Safe mode
Does the language implementation allow for a "safer mode" of execution? Usually termed Safe mode, a more realistic view is probably Safer mode or restricted mode. It is one thing to place restrictions on execution, and another thing entirely to allow execution of scripts from untrusted sources and assume nothing untoward will happen.
Along with a simple yes/no answer, describe what features are restricted when running in safe mode.
Jsish
The jsish interpreter allows a -s, --safe command line switch to restrict access to the file system.
For example, given safer.jsi:
<lang javascript>File.write('/tmp/safer-mode.txt', 'data line');</lang>
- Output:
prompt$ jsish safer.jsi prompt$ jsish -s safer.jsi /home/btiffin/lang/jsish/safer.jsi:2: error: write access denied by safe interp: /tmp/safer-mode.txt (at or near "data line") ERROR
The Jsish implementation borrows many ideas from , and also includes an Interp module. These sub interpreters can also be set to run in a safer mode.
prompt$ jsish # var si = new Interp({isSafe:true}); variable # si.source('safer.jsi'); error: read access denied: /home/btiffin/lang/jsish/safer.jsi ERROR
Some control is allowed over the restrictions provided by safer mode.
<lang javascript>var interp1 = new Interp({isSafe:true, safeWriteDirs:['/tmp'], , safeReadDirs:['/tmp']});</lang>
REXX
For running REXX on IBM mainframes, REXX supports the option Scan for the trace statement.
This allows the program to be processed (and be checked for syntax errors), but commands to the "host system" won't be executed.
However, not all REXXes support this option.