r3wp [groups: 83 posts: 189283]
  • Home
  • Script library
  • AltME Archive
  • Mailing list
  • Articles Index
  • Site search
 

World: r3wp

[Script Library] REBOL.org: Script library and Mailing list archive

Volker
24-Aug-2006
[338x3]
but i try.
Tried download again, there is a bug:
|| Downloading:   qml-ed.r/RS-NE+.gif
|| REQUEST FAILED...
|| REBOL.ORG says file not in package
seems the "+" is a problem? (Overlokked that message the first time)
Sunanda
24-Aug-2006
[341x2]
You replied as I was typing....
I think you may be right.......

Looks like the downloader has failed to find the files that contain 
a + in the file names.
Must be -- + in the HTTP spec is escaped to a space at times. So 
i suspecet we're creating or receiving badly escaped URLs in this 
case.
Volker
24-Aug-2006
[343x2]
that could be it.
but a "+" in a filename looks weird anyway :)
Sunanda
24-Aug-2006
[345]
That's true -- and it may recreate some cross-platform issues on 
platforms that don't allow + in a file name.
But we should be able to deal with it.
Thanks for the help in isolating the problem.
Volker
24-Aug-2006
[346]
A bigger errormessage would be good, and some hint on upload.
Sunanda
24-Aug-2006
[347x2]
The upload obviously worked -- "+" is not an illegal file name character 
on either Edgars machine or REBOL.org's current host

The notes on contributing a packgae do have detailed guidelines on 
file naming to avoind problems -- see 2.1 here:

http://www.rebol.org/cgi-bin/cgiwrap/rebol/boiler.r?display=cons-upload-package-files-help.html

But, of course, the author probably wrote the application before 
they had any idea it might be rolled out as a package for use on 
a wide range of platforms.
***

The bottom line is that + ought to work in a file name for the downloader, 
so I'll get it fixed.
Volker's problem fixed: both repack and repack-core not correctly 
handle files with "+" in their names.


DideC's problem remains: could you check the version of REBOL you 
are using and let me know? Thanks,
DideC
25-Aug-2006
[349]
View 1.3.2.3.1 on WinXP.
%user.r only contains set-net line.
%rebol.r contains nothing but the header.


I have tried it too from the WWR ( rebol.com / Public / library / 
scripts / type / package / calendar-install.r  cause qml-ed is not 
there for the moment).

Just click the [ ... ] button to choose a different folder, then 
clicked [ OK ] and it quits !!
Pekr
25-Aug-2006
[350]
hehe, Pierre Johnson (whom we excluded from ml), replies in Class 
R3 blog :-)
Gabriele
25-Aug-2006
[351x3]
the image and css files where not done by me, and although i had 
some suspects while uploading them to rebol.org i didn't want to 
change everything just for that :) i can add a zip file to the package 
if you need me to.
the images can also be downloaded from qtask.com/images/qwiki/
note that the script still works without them. you'll just miss them 
when viewing the html result in the browser if you used anything 
that needs them (rounded boxes for e.g.)
Sunanda
25-Aug-2006
[354]
DideC -- thanks......I see the problem with 1.3.2 too.  Let me get 
back to you.
***
qml-ed not there
 -- No reason it should have vanished. That's strange.


Meanwhile, repack-core.r should do the job of downloading it for 
for.
DideC
25-Aug-2006
[355]
Don't bother, I have loaded qml-ed already, wihtout changing the 
dir, and moving files manually after.

(I had the same error  than volker, but solved it by unchecking the 
images files in repack and downloaded only "template.html")

qml-ed not there
 -- How often is the Rebol index files rebuilded ?
Sunanda
25-Aug-2006
[356]
Gald you got it downloaded in the end.
The indexes are close to real time.

It is possible for the owner of a package like qmr-ed to make it 
unavailable while they are updating it. maybe that happened this 
time.
Gabriele
25-Aug-2006
[357x2]
i didn't change the availability of the package.
btw, if you have suggestions, please let me know. i'll be working 
a bit more on it on monday.
Maxim
14-Sep-2006
[359x5]
I am in the process of updating repack.
so far glayout (much better version) is able to open the same core 
without issues.  am doing download tests to be sure the code itself 
still works... hoping nothing changed within the lds service library...
sunanda I am getting an error from the server itself!  


whenever I press download, it starts the download (shows the green 
xfer data on the bottom, and then pops up an error:  here is the 
error as reported by LDS:

make object! [
    status: [100 20 "bad cgi data"]
    data: none
    diagnostics: [9003 9003 9.8084E-2]
    supply-date: 15-Sep-2006/1:03:41
]


have any idea where that is comming from?  I try to check it out 
from the code I have... but its been sooo long, A nudge on your part 
will help me get this working much faster IMHO!
well, I did all I can, for some reason, the same code under 1.2.10 
and 1.3.2 does not seem to request the same data on the server (I 
guess).
I am investigating further...
Maxim
15-Sep-2006
[364]
pow!  all fixed  :-)  just need to let slim  build the new release 
from all the included files and we are set !  I will also be doing 
just a little bit of cleanup within GLayout to reduce the size of 
the final single-file app...
Sunanda
15-Sep-2006
[365]
Thanks Maxim,

I'll try to get the new version up and live in a few days -- I'm 
pretty busy myself right now.
Maxim
20-Sep-2006
[366x4]
everyone, its all working now... sorry if we forgot to notify the 
list...
new version uses slick new version of GLayout which has gfx partly 
based on Henrik tests which he supplied a few weeks ago...
as usuall, type the following in a rebol console to open up the tool.

do http://www.rebol.org/library/public/repack.r


also note that I put a lot of effort in optimizing the file's final 
size, including a lot of hacking out, removing comments, compressing, 
etc... its now a much smaller download.
We are glad to announce that a newer version of the rebol.org package 
downloader is now available for people using REBOL|view 1.3.2

as usuall, type the following in a rebol console to open up the tool. 
it now adapts to version automatically
<pre>do http://www.rebol.org/library/public/repack.r</pre>

also note that I put a lot of effort in optimizing the file's final 
size, including a lot of hacking out, removing comments, compressing, 
etc... its now a much smaller download.
This version uses a slick 
new version of GLayout which has gfx largely based on Henrik's tests 
which he supplied a few weeks ago...
Gregg
21-Sep-2006
[370]
Thanks for doing that Maxim!
Maxim
21-Sep-2006
[371]
my pleasure... only strange that my last post ended up  here... it 
was a copy/paste of info in this group meant for another group.. 
hum...
Graham
21-Sep-2006
[372]
Glayout is LGPL.   What does that mean for any script using Glayout??
Maxim
21-Sep-2006
[373x2]
that you can use it without your tool being confined to GPL.  any 
changes to bring to GLayout must be made available on demand.
unless you include GLayout *within* your script, in which case, your 
app is now GPL.  but I'll be happy to give people non LGPL versions 
if they really need it  (if they ask nicely  ;-)
Graham
21-Sep-2006
[375]
So, if you are happy to give people non LGPL .. why not release as 
BSD or something else?
Maxim
21-Sep-2006
[376x3]
cause I'm a control freak  ;-)
I have been thinking about it.
(aren't we in the wrong group?  ;-)
Graham
21-Sep-2006
[379x2]
Basically my experience is that anything that has the word GPL in 
it does not get used.
If you want to promote your glayout .. you need to change the license 
.  See rebgui's license.
Maxim
21-Sep-2006
[381x2]
unless it also has the letters mysql  ;-)
LGPL does not impede useage by forcing your apps to become open source.
Graham
21-Sep-2006
[383]
I'm not sure people know the difference between the lesser GPL and 
GPL.
Henrik
22-Sep-2006
[384]
http://www.rebol.net/article/0243.html<--- some interesting points 
regarding GPL here.
Robert
22-Sep-2006
[385]
downloader: Download DIR should be create if it doesn't exist.
Maxim
22-Sep-2006
[386x2]
talking about newer repack.r?
the file browser allows you to create dirs... although I admit auto-creation 
of typed paths in the field is a good idea (maybe with a little req 
asking for confirmation)