mdbtools-dev

Re: [mdb-dev] Segfault reading JET3 mdb

[<br...@br...>]

This is a know problem, well sort of.  It happens with deleted columns
(create a table, populate it, drop a column).  If you compact/repair
the database in Access then all is well, I just haven't figured out
how Access handles it yet.  (Imagine create table, populate, delete
column, add different column....problems).

I had considered simply aborting on finding this condition, but I'd
really like to fix it.

The transition to crack_row still needs a bit of work I think.  It's
there to clean up that awful data.c stuff, and to support writing via
it's partner mdb_pack_row().

If you wouldn't mind, post your patches to Sourceforge so I can keep
track of them easier.

Cheers,

Brian

On Mon, 2 Feb 2004 09:32:11 +0000, Will Newton wrote:

> 
> 
> Currently I get a segfault when reading a JET3 database I have
> here. Unfortunately the file is rather large and proprietary, so I
> can't really make it available. It seems that "mdb_crack_row" is
> returning a "num_cols" different from the length of "fields" so
> subsequent iteration over "fields" runs past the end of the array.
The
> returned value of "num_cols" is not consistently greater or smaller
> than the length of "fields", it seems like the wrong value entirely
is
> being read. 0.5 didn't do this, but HEAD did.
> 
> The attached patch fixes the problem. Why, I'm not too sure.
> 
> Exporting my database still has problems:
> 
> 1) On certain large tables the text in some fields seems to be offset
> slightly i.e. the start of one field would like like the end of
> another (the last field?). I have these columns:
> 
> id,quote_number,declaration_number,new_renewal,agent_code,long_name
> 17829,"",37124,"41196NEWTUNICKWilliam N","ewbold & George..."
> 
> I have abbreviated the data because the table is rather large. The
> data should look more like:
> 
> 17829,37124,41196,"NEW","TUNICK","William Newbold & George..."
> 
> That is an approximation as I'm not entirely conversant with the
> schema of this legacy data, but it does illustrate the general
> appearance of the bug.
> 
> 2) Memo fields are not being read quite right. WHen exporting a large
> table with memo fields (in fact it is the same table as previously
> mentioned) I get error messages:
> 
> ...
> offset 731000832 is beyond EOF
> offset 419665920 is beyond EOF
> offset 733181952 is beyond EOF
> offset 1090744320 is beyond EOF
> ...
> 
> And ultimately a segfault. It seems the value of "lval_pg" when
> reading non-inline memo fields is incorrect - again it looks like a
> pseudo random value rather than any consistent error.
> 
> 
> These problems may be related - many of the simple tables in this
> database export cleanly. None of the tables with memo fields export
> correctly AFAICT.

Re: [mdb-dev] Segfault reading JET3 mdb (cont.)

[Petro V. <Pe...@nc...>]

On Mon, 02 Feb 2004 09:44:09 +0000
Will Newton <wi...@gb...> wrote:

> 
> The patch is really attached this time (good old Monday morning).
> 
> Heres some additional debugging output from reading memo fields:
> 
> Reading LVAL page 000001
> offset 926943232 is beyond EOF
> Reading LVAL page 028f72
> offset 343642112 is beyond EOF
> Reading LVAL page 000009
> offset 776472576 is beyond EOF
> Reading LVAL page 001770
> Reading LVAL page 100040
> offset 532154368 is beyond EOF
> Reading LVAL page 028f72
> offset 343642112 is beyond EOF
> Reading LVAL page 654d20
> offset 711524352 is beyond EOF
> 
> The wildly fluctuating values of lval_pg seem a bit odd to
> me.
> 

Hi, I've used your patch on the source I got from CVS a couple of days ago.
However, I am getting this error:

# isql MSAccess
+---------------------------------------+
| Connected!                            |
|                                       |
| sql-statement                         |
| help [tablename]                      |
| quit                                  |
|                                       |
+---------------------------------------+
SQL> select * from ctx
isql: relocation error: /usr/lib/libmdbodbc.so: undefined symbol: mdb_crack_row
# 

The interesting bit is that i can browse the tables in the database throught
gmdb2. The data shows up. When i try to run a query gmdb2 segfaults.

Any ideas?


--
Petro

[mdb-dev] Jumps in large JET3 tables

[Will N. <wi...@gb...>]

I think some of my problems are caused by mdb tools not supporting
"jumps" when reading the columns of my large table. I modified
mdb_crack_row to look a bit like data.c used to (the stuff in the #if
0 block) to support jumps. It seems to work, but there are some broken
bits so I'd rather someone who actually understood the code could
comment on this.

I've attached a patch that realigns at least some of my data, although
it may misalign yours.

[mdb-dev] Doubles read incorrectly

[Will N. <wi...@gb...>]

Double values seem to get read incorrectly. This patch looks like it
is probably correct but should be verified.

[mdb-dev] Segfault reading JET3 mdb (cont.)

[Will N. <wi...@gb...>]

The patch is really attached this time (good old Monday morning).

Heres some additional debugging output from reading memo fields:

Reading LVAL page 000001
offset 926943232 is beyond EOF
Reading LVAL page 028f72
offset 343642112 is beyond EOF
Reading LVAL page 000009
offset 776472576 is beyond EOF
Reading LVAL page 001770
Reading LVAL page 100040
offset 532154368 is beyond EOF
Reading LVAL page 028f72
offset 343642112 is beyond EOF
Reading LVAL page 654d20
offset 711524352 is beyond EOF

The wildly fluctuating values of lval_pg seem a bit odd to
me.

[mdb-dev] Segfault reading JET3 mdb

[Will N. <wi...@gb...>]

Currently I get a segfault when reading a JET3 database I have
here. Unfortunately the file is rather large and proprietary, so I
can't really make it available. It seems that "mdb_crack_row" is
returning a "num_cols" different from the length of "fields" so
subsequent iteration over "fields" runs past the end of the array. The
returned value of "num_cols" is not consistently greater or smaller
than the length of "fields", it seems like the wrong value entirely is
being read. 0.5 didn't do this, but HEAD did.

The attached patch fixes the problem. Why, I'm not too sure.

Exporting my database still has problems:

1) On certain large tables the text in some fields seems to be offset
slightly i.e. the start of one field would like like the end of
another (the last field?). I have these columns:

id,quote_number,declaration_number,new_renewal,agent_code,long_name
17829,"",37124,"41196NEWTUNICKWilliam N","ewbold & George..."

I have abbreviated the data because the table is rather large. The
data should look more like:

17829,37124,41196,"NEW","TUNICK","William Newbold & George..."

That is an approximation as I'm not entirely conversant with the
schema of this legacy data, but it does illustrate the general
appearance of the bug.

2) Memo fields are not being read quite right. WHen exporting a large
table with memo fields (in fact it is the same table as previously
mentioned) I get error messages:

...
offset 731000832 is beyond EOF
offset 419665920 is beyond EOF
offset 733181952 is beyond EOF
offset 1090744320 is beyond EOF
...

And ultimately a segfault. It seems the value of "lval_pg" when
reading non-inline memo fields is incorrect - again it looks like a
pseudo random value rather than any consistent error.


These problems may be related - many of the simple tables in this
database export cleanly. None of the tables with memo fields export
correctly AFAICT.

Re: [mdb-dev] cvs version trouble + seg faults in 0.5

[<br...@br...>]

Er, that should have read autoconf 2.57

On Sat, 31 Jan 2004 08:23:17 -0800 (PST), br...@br... wrote:

> 
> Unfortunately I suck at autoconf, it's a miracle the thing works at
> all.  I found this on the debian lists:
> 
> http://keithp.com/pipermail/nickle/2003-July/000266.html
> 
> which says AC_CONFIG_FILES requires autoconf 2.52 or above or exactly
> this problem happens.  Unfortuately, we need AC_CONFIG_FILES instead
> of AC_OUTPUT because of the conditional compile of odbc, sql engine,
> gui, etc....If somebody has another way to do it, I'm open.
> 
> Brian
> 
> PS sourceforge appears to be having problems with CVS.  
> 
> 
> On Fri, 30 Jan 2004 22:52:53 -0500, Petro Verkhogliad wrote:
> 
> > 
> > Hi,
> > 
> > still the same problem. I have these binaries:
> > automake      automake-1.4  automake-1.5 
> > automake-1.6  automake-1.7he 
> > 
> > i looked at the output from configure. it says that it is using
> > automake-1.4.
> > I've tried removing the brackets around that line. but still
> nothing. 
> > 
> > I am attaching the configure.in file. perhaps you can tell me what
is
> > wrong with
> > it.
> > 
> > at the same time, i cannot access the cvs repository anymore. Do
you
> > have any
> > idea why that is. i get the fowllowing: 
> > 
> > cvs [checkout aborted]: end of file from server (consult above
> > messages if any)
> > 
> > when i run:
> > 
> > cvs -z3 -d:pserver:ano...@cv...:/cvsroot/mdbtools
co
> > mdbtools
> > 
> > Thank you for your time,
> > 
> > Petro
> > 
> > On Fri, 30 Jan 2004 19:08:06 -0800 (PST)
> > br...@br... wrote:
> > 
> > > What version of automake do you have?  Apparently there are
> problems
> > > with AC_CONFIG_FILES and brackets prior to 1.6.3
> > > 
> > > If you have an older version, try removing the brackets from the
> > > AC_CONFIG_FILES line in configure.in and see if it takes.
> > > 
> > > Brian
> > > 
> > > On Fri, 30 Jan 2004 19:54:34 -0500, Petro Verkhogliad wrote:
> > > 
> > > > 
> > > > Hi,
> > > > 
> > > > I have a linux server that is running apache+php. From there i
am
> > > > connecting to
> > > > the Windows server that has the mdb files i need.
> > > > 
> > > > Problem 1.
> > > > 
> > > > I got mdbtools-0.5-1 and mdbtools-odbc-0.5-1 installed. I can
> > connect
> > > > to the
> > > > .mdb and run some simple queries. However, I am getting errors
> when
> > > > the file has
> > > > understores in it. I have tried the patched version. It is
still
> > > > giving me
> > > > problems.
> > > > 
> > > > Problem 2.
> > > > 
> > > > It seems that this problem has been fixed in cvs. Which leads
me
> to
> > > > problem two.
> > > > Yesterday I checked out the code from cvs and tryed to build
it.
> > the
> > > > ./configure
> > > > command quits with following message:
> > > > 
> > > > checking GNOME_CFLAGS... -DORBIT2=1 -pthread
> > > > -I/usr/include/libglade-2.0
> > > > -I/usr/include/gtk-2.0 -I/usr/include/libxml2
> > > > -I/usr/lib/gtk-2.0/include
> > > > -I/usr/include/atk-1.0 -I/usr/include/pango-1.0
> > -I/usr/X11R6/include
> > > > -I/usr/include/freetype2 -I/usr/include/glib-2.0
> > > > -I/usr/lib/glib-2.0/include
> > > > -I/usr/include/libgnomeui-2.0 -I/usr/include/libgnome-2.0
> > > > -I/usr/include/libgnomecanvas-2.0 -I/usr/include/libart-2.0
> > > > -I/usr/include/gconf/2 -I/usr/include/libbonoboui-2.0
> > > > -I/usr/include/orbit-2.0
> > > > -I/usr/include/libbonobo-2.0 -I/usr/include/gnome-vfs-2.0
> > > > -I/usr/lib/gnome-vfs-2.0/include
> > > -I/usr/include/bonobo-activation-2.0 
> > > > checking
> > > > GNOME_LIBS... -Wl,--export-dynamic -pthread -L/usr/X11R6/lib
> > > > -lglade-2.0
> > > > -lgnomeui-2 -lSM -lICE -lbonoboui-2 -lxml2 -lpthread -lz
> > > > -lgnomecanvas-2
> > > > -lgnome-2 -lpopt -lart_lgpl_2 -lpangoft2-1.0 -lgtk-x11-2.0
> > > > -lgdk-x11-2.0
> > > > -latk-1.0 -lgdk_pixbuf-2.0 -lpangoxft-1.0 -lpangox-1.0
> -lpango-1.0
> > > > -lbonobo-2
> > > > -lgconf-2 -lgnomevfs-2 -lbonobo-activation -lORBit-2
> -lgobject-2.0
> > > > -lgthread-2.0
> > > > -lm -lgmodule-2.0 -ldl -lglib-2.0  ./configure: line 6351:
syntax
> > > > error near
> > > > unexpected token `src/Makefile'./configure: line 6351:
> > > > `AC_CONFIG_FILES(src/Makefile)'
> > > > 
> > > > I read some the revelant posts on mailing list. I edited the
> > > > configure.in file
> > > > as specified in the 'installing mdbtools cvs on slackware 9.1'
> > post.
> > > > That
> > > > solution did not solve the problem.
> > > > 
> > > > I am running Gentoo Linux 1.4. I can post more information if
> > > required
> > > > (about
> > > > specific utility versions i have installed)
> > > > 
> > > > Please help.
> > > > 
> > > > Thank you.
> > > > 
> > > > -- 
> > 
> > 
> > -- 
> 
> 
> -------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> mdbtools-dev mailing list
> mdb...@li...
> https://lists.sourceforge.net/lists/listinfo/mdbtools-dev

Re: [mdb-dev] cvs version trouble + seg faults in 0.5

[<br...@br...>]

Unfortunately I suck at autoconf, it's a miracle the thing works at
all.  I found this on the debian lists:

http://keithp.com/pipermail/nickle/2003-July/000266.html

which says AC_CONFIG_FILES requires autoconf 2.52 or above or exactly
this problem happens.  Unfortuately, we need AC_CONFIG_FILES instead
of AC_OUTPUT because of the conditional compile of odbc, sql engine,
gui, etc....If somebody has another way to do it, I'm open.

Brian

PS sourceforge appears to be having problems with CVS.  


On Fri, 30 Jan 2004 22:52:53 -0500, Petro Verkhogliad wrote:

> 
> Hi,
> 
> still the same problem. I have these binaries:
> automake      automake-1.4  automake-1.5 
> automake-1.6  automake-1.7he 
> 
> i looked at the output from configure. it says that it is using
> automake-1.4.
> I've tried removing the brackets around that line. but still
nothing. 
> 
> I am attaching the configure.in file. perhaps you can tell me what is
> wrong with
> it.
> 
> at the same time, i cannot access the cvs repository anymore. Do you
> have any
> idea why that is. i get the fowllowing: 
> 
> cvs [checkout aborted]: end of file from server (consult above
> messages if any)
> 
> when i run:
> 
> cvs -z3 -d:pserver:ano...@cv...:/cvsroot/mdbtools co
> mdbtools
> 
> Thank you for your time,
> 
> Petro
> 
> On Fri, 30 Jan 2004 19:08:06 -0800 (PST)
> br...@br... wrote:
> 
> > What version of automake do you have?  Apparently there are
problems
> > with AC_CONFIG_FILES and brackets prior to 1.6.3
> > 
> > If you have an older version, try removing the brackets from the
> > AC_CONFIG_FILES line in configure.in and see if it takes.
> > 
> > Brian
> > 
> > On Fri, 30 Jan 2004 19:54:34 -0500, Petro Verkhogliad wrote:
> > 
> > > 
> > > Hi,
> > > 
> > > I have a linux server that is running apache+php. From there i am
> > > connecting to
> > > the Windows server that has the mdb files i need.
> > > 
> > > Problem 1.
> > > 
> > > I got mdbtools-0.5-1 and mdbtools-odbc-0.5-1 installed. I can
> connect
> > > to the
> > > .mdb and run some simple queries. However, I am getting errors
when
> > > the file has
> > > understores in it. I have tried the patched version. It is still
> > > giving me
> > > problems.
> > > 
> > > Problem 2.
> > > 
> > > It seems that this problem has been fixed in cvs. Which leads me
to
> > > problem two.
> > > Yesterday I checked out the code from cvs and tryed to build it.
> the
> > > ./configure
> > > command quits with following message:
> > > 
> > > checking GNOME_CFLAGS... -DORBIT2=1 -pthread
> > > -I/usr/include/libglade-2.0
> > > -I/usr/include/gtk-2.0 -I/usr/include/libxml2
> > > -I/usr/lib/gtk-2.0/include
> > > -I/usr/include/atk-1.0 -I/usr/include/pango-1.0
> -I/usr/X11R6/include
> > > -I/usr/include/freetype2 -I/usr/include/glib-2.0
> > > -I/usr/lib/glib-2.0/include
> > > -I/usr/include/libgnomeui-2.0 -I/usr/include/libgnome-2.0
> > > -I/usr/include/libgnomecanvas-2.0 -I/usr/include/libart-2.0
> > > -I/usr/include/gconf/2 -I/usr/include/libbonoboui-2.0
> > > -I/usr/include/orbit-2.0
> > > -I/usr/include/libbonobo-2.0 -I/usr/include/gnome-vfs-2.0
> > > -I/usr/lib/gnome-vfs-2.0/include
> > -I/usr/include/bonobo-activation-2.0 
> > > checking
> > > GNOME_LIBS... -Wl,--export-dynamic -pthread -L/usr/X11R6/lib
> > > -lglade-2.0
> > > -lgnomeui-2 -lSM -lICE -lbonoboui-2 -lxml2 -lpthread -lz
> > > -lgnomecanvas-2
> > > -lgnome-2 -lpopt -lart_lgpl_2 -lpangoft2-1.0 -lgtk-x11-2.0
> > > -lgdk-x11-2.0
> > > -latk-1.0 -lgdk_pixbuf-2.0 -lpangoxft-1.0 -lpangox-1.0
-lpango-1.0
> > > -lbonobo-2
> > > -lgconf-2 -lgnomevfs-2 -lbonobo-activation -lORBit-2
-lgobject-2.0
> > > -lgthread-2.0
> > > -lm -lgmodule-2.0 -ldl -lglib-2.0  ./configure: line 6351: syntax
> > > error near
> > > unexpected token `src/Makefile'./configure: line 6351:
> > > `AC_CONFIG_FILES(src/Makefile)'
> > > 
> > > I read some the revelant posts on mailing list. I edited the
> > > configure.in file
> > > as specified in the 'installing mdbtools cvs on slackware 9.1'
> post.
> > > That
> > > solution did not solve the problem.
> > > 
> > > I am running Gentoo Linux 1.4. I can post more information if
> > required
> > > (about
> > > specific utility versions i have installed)
> > > 
> > > Please help.
> > > 
> > > Thank you.
> > > 
> > > -- 
> 
> 
> -- 

[mdb-dev] cvs version trouble + seg faults in 0.5

[Petro V. <Pe...@nc...>]

Hi,

I have a linux server that is running apache+php. From there i am connecting to
the Windows server that has the mdb files i need.

Problem 1.

I got mdbtools-0.5-1 and mdbtools-odbc-0.5-1 installed. I can connect to the
.mdb and run some simple queries. However, I am getting errors when the file has
understores in it. I have tried the patched version. It is still giving me
problems.

Problem 2.

It seems that this problem has been fixed in cvs. Which leads me to problem two.
Yesterday I checked out the code from cvs and tryed to build it. the ./configure
command quits with following message:

checking GNOME_CFLAGS... -DORBIT2=1 -pthread -I/usr/include/libglade-2.0
-I/usr/include/gtk-2.0 -I/usr/include/libxml2 -I/usr/lib/gtk-2.0/include
-I/usr/include/atk-1.0 -I/usr/include/pango-1.0 -I/usr/X11R6/include
-I/usr/include/freetype2 -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include
-I/usr/include/libgnomeui-2.0 -I/usr/include/libgnome-2.0
-I/usr/include/libgnomecanvas-2.0 -I/usr/include/libart-2.0
-I/usr/include/gconf/2 -I/usr/include/libbonoboui-2.0 -I/usr/include/orbit-2.0
-I/usr/include/libbonobo-2.0 -I/usr/include/gnome-vfs-2.0
-I/usr/lib/gnome-vfs-2.0/include -I/usr/include/bonobo-activation-2.0  checking
GNOME_LIBS... -Wl,--export-dynamic -pthread -L/usr/X11R6/lib -lglade-2.0
-lgnomeui-2 -lSM -lICE -lbonoboui-2 -lxml2 -lpthread -lz -lgnomecanvas-2
-lgnome-2 -lpopt -lart_lgpl_2 -lpangoft2-1.0 -lgtk-x11-2.0 -lgdk-x11-2.0
-latk-1.0 -lgdk_pixbuf-2.0 -lpangoxft-1.0 -lpangox-1.0 -lpango-1.0 -lbonobo-2
-lgconf-2 -lgnomevfs-2 -lbonobo-activation -lORBit-2 -lgobject-2.0 -lgthread-2.0
-lm -lgmodule-2.0 -ldl -lglib-2.0  ./configure: line 6351: syntax error near
unexpected token `src/Makefile'./configure: line 6351:
`AC_CONFIG_FILES(src/Makefile)'

I read some the revelant posts on mailing list. I edited the configure.in file
as specified in the 'installing mdbtools cvs on slackware 9.1' post. That
solution did not solve the problem.

I am running Gentoo Linux 1.4. I can post more information if required (about
specific utility versions i have installed)

Please help.

Thank you.

-- 

Re: [mdb-dev] Currency data type

[Tony T. <tt...@mv...>]

At 11:02 AM 2004/01/21 -0600, you wrote:
>I found that mdbtools can not output currency data correctly. In currency 
>fields data have a $ sign and one or more commas, I am not sure if $ and , 
>are just for display or actually are part of stored data.

 From the A97 help "Currency variables are stored as 64-bit (8-byte) 
numbers in an integer format, scaled by 10,000 to give a fixed-point number 
with 15 digits to the left of the decimal point and 4 digits to the right. 
This representation provides a range of -922,337,203,685,477.5808 to 
922,337,203,685,477.5807. "

The $, Euro or other symbol as appropriate comes from your Windows Regional 
Settings.   Thus the $ and , are not stored as part of the data

Tony
-----
Tony Toews, Microsoft Access MVP
Microsoft Access Links, Hints, Tips & Accounting Systems at
    http://www.granite.ab.ca/accsmstr.htm

[mdb-dev] Currency data type

[john <joh...@am...>]

I found that mdbtools can not output currency data correctly. In currency 
fields data have a $ sign and one or more commas, I am not sure if $ and , 
are just for display or actually are part of stored data.

John

-- 
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/

[mdb-dev] where clause syntax?

[Matt M. <mm...@tc...>]

	Hi, I'm using a select statement in my php code to query my
access db with mdbtools and unixODBC. When I do purely a select
statement, it works fine, but it returns all 425 records. When I do a
select statement with a where clause, which I know should work, it
returns nothing. I haven't been able to find anything which describes
the syntax used by mdbtools to determine whether I've got something
wrong. Is there a way around this? I tried searching the archives, but
got an error message from sourceforge. 

TIA,

Matt

Re: [mdb-dev] Fixes to make binary fields sort of work

[Christophe P. <ch...@ch...>]

Le ven 09/01/2004 à 04:10, Philip Gladstone a écrit :
> Hi,

Hello,

> What is the protocol for submitting patches on this project?

I'd say the normal way is to use the "submit patch" function of
SourceForge, another is to post the diff file on this mailing-list.

HTH,

-- 
Christophe «CHiPs» PETIT <ch...@ch...> http://chips.free.fr/
« Défendre les logiciels libres, c'est défendre la liberté. »
Linux-Nantes: partagez votre savoir http://www.linux-nantes.fr.eu.org/
« Gentiment, la petite Cosette nous a offert son jus »

[mdb-dev] MDB Tools Roadmap

[<br...@br...>]

Hi all,

Well, after a lengthy hiatus, I've found some time and
the interest to put some time in on MDB Tools (note:
having a child really puts a crimp in your time to work
on open source projects).

I have merged all the patches that looked straight
forward, there are some that are a bit messy or just
didn't make it on sourceforge (you know who you are). 
But, for the most part my queue is a little emptier.

Anyway, I wanted to outline my plans for the next few
releases.  I plan on doing some -pre releases between
these major milestones but what will be in those is not
so clear yet.

Release 0.6-pre1
----------------

Basically, I intend to tidy up the patches, review and
merge a few more.  In addition mdb_read_row was
reworked to make way for write support and now calls
mdb_crack_row to split the row into separate fields
(the writing routines use mdb_pack_row to reverse the
process).  It seems to be working for the most part,
but I'm having trouble with deleted columns in JET3
databases, and there are probably some other corner
cases that are failing.   Need lots of testing on this
prior to the release! 

I hope to cut down the bug list a bit more for this
release too.

Release 0.6-pre2 to preX
------------------------

I'd like to support iconv for JET4 databases (they use
UCS-2 internally and we basically punt on non-western
languages). I've got a patch along these lines, but I
wanted to rework it a bit, needs autoconf detection and
 some thread safety things.

I'll entertain other small features (especially if they
come with a patch!) and keep the ball rolling while the
big stuff gets worked on.

Release 0.6
-----------

Concurrent with all this we'll be working towards 0.6. 
The 0.6 release will be our first release to feature
writable mdb databases.  It will not be full write
capability; we will consider this release feature
complete when mdb-import works properly.  (mdb-import
takes a CVS file and loads it into a predefined table,
e.g. the opposite of mdb-export).

Pieces of this are already done.  updates in place,
that is updates on non-indexed columns and that do not
require a page split, are curently working see updrow.c
for an example.  mdb-import will also currently follow
indexes, find the proper data page and insert the row
on it.

What is left to do is:

1. Updating indexes.  For each index, build a
IndexChain from root to current row, and follow it back
up modifying indexes as necessary.
2. Page splits.  Both on the data pages and index
pages.  This needs to update the usage maps too, which 
gets a bit hairy.

Alot of the index stuff has been done on the read side.
 If you set MDBOPTS environment variable to "use_index"
you'll turn on index evaluation.  libmdb uses a cost
based routine to choose the best access strategy.  I'd
like feedback on this one, I've tested it on JET3, and
I'm not sure how well it works on JET4 (it is on my
todo list).

Release 0.7
-----------

In 0.7 we'll add insert/update/delete capabilities to
the SQL Engine, and thus to the ODBC driver.

The ODBC driver is in dire need of some attention too,
I'm hoping to clean it up gradually culminating in this
release.

gmdb2 will also gain editable grid/update capability
during this release.  So you can select a table, click
the data button, edit the grid and hit Save and your
changes will be saved to the database.  Should make
this really nice for colabrative work (receive an mdb
in email,edit it, send it back).

Release 0.8
-----------

DDL for table create/alter/drop will likely make it in
this release. It would be nice to be able to create a
database from scratch, but that is indeed an ambitious
goal.

Around this release we'll also start focusing on some
of the non-table objects in the mdb files. 
Specifically, Forms and VBA code.  See mdb-prop.c for
the first hints of this.

I'd like eventually to be able to extract an Access
Form and convert it to a Glade XML file.  This would
really help folks converting Access to something else.

Release 0.9
-----------

This is the last release I'm going to try to predict. 
I'd like to have compact/repair and data recovery tools
in place.  Keep an eye on mdb-check.c for the beginings
of this.

Join support in the SQL Engine should make it in by
here.  I've already started some work on temp
table/worktables which will be needed by joins. We
should be able to join accross mdb files which will be
a neat featue (essentially worktables are seperate
stripped down mdb files).  BTW, these worktables are
currently used to support the "list tables" "describe
table" commands as proper result sets instead of
dumping to stdout, needs more work still but a kinda
kludgy solution is now in place.

Also possible is MSDE databases. While not wildly
popular currently they could be a force to be reckoned
with.

[mdb-dev] Fixes to make binary fields sort of work

[Philip G. <ph...@gl...>]

Hi,

I have a set of patches that make mdbtools be able to handle a database 
that has a binary column in it (this is actually the Adobe Photoshop 
Album database). It is definitely not a complete solution to the 
problem, but it is a step in the right direction.

What is the protocol for submitting patches on this project?

Philip

-- 
Philip Gladstone
* Check out the live pondcam at http://pond.gladstonefamily.net

Re: [mdb-dev] Standalone version

[<br...@br...>]

You can set the LD_LIBRARY_PATH (or LIBPATH depending
on platform) variable to include the directory that
contains the .so files.  For instance

export LD_LIBARY_PATH=${LD_LIBRARY_PATH}:/path/lib

You can then run "ldd" (if you're on linux) to check to
see if it resolves all the dependencies.

Cheers,

Brian

On Wed, 7 Jan 2004 13:42:21 +1100 (EST), "Tudor" wrote:

> 
> Hello,
> 
> Well done on mdbtools.   I just found out about it and
> I think the idea
> and implementation is great!
> 
> I have downloaded and installed it onto my home pc
> successfully and
> everything works OK.  Then I tried uploading it to my
> web server.   I can
> run applications on my web server, currently I have a
> script that uses an
> uploaded version of unzip to unzip uploaded files
> before processing, so I
> know that running standalone apps on the server works
> ok.
> 
> I wish to do a similar thing with mdbtools, however,
it
> complains that
> it's missing a shared object.   I tried shifting the
> shared object into
> the same directory as the app on the server but it
> didn't work.  My web
> server runs jailshell which means I can't copy the
> shared object into the
> location it's looking for.  Is there a way I can
> specify that the programs
> should be compiled as standalone, or alternatively
tell
> it to look for the
> shared object in a different loaction?
> 
> Cheers,
> Tudor.
> 
> 
> 
> 
> 
>
-------------------------------------------------------
> This SF.net email is sponsored by: IBM Linux
Tutorials.
> Become an expert in LINUX or just sharpen your
skills. 
> Sign up for IBM's
> Free Linux Tutorials.  Learn everything from the bash
> shell to sys admin.
> Click now!
> http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click
> _______________________________________________
> mdbtools-dev mailing list
> mdb...@li...
>
https://lists.sourceforge.net/lists/listinfo/mdbtools-dev

[mdb-dev] Standalone version

[Tudor <tu...@tu...>]

Hello,

Well done on mdbtools.   I just found out about it and I think the idea
and implementation is great!

I have downloaded and installed it onto my home pc successfully and
everything works OK.  Then I tried uploading it to my web server.   I can
run applications on my web server, currently I have a script that uses an
uploaded version of unzip to unzip uploaded files before processing, so I
know that running standalone apps on the server works ok.

I wish to do a similar thing with mdbtools, however, it complains that
it's missing a shared object.   I tried shifting the shared object into
the same directory as the app on the server but it didn't work.  My web
server runs jailshell which means I can't copy the shared object into the
location it's looking for.  Is there a way I can specify that the programs
should be compiled as standalone, or alternatively tell it to look for the
shared object in a different loaction?

Cheers,
Tudor.

[mdb-dev] MDBTools ODBC driver

[Yves <yve...@wa...>]

Hello,
I'm trying to use the MDBTools ODBC driver to read a msaccess97 file in 
OpenOffice.org.
I've configured the driver in unixODBC, here are my odbc.ini and odbcinst.ini  
:

-odbc.ini
[Gestiondesbadges]
Description = Gestion des badges
Driver = MDB
Servername = localhost
Database = /home/yves/Gestion_des_badges.mdb
Port = 5432

-odbcinst.ini
[MDB]
Description		= ODBC pour MDB
Driver		= /usr/lib/libmdbodbc.so
Setup		= 
FileUsage		= 1
CPTimeout		= 
CPReuse		= 

I can configure the datasource in openoffice, but when I want to see the 
tables I get the following error message : [unixODBC]Could not find DSN in 
odbc.ini.
I have the same error message in the ODBC DataManager.

Could someone help me ?
Thanks.

Re: [mdb-dev] Getting the CVS to build and seg faults

[<dav...@ya...>]

 --- Tom Wright <mdb...@ma...> wrote: 
> Thanks for this David,
> I've tried applying the patches to the rc2 version
> of mdbtools however I 
> am having another problem getting it to compile as
> it is complaining about 
> an undefined reference to `log10'
> Does anyone have ideas how to solve this one?
> Thanks
> Tom
> 

Did you apply David Mansfield's combined patches
before applying my patches?  The log10 bit seems to
come from his patch to data.c.  I wrote my patches
after applying his and did not really delve too deeply
into his changes and am therefore not sure how
dependent my changes are on his changes -- although it
seems that they could well be.  I included David
Mansfield's combined patch in the tar.gz file.  Take a
look at the README for more details.

HTH.

Dave

=====
David Whiting
Adult Morbidity and Mortality Project (AMMP)
Box 65243, Dar es Salaam, Tanzania

Tel: +255 22 215 33 88         Fax: +255 22 215 33 85

________________________________________________________________________
Yahoo! Messenger - Communicate instantly..."Ping" 
your friends today! Download Messenger Now 
http://uk.messenger.yahoo.com/download/index.html

Re: [mdb-dev] Getting the CVS to build and seg faults

[Tom W. <mdb...@ma...>]

Thanks for this David,
I've tried applying the patches to the rc2 version of mdbtools however I 
am having another problem getting it to compile as it is complaining about 
an undefined reference to `log10'
Does anyone have ideas how to solve this one?
Thanks
Tom


On Fri, 19 Dec 2003 10:47:20 +0000 (GMT), David Whiting 
<dav...@ya...> wrote:

>  --- Tom Wright <mdb...@ma...> wrote:
> [...]
>>
>> Unfortunatly my data base has carrige returns in
>> memo fields which cause
>> mdb-export to seg fault every time. If anyone has a
>> fix for this can they
>> please let me know.
>
> I submitted a patch for mdb-export.c to:
>
> http://sourceforge.net/tracker/?atid=302294&group_id=2294&func=browse
>
> [I think I might have got the URL wrong in a previous
> post]
>
> that allows you to specify the row delimiter for
> exactly this reason.  I was then able to export using
> '|' to mark the end of the row and then tell MySQL to
> import the data with LINES TERMINATED BY '|'.  It was
> part of some changes I made to do something like
> mysqldump.  I have not tested this on any other
> backend.
>
> Dave
>
> =====
> David Whiting
> Adult Morbidity and Mortality Project (AMMP)
> Box 65243, Dar es Salaam, Tanzania
>
> Tel: +255 22 215 33 88         Fax: +255 22 215 33 85
>
> ________________________________________________________________________
> Yahoo! Messenger - Communicate instantly..."Ping"
> your friends today! Download Messenger Now
> http://uk.messenger.yahoo.com/download/index.html
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: IBM Linux Tutorials.
> Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
> Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
> Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click
> _______________________________________________
> mdbtools-dev mailing list
> mdb...@li...
> https://lists.sourceforge.net/lists/listinfo/mdbtools-dev

Re: [mdb-dev] Getting the CVS to build and seg faults

[<dav...@ya...>]

 --- Tom Wright <mdb...@ma...> wrote: 
[...]
> 
> Unfortunatly my data base has carrige returns in
> memo fields which cause 
> mdb-export to seg fault every time. If anyone has a
> fix for this can they 
> please let me know.

I submitted a patch for mdb-export.c to:

http://sourceforge.net/tracker/?atid=302294&group_id=2294&func=browse

[I think I might have got the URL wrong in a previous
post]

that allows you to specify the row delimiter for
exactly this reason.  I was then able to export using
'|' to mark the end of the row and then tell MySQL to
import the data with LINES TERMINATED BY '|'.  It was
part of some changes I made to do something like
mysqldump.  I have not tested this on any other
backend.

Dave

=====
David Whiting
Adult Morbidity and Mortality Project (AMMP)
Box 65243, Dar es Salaam, Tanzania

Tel: +255 22 215 33 88         Fax: +255 22 215 33 85

________________________________________________________________________
Yahoo! Messenger - Communicate instantly..."Ping" 
your friends today! Download Messenger Now 
http://uk.messenger.yahoo.com/download/index.html

[mdb-dev] Getting the CVS to build and seg faults

[Tom W. <mdb...@ma...>]

Here is how I got the CVS download to compile on my system (Slackware 9.1)

Edit the mdbtools/configure.ini file

Delete the line:
AC_CONFIG_FILES([src/Makefile])

Edit the line:
AC_OUTPUT(src/util/Makefile src/extras/Makefile Makefile include/Makefile 
src/Makefile src/libmdb/Makefile include/Makefile src/sql/Makefile 
src/odbc/Makefile doc/Makefile src/gmdb2/Makefile 
src/gmdb2/gladefiles/Makefile src/gmdb2/pixmaps/Makefile 
src/gmdb2/help/Makefile src/gmdb2/help/C/Makefile mdbtools.spec)

To remove the duplicate include/Makefile

export SED=sed

The last export command also needs to be carried out as root for 'make 
install' to work.

Unfortunatly my data base has carrige returns in memo fields which cause 
mdb-export to seg fault every time. If anyone has a fix for this can they 
please let me know.

Re: [mdb-dev] troubles with mdb odbc

[Adam W. <awi...@wh...>]

> I've got a web server based on linux redhat9 and a windows2000 machine with 
> and mdb access2000 file which i would like to query from the linux machine.
> So I installed unixODBC 2.2.5 and mdbtools from these rpm packets:
> mdbtools-0.5-1.i386.rpm
> mdbtools-odbc-0.5-1.i386.rpm
> Now I've mounted the windows directory with the mdb file via samba so I've 
> the mdb file in /home/lc/be.mdb.
> ...
> If I try mdb-sql it works fine and I can query tables on the mdb file; but 
> if I try to connect to the source via odbc using OpenOffice Data source it 
> doesn't work. OpenOffice finds the odbc source but can't read it.
> Here is a java code written to test the odbc connection. This code works 
> fine on a windows machine with the same odbc source configured but it 
> doesn't on my linux box ??
> Anyone can help please? I'm going crazy!!!

We get very much the same thing.  unixODBC works, MDBtools quasi-works,
MDB+ODBC goes down in flames.

[mdb-dev] troubles with mdb odbc

[Luca C. <jo...@ti...>]

Hello All,
I've got a web server based on linux redhat9 and a windows2000 machine with 
and mdb access2000 file which i would like to query from the linux machine.
So I installed unixODBC 2.2.5 and mdbtools from these rpm packets:
mdbtools-0.5-1.i386.rpm
mdbtools-odbc-0.5-1.i386.rpm
Now I've mounted the windows directory with the mdb file via samba so I've 
the mdb file in /home/lc/be.mdb.

here is my odbc.ini and odbcinst.ini.

===CUT=================================
[lc@einstein lc]$ cat /etc/odbc.ini
[prova]
Description = Microsoft cazzi
Driver = MDBToolsODBC
Database = /home/lc/be.mdb
Servername = localhost
UserName =
Password =
port = 5432

[lc@einstein lc]$ cat /etc/odbcinst.ini
[PostgreSQL]
Description             = ODBC for PostgreSQL
Driver          = /usr/lib/libodbcpsql.so
Setup           = /usr/lib/libodbcpsqlS.so
FileUsage               = 1

[MDBToolsODBC]
Description             = MDB Tools ODBC drivers
Driver          = /usr/lib/libmdbodbc.so.0
Setup           =
FileUsage               = 1
CPTimeout               =
CPReuse         =
===================================================

If I try mdb-sql it works fine and I can query tables on the mdb file; but 
if I try to connect to the source via odbc using OpenOffice Data source it 
doesn't work. OpenOffice finds the odbc source but can't read it.
Here is a java code written to test the odbc connection. This code works 
fine on a windows machine with the same odbc source configured but it 
doesn't on my linux box ??
Anyone can help please? I'm going crazy!!!

Regards
Luca Cervasio

======CUT test.java =====================
import java.sql.*;

class test {

         public static void main (String args[]) {
                 try {
                         Connection conn = null;
                         Statement st = null;
                         ResultSet rs = null;
                         Class.forName("sun.jdbc.odbc.JdbcOdbcDriver");
                         conn = DriverManager.getConnection("jdbc:odbc:prova");
                         st = conn.createStatement();
                         rs = st.executeQuery("select * from tblarticolo");
                         rs.close();
                         st.close();
                         conn.close();
                         System.out.println("GREAT it works!");
                 } catch (Exception e) {
                         System.out.println("it doesn't works...");
                 }
         }

}

Re: [mdb-dev] mysql backend; mdbdump (like mysqldump)

[<dav...@ya...>]

Bother.  I forgot to include the mdbdump script when I
uploaded the patches.  I have uploaded the patches
again and included the script this time.  Here's the
script on it's own:

# -----------------------
# mdbdump shell script.
# -----------------------
# Shell script to run mdb-schema and mdb-export giving
# an overall effect similar to mysqldump, i.e.
creating
# a table schema and a dump of SQL INSERT statements 
# that can be redirected to a MySQL process or a file.
# DW, 2003-12-09
                                                      
 
# Dump the schema
# option -S replaces blanks in names with _
# -T allows us to choose one table
# mysql specifies the mysql backend
mdb-schema -S -T "$2" "$1" mysql
                                                      
 
# Dump the data
# -I exports as INSERT statements instead of 
# comma-separated.
# -H suppresses the creation of a list of field names
# -D allows us to specify the date format
mdb-export -I -H -D %Y-%m-%d "$1" "$2"




 --- David Whiting <dav...@ya...>
wrote: > I have submitted some patches to:
> 
> http://sourceforge.net/tracker/index.php


=====
David Whiting
Adult Morbidity and Mortality Project (AMMP)
Box 65243, Dar es Salaam, Tanzania

Tel: +255 22 215 33 88         Fax: +255 22 215 33 85

________________________________________________________________________
BT Yahoo! Broadband - Save £80 when you order online today. Hurry! Offer ends 21st December 2003. The way the internet was meant to be. http://uk.rd.yahoo.com/evt=21064/*http://btyahoo.yahoo.co.uk