Debugger for GnuCOBOL
An extension to debug or execute GnuCOBOL code. Forked from COBOL Degub.
- Features
- Requirements
- Usage
- Code coverage
- Attaching to a running process
- Display application output in a separate window
- Documentation
- Troubleshooting
- Development
- Setting breakpoints
- Continue, Stop, Restart, Step Over, Step Into, Step Out
- Variables pane, including Copy Value, Copy as Expression and Add to Watch
- Watch pane with expressions
- Code coverage
- No mainframe required
This extension is meant to work in combination with the SuperBOL Studio extension for COBOL.
- GnuCOBOL
cobc
3.1+ installed. - GNU Debugger
gdb
13.0+ installed.
When your launch.json
config is set up, you can debug or execute your COBOL program. If you debug a Compilation Group (main- and sub- programs), you need to list sub-programs inside group
property. Here's an example:
{
"version": "0.2.0",
"configurations": [
{
"name": "COBOL debugger",
"type": "gdb",
"request": "launch",
"cobcargs": ["-free", "-x"],
"group": ["subsample.cbl", "subsubsample.cbl"]
}
]
}
Pick COBOL debugger
from the dropdown on the Debug pane in VS Code. Press the Play button or F5
to debug or Ctrl+F5
to execute.
The debugger uses C sourcecode generated by the compiler upon each debugging session. If the sourcemap isn't accurate or you see any other issues, please make a bug-report.
You can estimate an execution flow of your COBOL program.
Set coverage
property to true
in your launch.json
and start debugging session. Here's an example:
{
"version": "0.2.0",
"configurations": [
{
"name": "COBOL debugger",
"type": "gdb",
"request": "launch",
"cobcargs": ["-free", "-x"],
"coverage": true
}
]
}
The extension decodes the code-coverage files in gcov
format generated by the compiler.
You may debug your COBOL program attaching to a running process. In order to achieve that, you have two options:
Add pid
property to your launch.json
and start debugging session (you can use a input variable to help like the sample below).
Here's an example:
{
"version": "0.2.0",
"configurations": [
{
"name": "COBOL debugger attach",
"type": "gdb",
"request": "attach",
"cobcargs": ["-free", "-x"],
"pid": "${input:pid}"
}
],
"inputs": [
{
"id": "pid",
"type": "promptString",
"description": "PID to attach"
}
]
}
Add remoteDebugger
property to your launch.json
.
Here's an example:
{
"version": "0.2.0",
"configurations": [
{
"name": "COBOL debugger attach",
"type": "gdb",
"request": "attach",
"cobcargs": ["-free", "-x"],
"remoteDebugger": "localhost:5555"
}
]
}
Add gdbtty
property to your launch.json
. Here’s an example:
{
"name": "COBOL debugger",
"type": "gdb",
"request": "launch",
"cobcargs": [
"-free",
"-x"
],
"coverage": false,
"gdbtty": true
}
On Linux you can see the output of the application in Vs Code itself. Add gdbtty
property with vscode
value to your launch.json
. Here is an example:
{
"name": "COBOL debugger",
"type": "gdb",
"request": "launch",
"cobcargs": [
"-free",
"-x"
],
"coverage": false,
"gdbtty": "vscode"
}
You can also use these options with gdbtty
: xterm
, gnome-terminal
, konsole
and xfce4-terminal
.
For a more in depth documentation please check the SuperBOL Documentation
Add verbose
property to your launch.json
and start debugging session. In DEBUG CONSOLE
you will see complete communication log between gdb
and VS Code. Here's an example:
{
"version": "0.2.0",
"configurations": [
{
"name": "COBOL debugger",
"type": "gdb",
"request": "launch",
"cobcargs": ["-free", "-x"],
"verbose": true
}
]
}
- Fork the repository.
- Clone it to your machine and open with VS Code.
- Install dependencies by
yarn install
command in the terminal. - Pick
Extension
from the dropdown on the Debug pane and pressF5
. This will open new VS Code instance with your cloned extension in debugging mode. - Follow Requirements and Usage sections above.
- In the first VS Code instance you may put breakpoints to explore the functionality.
- Stop the second VS Code instance and implement your idea in TypeScript.
- Pick
Tests
from the dropdown on the Debug pane and pressF5
. Keep them green. - Push your changes and create Pull Request to the original repository.