Release Notes for 7.3C UNIX
Version 7 Release Notes
Release: 7.3C
Ports: UNIX
Products: PROGRESS Version 7.3C for UNIX
Date: August 1995
1 Installation Issues
There are no installation issues with this release.
2 New Features and Functionality
This section lists new features or functionality provided in this
release.
2.1 Providing Help for PROGRESS Applications
PROGRESS provides two ways of supplying on-line help to application
users through the use of help strings and the development of a help
system using the PROGRESS Help Development utilities. A help string
is a string that appears in the status bar of a window and describes
the function of the field-level widget that has input focus. A help
system is a multi-component system that delivers large units of help
information to application users and provides tools that allow users
to navigate through help information.
It is our intention to continue to offer on-line help through the
use of help strings on all platforms. It is also our intention to
continue to provide the PROGRESS Help Development utilities only
on those platforms that offer a graphical user interface. Support
for the PROGRESS Help Development utilities on character platforms
is planned to be excluded from these platforms beginning with the
release of PROGRESS Version 8.
2.2 ORACLE and CISAM Certifications
ORACLE 7.1.6 and CISAM 6.0 have been successfully certified for this
release. No compatibility issues have been found.
3 Known Issues
This section lists known problems and other issues with this release.
3.1 Problem in Xtoolkit Layer of X11R5
There is a known problem within the Xtoolkit layer of X11R5 which
results in intermittent core dumps. This happens when a user selects
keys which cause a window to be destroyed and a new one to be created.
The problem is caused by a bug in the Xt layer; specifically in
Keyboard.c.
Customers who are experiencing the problem on any platform should
contact their X11R5 vendor and request a patch.
3.2 IBM RS/6000 - PROGRESS and SYBASE SQL Server
A TCP/IP open failure occurs on the IBM RS/6000 when you attempt
to run a PROGRESS to SYBASE application from the same window where
the SYBASE SQL Server was started. To avoid this error, start up
a SYBASE SQL Server, then close the window.
3.3 Solaris - PROBUILD and the C Compiler
When you run PROBUILD to generate a link script, the link script
contains references to the C compiler (CC). If the C compiler has
not been installed (the default), and you run the link script, you
will get errors. If you set cc to /usr/ccs/bin/ld, the link script
will run, but the resulting executable will dump core.This problem
occurs because the SparcCompiler front-end (cc) adds additional para-
meters to the loader. Progress Software is working with Sun Micro-
systems to find a solution which would allow our customers to
PROBUILD with the loader provided by Sun.
The SparcCompiler (3.0.1) adds the following parameters just after
"ld":
/opt/SUNWspro/SC3.0.1/lib/crti.o
/opt/SUNWspro/SC3.0.1/lib/crtl.o
/opt/SUNWspro/SC3.0.1/lib/_fstd.o
/opt/SUNWspro/SC3.0.1/lib/values-xs.o
These parameters are followed by all of Progress´ arguments to "ld´,
and the following is appended to the end:
-Y P,/opt/SUNWspro/SC3.0.1/lib:/usr/ccs/lib:/usr/lib
-Qy -lc /opt/SUNWspro/SC3.0.1/lib/crtn.o
NOTE:
The path /opt/SUNWspro/SC3.0.1 is the default directory for the Sun
SparcCompiler, and includes a version-dependent subdirectory
"SC3.0.1" corresponding to the version of the compiler. A customer
may have a different installation location and/or version of the
compiler.
If you do not have the C compiler installed, but still want to use
PROBUILD, run ld.setup in the dlc/probuild/eucapp directory before
running the link script generated by PROBUILD.
3.4 TTY Dictionary Invalidates Schema Cache
Whenever you go into the TTY dictionary table, field, or index editor,
the corresponding binary schema cache is invalidated, if one exists
and is in use as a connection parameter.
The following error is displayed:
The time stamp in the database does not match the time stamp in the
-cache file: /usr/rdl/devdbs6/bug.cf
WARNING: The -cache /usr/rdl/devdbs6/bug.cf parameter was used.
An error occurred while attempting to read the schema cache from
the named file.
The schema cache will be read from the database instead.
3.5 TTY Help May Not Work Without Environment Variables
TTY Help may not work if the following environment variables are
not set:
* HHPATH - should point to the directory containing the ttyhelp
module, for example:
HHPATH=$DLC/bin;export HHPATH
* HOHPATH should point to the directory containing the Help files,
for example:
HOHPATH=$DLC/prohelp;export HOHPATH
Also, PROTERMCAP must be set to $DLC/protermcap for help to work
properly.
3.6 Redirect Output to a File to Avoid Problems
Do not start PROGRESS processes with stdin, stdout, or stderr closed,
for example:
_mprshut dbname -C apw <&- >&- 2>&- &
This could result in database corruption. Instead, redirect stderr
and stdout to a file.
3.7 PROBKUP -Estimate Option Not For On-line Use
The PROBKUP -Estimate option is intended for off-line backup only,
and does not work when running on-line.
3.8 Index Check and Rebuild With Long Words
If you word-index a field that has words longer than 184 characters
without spaces, tab, or other special characters with ASCII codes
of 32 or less, they are handled differently than fixed versions by
Version 7.3A and earlier, and by Version 7.3B and 7.3C index check
and repair utilities. The fixed versions recognize the words
correctly, and truncate any words longer than 184 characters
(max word size). The earlier versions chop the long word every
184 characters, and create multiple keys.
3.9 INCR Dump Not Case-Sensitive for Labels
When you want to make case changes to a label for a particular field,
the change cannot be applied to other databases by creating a
delta.df file. For example,
current label Name
new label NAME
When creating the delta.df file, this schema change from lower- to
upper-case letters will not be recognized, and therefore cannot be
ported to another database in this manner.
3.10 TLI Not Supported in Version 7.3A or Later
Although TLI networking is listed in Appendix B of the Progress
System Administration Guide as a configurable element in PROBUILD,
it is not currently supported.
3.11 Differences in Collation for CONVMAP.CP
The collation names referred to here are those used in the convmap.dat
file.
1. V6-Danish/ISO8859-1
Character 158 which is non-printing is 0 in the version 6 Danish
collation.
It is 254 in 7.2 Danish collations and the 7.3A V6-Danish/ISO8859-1
collation
table in convmap.dat.
2. V6-Swedish/ISO8859-1
Character 158 which is non-printing is 0 in the version 6 Swedish
collation. It is 254 in 7.2 Swedish collations and the 7.3A
V6-Swedish/ISO8859-1 collation table in convmap.dat.
3. V6-Basic/ISO8859-1
4. V7.3C Basic/ISO8859-1 differs from the 7.2x Basic as follows:
If you would like the collation available in 7.2, you need to use
the 72-Basic/ISO8859-1 table.
Case-sensitive Sort
A umlaut now maps with the rest of the A´s
O umlaut now maps with the rest of the O´s
U umlaut now maps with the rest of the U´s
a umlaut now maps with the rest of the a´s
o umlaut now maps with the rest of the o´s
u umlaut now maps with the rest of the u´s
The above letters used to appear after the rest of the letters as
follows:
ABC abc u-umlaut a-umlat A-umlaut o-umlaut O-umlaut U-umlaut
Changed character 158 to be 254
U-^ maps with the rest of the U´s instead of O
o-tilde now maps with o´s
O-tilde now maps with O´s
Case-insensitive Sort
A umlaut now maps with the rest of the A´s
O umlaut now maps with the rest of the O´s
U umlaut now maps with the rest of the U´s
a umlaut now maps with A
o umlaut now maps with O
u umlaut now maps with U
3.12 R-Code Codepage Mismatch With -CPINTERNAL
There is a potential problem if a user has an r-code file that was
compiled with a -cpinternal codepage, but it is run later in another
session with a different -cpinternal codepage. It is possible that
the r-code codepage cannot be mapped to the current -cpinternal
codepage.
If the r-code contains strings which are used to control program
logic or values to be put in the database, the application might
not run correctly or the database might have scrambled data in it
if a conversion from r-code codepage to -cpinternal is not possible.
We recommend that you use rcode-info:codepage and session:charset
to determine if the two codepages are compatible.
3.13 Version 7 Product Update Incorrect on CONV67 Procedure
There are two ways to convert existing Version 6 databases to version
7.3C. You can either dump the Version 6 database and then load the
data into an empty Version 7 database, or you can use the CONV67
utility. Version 7.3C has more support for different code pages
than Version 6, and it is therefore important that Version 7 databases
have the appropriate code page label for the data they contain.
Version 7 also offers a number of different sort orders, so you may want
to load a particular collation table into your new database.The sections
in the Product Update Description that discuss upgrading databases
from Version 6 to Version 7 are replaced with the corresponding sections
below.
* The CONV67 Steps
* The CONVCHAR Process
* Using Dump/Reload for Database Conversion
The section "Loading the Collation Table" is added and follows "The
CONVCHAR Process".
The CONV67 Steps
1. Install Version 7 using the instructions in the "Progress Installation
Notes".
2. Determine what codepage you want the data to be in, as described
in the "Character Set Processing" chapter in the "PROGRESS System
Administration Guide." These
Release Notes contain the most current list of codepages and collations.
3. Execute the PROUTIL CONV67 utility. PROUTIL CONV67 has this
syntax:
proutil <db-name> -C conv67 [ibm850 | undefined]
If your codepage is to be anything other than "ibm850" or "iso8859-1",
specify "undefined" at this step. Once PROUTIL CONV67 finishes, the
database can support Version 7 functionality.
4. If the codepage determined in Step 2 is ibm850 or undefined,
back up your newly converted Version 7 database and review the Database
Chapter of the Product Update Description to ensure that your data
definitions handle the incompatibilities between Version 6 and Version
7 data definitions. If you decide to change the codepage or collation,
continue to the next section.
The CONVCHAR Process
The CONVCHAR qualifier on the PROUTIL utility converts the data in
the database to another character set, and updates the internal codepage
identifier. Here is a partial syntax for CONVCHAR. Refer to the
"PROGRESS System Administration Reference" for the full syntax.
proutil <db-name> -C convchar convert <codepage>
where <codepage> is the codepage name you determined in "The CONV67
Steps" above.
Loading the Collation Table
After you have set the database codepage with CONVCHAR you can change
the database collation table. This table determines the sequence
in which data is sorted.
You load the collation table into the database using the Data
Administration tool.
1. Determine the appropriate collation table (.df file) to load
from $DLC/prolang/<lang>.
2. Be sure to use the correct "-cpinternal", "-cpstream" and "-cpcoll"
parameters when starting PROGRESS.
3. In the Data Administration tool, select Admin->Load Data and
Definitions->Data Definitions (.df file) and enter the name of the
.df file.
4. After loading the collation table, you MUST rebuild the indexes.
The command to do this is:
proutil <db-name> -C idxbuild all
Example:
Convert a Version 6 database with Russian data (codepage 1251) to
the Version 7 Russian database. The codepage we will use is "1251";
the collation is "russian". The Russian collation table for codepage
1251 is in the file "rus1251.df".
1. proutil myrus -C conv67 undefined
2. proutil myrus -C convchar convert 1251
3. _progres myrus -p _admin.p -cpinternal 1251 -cpstream 1251 -cpcoll
russian
4. Admin->Load Data and Definitions->Data Definitions (.df file)
5. Enter the name $DLC/prolang/rus/rus1251.df
6. proutil myrus -C idxbuild all
Using Dump/Reload for Database Conversion
If you choose to dump your Version 6 database and then reload it
into Version 7 as your conversion method, here are some key points
to consider. When you create the Version 6 dump files, make sure
the data (.d) and data definition (.df) filenames are no longer than
8 characters. If your Version 6 database contains ibm850 data, the
.d files contain ibm850 data. In this case, during the load process
you must use -cpstream ibm850. Since Version 7 databases are by default
iso8859-1, the load process will convert the ibm850 data to iso8859-1
if you do not specify -cpstream ibm850. If the data in your Version
6 database uses a code page other than ibm850 or iso8859-1, perform
these steps to load your data into a Version 7 database:
1. Create an empty Version 7 database.
prodb <db-name> $DLC/prolang/<lang>/empty
2. Use the Data Admin tool to load the data. Be sure to specify
the -cpstream, -cpinternal and, if necessary, -cpcoll parameters
on the command line so that the data is loaded with the appropriate
character set.
3. Since files dumped from Version 6 do not have a codepage defined
in them, you will get a message saying so. Then you will be asked
to enter the codepage to use for this load. The fill-in will begin
with the name of the codepage of the empty.db you used in the prodb
command.
Supported Conversion Tables
The following table lists the code pages that PROGRESS supports and
their corresponding collation names and languages.
Code Page Collation Languages
Name Name
____________________________________________________________________
1250 Czech Czech, Slovak
1250 Hungarian Hungarian
1250 POLISH Polish
1250 ROMANIAN Romanian
1251 RUSSIAN Russian
1253 GREEK Greek
1254 TURKISH Turkish
1256 BASIC Arabic
620-2533 THAI Thai
708 BASIC Arabic
709 BASIC Arabic
710 BASIC Arabic
711 BASIC Arabic
714 BASIC Arabic
720 BASIC Arabic
721 BASIC Arabic
786 BASIC Arabic
BIG-5 BASIC Traditional Chinese
GB2312 BASIC Simplified Chinese
ibm437 BASIC Basic
ibm850 72-BASIC Version 7.2 Basic
ibm850 BASIC Western Europe and
the Americas
ibm850 DANISH Danish
ibm850 FINNISH Finnish
ibm850 GERMAN-LIBRARY German
ibm850 ICELANDIC Icelandic
ibm850 NORWEGIAN Norwegian
ibm850 SWEDISH Swedish
ibm851 GREEK Greek
ibm852 CZECH Czech, Slovak
IBM852 HUNGARIAN Hungarian
IBM852 POLISH Polish
IBM852 ROMANIAN Romanian
ibm857 TURKISH Turkish
IBM861 ICELANDIC Icelandic
ibm862 HEBREW Hebrew
ibm866 RUSSIAN Russian
iso8859-1 72-BASIC Version 7.2 Basic
iso8859-1 BASIC Basic
ISO8859-1 CZECH Czech, Slovak
iso8859-1 DANISH Danish
iso8859-1 FINNISH Finnish
iso8859- GERMAN-LIBRARY German
iso8859-1 ICELANDIC Icelandic
iso8859-1 NORWEGIAN Norwegian
iso8859-1 SWEDISH Swedish
iso8859-1 V6-BASIC Version 6 Basic
iso8859-1 V6-DANISH Version 6 Danish
iso8859-1 V6-SWEDISH Version 6 Swedish
iso8859-2 CZECH Czech, Slovak
ISO8859-2 HUNGARIAN Hungarian
ISO8859-2 POLISH Polish
ISO8859-2 ROMANIAN Romanian
iso8859-5 RUSSIAN Russian
iso8859-8 HEBREW Hebrew
koi8-r RUSSIAN Russian
KSC5601 BASIC Korean
MAZOVIA POLISH Polish
ROMAN-8 BASIC Basic
ROMAN-8 CZECH Czech, Slovak
SHIFT-JIS BASIC Japanese
undefined BASIC Basic
undefined V6-BASIC Version 6 Basic
undefined V6-DANISH Version 6 Danish
undefined V6-SWEDISH Version 6 Swedish
The following table lists the conversions that the convmap.cp file
supports. It is an up-to-date edition of Table A-2 in the "PROGRESS
System Administration Guide."
Source Target Comments
Code Page Code Page
___________________________________________________________________________________
1250 IBM852 Czech Republic,
Slovakia, Hungary,
Poland, Romania
1250 ISO8859-2 Czech Republic,
Slovakia, Hungary,
Poland, Romania
1250 MAZOVIA Poland
1251 IBM866 Cyrillic
1251 ISO8859-5 Cyrillic
1251 KOI8-r Cyrillic
1253 IBM851 Greece
1254 IBM857 Turkey
1256 708 Arabic countries
1256 709 Arabic countries
1256 710 Arabic countries
1256 711 Arabic countries
1256 714 Arabic countries
1256 720 Arabic countries
1256 721 Arabic countries
1256 786 Arabic countries
708 1256 Arabic countries
709 1256 Arabic countries
710 1256 Arabic countries
711 1256 Arabic countries
714 1256 Arabic countries
720 1256 Arabic countries
721 1256 Arabic countries
786 1256 Arabic countries
german-7-bit ISO8859 Western Europe
and Americas
For 7-bit terminals
and -cpterm only
ibm037 IBM850 AS/400
ibm037 ISO8859-1 AS/400
ibm278 IBM850 AS/400
ibm278 ISO8859-1 AS/400
ibm437 IBM850 Western Europe
and Americas
ibm437 ISO8859-1 Western Europe
and Americas
ibm500 IBM850 AS/400
ibm500 ISO8859-1 AS/400
ibm850 IBM037 AS/400
ibm850 IBM278 AS/400
ibm850 IBM437 Western Europe
and Americas
ibm850 IBM500 AS/400
IBM850 IBM861 Iceland
ibm850 ISO8859-1 Western Europe
and Americas
ibm851 1253 Greece
ibm852 1250 Czech Republic,
Slovakia,Hungary,
Poland, Romania
ibm852 ISO8859-2 Czech Republic,
Slovakia, Hungary,
Poland, Romania
IBM852 MAZOVIA Poland
ibm857 1254 Turkey
IBM861 IBM850 Iceland
IBM861 ISO8859-1 Iceland
ibm862 ISO8859-8 Israel
ibm866 1251 Cyrillic
ibm866 ISO8859-5 Cyrillic
ibm866 KOI8-r Cyrillic
iso8859-1 GERMAN-7-bit Western Europe
and Americas
For 7-bit terminals
and -cpterm only
iso8859-1 IBM037 AS/400
iso8859-1 IBM278 AS/400
iso8859-1 IBM437 Western Europe
and Americas
iso8859-1 IBM500 AS/400
iso8859-1 IBM850 Western Europe
and Americas
ISO8859-1 IBM861 Iceland
iso8859-1 ISO8859-1 Western Europe
and Americas
ISO8859-1 ROMAN-8 Western Europe
and Americas
iso8859-1 SWEDISH-7-bit Western Europe
and Americas
For7-bit terminals
and -cpterm only
iso8859-2 1250 Czech Republic,
Slovakia, Hungary,
Poland, Romania
iso8859-2 IBM852 Czech Republic,
Slovakia, Hungary,
Poland, Romania
ISO8859-2 MAZOVIA Poland
iso8859-5 1251 Cyrillic
iso8859-5 IBM866 Cyrillic
iso8859-5 KOI8-r Cyrillic
iso8859-8 IBM862 Israel
koi8-r 1251 Cyrillic
koi8-r IBM866 Cyrillic
koi8-r ISO8859-5 Cyrillic
MAZOVIA 1250 Poland
MAZOVIA IBM852 Poland
MAZOVIA ISO8859-2 Poland
ROMAN-8 ISO8859-1 Western Europe
and Americas
swedish-7-bit ISO8859-1 Western Europe
and Americas
For7-bit terminals
and -cpterm only
3.14 Demo and Sports Database Code Page Issues
The demo and sports database code pages are undefined. This is to
allow ease of use by all languages and nationalities.
If you want to create a "native" database or application, you can
first convert these databases to the "native" code page and collation
by following the steps in Appendix A of the Progress System
Administration Guide.
3.15 Temp-Table and Work-Table Enhancements
Two enhancements for temporary tables and work tables are the following:
* The option to include dictionary validation for the fields of a
TEMP-TABLE or WORK-TABLE that is defined like a database table. For
example:
DEFINE TEMP-TABLE xxx LIKE customer VALIDATE.
* Support for joins between a TEMP-TABLE or WORK-TABLE and any other
table. As with joins between database tables, the two tables must
contain a commonly named field that participates in a unique index
for at least one of the tables.
3.16 ADD KEEP-FRAME-Z-ORDER Attribute
Setting the KEEP-FRAME-Z-ORDER to TRUE for a window ensures that
any overlapping frames in the window that are not in the foreground
do not automatically move to the foreground when they receive focus.
3.17 New Session Attributes
The following are new session attributes:
SESSION:CPINTERNAL Determines the code page used in memory.
SESSION:CPSTREAM Determines the code page used for stream I/O.
SESSION:CPTERM Specifies the code page of your character
terminals.
SESSION:CPPRINT Specifies the code page used for printer output.
SESSION:CPRCODEIN Specifies the code page to convert text instead
of the code page with which the r-code was written.
SESSION:CPRCODEOUT Specifies the code page to mark and convert text.
If not specified,the code page specified with
-cpinternal is used.
For more information, see Chapter 4, Startup Parameters, in the System
Administration Guide.
3.18 Frame Phrase Attr-Space Option/Attribute
The ATTR-SPACE and the NO-ATTR-SPACE of the Frame Phrase are supported
for backward compatibility only. The only effect of the ATTR-SPACE
option is to set the ATTR-SPACE attribute for each fill-in or text
widget to TRUE, and the only effect of the NO-ATTR-SPACE option is
to set the ATTR-SPACE attribute for each fill-in and text widget
to FALSE. Otherwise, they are ignored, since Version 7 does not support
space-taking terminals.
3.19 Support for Word Indexes
The CONTAINS and MATCHES operators are not supported in PROGRESS/SQL
for Version 7.3C, but will be supported in a future release.
3.20 FastTrack Runtime
As of PROGRESS Version 7, FastTrack is no longer supported nor shipped
as a development tool. However, a FastTrack runtime product has been
provided to enable customers to run FastTrack applications in Version
7.
3.21 Bleeding Record Lock Handling
In PROGRESS releases prior to Version 7.3B, a SHARE-LOCK can be held
on a record accessed by two or more buffers even after the only buffer
that may have accessed the record with SHARE-LOCK has been disconnected
from the record.
The following code segment for the sports database illustrates a
bleeding record lock situation in releases prior to Version 7.3B:
DEFINE BUFFER acust FOR customer.
DEFINE BUFFER bcust FOR customer.
FIND acust 1 WHERE acust.cust-num = 1 NO-LOCK.
FIND bcust 1 Where bcust.cust-num = 1 SHARE-LOCK.
RELEASE bcust.
/*At this point, customer record 1 still holds a SHARE-LOCK*/
After the RELEASE statement, there is a SHARE-LOCK on customer record
1, even though this record has been released from the bcust buffer.
This is because the acust buffer still references customer 1, and
PROGRESS retains the SHARE-LOCK for customer 1, even though the record
was originally requested for acust with NO-LOCK. Thus, the SHARE-LOCK
has bled from the bcust buffer to the acust buffer.
Starting with Version 7.3B, PROGRESS prevents bleeding record locks
from occurring. Any time a transaction ends or a record is disconnected
from a buffer, PROGRESS checks if any other buffer requires a SHARE-LOCK
on that record. If not, any SHARE-LOCK for that record is downgraded
to a NO-LOCK. In the previous example, customer 1 is downgraded to
NO-LOCK after the RELEASE statement.
Records are typically disconnected from a buffer for the following
reasons:
* The buffer has gone out of scope.
* The procedure explicitly releases the record.
* The procedure reads a new record to replace the current one in
the buffer.
To retain bleeding record lock behavior in Release 7.3B for downward
compatibility, use the Bleeding Record Lock (-brl) parameter for
database connection or server startup. For any application written
prior to Release 7.3B (or using equivalent semantics), you can tell
whether you need the -brl parameter. PROGRESS generates the following
run-time messages when a record you are trying to update (customer,
in this case) is unexpectedly downgraded to NO-LOCK:
**customer record has NO-LOCK status, update to field not allowed.
**Unable to update customer Field.
3.22 Identity Fields
A column defined in SYBASE Version 10 with the IDENTITY property
is treated by PROGRESS as a read-only field. However, if the field
is on a form, a user can still land on the field and add/delete/change
the data for that field. Since the field is read-only, any changes
the user makes are ignored. On forms or screens in end-user
applications, make the field read-only.
3.23 SYBASE Stored Procedure with Defaulted Parameter
PROGRESS passes something to a SYBASE stored procedure which SYBASE
interprets as a parameter that is not null. However, PROGRESS is
not passing an actualparameter.
3.24 Verify in SYBASE Version 10
The "Verify SYBASE Table Definition" function in the Data Administra-
tion tool finds a difference, and allows the user to select OK to fix
the problem, but the fix does not take effect.
3.25 ORACLE Procedure Parameter Date
A stored procedure with "in a param date" cannot select rows using
that date. The ORACLE date contains a time component, but the PROGRESS
date does not.
For a workaround, use the ORACLE TRUNC function when testing a date
parameter for equality with a column in the stored procedure. For
example:
SELECT ... from <table> WHERE TRUNC (<column>) =
TRUNC (<parameter>)
instead of:
WHERE <column> = <parameter
3.26 Password Parameter displays on SVR4 Machines
When you do a ps -deaf on a machine that is running a spawned server,
you can see the -P (password) parameter.
3.27 Lowercase _buffer_ Broken in SYBASE Version 10
When creating a view In SYBASE Version 10, if lowercase is used for
the word "_buffer_" it will cause error 2236 an 2237 when the stored
procedure is run. This does not occur in SYBASE Version 4. Therefore,
when you upgrade from SYBASE Version 4 to SYBASE Version 10, you
will have a problem.
The error messages displayed are:
Cannot execute a SQL statement while a stored procedure, a SEND-SQL
cursor or a VIEW cursor is open. (2236)
You may use the -Dsrv PRGRS PROC TRAN,1 startup parameter for multiple
database connections. (2237)
For a workaround, change the case of _buffer_ in .p to _BUFFER_.
Drop the SYBASE stored procedure, and add the the SYBASE stored procedure
that has the uppercase word.
3.28 Cannot Override Schema Holders
You cannot override the schema holder or the "other" parameters in
the Print Engine. The connect string specified in the RB-DB CONNECTION
parameter is parsed by Report Builder before being passed to the
SQL Engine. Report Builder only looks for certain connect parameters.
Therefore you cannot override the database type, schema holder, or
other connect parameters.
To run a report against an Oracle or SYBASE database using the Report
Engine, the report must have been saved with the correct database
type and schema holder.
3.29 Core Dump Call Stored Procedure with Datetime Parameter
You cannot call stored procedures in SYBASE Version 4 which have
parameters of type "datetime".
3.30 ORACLE DataServer ORALIB environment variable
To set the ORALIB environment variable for ORACLE Version 7.1, follow
the steps in Chapter 3 of the PROGRESS DataServer for ORACLE Guide,
"Creating the DataServer Executables on UNIX" with the following
substitutions for Steps 2 and 4:
2> Copy the files from $ORACLE_HOME/rdbms/demo to another directory.
4> Enter the following command to use the make file to link the
ORACLE sample
C program and direct the output to a temporary file:
make -f $ORACLE_HOME/rdbms/lib/oracle.mk > tmpfile
Progress Software Technical Support Note # 15109