Issue 5952 in pharo: UnicodeSpec is 25.000 lines of Code...

Status: Workneeded
Owner: [hidden email]
CC: [hidden email]
Labels: Type-Cleanup Milestone-2.0

New issue 5952 by [hidden email]: UnicodeSpec is 25.000 lines of  
Code...
http://code.google.com/p/pharo/issues/detail?id=5952

... we need a way to store data on disk.

UnicodeSpec linesOfCode 25719

the class takes up nearly 1 MB in space in the image... and that with the  
code itself living in the .changes.
(string literals are in the image).





_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #1 on issue 5952 by [hidden email]: UnicodeSpec is  
25.000 lines of Code...
http://code.google.com/p/pharo/issues/detail?id=5952

Yeap, this code should be somewhere, not just removed and forgotten.  Now,  
we have to achieve to take it outside the image in a smart way :).


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #2 on issue 5952 by [hidden email]: UnicodeSpec is 25.000 lines  
of Code...
http://code.google.com/p/pharo/issues/detail?id=5952

You don't need to keep 2 copies of same data in image
- first in form of literals
- second in form of initialized data

See SqS/NBOpenGL-Core package,
look at
NBGLConstants class>>data ,
NBGLConstants class>>initialize
for example how it can be done.

Second, you can also avoid having this data unless someone using it
- just make stuff be lazily initialized, and also add it to a cleanup  
method to
wipe this data out of image.

Like that we don't need to throw out anything, we have functionality,
but we save memory in cases if nobody using it.


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #3 on issue 5952 by marianopeck: UnicodeSpec is 25.000 lines of  
Code...
http://code.google.com/p/pharo/issues/detail?id=5952

What I would do personally, Is to remove the class UnicodeSpec. Previously  
there was a dependecy with Zinc and the table was download from internet.
This class side #initialize is only needed when bootstaping right? in each  
other case would I need to change the table? even if the class changes and  
Monticello sends #initialize, you don't want to set the table again.
So...what about puttings this initialize in the "boostraper" code?  
so...those who care about being able to initialize, then they do it. anway,  
apart from that, what about storing the table in an external file?  for me  
it makes complete sense to have a folder of "resources" when boostrapping.  
So such table can be there. As a plain txt that you read, or if you want  
using DataStream (if you don't want to depend on FUel).


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #4 on issue 5952 by [hidden email]: UnicodeSpec is 25.000 lines  
of Code...
http://code.google.com/p/pharo/issues/detail?id=5952

we already having an external files: .sources , .changes :)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Comment #5 on issue 5952 by [hidden email]: UnicodeSpec is  
25.000 lines of Code...
http://code.google.com/p/pharo/issues/detail?id=5952

Marcus, after including this slice, the package  
Multilingual-UnicodeInitialization can be safely removed.
I've learned the lesson and put the initialization stuff in the bootstrap  
code :)

Name: SLICE-Issue-5952-UnicodeSpec-is-25000-lines-of-Code-GuillermoPolito.1
Author: GuillermoPolito
Time: 29 June 2012, 6:02:18.729 pm
UUID: 703409c8-0c7f-4907-ab2a-b97cf4510854
Ancestors:
Dependencies: Multilingual-Encodings-GuillermoPolito.30

Better comments to Unicode initialization


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: FixReviewNeeded

Comment #6 on issue 5952 by [hidden email]: UnicodeSpec is  
25.000 lines of Code...
http://code.google.com/p/pharo/issues/detail?id=5952

(No comment was entered for this change.)


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: MonkeyIsChecking

Comment #7 on issue 5952 by [hidden email]: UnicodeSpec is 25.000  
lines of Code...
http://code.google.com/p/pharo/issues/detail?id=5952#c7

The Monkey is currently checking this issue. Please don't change it!


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: FixToInclude

Comment #8 on issue 5952 by [hidden email]: UnicodeSpec is 25.000  
lines of Code...
http://code.google.com/p/pharo/issues/detail?id=5952#c8

This Issue has been checked by Ulysse the Monkey
       
5163 tests passed in 00:00:13s:
===============================
        CollectionsTests-Arrayed (562)
        CollectionsTests-Atomic (12)
        CollectionsTests-Sequenceable (907)
        CollectionsTests-SplitJoin (27)
        CollectionsTests-Stack (16)
        CollectionsTests-Streams (37)
        CollectionsTests-Strings (592)
        CollectionsTests-Support (12)
        CollectionsTests-Text (45)
        CollectionsTests-Unordered (1951)
        CollectionsTests-Weak (739)
        CompilerTests (179)
        SUnit-Core-Extensions (3)
        SUnit-Core-Utilities (3)
        SUnit-Tests-Core (78)

----------------------------------------------------------
Loaded Source:  
SLICE-Issue-5952-UnicodeSpec-is-25000-lines-of-Code-GuillermoPolito.1 from  
http://ss3.gemstone.com/ss/PharoInbox
Tested using Pharo-2.0-20172-a on CoInterpreter  
VMMaker-oscog-EstebanLorenzano.160 uuid:  
bec8cdf0-4e06-4975-8c02-e882fadf4df3 Jun 22 2012,  
StackToRegisterMappingCogit VMMaker-oscog-EstebanLorenzano.160 uuid:  
bec8cdf0-4e06-4975-8c02-e882fadf4df3 Jun 22 2012,  
https://git.gitorious.org/cogvm/blessed.git Commit:  
744bfe905c78a1a5d408680a8780367ea77e0549 Date: Fri Jun 1 15:17:41 2012  
+0200 By: Esteban Lorenzano <[hidden email]>


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker

Updates:
        Status: Integrated

Comment #9 on issue 5952 by [hidden email]: UnicodeSpec is 25.000  
lines of Code...
http://code.google.com/p/pharo/issues/detail?id=5952

in 2.0 174


_______________________________________________
Pharo-bugtracker mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-bugtracker