mdbtools-dev

Re: [mdb-dev] no data from jet 3 ?

[Rob <ro...@ir...>]

Thanks for the response!

I tried converting the database back to prior version through Access 2000, 
and get the same result as before; mdb-tables doesn't list anything for the 
converted db.  The Access2000 version of the database lists fine.

The App is Shooting Star http://www.foodogsoftware.com/ - an eBay auction 
management utility.  I'd rather not change this, as it does everything I 
need it to do, and I've got rather a lot of historical data stored in 
it.  I'm just trying to write a routine to track our ebay sales and tie 
them into our own web store ( http://www.russianglass.co.uk ).

If I give it the upconverted database, I get the error:
Error #3343 Unrecognized database format 'G:\blah\userdata.mdb'.
      Jet Version: 3.51
      Source: DAO.Workspace
      Help Topic 5003343 in jeterr35.hlp

It is quite happy, however, with having it's database password removed, and 
didn't seem to mind the db which had been converted up and back down again.

I presume that despite my having Access 2000 installed on this machine, 
it's using a lesser version runtime routines?  (It's much the same error if 
I try to open the later version db in Access 97.)

Thanks,

Rob


At 03:43 15/11/2004, Sam Moffatt wrote:

>In Access 2000, there is an option under the tools menu option to save
>it as an older version of the database, so you don't need access 97 to
>save databases in that format. Perhaps try using that. Also, what
>language is your windows app running? if you've got access 2000
>running, it shouldn't have any issues. and can you change the windows
>app?
>
>Sam
>
>
>On Mon, 15 Nov 2004 01:58:46 +0000, Rob <ro...@ir...> wrote:
> > Hi!
> >
> > I've been working on an ap in php, accessing an access database via odbc &
> > mdbtools.
> >
> > The database is actually from windows application, and appears to be using
> > the jet 3.51 engine for access.   As it was password protected (but I know
> > the password) I loaded it into Access 2000, converted to new format,
> > cleared the password, and saved it out as a temporary copy.  mdb tools
> > works fine on this copy, but I guess it's now in Jet 4 format.  The windows
> > app certainly objects to it anyway.
> >
> > I finally got around to installing office 97 on a pc so I could remove the
> > password from the database without converting the format.  This I managed
> > to do, and the windows app is happy with it, however mdb tools refuses to
> > return any data from it.  Even mdb-tables just prints a blank line and 
> exits.
> >
> > mdb-tables does however list the tables in Northwind.mdb samed from the
> > office 97 install.
> >
> > I'm using 0.6pre1, downloaded Nov 1st, running on FreebSD 4.10-STABLE #0.
> >
> > Is there anything else I can try, do, tell you, to solve this?
> >
> > Thanks,
> >
> > Rob.

Re: [mdb-dev] no data from jet 3 ?

[Sam M. <pa...@gm...>]

In Access 2000, there is an option under the tools menu option to save
it as an older version of the database, so you don't need access 97 to
save databases in that format. Perhaps try using that. Also, what
language is your windows app running? if you've got access 2000
running, it shouldn't have any issues. and can you change the windows
app?

Sam


On Mon, 15 Nov 2004 01:58:46 +0000, Rob <ro...@ir...> wrote:
> Hi!
> 
> I've been working on an ap in php, accessing an access database via odbc &
> mdbtools.
> 
> The database is actually from windows application, and appears to be using
> the jet 3.51 engine for access.   As it was password protected (but I know
> the password) I loaded it into Access 2000, converted to new format,
> cleared the password, and saved it out as a temporary copy.  mdb tools
> works fine on this copy, but I guess it's now in Jet 4 format.  The windows
> app certainly objects to it anyway.
> 
> I finally got around to installing office 97 on a pc so I could remove the
> password from the database without converting the format.  This I managed
> to do, and the windows app is happy with it, however mdb tools refuses to
> return any data from it.  Even mdb-tables just prints a blank line and exits.
> 
> mdb-tables does however list the tables in Northwind.mdb samed from the
> office 97 install.
> 
> I'm using 0.6pre1, downloaded Nov 1st, running on FreebSD 4.10-STABLE #0.
> 
> Is there anything else I can try, do, tell you, to solve this?
> 
> Thanks,
> 
> Rob.
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: InterSystems CACHE
> FREE OODBMS DOWNLOAD - A multidimensional database that combines
> robust object and relational technologies, making it a perfect match
> for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8
> _______________________________________________
> mdbtools-dev mailing list
> mdb...@li...
> https://lists.sourceforge.net/lists/listinfo/mdbtools-dev
>

[mdb-dev] no data from jet 3 ?

[Rob <ro...@ir...>]

Hi!

I've been working on an ap in php, accessing an access database via odbc & 
mdbtools.

The database is actually from windows application, and appears to be using 
the jet 3.51 engine for access.   As it was password protected (but I know 
the password) I loaded it into Access 2000, converted to new format, 
cleared the password, and saved it out as a temporary copy.  mdb tools 
works fine on this copy, but I guess it's now in Jet 4 format.  The windows 
app certainly objects to it anyway.

I finally got around to installing office 97 on a pc so I could remove the 
password from the database without converting the format.  This I managed 
to do, and the windows app is happy with it, however mdb tools refuses to 
return any data from it.  Even mdb-tables just prints a blank line and exits.

mdb-tables does however list the tables in Northwind.mdb samed from the 
office 97 install.

I'm using 0.6pre1, downloaded Nov 1st, running on FreebSD 4.10-STABLE #0.

Is there anything else I can try, do, tell you, to solve this?

Thanks,

Rob.

Re: [mdb-dev] Compiling and glib

[Jeff S. <why...@ya...>]

Hm..., -lreadline should have come before -lncurses in READLINE_LIBS.
That could be the cause of the linker error.  How it got in that order
I do not know.  You might get past this problem by editing the line in
src/util/Makefile that says
   READLINE_LIBS=' -lncurses -lreadline'
to
   READLINE_LIBS=' -lreadline -lncurses'
and then start up make again.

Also note that some readline problems have been corrected recently,
so you may want to try the version of mdbtools in cvs.

 -- Jeff Smith


--- Martin Trautmann <tr...@gm...> wrote:
> On 2004-11-09 12:42, Martin Trautmann wrote:
> > the archives told me about a solved problem for glib support.
> > 
> > I just tried both 0.5 and 0.6pre. Both share the same problem:
>  
> >  checking for GLIB - version >= 2.0.0... no
> >  *** Could not run GLIB test program, checking why...
> >  *** The test program failed to compile or link. See the file config.log for the
> >  *** exact error that occured. This usually means GLIB is incorrectly installed.
> >  
> >  glib 2.0 is required by MDB Tools.
> >  It can be downloaded at www.gtk.org.
> 
> I ended up with a copy of glib.h in $PREFIX/include/
> and ran successfully til mdb-sql.c:23:31: readline/readline.h
> 
> ac_cv_lib_readline_add_history=yes
> READLINE_LIBS=' -lncurses -lreadline'
> 
>     undefined reference to `tputs'
>     undefined reference to `tgetflag'
> 
> yet another problem prevented to compile readline-5.0, while I still use
> readline 4.3


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

Re: [mdb-dev] Compiling and glib

[Martin T. <tr...@gm...>]

On 2004-11-09 12:42, Martin Trautmann wrote:
> the archives told me about a solved problem for glib support.
> 
> I just tried both 0.5 and 0.6pre. Both share the same problem:
 
>  checking for GLIB - version >= 2.0.0... no
>  *** Could not run GLIB test program, checking why...
>  *** The test program failed to compile or link. See the file config.log for the
>  *** exact error that occured. This usually means GLIB is incorrectly installed.
>  
>  glib 2.0 is required by MDB Tools.
>  It can be downloaded at www.gtk.org.

I ended up with a copy of glib.h in $PREFIX/include/
and ran successfully til mdb-sql.c:23:31: readline/readline.h

ac_cv_lib_readline_add_history=yes
READLINE_LIBS=' -lncurses -lreadline'

    undefined reference to `tputs'
    undefined reference to `tgetflag'

yet another problem prevented to compile readline-5.0, while I still use
readline 4.3

[mdb-dev] Compiling and glib

[Martin T. <tr...@gm...>]

Hi all,

the archives told me about a solved problem for glib support.

I just tried both 0.5 and 0.6pre. Both share the same problem:

My glib.h (from version 2.5.5) is located in

  $PREFIX/include/glib-2.0

./configure --prefix=$PREFIX

... will complain:

 checking for GLIB - version >= 2.0.0... no
 *** Could not run GLIB test program, checking why...
 *** The test program failed to compile or link. See the file config.log for the
 *** exact error that occured. This usually means GLIB is incorrectly installed.
 
 glib 2.0 is required by MDB Tools.
 It can be downloaded at www.gtk.org.


I tried both --disable-glibtest and configuring PKG_CONFIG_PATH (or
PKG_CONFIG) - it does not accept a common $PATH format

ok, installing pkgconfig-0.15.0... no successs either

Which version is required here?

Thanks,
Martin

[mdb-dev] Weird behaviour (cleanup) when providing a python wrapper

[Christian T. <ct...@go...>]

Hi,

I'm having some problems with cleanup and sane results. I created a
python module using SWIG and added a couple of helper methods to handle
the GPtrArray and GArray constructs which aren't mapped/wrapped by
default.

The behaviour is quite confusing and let's me think if there is some
weird interaction during python/c-mode switches.

Example: When I open multiple database files after a sequence of=20

mdb_init()
mdb_open()
.
.
mdb_free_handle()
mdb_exit()

Later on, calls to mdb_read_catalog() will return all tables from the
earlier mdb files as well.

The same happens during one session when touching multiple tables
regarding the columns.

It does not happen during later calls to mdb_read_columns() that are
unwrapped by SWIG.

Any hint or idea to special behaviour I should look out for would be
helpful.

Thanks,
Christian

--=20
gocept gmbh & co. kg - schalaunische str. 6 - 06366 koethen - germany
www.gocept.com - ct...@go... - phone +49 3496 30 99 112 -
fax +49 3496 30 99 118 - zope and plone consulting and development

Re: [mdb-dev] Re: Writing

[Avner P. <av...@in...>]

So the update_row() function in write.c only works on Jet3?
Or is the problem in the util?
I know update_row uses mdb_crack_row to get the fields which works with 
Jet4 and then just updates the field's value and size according to the 
bind pointers.. but the value it upates doesn't contain those extra 
bytes .. can you further explain the differences and how to fix it?

Btw, on a side note, I have updated the parser and sql library to 
implement basic UPDATE and INSERT commands with update_row and 
insert_row. the UPDATE works good except the problem I told you. and 
INSERT inserts but the table doesn't get updated..

Jeff Smith wrote:

>Sorry for not responding to your first message.  In short, the updrow tool is quite
>primitive, and currently only works on text fields in Jet 3 (Access 97) databases.  I had
>never even taken a look at the tool before receiving your first message.
>
> -- Jeff Smith
>
>
>--- Avner Peled <av...@in...> wrote:
>  
>
>>Has no one else came across these issues? or no one is crazy enough to try..
>>I tried insert now too..and it does insert but it seems to insert it on 
>>the top and not update the table def correctly, the table doesn't show 
>>with the new row
>>
>>Avner Peled wrote:
>>
>>    
>>
>>>Hello!
>>>
>>>I'm having problems with writing (for example using the updrow 
>>>application) in latest mdbtools cvs.
>>>
>>>when changing a numerical value, the value and the value next to it 
>>>turn into garbage.
>>>
>>>when changing a string value, the following things happens:
>>>
>>>if my table, t_people, looks like this
>>>
>>>id      first_name       last_name      age    zipcode
>>>1       avner               peled              21       9540
>>>2       ...                      ....                   ...
>>>
>>>and I enter
>>>
>>>updrow avner1.mdb t_people "id = 1" last_name=shalom
>>>
>>>then this:
>>>
>>>0fdb  05 00 01 00 00 00 15 00  00 00 44 25 00 00 ff fe   ..........D%....
>>>0feb  61 76 6e 65 72 ff fe 70  65 6c 65 64 1c 00 15 00   avner..peled....
>>>0ffb  0e 00 02 00 
>>>1f                                                         .....
>>>
>>>turns into this:
>>>
>>>0000  05 00 01 00 00 00 15 00  00 00 44 25 00 00 ff fe      
>>>..........D%....
>>>0010  61 76 6e 65 72 73 68 61  6c 6f 6d 1b 00 15 00 0e   avnershalom.....
>>>0020  00 02 00 1f 
>>>when viewing the table after the update, last_name appears as "sao" 
>>>instead of "shalom"
>>>
>>>it seems the update ignores the "ff fe" in the start of the string, 
>>>what is that anwway?
>>>
>>>Any ideas? any help would be appreciated.
>>>If anyone wants I will also post the nuber update which turns small 
>>>numbers into negative and very big numbers.
>>>
>>>Thanks!
>>>
>>>Avner Peled
>>>av...@in...
>>>      
>>>
>
>
>
>		
>__________________________________
>Do you Yahoo!?
>Yahoo! Mail Address AutoComplete - You start. We finish.
>http://promotions.yahoo.com/new_mail 
>
>
>-------------------------------------------------------
>This Newsletter Sponsored by: Macrovision 
>For reliable Linux application installations, use the industry's leading
>setup authoring tool, InstallShield X. Learn more and evaluate 
>today. http://clk.atdmt.com/MSI/go/ins0030000001msi/direct/01/
>_______________________________________________
>mdbtools-dev mailing list
>mdb...@li...
>https://lists.sourceforge.net/lists/listinfo/mdbtools-dev
>  
>

Re: [mdb-dev] Re: Writing

[Jeff S. <why...@ya...>]

Sorry for not responding to your first message.  In short, the updrow tool is quite
primitive, and currently only works on text fields in Jet 3 (Access 97) databases.  I had
never even taken a look at the tool before receiving your first message.

 -- Jeff Smith


--- Avner Peled <av...@in...> wrote:
> Has no one else came across these issues? or no one is crazy enough to try..
> I tried insert now too..and it does insert but it seems to insert it on 
> the top and not update the table def correctly, the table doesn't show 
> with the new row
> 
> Avner Peled wrote:
> 
> > Hello!
> >
> > I'm having problems with writing (for example using the updrow 
> > application) in latest mdbtools cvs.
> >
> > when changing a numerical value, the value and the value next to it 
> > turn into garbage.
> >
> > when changing a string value, the following things happens:
> >
> > if my table, t_people, looks like this
> >
> > id      first_name       last_name      age    zipcode
> > 1       avner               peled              21       9540
> > 2       ...                      ....                   ...
> >
> > and I enter
> >
> > updrow avner1.mdb t_people "id = 1" last_name=shalom
> >
> > then this:
> >
> > 0fdb  05 00 01 00 00 00 15 00  00 00 44 25 00 00 ff fe   ..........D%....
> > 0feb  61 76 6e 65 72 ff fe 70  65 6c 65 64 1c 00 15 00   avner..peled....
> > 0ffb  0e 00 02 00 
> > 1f                                                         .....
> >
> > turns into this:
> >
> > 0000  05 00 01 00 00 00 15 00  00 00 44 25 00 00 ff fe      
> > ..........D%....
> > 0010  61 76 6e 65 72 73 68 61  6c 6f 6d 1b 00 15 00 0e   avnershalom.....
> > 0020  00 02 00 1f 
> > when viewing the table after the update, last_name appears as "sao" 
> > instead of "shalom"
> >
> > it seems the update ignores the "ff fe" in the start of the string, 
> > what is that anwway?
> >
> > Any ideas? any help would be appreciated.
> > If anyone wants I will also post the nuber update which turns small 
> > numbers into negative and very big numbers.
> >
> > Thanks!
> >
> > Avner Peled
> > av...@in...



		
__________________________________
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
http://promotions.yahoo.com/new_mail 

[mdb-dev] Re: Writing

[Avner P. <av...@in...>]

Has no one else came across these issues? or no one is crazy enough to try..
I tried insert now too..and it does insert but it seems to insert it on 
the top and not update the table def correctly, the table doesn't show 
with the new row

Avner Peled wrote:

> Hello!
>
> I'm having problems with writing (for example using the updrow 
> application) in latest mdbtools cvs.
>
> when changing a numerical value, the value and the value next to it 
> turn into garbage.
>
> when changing a string value, the following things happens:
>
> if my table, t_people, looks like this
>
> id      first_name       last_name      age    zipcode
> 1       avner               peled              21       9540
> 2       ...                      ....                   ...
>
> and I enter
>
> updrow avner1.mdb t_people "id = 1" last_name=shalom
>
> then this:
>
> 0fdb  05 00 01 00 00 00 15 00  00 00 44 25 00 00 ff fe   ..........D%....
> 0feb  61 76 6e 65 72 ff fe 70  65 6c 65 64 1c 00 15 00   avner..peled....
> 0ffb  0e 00 02 00 
> 1f                                                         .....
>
> turns into this:
>
> 0000  05 00 01 00 00 00 15 00  00 00 44 25 00 00 ff fe      
> ..........D%....
> 0010  61 76 6e 65 72 73 68 61  6c 6f 6d 1b 00 15 00 0e   avnershalom.....
> 0020  00 02 00 1f 
> when viewing the table after the update, last_name appears as "sao" 
> instead of "shalom"
>
> it seems the update ignores the "ff fe" in the start of the string, 
> what is that anwway?
>
> Any ideas? any help would be appreciated.
> If anyone wants I will also post the nuber update which turns small 
> numbers into negative and very big numbers.
>
> Thanks!
>
> Avner Peled
> av...@in...
>
>
>
>

Re: [mdb-dev] compilation issues

[Jeff S. <why...@ya...>]

1. I fixed the GLib issue in CVS several days ago.
2. As Christian said, autogen.sh will generate all files necessary to run configure. 
Note that it is best to have fairly up-to-date versions of automake, autoconf, and
libtool.

 -- Jeff Smith

--- Peter Scheer <phm...@st...> wrote:
> Hi,
> 
> I am following the discussions on this list for a while now but I am
> still using the 0.5 version. But unfortunately I need to upgrade to 0.6
> because BOOL fields are not handled properly in my current version
> (always returns NULL with mdb-export).
> 
> But 0.6 does not compile on my system because of the GLIB dependency.
> When do you think this will be fixed?
> 
> 
> And as a second question, I just did a
>     cvs -z3
> -d:pserver:ano...@cv...:/cvsroot/mdbtools co
> mdbtools
> but that downloads all files except for the configure, config.guess and
> config.sub
> 
> Did I do something wrong here?
> Please tell me because I have never used cvs before...
> 
> Thanks in advance
> 
> Peter
> 
> Jeff Smith wrote:
> 
> >Sorry about that.  I did not mean to incorporate a dependency on GLib 2.4.
> >I will correct this shortly.
> >
> > -- Jeff
> ><snip>
> >  
> >
> >>>>>>- g_ptr_array_foreach not in my glibc 2.2.3 version => changed
> >>>>>>            
> >>>>>>
> >>>>                                    ^^^^^
> >>>>I assume you mean glib
> >>>>        
> >>>>
> >>>Oops ! yes of course.
> >>>      
> >>>



		
_______________________________
Do you Yahoo!?
Express yourself with Y! Messenger! Free. Download now. 
http://messenger.yahoo.com

Re: [mdb-dev] compilation issues

[Christian L. <chr...@la...>]

On mer, 2004-10-27 at 16:49, Peter Scheer wrote:

> And as a second question, I just did a
>     cvs -z3
> -d:pserver:ano...@cv...:/cvsroot/mdbtools co
> mdbtools
> but that downloads all files except for the configure, config.guess and
> config.sub
they must be generated by aclocal, using autogen.sh script which is
in the cvs

Bye, Pif.

Re: [mdb-dev] compilation issues

[Peter S. <phm...@st...>]

Hi,

I am following the discussions on this list for a while now but I am
still using the 0.5 version. But unfortunately I need to upgrade to 0.6
because BOOL fields are not handled properly in my current version
(always returns NULL with mdb-export).

But 0.6 does not compile on my system because of the GLIB dependency.
When do you think this will be fixed?


And as a second question, I just did a
    cvs -z3
-d:pserver:ano...@cv...:/cvsroot/mdbtools co
mdbtools
but that downloads all files except for the configure, config.guess and
config.sub

Did I do something wrong here?
Please tell me because I have never used cvs before...

Thanks in advance

Peter

Jeff Smith wrote:

>Sorry about that.  I did not mean to incorporate a dependency on GLib 2.4.
>I will correct this shortly.
>
> -- Jeff
><snip>
>  
>
>>>>>>- g_ptr_array_foreach not in my glibc 2.2.3 version => changed
>>>>>>            
>>>>>>
>>>>                                    ^^^^^
>>>>I assume you mean glib
>>>>        
>>>>
>>>Oops ! yes of course.
>>>      
>>>

[mdb-dev] Re: mdb-export producing bad results on one particular table

[Calvin S. <cal...@ho...>]

Using current cvs and changing the 16 back to a 12 also fixes bug #1049013 
which has an attachment of a database that demonstrates the problem.  Also 
attached to the bug is the same database after 'compressing' it from withing 
access itself.  After compression then it works with both 16 and 12

I'm not familar with the mdb file structure itself but it seems there needs 
to be a check for when to use 12 and when to use 16

Calvin

--- Original Message ---
Well .. I"ve got it !

   As 0.5 version worked fine for me, I tried to find from which version
it became bad, and i found this :

revision 1.28
date: 2004/01/06 00:42:08;  author: brianb;  state: Exp;  lines: +2 -2
patches 00-14 from David Mansfield

RCS file: /cvsroot/mdbtools/mdbtools/src/libmdb/file.c,v
retrieving revision 1.28
retrieving revision 1.27
diff -b -r1.28 -r1.27
27c27
< 	4096, 0x0c, 16, 45, 47, 51, 55, 56, 63, 12, 15, 23, 5, 25, 59
---
 > 	4096, 0x0c, 12, 45, 47, 51, 55, 56, 63, 12, 15, 23, 5, 25, 59

   Without this change, it works; with it, it stop working.
   I put back the "12" value in the last cvs version, and it works.
   I don"t know what is this tab_num_rows_offset, but it seems buggy
with my file. it"s a Jet4 file, but perhaps there are different
subversions ?

_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/

Re: [mdb-dev] mdbtools with tcl

[Christian L. <chr...@la...>]

On sam, 2004-10-16 at 15:43, Christian Lefebvre wrote:
>   I wrote some code to use mdbsql from tcl.
  Here it is ...

  You just have to untar it from mdbtools source dir, goto to
src/tcl and run make.

  Just tested under linux, but should work under other unixes
and windows. I do know how to make a automake file => Makefile should
be adapted to your install.

Bye, Pif.

Re: [mdb-dev] mdb-export producing bad results on one particular table

[Christian L. <chr...@la...>]

On sam, 2004-10-16 at 15:27, Christian Lefebvre wrote:
>   I've got the same problem than Max Bowsher
> (http://sourceforge.net/mailarchive/message.php?msg_id=9588463)

Well .. I've got it !

  As 0.5 version worked fine for me, I tried to find from which version
it became bad, and i found this :

revision 1.28
date: 2004/01/06 00:42:08;  author: brianb;  state: Exp;  lines: +2 -2
patches 00-14 from David Mansfield

RCS file: /cvsroot/mdbtools/mdbtools/src/libmdb/file.c,v
retrieving revision 1.28
retrieving revision 1.27
diff -b -r1.28 -r1.27
27c27
< 	4096, 0x0c, 16, 45, 47, 51, 55, 56, 63, 12, 15, 23, 5, 25, 59
---
> 	4096, 0x0c, 12, 45, 47, 51, 55, 56, 63, 12, 15, 23, 5, 25, 59

  Without this change, it works; with it, it stop working.
  I put back the "12" value in the last cvs version, and it works.
  I don't know what is this tab_num_rows_offset, but it seems buggy
with my file. it's a Jet4 file, but perhaps there are different
subversions ?

Re: [mdb-dev] compilation issues

[Jeff S. <why...@ya...>]

Sorry about that.  I did not mean to incorporate a dependency on GLib 2.4.
I will correct this shortly.

 -- Jeff


--- Brian Bruns <bri...@gm...> wrote:

> Oops, this should have gone to list
> --
> 
> g_ptr_array_foreach was introduced in glib 2.4 AFAICT.  The offending
> checkin seems to be
> 
> ------------
> revision 1.31
> date: 2004/08/25 02:12:59;  author: whydoubt;  state: Exp;  lines: +46 -71
> Tidy up a few SQL-related functions
> --------------
> 
> Seems we need to go back to using a plain old for loop.
> 
> Brian
> 
> 
> On Mon, 18 Oct 2004 22:12:39 +0200, Christian Lefebvre
> <chr...@la...> wrote:
> > On lun, 2004-10-18 at 09:47, Michael Wood wrote:
> > > On Sat, Oct 16, 2004 at 11:37:31PM +0200, Christian Lefebvre wrote:
> > > > On sam, 2004-10-16 at 20:38, Christian Lefebvre wrote:
> > > >
> > > >   but those ones are still there ;-)
> > > > > - Makefile contains "READLINE_LIBS = no" instead of
> > > > >   "READLINE_LIBS =" => modified dyrectly in generated
> > > > >   Makefiles cause I don't know how it's generated.
> > >
> > > I would guess this is a bug in the autoconf/configure stuff.
> > > Haven't looked, though, and I am no autoconf expert :)
> > neither I am !
> > 
> > > > > - g_ptr_array_foreach not in my glibc 2.2.3 version => changed
> > >                                     ^^^^^
> > > I assume you mean glib
> > Oops ! yes of course.
> > 
> > > $ nm /usr/lib/libglib-2.0.a | grep g_ptr_array_foreach
> > > 00001220 T g_ptr_array_foreach
> > 
> > $ nm /usr/lib/libglib-2.0.a | grep g_ptr_array_
> > 00000e10 T g_ptr_array_add
> > 00000970 T g_ptr_array_free
> > 00000a60 t g_ptr_array_maybe_expand
> > 00000860 T g_ptr_array_new
> > 00000cf0 T g_ptr_array_remove
> > 00000d80 T g_ptr_array_remove_fast
> > 00000b50 T g_ptr_array_remove_index
> > 00000c30 T g_ptr_array_remove_index_fast
> > 00000ab0 T g_ptr_array_set_size
> > 00000880 T g_ptr_array_sized_new
> > 00000e90 T g_ptr_array_sort
> > 00000f00 T g_ptr_array_sort_with_data
> >          U g_ptr_array_add
> >          U g_ptr_array_free
> >          U g_ptr_array_new
> >          U g_ptr_array_set_size
> > 
> > no foreach
> > 
> > $ rpm -qif /usr/lib/libglib-2.0.a
> > Name        : libglib2.0_0-devel           Relocations: (not relocateable)
> > Version     : 2.2.3                             Vendor: MandrakeSoft
> > Release     : 1mdk                          Build Date: Wed 27 Aug 2003 11:00:42 AM
> CEST
> > Install Date: Mon 11 Oct 2004 09:58:56 PM CEST      Build Host: hp6.mandrakesoft.com
> > Group       : Development/C                 Source RPM: glib2.0-2.2.3-1mdk.src.rpm
> > 
> > -------------------------------------------------------
> > This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
> > Use IT products in your business? Tell us what you think of them. Give us
> > Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
> > http://productguide.itmanagersjournal.com/guidepromo.tmpl
> > _______________________________________________
> > mdbtools-dev mailing list
> > mdb...@li...
> > https://lists.sourceforge.net/lists/listinfo/mdbtools-dev
> >
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
> Use IT products in your business? Tell us what you think of them. Give us
> Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
> http://productguide.itmanagersjournal.com/guidepromo.tmpl
> _______________________________________________
> mdbtools-dev mailing list
> mdb...@li...
> https://lists.sourceforge.net/lists/listinfo/mdbtools-dev
> 



		
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com

Re: [mdb-dev] compilation issues

[Brian B. <bri...@gm...>]

Oops, this should have gone to list
--

g_ptr_array_foreach was introduced in glib 2.4 AFAICT.  The offending
checkin seems to be

------------
revision 1.31
date: 2004/08/25 02:12:59;  author: whydoubt;  state: Exp;  lines: +46 -71
Tidy up a few SQL-related functions
--------------

Seems we need to go back to using a plain old for loop.

Brian


On Mon, 18 Oct 2004 22:12:39 +0200, Christian Lefebvre
<chr...@la...> wrote:
> On lun, 2004-10-18 at 09:47, Michael Wood wrote:
> > On Sat, Oct 16, 2004 at 11:37:31PM +0200, Christian Lefebvre wrote:
> > > On sam, 2004-10-16 at 20:38, Christian Lefebvre wrote:
> > >
> > >   but those ones are still there ;-)
> > > > - Makefile contains "READLINE_LIBS = no" instead of
> > > >   "READLINE_LIBS =" => modified dyrectly in generated
> > > >   Makefiles cause I don't know how it's generated.
> >
> > I would guess this is a bug in the autoconf/configure stuff.
> > Haven't looked, though, and I am no autoconf expert :)
> neither I am !
> 
> > > > - g_ptr_array_foreach not in my glibc 2.2.3 version => changed
> >                                     ^^^^^
> > I assume you mean glib
> Oops ! yes of course.
> 
> > $ nm /usr/lib/libglib-2.0.a | grep g_ptr_array_foreach
> > 00001220 T g_ptr_array_foreach
> 
> $ nm /usr/lib/libglib-2.0.a | grep g_ptr_array_
> 00000e10 T g_ptr_array_add
> 00000970 T g_ptr_array_free
> 00000a60 t g_ptr_array_maybe_expand
> 00000860 T g_ptr_array_new
> 00000cf0 T g_ptr_array_remove
> 00000d80 T g_ptr_array_remove_fast
> 00000b50 T g_ptr_array_remove_index
> 00000c30 T g_ptr_array_remove_index_fast
> 00000ab0 T g_ptr_array_set_size
> 00000880 T g_ptr_array_sized_new
> 00000e90 T g_ptr_array_sort
> 00000f00 T g_ptr_array_sort_with_data
>          U g_ptr_array_add
>          U g_ptr_array_free
>          U g_ptr_array_new
>          U g_ptr_array_set_size
> 
> no foreach
> 
> $ rpm -qif /usr/lib/libglib-2.0.a
> Name        : libglib2.0_0-devel           Relocations: (not relocateable)
> Version     : 2.2.3                             Vendor: MandrakeSoft
> Release     : 1mdk                          Build Date: Wed 27 Aug 2003 11:00:42 AM CEST
> Install Date: Mon 11 Oct 2004 09:58:56 PM CEST      Build Host: hp6.mandrakesoft.com
> Group       : Development/C                 Source RPM: glib2.0-2.2.3-1mdk.src.rpm
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
> Use IT products in your business? Tell us what you think of them. Give us
> Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
> http://productguide.itmanagersjournal.com/guidepromo.tmpl
> _______________________________________________
> mdbtools-dev mailing list
> mdb...@li...
> https://lists.sourceforge.net/lists/listinfo/mdbtools-dev
>

Re: [mdb-dev] compilation issues

[Christian L. <chr...@la...>]

On lun, 2004-10-18 at 09:47, Michael Wood wrote:
> On Sat, Oct 16, 2004 at 11:37:31PM +0200, Christian Lefebvre wrote:
> > On sam, 2004-10-16 at 20:38, Christian Lefebvre wrote:
> > 
> >   but those ones are still there ;-)
> > > - Makefile contains "READLINE_LIBS = no" instead of
> > >   "READLINE_LIBS =" => modified dyrectly in generated
> > >   Makefiles cause I don't know how it's generated.
> 
> I would guess this is a bug in the autoconf/configure stuff.
> Haven't looked, though, and I am no autoconf expert :)
neither I am !

> > > - g_ptr_array_foreach not in my glibc 2.2.3 version => changed
>                                     ^^^^^
> I assume you mean glib
Oops ! yes of course.

> $ nm /usr/lib/libglib-2.0.a | grep g_ptr_array_foreach
> 00001220 T g_ptr_array_foreach

$ nm /usr/lib/libglib-2.0.a | grep g_ptr_array_
00000e10 T g_ptr_array_add
00000970 T g_ptr_array_free
00000a60 t g_ptr_array_maybe_expand
00000860 T g_ptr_array_new
00000cf0 T g_ptr_array_remove
00000d80 T g_ptr_array_remove_fast
00000b50 T g_ptr_array_remove_index
00000c30 T g_ptr_array_remove_index_fast
00000ab0 T g_ptr_array_set_size
00000880 T g_ptr_array_sized_new
00000e90 T g_ptr_array_sort
00000f00 T g_ptr_array_sort_with_data
         U g_ptr_array_add
         U g_ptr_array_free
         U g_ptr_array_new
         U g_ptr_array_set_size

no foreach

$ rpm -qif /usr/lib/libglib-2.0.a
Name        : libglib2.0_0-devel           Relocations: (not relocateable)
Version     : 2.2.3                             Vendor: MandrakeSoft
Release     : 1mdk                          Build Date: Wed 27 Aug 2003 11:00:42 AM CEST
Install Date: Mon 11 Oct 2004 09:58:56 PM CEST      Build Host: hp6.mandrakesoft.com
Group       : Development/C                 Source RPM: glib2.0-2.2.3-1mdk.src.rpm

Re: [mdb-dev] mdb-export producing bad results on one particular table

[Michael W. <mw...@ic...>]

On Sat, Oct 16, 2004 at 11:37:31PM +0200, Christian Lefebvre wrote:
> On sam, 2004-10-16 at 20:38, Christian Lefebvre wrote:
> 
>   but those ones are still there ;-)
> > - Makefile contains "READLINE_LIBS = no" instead of
> >   "READLINE_LIBS =" => modified dyrectly in generated
> >   Makefiles cause I don't know how it's generated.

I would guess this is a bug in the autoconf/configure stuff.
Haven't looked, though, and I am no autoconf expert :)

> > - g_ptr_array_foreach not in my glibc 2.2.3 version => changed
                                    ^^^^^
I assume you mean glib (which is completely different from
glibc.)  mdbtools probably just depends on a different version
of glib than the one you have installed.  glib-2.0 does have
g_ptr_array_foreach.  I haven't looked at other versions.

$ nm /usr/lib/libglib-2.0.a | grep g_ptr_array_foreach
00001220 T g_ptr_array_foreach

I've just had a look in configure.in and it says:

if test "$no_glib" = "yes"; then
        echo
        echo "glib 2.0 is required by MDB Tools."
        echo "It can be downloaded at www.gtk.org."
        echo
        exit 1
fi

> >   src/sql/mdbsql.c like before 1.31 version (thanks to "cvs diff"  ;-)
> 
>   About the missing rows, I tried different variations, but
> i've still not found why it doesn't works sometimes.

Do you have old libmdb stuff lying around?  That might somehow
be interfering?

I hope this helps.

-- 
Michael Wood <mw...@ic...>

Re: [mdb-dev] mdb-export producing bad results on one particular table

[Christian L. <chr...@la...>]

On sam, 2004-10-16 at 20:38, Christian Lefebvre wrote:

  Sorry, forget the 3 first points, I upgraded autoconf and it works
fine ...

> - unknown AC_PROG_LIBTOOL => changed to AM_PROG_LIBTOOL
> - unknown AC_CONFIG_FILES => deleted and added src/Makefile in
>   AC_OUTPUT list
> - unknown AC_HELP_STRING => ignored because I couldn't find from
>   where it appears

  but those ones are still there ;-)
> - Makefile contains "READLINE_LIBS = no" instead of
>   "READLINE_LIBS =" => modified dyrectly in generated Makefiles cause I
>   don't know how it's generated.
> - g_ptr_array_foreach not in my glibc 2.2.3 version => changed
>   src/sql/mdbsql.c like before 1.31 version (thanks to "cvs diff"  ;-)

  About the missing rows, I tried different variations, but i've still
not found why it doesn't works sometimes.

  To be continued (it's time to bed here ;-)
  Bye, Pif.

Re: [mdb-dev] mdb-export producing bad results on one particular table

[Christian L. <chr...@la...>]

On sam, 2004-10-16 at 16:02, Jeff Smith wrote:
> Please try the CVS version
  Well .. I tried ;-)

  I don't know exactly how autoconf & co work, but I had to change
some stuffs in it to make it work on my box :
- unknown AC_PROG_LIBTOOL => changed to AM_PROG_LIBTOOL
- unknown AC_CONFIG_FILES => deleted and added src/Makefile in
  AC_OUTPUT list
- unknown AC_HELP_STRING => ignored because I couldn't find from
  where it appears
- Makefile contains "READLINE_LIBS = no" instead of
  "READLINE_LIBS =" => modified dyrectly in generated Makefiles cause I
  don't know how it's generated.
- g_ptr_array_foreach not in my glibc 2.2.3 version => changed
  src/sql/mdbsql.c like before 1.31 version (thanks to "cvs diff"  ;-)

  After this, everything compiled fine, but the result is really weird :
- if I do this :
  export LD_LIBRARY_PATH=src/libmdb/.libs:src/sql/.libs
  ./src/util/mdb-export ../liste.mdb UNITE

  I got an empty result

- if I do this :
  make install
  cd "where the --prefix option points to"
  export LD_LIBRARY_PATH=lib
  bin/mdb-export ../liste.mdb UNITE

  I got the correct result !!!

  What's the difference ?
  What did "make install" do on files to make them work ?!

  Last test: after a "make install" on 0.6pre1 tar, I still got
the fault.
  It really seems to be a "twilight zone" story :-)
  So, for the moment, I've a version which works, but I'd like to
understand why !

Bye, Pif.

Re: [mdb-dev] don't compile if no readline.h

[Christian L. <chr...@la...>]

On sam, 2004-10-16 at 15:57, Jeff Smith wrote:
> It looks like your patch is not against the CVS version.
Sorryn I use the 0.6pre1 tar from source forge.

>   For future reference,
> please create patches using 'diff -u'.
Ok.

Pif.

Re: [mdb-dev] mdb-export producing bad results on one particular table

[Jeff S. <why...@ya...>]

Please try the CVS version (contact the board again if you need help
doing so), and if it still gives you these problems, you can send me
you db to look at.

 -- Jeff Smith


--- Christian Lefebvre <chr...@la...> wrote:
>   I've got the same problem than Max Bowsher
> (http://sourceforge.net/mailarchive/message.php?msg_id=9588463)
> but I can't "Compact Database" because I work under linux, on a single
> jet4 file that is sent "as is" by an external export routine.
> 
>   In fact, mdb-export and "select *" with mdb-sql give the same
> result : some rows of some tables are missing (all of them for one
> table).
>   The same requests with 0.5 work fine.
>   I don't know anything about mdb structure, so I can't help anyway,
> but I can send the mdb file in private if it helps (2.5Mb).
> 
> Bye, Pif.



		
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com

Re: [mdb-dev] don't compile if no readline.h

[Jeff S. <why...@ya...>]

It looks like your patch is not against the CVS version.  CVS already
contains a (not so nice) fix for this problem.  For future reference,
please create patches using 'diff -u'.

 -- Jeff Smith


--- Christian Lefebvre <chr...@la...> wrote:

> Hello,
>   I use mdbtools 0.5 for a personal project since last week.
>   I just downloaded 0.6pre version, and "make" failed.
>   "configure" correctly found that readline.h is not installed, but
> mdb-sql.c doesn't nest "*_history" function calls with
> "ifdef HAVE_READLINE".
> 
>   Here is a patch to apply to mdb-sql.c file to add these ifdef
> 
>   Hope this helps,
> 
> Bye, Pif.



		
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com