objectscript-devel Mailing List for ObjectScript (Page 32)
Brought to you by:
rob_d_clark
Menu
▾
▴
objectscript-devel — devel related msgs, including mail from scarab issue tracking system
You can subscribe to this list here.
| 2003 |
Jan
(47) |
Feb
(61) |
Mar
(73) |
Apr
(149) |
May
(202) |
Jun
(219) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(13) |
Dec
(65) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2004 |
Jan
(81) |
Feb
(62) |
Mar
(36) |
Apr
(48) |
May
(50) |
Jun
(9) |
Jul
(10) |
Aug
(9) |
Sep
(30) |
Oct
(6) |
Nov
|
Dec
|
| 2006 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
(2) |
Jun
(1) |
Jul
|
Aug
(1) |
Sep
(6) |
Oct
(4) |
Nov
(1) |
Dec
(3) |
| 2007 |
Jan
|
Feb
(5) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(10) |
Sep
|
Oct
(1) |
Nov
|
Dec
|
| 2008 |
Jan
(2) |
Feb
|
Mar
(1) |
Apr
(1) |
May
(2) |
Jun
(4) |
Jul
(14) |
Aug
(1) |
Sep
(10) |
Oct
(7) |
Nov
(47) |
Dec
(38) |
|
From: Rob C. <rob...@ma...> - 2003-05-16 18:31:30
|
Issue OBJS126 (interrupt() a console thread) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS126>
The following modifications were made to this issue:
A user has mentioned that they would like this functionality to be bound to ctrl-c... try adding a KeyListener to the console component to intercept the ctrl-c. I am not sure if it will be possible to prevent the console key listener itself from also receiving ctrl-c event as well
|
|
From: Rob C. <rob...@ma...> - 2003-05-16 17:44:40
|
Issue OBJS126 (interrupt() a console thread) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS126>
The following modifications were made to this issue:
Status changed from 'New' to 'Assigned'
Comment: move to 'assigned' state
|
|
From: Rob C. <rob...@ma...> - 2003-05-16 17:44:23
|
User rob...@ma... has added user rob...@ma... to Assigned To. You can view the issue detail at the following URL: <http://icandy.homeunix.org:443/scarab/issues/id/OBJS126> |
|
From: Rob C. <rob...@ma...> - 2003-05-16 17:44:05
|
You can view the issue detail at the following URL: <http://icandy.homeunix.org:443/scarab/issues/id/OBJS126> Type : Enhancement Issue Id : OBJS126 Reported by: Rob Clark rob...@ma... - (rob...@ma...) Details: Headline: interrupt() a console thread Description: add a way to interrupt() the thread created to read-eval-print for a given console. This way the users have some way to recover if something hangs. Severity: Minor Functional Area: chimera Status: New |
|
From: Brad H. <bh...@ti...> - 2003-05-14 16:08:37
|
cool by me Rob Clark wrote: > Issue OBJS120 (enhanced for loop) has just been modified by user rob...@ma... > > You can view the issue detail at the following URL: > <http://icandy.homeunix.org:443/scarab/issues/id/OBJS120> > > The following modifications were made to this issue: > > note... can't use "in" as a keyword, since it is too likely to already be used as a variable name, so perhaps: > > for( var foo : collection ) > > would be ok?? this is similar to the java syntax proposed for v1.5 > > > ------------------------------------------------------- > Enterprise Linux Forum Conference & Expo, June 4-6, 2003, Santa Clara > The only event dedicated to issues related to Linux enterprise solutions > www.enterpriselinuxforum.com > > _______________________________________________ > Objectscript-devel mailing list > Obj...@li... > https://lists.sourceforge.net/lists/listinfo/objectscript-devel |
|
From: Rob C. <rob...@ma...> - 2003-05-14 16:01:47
|
Issue OBJS122 (editor problem, multiple function tree tabs...) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS122>
The following modifications were made to this issue:
Status changed from 'Assigned' to 'Resolved'
Comment: move to 'resolved' state
|
|
From: Rob C. <rob...@ma...> - 2003-05-14 14:45:15
|
Issue OBJS123 (create "pkg" in base.os) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS123>
The following modifications were made to this issue:
Status changed from 'Assigned' to 'Resolved'
Comment: move to 'resolved' state
|
|
From: Rob C. <rob...@ma...> - 2003-05-14 14:30:41
|
Issue OBJS25 (PopupTrigger should intercept key & mouse events) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS25>
The following modifications were made to this issue:
Status changed from 'Assigned' to 'Resolved'
Comment: move to 'resolved' state
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 08:24:46
|
Issue OBJS95 (when .jar files are not in current dir) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS95>
The following modifications were made to this issue:
Status changed from 'New' to 'Closed'
Comment: duplicate of OBJS105, since it changed how classes are loaded from cache.jar, by creating our own ClassLoader, we don't need cache.jar in the classpath, so it can be created in the current directory, user's home directory, whatever
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 08:23:03
|
Issue OBJS124 (performance: Symbol type) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS124>
The following modifications were made to this issue:
note, see OBJS39
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 08:22:36
|
Issue OBJS39 (closed scope optimizations) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS39>
The following modifications were made to this issue:
on approach would be to cache the Reference that is statically resolved as an instance member of the NodeEvaluator, rather than as a local variable of the method.
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 08:19:51
|
Issue OBJS124 (performance: Symbol type) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS124>
The following modifications were made to this issue:
... but another alternative to createMemberBySlot(), and getMemberBySlot() is more agressive caching... when referencing a member of a scope that contains the NodeEvaluator, we can take advantage of the face that Scope uses a table of Reference which never changes, we can cache looked up values (and perhaps hoisted values?) as instance variables, rather than local variables.
I should probably split Scope out of CompilerVisitor, to reduce the size of the main part of the compiler, and also so there can be several implementations of Scope which generate the different sort of code to cache in either the NodeEvaluator instance itself, or as local variables
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 08:03:11
|
Issue OBJS20 (stack trace line # problem) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS20>
The following modifications were made to this issue:
hmm, that was *supposed* to say 'include', not 'intrude'
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 08:02:25
|
Issue OBJS20 (stack trace line # problem) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS20>
The following modifications were made to this issue:
intrude a Stack (ScriptContext) ? object, with setScope(), setLineNumber() methods. Use hashtable to map current thread to ScriptContext (which can possibly also be passed on stack to avoid hashtable lookup, via varients on evalNode(), callAsFunction(), etc.) By using setScope() and setLineNumber(), as separate methods, we can reduce the instruction count for the more frenquent act of setting the line number, by not having to push the scope onto the stack (and store it in setLineNumber())
One idea, that could improve on performance is, if there is no breakpoint in the current NodeEvaluator (?) use a ScriptContext implementation that does not call the debugger. This would minimize the overhead of the debugging support when debugging is not enabled.
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:48:31
|
Issue OBJS106 (ObjectScript performance) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS106>
The following modifications were made to this issue:
Added 'blocking' child dependency on issue OBJS39
Added 'blocking' parent dependency on issue OBJS106
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:48:31
|
Issue OBJS39 (closed scope optimizations) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS39>
The following modifications were made to this issue:
Added 'blocking' child dependency on issue OBJS39
Added 'blocking' parent dependency on issue OBJS106
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:46:51
|
Issue OBJS106 (ObjectScript performance) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS106>
The following modifications were made to this issue:
Added 'blocking' child dependency on issue OBJS125
Added 'blocking' parent dependency on issue OBJS106
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:46:51
|
Issue OBJS125 (performance: hoisting members) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS125>
The following modifications were made to this issue:
Added 'blocking' child dependency on issue OBJS125
Added 'blocking' parent dependency on issue OBJS106
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:46:25
|
You can view the issue detail at the following URL: <http://icandy.homeunix.org:443/scarab/issues/id/OBJS125> Type : Enhancement Issue Id : OBJS125 Reported by: Rob Clark rob...@ma... - (rob...@ma...) Details: Headline: performance: hoisting members Description: if there was a way to indicate to the compiler that an object's getMember(symbol) will return the same result for the same symbol parameter, the the compiler could cache the result in order to avoid the overhead of repeated getMember() calls... Severity: Major Functional Area: compiler Status: New |
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:37:23
|
Issue OBJS124 (performance: Symbol type) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS124>
The following modifications were made to this issue:
if each symbol has a unique-id, then we can use and int<->int hashtable to map member symbol to slot.
Or, another interesting possibility is for the compiler to generate (perhaps as inner classes) classes implementing a MemberTable (?) interface (see OBJS116), which could use switch statements instead of actual tables... need to investigate the performance of this vs. hashtables, and verify that the overhead of the extra classes created isn't too bad. Also, with switch statements, you don't need to create members in the hashtable, so the compiled code can createMemberBySlot() and getMemberBySlot(), without touching the table at all (because the compiled code already knows the contents of the table... which is just there for the benefit of resolving variables which can't be statically resolved.
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:25:29
|
Issue OBJS106 (ObjectScript performance) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS106>
The following modifications were made to this issue:
Added 'blocking' child dependency on issue OBJS124
Added 'blocking' parent dependency on issue OBJS106
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:25:20
|
Issue OBJS124 (performance: Symbol type) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS124>
The following modifications were made to this issue:
Added 'blocking' child dependency on issue OBJS124
Added 'blocking' parent dependency on issue OBJS106
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 07:23:10
|
You can view the issue detail at the following URL: <http://icandy.homeunix.org:443/scarab/issues/id/OBJS124> Type : Enhancement Issue Id : OBJS124 Reported by: Rob Clark rob...@ma... - (rob...@ma...) Details: Headline: performance: Symbol type Description: perhaps it would be good to introduct a Symbol type... which could be a lightweight wrapper for an OString (or a String), with the added characteristic that each Symbol has a unique integer id. See comments for more... Severity: Major Functional Area: data Status: New |
|
From: Rob C. <rob...@ma...> - 2003-05-13 01:25:26
|
Issue OBJS38 (scope tweak) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS38>
The following modifications were made to this issue:
added file 'check-scope.os' at mod100/0/OBJS38_809_check-scope.os
|
|
From: Rob C. <rob...@ma...> - 2003-05-13 01:13:07
|
Issue OBJS120 (enhanced for loop) has just been modified by user rob...@ma...
You can view the issue detail at the following URL:
<http://icandy.homeunix.org:443/scarab/issues/id/OBJS120>
The following modifications were made to this issue:
note... can't use "in" as a keyword, since it is too likely to already be used as a variable name, so perhaps:
for( var foo : collection )
would be ok?? this is similar to the java syntax proposed for v1.5
|
120 messages has been excluded from this view by a project administrator.
