[Bugzilla] [Bug 13840] New: execstr should have an execution mode where it hide scope inherited variables

[bugzilla.admin at scilab.org]

http://bugzilla.scilab.org/show_bug.cgi?id=13840
--------- Bug Summary -----------
execstr should have an execution mode where it hide scope inherited variables

---------- Changed by ------------
clement.david at scilab-enterprises.com

---------- What changed ----------
bug_id, short_desc, product, version, rep_platform, op_sys, bug_status, bug_severity, priority, component, assigned_to, reporter


            Bug ID: 13840
           Summary: execstr should have an execution mode where it hide
                    scope inherited variables
           Product: Scilab software
           Version: YaSp GIT version (Scilab 6 dev.)
          Hardware: PC
                OS: All OS
            Status: NEW
          Severity: Wishlist
          Priority: P5
         Component: Scilab
          Assignee: bugzilla.admin at scilab.org
          Reporter: clement.david at scilab-enterprises.com

BUG DESCRIPTION:
----------------

When checking for variable existence or script validation, I use execstr with "errcatch". However I want to protect the executed string to access
pre-defined variable or to push variables to the calling scope (using resume). Currently this is not possible but a scope-restricted
global-restricted execution mode can be easily added in YaSp to handle that situation.

Note: currently some Scilab included script use 'who' to compare existing variables before and after the execution but this is not really safe as it
is possible to modify "global" variables or resume-ing existing variables to push them.



HOW TO REPRODUCE THE BUG:
-------------------------

A=1

txt="V=3 ; disp(A,V)" // this is a user entered text to validate
ierr=execstr(txt, ["errcatch" "clear"])



OTHER INFORMATION:
------------------

-- 
You are receiving this mail because:
You are watching all bug changes.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/bugzilla/attachments/20150427/a9713ea4/attachment.htm>