Menu
▾
▴
narf-lib-devel — Developers and Users of NARF
You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(3) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
(28) |
Feb
(8) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(2) |
Sep
|
Oct
(1) |
Nov
(58) |
Dec
(9) |
| 2004 |
Jan
(2) |
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
(2) |
Jul
|
Aug
|
Sep
|
Oct
(2) |
Nov
|
Dec
(2) |
| 2005 |
Jan
|
Feb
|
Mar
(2) |
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Eutimio E. <Tho...@ke...> - 2005-04-07 12:55:51
|
Hello, father. pimp sent out by the Secretary of State to offer the Captain the round to face his lordship with that question. Later he was to He rose, tall, active, and arrestingly elegant in a scarlet, the seaward side. that she might escape the insufferable heat of Jamaica in that of a natural petulance aggravated by the dulness of life in Barba delivered it as truculently as the Admiral could have desired. Good-day to you, sir, Blood hailed him pleasantly. I have the Have a nice day. |
|
From: Patrick M. <pa...@he...> - 2005-03-17 05:20:12
|
Errors in browser:
http://www.narf-lib.org/cgi-bin/error.rb
I will be out of commission for the next month, but things should be
heating up towards the end of April.
Cheers,
Patrick
|
|
From: Nkiruka B. <Xa...@ja...> - 2005-03-15 12:08:09
|
Hello, opened the sluices of his rage. And M. de Cussy himself invited You said you vould show us zome vine dings. Vhere are dese vine him with a question: If you will deliver the message, it may enable me to judge. his compatriot: ourselves; we must depend entirely upon our small arms, and how Blood. I beg your pardon for having misunderstood you. and by his methods of bullying and coercing the jury into bloodsh Peter Blood stiffened, and his vivid blue eyes stabbed the bloate her anger gave him concern. It could not have been so yesterday. Have a good day. |
|
From: Adam W. <ad...@ta...> - 2004-12-24 04:10:43
|
At 10:28 PM -0500 12/23/04, Patrick May wrote: >What do you think of this rough roadmap? Easy to install, easy to use. It sounds grand to me. I'd add easy to learn. >p.s. I locked down the narf-lib.org wiki b/c the only edits were >spammers and spam restores. Good idea. -- Adam Wildavsky Extreme Programmer Tameware, LLC ad...@ta... http://www.tameware.com |
|
From: Patrick M. <pa...@he...> - 2004-12-24 03:29:06
|
Hello, Narf now has lots of competition. We aren't one of the few kids on the ruby web library block anymore. I'd like to make Narf relevant. Relevant, like how say php is relevant. As in, php can win at my job, and Narf can't. So that sucks. I was thinking of this roadmap: + When I imported WebUnit, I forked it into Web::Unit inside narf. Time to undo this fork, so that it is easier to integrate code. + Undo the fork, then re-write Narf's cgi testing to overlap with WebUnit. This way, some of the useful methods like #assert_form_includes will be implemented in a way that can be merged back into WebUnit. + One of the happy by products of forking Web::Unit was the discovery that the WebUnit test suite can be used to "validate" Narf running on a particular server configuration. I'd like to re-write these tests, so that Narf depends on a test script instead of a site. + There's a weird quality of templating libraries. First programmers throw away the functionality of a programming language, muttering about separating "Display Logic" and "Code". Then, as they discover that "Display Logic" could use some of the nice features of a programming language, they start implementing a wierd half language on the side. Since Ruby is perfect for everything, I'd like to make Narflates more like eruby+. Ruby code in your narflates, in addition to the nice form stuff like <narf:input ....> etc. Oh yeah, and the different tags will get a Ruby api for creating them, so you can do the in code generation of forms too. And finally, we should support <? ?> in addition to <% %> because the % is a pain in the @$$ to type. this is all rolling up to a drum roll.... + I think the reason php is winning is b/c it is easy to install, and easy to write. I like code in html environments that are easy to install on every web server. So once narflates can contain ruby code, I want to rebuild narf as eruby++. Easy installation, ala php. And yea, cgi scripts, IOWA, Rails, and everything else will still run in narflates. Because we will be a sneaky bastards. What do you think of this rough roadmap? Cheers, Patrick p.s. I locked down the narf-lib.org wiki b/c the only edits were spammers and spam restores. seemed like a waste of time to me. Email me if you want to edit the pages, and I'll send you the link / password. |
|
From: Patrick M. <pa...@he...> - 2004-10-09 07:36:59
|
Hello, On Friday, October 8, 2004, at 08:41 PM, Brian Wisti wrote: > I'm just poking my way through the main Web Development libraries. I > hit on Narf first because it looked like it required the least > commitment from me to get it installed :-) > > If I have any brilliant questions, I'll be posting again. Let me know if you need help with anything. Cheers, Patrick |
|
From: Brian W. <bri...@ya...> - 2004-10-09 00:41:47
|
Hi all, I'm just poking my way through the main Web Development libraries. I hit on Narf first because it looked like it required the least commitment from me to get it installed :-) If I have any brilliant questions, I'll be posting again. Kind Regards, Brian Wisti http://coolnamehere.com/ |
|
From: Patrick M. <pa...@he...> - 2004-06-27 23:01:17
|
Hello,
I'm working on the documentation for NARF. I thought that if I pruned
down the docs, and focused on a simple "published" api, that it would
be easier for people to download NARF and make use of it.
I'm trying to focus on the things that I don't think will change in
NARF.
You can see the stuff I'm doing in the svn repository, or you can look
online at:
http://www.narf-lib.org/doc/
I'm still working on this, so online might be a few hours behind me.
Cheers,
Patrick
|
|
From: May, P. <Pat...@Th...> - 2004-06-21 14:36:36
|
Hello
The latest releases of narf integrate with htmlarea for cross-platform rich
text editing, but there still are lovely browser tweaks to hash out with the
new narf. Also, I'd like to inform you of another important change. I
changed one important pref, the default location of the page directory. You
can restore the old behaviour with:
Web::Wiki::set_pref( "store_dir" => "../pages/" )
you can set multiple prefs with this, ie
Web::Wiki::set_pref( "store_dir" => "../pages/",
"resourceurl" => "/resources/" )
The prefs need to be organized a bit, but we at least are evolving a config
mechanism for the wiki. Maybe I'll add a method to load the pref's from a
yaml file.... :-)
~ Patrick
"This communication is intended solely for the addressee and is
confidential and not for third party unauthorized distribution."
|
|
From: May, P. <Pat...@Th...> - 2004-05-26 21:54:31
|
Hello,
Sorry for the delay.
> Recently ran into some free time so I decided to check the status of
> narf again. However I am having trouble with the check out.
Thanks for checking things out! Narf's subversion server melted (magic
smoke was spotted). The new url is:
http://svn.narf-lib.org/svn/narf/trunk/
everything you were doing looked correct, your co command would just be:
svn co http://svn.narf-lib.org/svn/narf/trunk/ narf
Cheers,
Patrick
>
> I'm a total subversion newbie so I would like to check that everything
> is up and running before I start digging in too deep. I get the
> following output when attempting to check out the latest narf.
>
> emiel@beastie:~> svn --version
> svn, version 0.32.1 (r7497)
> compiled Jan 15 2004, 13:05:29
>
> ...
>
> emiel@beastie:~> svn co http://os.u2i.com/svn/narf/trunk/ narf
> svn: No such file or directory
> svn: PROPFIND request failed on '/svn/narf/trunk'
> svn:
> Could not open the requested SVN filesystem
>
> Cheers,
>
> Emiel
> --
> E F van de Laar
> www.il.fontys.nl/~emiel
> PGP pubkey: %finger em...@il...
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Perforce Software.
> Perforce is the Fast Software Configuration Management System offering
> advanced branching capabilities and atomic changes on 50+ platforms.
> Free Eval! http://www.perforce.com/perforce/loadprog.html
> _______________________________________________
> Narf-lib-devel mailing list
> Nar...@li...
> https://lists.sourceforge.net/lists/listinfo/narf-lib-devel
>
"This communication is intended solely for the addressee and is
confidential and not for third party unauthorized distribution."
|
|
From: Tom C. <to...@u2...> - 2004-01-15 16:40:41
|
Your timing is pretty bad :-) You picked the day I was migrating the subversion repository to the new version. Anyway, you should have no problems now, -Tom > Hello all, > > Recently ran into some free time so I decided to check the status of > narf again. However I am having trouble with the check out. > > I'm a total subversion newbie so I would like to check that everything > is up and running before I start digging in too deep. I get the > following output when attempting to check out the latest narf. > > emiel@beastie:~> svn --version > svn, version 0.32.1 (r7497) > compiled Jan 15 2004, 13:05:29 > > ... > > emiel@beastie:~> svn co http://os.u2i.com/svn/narf/trunk/ narf > svn: No such file or directory > svn: PROPFIND request failed on '/svn/narf/trunk' > svn: > Could not open the requested SVN filesystem > > Cheers, > > Emiel > -- > E F van de Laar > www.il.fontys.nl/~emiel > PGP pubkey: %finger em...@il... > > > ------------------------------------------------------- > This SF.net email is sponsored by: Perforce Software. > Perforce is the Fast Software Configuration Management System offering > advanced branching capabilities and atomic changes on 50+ platforms. > Free Eval! http://www.perforce.com/perforce/loadprog.html > _______________________________________________ > Narf-lib-devel mailing list > Nar...@li... > https://lists.sourceforge.net/lists/listinfo/narf-lib-devel > > > |
|
From: E F v. de L. <em...@il...> - 2004-01-15 12:43:56
|
Hello all, Recently ran into some free time so I decided to check the status of narf again. However I am having trouble with the check out. I'm a total subversion newbie so I would like to check that everything is up and running before I start digging in too deep. I get the following output when attempting to check out the latest narf. emiel@beastie:~> svn --version svn, version 0.32.1 (r7497) compiled Jan 15 2004, 13:05:29 ... emiel@beastie:~> svn co http://os.u2i.com/svn/narf/trunk/ narf svn: No such file or directory svn: PROPFIND request failed on '/svn/narf/trunk' svn: Could not open the requested SVN filesystem Cheers, Emiel -- E F van de Laar www.il.fontys.nl/~emiel PGP pubkey: %finger em...@il... |
|
From: Patrick M. <pa...@he...> - 2003-12-19 02:53:57
|
Hello,
Big news for the day -- I've forked webunit:
http://www.xpenguin.biz/download/webunit/
First things first. Lots of thanks are due to the authors of webunit:
yuichi TAKAHASHI
Kouhei Sutou
Narushima Hironori
Torsten Rueger
Kazuhiro NISHIYAMA
IKAGAWA, Kauztomo
Takahiro Maebashi
MoonWolf
I am now integrating webunit into narf. I'm trying to keep the owner
of webunit, yuichi TAKAHASHI, in the loop. I sent an email to him,
but haven't heard back from him yet. I certainly don't speak Japanese,
so there is probably a communication problem.
I noticed that the tests of webunit don't currently run, so I made the
tests work with the narf server, pulled the code down and dumped it
into the narf tree. I'll be migrating the WebUnit constant to
Web::Unit, and then I will work on integrating Web::Unit's form
handling and html testing methods with Narf. Fun fun!
If webunit turns out to be an active project I have no problem with
pulling it back out of narf. For now, if I'm going to maintain this
code I'm going to put it in my tree.
Anyways. I'm excited to start playing with the code! My goals are to
continue to support the use of webunit as standalone component, while
integrating its features with Narf's web testing apis. Something fun
should come out of here :-)
~ Patrick
|
|
From: William L. A. <ba...@ac...> - 2003-12-10 14:18:27
|
Patrick: I thought so, but I'm not sure how to find it. I did read through some and actually count the tags and couldn't find any missing. But it's easy to miss when counting manually. I'll be away 'til the 18th so won't be able to do much. I did start to simplify the descriptions in the cookbook.rdoc. Bill Patrick May wrote: > Bill, > > I get this error too. I think it's a problem in one of the *.html > templates in > > lib/web/wiki/resources/ > > This should be an easy one to fix, once we find it. > > ~ Patrick > > > On Thursday, December 4, 2003, at 10:48 PM, William L. Anderson wrote: > >> Patrick: here's the only error I get on OS X and Windows (cygwin): >> >> $ ruby build.rb testwiki >> Target: testwiki | test narf wiki >> Loaded suite Unnamed TestSuite >> Started >> ...........................................warning: extra end tag </a> >> . >> Finished in 23.90606 seconds. >> >> I've tried to look around for the extra tag, but really have no idea >> how to discover which test the warning arises from. It looks like the >> next to last, but I don't know how to find it. >> >> Hints appreciated. >> >> Bill >> <mime-attachment> > > > > > ------------------------------------------------------- > This SF.net email is sponsored by: SF.net Giveback Program. > Does SourceForge.net help you be more productive? Does it > help you create better code? SHARE THE LOVE, and help us help > YOU! Click Here: http://sourceforge.net/donate/ > _______________________________________________ > Narf-lib-devel mailing list > Nar...@li... > https://lists.sourceforge.net/lists/listinfo/narf-lib-devel > |
|
From: Patrick M. <pa...@he...> - 2003-12-10 14:11:31
|
Bill,
I get this error too. I think it's a problem in one of the *.html
templates in
lib/web/wiki/resources/
This should be an easy one to fix, once we find it.
~ Patrick
On Thursday, December 4, 2003, at 10:48 PM, William L. Anderson wrote:
> Patrick: here's the only error I get on OS X and Windows (cygwin):
>
> $ ruby build.rb testwiki
> Target: testwiki | test narf wiki
> Loaded suite Unnamed TestSuite
> Started
> ...........................................warning: extra end tag </a>
> .
> Finished in 23.90606 seconds.
>
> I've tried to look around for the extra tag, but really have no idea
> how to discover which test the warning arises from. It looks like the
> next to last, but I don't know how to find it.
>
> Hints appreciated.
>
> Bill
> <mime-attachment>
|
|
From: William L. A. <ba...@ac...> - 2003-12-05 03:49:19
|
Patrick: here's the only error I get on OS X and Windows (cygwin): $ ruby build.rb testwiki Target: testwiki | test narf wiki Loaded suite Unnamed TestSuite Started ...........................................warning: extra end tag </a> . Finished in 23.90606 seconds. I've tried to look around for the extra tag, but really have no idea how to discover which test the warning arises from. It looks like the next to last, but I don't know how to find it. Hints appreciated. Bill |
|
From: William L. A. <ba...@ac...> - 2003-12-04 21:34:10
|
Patrick: I agree completely with getting rid of the warnings.
It's a pain in the butt on Windows; that OS really needs a "sudo"
command.
I commend your effort in making it really work on Windows. I don't
have a pure windows system here to help.
Bill
On Thu, Dec 04, 2003 at 10:07:31AM -0500, May, Patrick wrote:
> Bill,
>=20
> > Patrick: on my cygwin seup Narf continues to be free of errors. I=20
> > continue to filter the directory permission warnings from the Ruby=20
> > system() command using ruby -W0
>=20
> I'm going to try and fix all of these. I think most of these system calls
> are like this:
>=20
> system("rm -Rf testdir")
>=20
> Another problem besides the permission warnings is that pure windows
> installs won't be able to test ruby. I realized that I still had cygwin,
> and thus rm, on my machine. These aren't big things to test, but I don't
> want someone to download narf to check it out and be scared away by a bun=
ch
> of failures that could have been easily fixed.
>=20
> Cheers,
>=20
> Patrick
>=20
>=20
> "This communication is intended solely for the addressee and is
> confidential and not for third party unauthorized distribution."
>=20
|
|
From: May, P. <Pat...@Th...> - 2003-12-04 15:07:45
|
Bill,
> Patrick: on my cygwin seup Narf continues to be free of errors. I
> continue to filter the directory permission warnings from the Ruby
> system() command using ruby -W0
I'm going to try and fix all of these. I think most of these system calls
are like this:
system("rm -Rf testdir")
Another problem besides the permission warnings is that pure windows
installs won't be able to test ruby. I realized that I still had cygwin,
and thus rm, on my machine. These aren't big things to test, but I don't
want someone to download narf to check it out and be scared away by a bunch
of failures that could have been easily fixed.
Cheers,
Patrick
"This communication is intended solely for the addressee and is
confidential and not for third party unauthorized distribution."
|
|
From: William L. A. <ba...@ac...> - 2003-12-04 14:25:07
|
Patrick: on my cygwin seup Narf continues to be free of errors. I continue to filter the directory permission warnings from the Ruby system() command using ruby -W0 Bill Patrick May wrote: > Hi folks, > > I fixed up the windows tests tonight. Narf should be failure free on > the Prag Windows Installer version of ruby. Next target will be cygwin > ruby. > > Cheers, > > Patrick > > > > ------------------------------------------------------- > This SF.net email is sponsored by: SF.net Giveback Program. > Does SourceForge.net help you be more productive? Does it > help you create better code? SHARE THE LOVE, and help us help > YOU! Click Here: http://sourceforge.net/donate/ > _______________________________________________ > Narf-lib-devel mailing list > Nar...@li... > https://lists.sourceforge.net/lists/listinfo/narf-lib-devel > |
|
From: Patrick M. <pa...@he...> - 2003-12-04 05:14:34
|
Hi folks, I fixed up the windows tests tonight. Narf should be failure free on the Prag Windows Installer version of ruby. Next target will be cygwin ruby. Cheers, Patrick |
|
From: Patrick M. <pa...@he...> - 2003-12-03 04:57:35
|
Hello,
Sorry I wasn't at Xp-NewYorkCity Monday night. I choose to be
disappointed instead (I watched my favorite football team lose :-).
Anyways, I made committed some changes to the Upload api.
Web::Upload's aren't a SimpleDispatcher that delegates to a Tempfile
anymore, and they have some new methods:
anUpload.read
returns the content of an uploaded file.
anUpload.open{ |io|
io.read ... etc....
}
yields an io object so you can read io style if you want.
anUpload.write_to_file( filename )
writes the contents of anUpload to filename. Also, I removed one
method from the Web::Upload interface:
anUpload.path
I think that the actual path of an upload's tempfile (or even the fact
that it is a tempfile in the first place) is an implementation detail
that users shouldn't really play with. Plus, using this to move
uploads seemed to cause some problems with narf wiki upload tests on
Windows.
Cheers,
Patrick
|
|
From: William L. A. <ba...@ac...> - 2003-11-29 14:37:11
|
Patrick: here are the error msgs from the system command in the following 4 files. This is from running ruby build.rb test The messages result from the system() call. Bill |
|
From: Patrick M. <pa...@he...> - 2003-11-27 04:25:18
|
What files / lines are causing problems? ~ pat On Wednesday, November 26, 2003, at 05:52 PM, William L. Anderson wrote: > I've tracked it down to a warning that comes from using the system() > command. (irb is my friend). > > Is there a way to "advise" Ruby methods, functions, etc.? like you can > in Lisp? It's one thing to make sure that /usr/local/bin and those > types of directories have appropriate permissions (0755). But it's > another when I'm running the build.rb tests in my own src directory > and system() complains about those permissions. > > Bill > > Tom Clarke wrote: > >> That seems to be a ruby-on-windows thing. One way to fix that is to >> fix >> the permissions on those directories. I've not discovered how to fix >> it in >> any other way, though I would like to find out how.... >>> Patrick: if you want to share the load I can set up a Windows Ruby >>> install on my NT4 box and see what happens. There is no cygwin on >>> that >>> machine. >>> >>> And, by the way, I also get no errors running the wikitest. >>> >>> The warnings are all about having 040777 permissions on the document >>> root and cgi-bin directories (but I'll check and send you some >>> output). >>> >>> Bill >>> >>> May, Patrick wrote: >>> >>> >>>> Bill, >>>> >>>> duh, you just showed me. (ruby -W0 build.rb test) Anyways, I'll >>>> need to >>>> setup a cygwin install and see what these warnings are all about. >>>> This >>>> is a different issue than the problems I'm having with the >>>> Web::Upload's >>>> and 'ruby build.rb testwiki' >>>> >>>> Cheers, > > > > > ------------------------------------------------------- > This SF.net email is sponsored by: SF.net Giveback Program. > Does SourceForge.net help you be more productive? Does it > help you create better code? SHARE THE LOVE, and help us help > YOU! Click Here: http://sourceforge.net/donate/ > _______________________________________________ > Narf-lib-devel mailing list > Nar...@li... > https://lists.sourceforge.net/lists/listinfo/narf-lib-devel > |
|
From: William L. A. <ba...@ac...> - 2003-11-26 22:53:53
|
I've tracked it down to a warning that comes from using the system() command. (irb is my friend). Is there a way to "advise" Ruby methods, functions, etc.? like you can in Lisp? It's one thing to make sure that /usr/local/bin and those types of directories have appropriate permissions (0755). But it's another when I'm running the build.rb tests in my own src directory and system() complains about those permissions. Bill Tom Clarke wrote: > That seems to be a ruby-on-windows thing. One way to fix that is to fix > the permissions on those directories. I've not discovered how to fix it in > any other way, though I would like to find out how.... > > >>Patrick: if you want to share the load I can set up a Windows Ruby >>install on my NT4 box and see what happens. There is no cygwin on that >>machine. >> >>And, by the way, I also get no errors running the wikitest. >> >>The warnings are all about having 040777 permissions on the document >>root and cgi-bin directories (but I'll check and send you some output). >> >>Bill >> >>May, Patrick wrote: >> >> >>>Bill, >>> >>>duh, you just showed me. (ruby -W0 build.rb test) Anyways, I'll need to >>>setup a cygwin install and see what these warnings are all about. This >>>is a different issue than the problems I'm having with the Web::Upload's >>>and 'ruby build.rb testwiki' >>> >>>Cheers, > |
|
From: Tom C. <to...@u2...> - 2003-11-26 22:43:21
|
That seems to be a ruby-on-windows thing. One way to fix that is to fix the permissions on those directories. I've not discovered how to fix it in any other way, though I would like to find out how.... > Patrick: if you want to share the load I can set up a Windows Ruby > install on my NT4 box and see what happens. There is no cygwin on that > machine. > > And, by the way, I also get no errors running the wikitest. > > The warnings are all about having 040777 permissions on the document > root and cgi-bin directories (but I'll check and send you some output). > > Bill > > May, Patrick wrote: > >> Bill, >> >> duh, you just showed me. (ruby -W0 build.rb test) Anyways, I'll need to >> setup a cygwin install and see what these warnings are all about. This >> is a different issue than the problems I'm having with the Web::Upload's >> and 'ruby build.rb testwiki' >> >> Cheers, >> >> Patrick >> >> -----Original Message----- >> *From:* nar...@li... >> [mailto:nar...@li...]*On Behalf Of >> *May, Patrick >> *Sent:* Wednesday, November 26, 2003 5:03 PM >> *To:* 'nar...@li...' >> *Subject:* RE: [Narf-lib-devel] file upload api change >> >> Bill, >> >> Which test target are you running? >> >> Cheers, >> >> Patrick >> >> > -----Original Message----- >> > From: nar...@li... >> > [mailto:nar...@li...]On Behalf >> > Of William >> > L. Anderson >> > Sent: Wednesday, November 26, 2003 5:00 PM >> > To: nar...@li... >> > Subject: Re: [Narf-lib-devel] file upload api change >> > >> > >> > Patrick, OK this makes some sense. >> > >> > I had tons of permissions warnings but if I run >> > >> > ruby -W0 build.rb test >> > >> > then there are no warnings. >> > >> > I'd really like to understand the issues here. >> > >> > Bill >> > >> > May, Patrick wrote: >> > >> > > Bill, >> > > >> > > I'm running the Windows installer ruby, version 1.8.0-10 >> > (not cygwin). >> > > I think the permission warnings are failing tests re: >> > uploading images. >> > > On my box, this was the case. >> > > >> > > ~ Patrick >> > > >> > > > -----Original Message----- >> > > > From: nar...@li... >> > > > [mailto:nar...@li...]On Behalf >> > > > Of William >> > > > L. Anderson >> > > > Sent: Wednesday, November 26, 2003 4:36 PM >> > > > To: nar...@li... >> > > > Subject: Re: [Narf-lib-devel] file upload api change >> > > > >> > > > >> > > > Patrick: what version did you test. I tested Revision 30 on >> > > > Win2000 and >> > > > aside from permission warnings all the tests ran. So, I'm >> > > > confused about >> > > > what's broken. >> > > > >> > > > Bill >> > > > >> > > > May, Patrick wrote: >> > > > >> > > > > Hello, >> > > > > >> > > > > I think we may need to change the api for file uploads. >> > > > Recently, I ran >> > > > > all the narf tests on a Windows 2000 box. Besides the >> > > > expected sorts of >> > > > > failures (path issues / etc.) there were also some >> > disturbing ones >> > > > > related to moving files. >> > > > > >> > > > > I think the core of the issue is that when Narf processes >> > > > an uploaded >> > > > > file, it returns an opened Tempfile. A common thing >> > to do with an >> > > > > uploaded file is to move it to somewhere else. For some >> > > > reason, this >> > > > > works on unix's, but on windows you get permission denied >> > > > errors. I >> > > > > think it's because the file is still open. >> > > > > >> > > > > I will investigate the problem more, but I think I may >> need >> > > > to change >> > > > > the api so that the Web::Upload contains the path info >> > and helpful >> > > > > methods re: the Tempfile, instead of being literally an >> > > > Tempfile. (It >> > > > > currently is a SimpleDelegator delegating to the >> Tempfile). >> > > > > >> > > > > Cheers, >> > > > > >> > > > > Patrick >> > > > > >> > > > > >> > > > > "This communication is intended solely for the addressee >> and is >> > > > > confidential and not for third party unauthorized >> > distribution." >> > > > > >> > > > >> > > > >> > > > >> > > > >> > > > ------------------------------------------------------- >> > > > This SF.net email is sponsored by: SF.net Giveback Program. >> > > > Does SourceForge.net help you be more productive? Does it >> > > > help you create better code? SHARE THE LOVE, and help us >> help >> > > > YOU! Click Here: http://sourceforge.net/donate/ >> > > > _______________________________________________ >> > > > Narf-lib-devel mailing list >> > > > Nar...@li... >> > > > https://lists.sourceforge.net/lists/listinfo/narf-lib-devel >> > > > >> > > >> > > < >> > > >> > > >> > > "This communication is intended solely for the addressee and is >> > > confidential and not for third party unauthorized >> distribution." >> > > >> > > /BODY> >> > >> > >> > >> > >> > ------------------------------------------------------- >> > This SF.net email is sponsored by: SF.net Giveback Program. >> > Does SourceForge.net help you be more productive? Does it >> > help you create better code? SHARE THE LOVE, and help us help >> > YOU! Click Here: http://sourceforge.net/donate/ >> > _______________________________________________ >> > Narf-lib-devel mailing list >> > Nar...@li... >> > https://lists.sourceforge.net/lists/listinfo/narf-lib-devel >> > >> >> "This communication is intended solely for the addressee and is >> confidential and not for third party unauthorized distribution." >> >> BODY> >> >> "This communication is intended solely for the addressee and is >> confidential and not for third party unauthorized distribution." BODY> > > > > > ------------------------------------------------------- > This SF.net email is sponsored by: SF.net Giveback Program. > Does SourceForge.net help you be more productive? Does it > help you create better code? SHARE THE LOVE, and help us help > YOU! Click Here: http://sourceforge.net/donate/ > _______________________________________________ > Narf-lib-devel mailing list > Nar...@li... > https://lists.sourceforge.net/lists/listinfo/narf-lib-devel > > > |
2 messages has been excluded from this view by a project administrator.
