[Tcl-bugs] [ tcl-Feature Requests-1553630 ] standardize [subst] exception treatment

[SourceForge.net <no...@so...>]

Feature Requests item #1553630, was opened at 2006-09-06 15:44
Message generated for change (Settings changed) made by dgp
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=360894&aid=1553630&group_id=10894

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: 45. Parsing and Eval
Group: None
>Status: Closed
>Resolution: Rejected
Priority: 5
Private: No
Submitted By: Don Porter (dgp)
Assigned to: Don Porter (dgp)
Summary: standardize [subst] exception treatment

Initial Comment:

Question arose on the chat:

% catch {subst {0[return -level 0 -code 6 foo]1}} m o  
0
% set m
0foo1
% set o
-code 0 -level 0

Should [subst] really be swallowing
up extension exception codes like that?



----------------------------------------------------------------------

Comment By: Don Porter (dgp)
Date: 2007-02-22 12:48

Message:
Logged In: YES 
user_id=80530
Originator: YES


Doc review also reveals that
the behavior noted in the
original report here did get
documented in Tcl 8.4, so 
making changes is a matter of
a (mildly incompat) Feature
Request, not a bug report.
Shifting report to the RFE Tracker...

----------------------------------------------------------------------

Comment By: Don Porter (dgp)
Date: 2007-02-22 12:45

Message:
Logged In: YES 
user_id=80530
Originator: YES


upon further review, the feasibility/ease
of bytecompiling [subst] is mostly
independent of this issue.  Even if
[subst] had always handled extension
exception codes in a standard way, a
byte compiled version of [subst] would
still need special crafting to duplicate
[subst]'s handling of the break, continue,
and return codes from Tcl's built-in set.



----------------------------------------------------------------------

Comment By: Donal K. Fellows (dkf)
Date: 2007-02-21 17:42

Message:
Logged In: YES 
user_id=79902
Originator: NO

The current non-standard exception code handling in [subst] is probably
"because it happened to be convenient to implement that way". I'd not stand
in the way of any represenation that it is a bug that needs fixing...

----------------------------------------------------------------------

Comment By: Don Porter (dgp)
Date: 2007-02-21 14:20

Message:
Logged In: YES 
user_id=80530
Originator: YES


Note that this (mis-)behavior
appears to be an obstacle to
bytecompiling [subst], since
any compiled version would have
to match these non-standard
treatments of exception codes.

----------------------------------------------------------------------

Comment By: Don Porter (dgp)
Date: 2006-09-06 16:00

Message:
Logged In: YES 
user_id=80530


It appears that the interface
for Tcl_SubstObj() is a contributor
to this "design".

Since it returns either NULL, or
the substituted result, it has no
way to indicate other types of
exception conditions.  Thus, as
implemented, [subst] is similarly
constrained.

Seems similar in spirit to the
problem that Tcl_EvalTokens()
had.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=360894&aid=1553630&group_id=10894