbx.so Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading ./opendbx.dylib Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading ./libopendbx Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading ./libopendbx.so Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading ./libopendbx.dylib Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading /Users/Development/Smalltalk/Pharo/Pharo-1.1-OneClick.app/Contents/Resources/opendbx.bundle/Contents/MacOS/opendbx Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: tryLoading /Users/Development/Smalltalk/Pharo/Pharo-1.1-OneClick.app/Contents/Resources/opendbx Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: ioFindExternalFunctionIn(odbx_init, 2155408) Jul 31 20:57:47 MacBook Squeak VM Opt[22081]: Squeak VM Opt(22081,0xa0bf2500) malloc: *** error for object 0xf1f2f3f4: pointer being freed was not allocated\n*** set a breakpoint in malloc_error_break to debug Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: Squeak VM Opt(22081,0xa0bf2500) malloc: *** error for object 0xf1f2f3f4: pointer being freed was not allocated Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: *** set a breakpoint in malloc_error_break to debug Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: ioFindExternalFunctionIn(odbx_error, 2155408) Jul 31 20:57:47 MacBook [0x0-0x39039].org.squeak.pharo[22081]: ioFindExternalFunctionIn(odbx_error_type, 2155408) Jul 31 21:01:49 MacBook Safari[8492]: INSERT-HANG-DETECTED: Tx time:18.496361, # of Inserts: 0, # of bytes written: 0, Did shrink: YES Jul 31 21:05:31 MacBook [0x0-0x39039].org.squeak.pharo[22081]: VM: MouseModifierStateCarbon buttonStateBits 1 modifier I've also tried compiling a 64 bit version which worked and used the experimental 64 bit squeak vm, this doesn't work but generates no errors. Any ideas on how to fix the 32 bit out of memory issue ? Thanks

2 17

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

SqueakDBX August 2010