To display error origins, which file and line it originates, you could "compile" the weblang code into an AST before running it. This will also allow sending programs over the internet or storing them in a database. The compiled version will be faster than the YAML version since there is no YAML parsing overhead.
The AST could be JSON format and human readable.
Function invocations could be tracked through an two-way hash, like JWT, that contains the line number and file name and maybe more as part of the call id: filename.wl,34,abcdef3432df
These are then universal and can be passed into central error loggers.
To display error origins, which file and line it originates, you could "compile" the weblang code into an AST before running it. This will also allow sending programs over the internet or storing them in a database. The compiled version will be faster than the YAML version since there is no YAML parsing overhead.
The AST could be JSON format and human readable.
Function invocations could be tracked through an two-way hash, like JWT, that contains the line number and file name and maybe more as part of the call id:
filename.wl,34,abcdef3432df
These are then universal and can be passed into central error loggers.