FWD

Data Export¶

This input is not needed during the code conversion process. But to put an application into testing or into production, it will be necessary to migrate each database instance which is needed to support such installations. To do this, each database instance to be migrated must have its tables exported in the .d (EXPORT) format.

Please see Data Migration for details on how to export the data. The result would be a dump file for each table of each database instance being migrated. The base filename for each export file must match the DUMP-NAME attribute in the .df file ADD TABLE entry for the corresponding table.

What is the encoding of the data export files (.d data dump files)? This must match the encoding specified in the FWD tools for data import.

Internationalization (I18N)¶

Applications which use the Progress 4GL default settings for internationalization (I18N) can ignore this step. For other applications it is important to have an understanding of the internationalization requirements. I18N can effect everything from conversion processing to runtime support, from the client to the application server and the database server.

See the Internationalization chapter for details on how to encode these values for conversion.

Conversion Inputs Encoding¶

What is the encoding of the source files for the application (include files, procedures, classes)? This must be the same value for all files and it must match the encoding used by the FWD tools that are used to parse those files.

What is the encoding of the exported schemata? This should match the encoding of the source files because the same encoding type will be used to process the schemata as is used for the source files.

Translation Manager¶

TBD: Provide details here about what needs to be gathered for Converting a Translated Project.

Runtime Values¶

Does the application use any of the following?

• DBCS
• UNICODE
• Translation Manager
• country/locale-specific empty databases
• country/locale-specific Progress message files
• code page conversion features
• customized character processing tables - character attribute tables, case tables, collation tables, code-page conversion tables (convmap.dat)
• customized word break tables

What does OpenEdge report for the following:

• code-page(s) used in the Progress database
• SESSION:CPSTREAM
• SESSION:CPINTERNAL
• SESSION:CPTERM
• SESSION:CPPRINT
• SESSION:CPCASE
• SESSION:CPLOG
• SESSION:CPRCODEIN
• SESSION:CPRCODEOUT
• SESSION:CPCOLL
• SESSION:DATE-FORMAT
• SESSION:NUMERIC-DECIMAL-POINT
• SESSION:NUMERIC-FORMAT
• SESSION:NUMERIC-SEPARATOR

Process Environment¶

Some of the configuration or state of a Progress 4GL application is derived from the environment or command line used when the Progress runtime is started.

The most important of these values is the PROPATH. This specifies how external procedures and include files are to be found, so it has a huge impact on how the source code is preprocessed and how procedures link to one another (e.g. via the RUN statement). This is especially important when all references to include files or procedures are not relative to the root of the project's source code tree. In such a case, the PROPATH is essential for resolving otherwise ambiguous filenames. For example, the preprocessor may not find all include files if the PROPATH is not right, which could make the 4GL syntactically incorrect or logically incomplete (even if the code does syntactically parse properly). See above for the requirements about the PROPATH.

In addition, the default processing of the Progress environment can be modified using command line options (or a profile or .pf file with options). The list of these options needs to be known. These values can normally be found by reviewing all startup scripts or command lines used to launch the environment. .pf files can be "nested" to an arbitrary level. It is important that the entire "graph" of .pf files is gathered.

In a GUI application, any .ini files should be gathered. These usually contain font definitions, color definitions, as well as defining other options or configuration.

A list of all Windows Progress-specific or application-specific registry keys/values that are set.

All scripts, batch files or other launcher programs should be gathered. Consider that the application may have batch programs, appserver or WebSpeed brokers/agents and multiple kinds of clients, all of which might use different launching mechanisms.

Appserver, WebSpeed and DataServer configuration files (e.g. the ubroker.properties).

Finally, any environment variables that are read from 4GL code should be known and the possible values should be enumerated. This is also often something that can be found by reviewing startup scripts, command lines and user/global profile scripts. These variables and values are not likely to be needed for properly converting the source code, but it is very likely that they will be needed when configuring the runtime environment for executing the converted application.

Terminal Configuration¶

For character user interface (ChUI) applications, the terminal types used and the terminal configuration have an important effect on how the application will behave. In a graphical user interface (GUI) environment, these can be ignored.

Gather the set of supported protermcap files. These files provide a mapping between each terminal type's features and such things as colors or the keys that will be reported to the application as input.

Make a list of all the supported terminal types that must be certified to work in the converted environment.

Neither of these are needed for running the code conversion or for handling data migration. Instead, these are used to customize the FWD client's runtime environment to match the expected behavior for the supported terminals.

.wrx Files¶

Any use of OCX controls (a.k.a. Active-X controls) in 4GL code requires a corresponding .wrx file. To convert "live" OCX usage properly, these .wrx files are needed in FWD as well.

Static Resources, Templates and Configuration Files¶

GUI applications usually have images/bitmaps, icons, fonts that need to be present for the application to run. Likewise most applications have other dependencies such as document templates, configuration or data files. These will be needed at runtime and some of them may be best included at conversion time so that they can be stored/distributed in the converted application .jar file(s). This is especially useful for graphical resources such as images or icons, since these are statically known at compile/conversion time and they should always be present with the application. Distributing them separately is not a good practice.

HLC Source Code¶

If the application uses HLC (the High Level Call interface) to integrate 4GL code with native (usually C language functions) code, then it is important to review the HLC source code. The HLS source code is compiled into a static object using Progress 4GL header files and calling 4GL services to read and write variables. This resulting compiled object code is linked to create a custom Progress 4GL executable which can dispatch CALL language statements to the linked in HLC functions. Everything runs in the same process.

Such an approach will not work in FWD, however that code can be easily migrated to a shared library and the 4GL CALL statements can be rewritten to invoke the shared library functions. This change has the advantage that it can be automatically converted without losing any functionality. Three key points: the HLC code must be available for review; the HLC/4GL code will have to be modified by hand to use the shared library approach; and the FWD conversion/runtime does not yet have support for generic shared library usage.

Checklist¶

The following checklist can be used to ensure that everything is available:

Input Type	Checklist Item	Available
Source Code	All Progress 4GL external procedure files which are in use or are referenced.
Source Code	WebSpeed HTML files which are in use or are referenced.
Source Code	TRIGGER-PROCEDURE files referenced by the schema.
Source Code	OO class files referenced in the application.
Source Code	ADM/ADM2 customized templates or application-specific extensions.
Source Code	3 rd party procedures or classes used in the application AND which are expected to be converted (must be appropriately licensed for this to be legal).
Source Code	All application code is not in r-code or procedure library form.
Source Code	All referenced Progress 4GL include files.
Source Code	WebSpeed HTML files that are used as include files by 4GL source code.
Source Code	Any include files referenced in the database schema.
Source Code	3 rd party include files used in the application AND which are expected to be converted (must be appropriately licensed for this to be legal).
Database Schemata	An exported .df schema for each AND every database (with unique data definitions) that is accessed by the application.
Database Schemata	The logical database name for each .df file.
Database Schemata	The database metadata schema exported in a .df format. This is necessary if the application has ANY references to metadata tables or fields.
Database Schemata	The list of databases that are connected by default when the application starts.
REST and SOAP Artifacts	TBD
PROPATH	For each compilation set, provide the exact list of 4GL program names (external procedures, trigger procedures or classes) which are compiled in the set.
PROPATH	For each compilation set, provide the PROPATH used for compilation.
PROPATH	For each compilation set, provide the platform on which the compilation is done (e.g. Windows or Linux or UNIX).
PROPATH	For each unique 4GL session executed, provide the 4GL program name (external procedures) used as the startup program for the session.
PROPATH	For each unique 4GL session executed, provide the PROPATH used by the session.
PROPATH	For each unique 4GL session executed, provide the platform on which the session is run (e.g. Windows or Linux or UNIX).
Keyword Ignore List	Keyword ignore list.
Data Export	Data dump files in .d format for each table of each database instance to be supported as a test or production application environment.
Data Export	What is the encoding of the database dump (.d) files? This is the codepage used to read these files.
Conversion Inputs Encoding (I18N)	What is the encoding of the source files and schemata? This is the codepage used to read these files.
Translation Manager (I18N)	TBD
Runtime Values (I18N)	List of I18N features used.
Runtime Values (I18N)	Customized I18N data files.
Runtime Values (I18N)	List of code pages used at runtime.
Runtime Values (I18N)	Date format(s) used.
Runtime Values (I18N)	Numeric format(s) used.
Process Environment	The command line or .pf files in use to startup the Progress session.
Process Environment	All .ini files used by the application (GUI only).
Process Environment	A list of all Windows Progress-specific or application-specific registry keys/values that are set.
Process Environment	Any scripts, batch files or other launching programs for the application. Consider that the application may have batch programs, appserver or WebSpeed brokers/agents and multiple kinds of clients, all of which might use different launching mechanisms.
Process Environment	Appserver, WebSpeed and DataServer configuration files (e.g. the ubroker.properties).
Process Environment	The list of ENVIRONMENT variables read by the application and their possible values.
Terminal Configuration	The set of supported and customized protermcap files.
Terminal Configuration	The list of supported terminal types.
.wrx Files	.wrx file for any OCX that is in use.
Static Resources, Templates and Configuration Files	Images, icons and other resources or critical data/configuration files.
HLC	HLC source code if used.

Test and Development Environment¶

The developers working on the conversion must have access to an OpenEdge full 4GL development environment. This is a test/development system where the original application can be run in the 4GL against representative test databases.

At the beginning of the conversion project, source code can be compared to determine if it is broken in the 4GL, the environment can be investigated, 4GL compiles and listings can be generated to understand the code.

Once the code is converted, access is necessary as a baseline to compare against the converted code (to compare the look/feel, behavior and functionality of the 4GL and FWD environments). Accordingly, there are important uses of the comparison environment throughout the entire conversion project, so it is imperative that such an environment is available. That environment must also be as close to a production environment as is possible.

This development environment must be a full 4GL development license which provides the ability to write 4GL code and run it and any 4GL development tools needed. This is often needed to explore how a particular 4GL feature works and ensure that various forms of that feature can be tested.