A new link Informations (to translate in po files) is present under
the title. On mouse over the popup opens in the right half of the
screen
thierry: nice concept, but instead of just a link, I wonder if you could
configure it so that if the banner is short enough, it displays
I'd rather see the icon in the banner. It gives the banner some color and
definition.
It's probably something in the library rather than the script. I'm still
finding my way around the library sources.
Yes, but remember what we're trying to do here -- we're trying to
compensate for a malfunctioning client browser.
Already got it ... and yes, it's a great tool.
Security isn't annoying. Time zones are annoying. Time zones are VERY
annoying. And for some reason, time zone bugs, once fixed, have a
tendency to un-fix themselves.
Attached is the new Dutch translation.
Day-to-day experience in working with Webcit has much improved the
translation.
Thank you! It's been committed to svn and will appear in the next
release.
It would be a pleasure to have you join the project. We try to keep the
code clean and profusely commented, but if you have any questions about
it, ask away.
Ironically, the problem that caused me to make that comment fixed itself.
On our mail server at the Big Blue X, webcit suddenly started displaying
timestamps in GMT. I was not looking forward to figuring out why.
As it turns out ... grelf updated the tzdata on our mail server but
That would be work. :)
Good news! I have been given maintainership of libical. Whatever we do
with it will now be the official version.
I don't intend to do active development on it, but I'm hoping we can at
least consolidate the various fragments of 0.24 that have been floating
around and get everyone running directly from the upstream code base
again. I'll periodically update the included tzdata as well. And of
course,
Yes, but it's done by Someone Else. It's basically just a recompile of government-supplied data a few times a year.
on fleebs question in the support room, shouldn't we be able to fully
compose an email on all private folders? At least the use of sieve
implies this, right?
Yes, but we have to figure out the semantics of it. We can't just send
mail all the time, because we might end up sending
As a mailinglist post indicates, there needs to be some handwork be
done so older dates are displayed correct?
It would be nice if this were done, but it would require a *lot* of work.
You would basically have to bring in every old version of tzdata, know
when they were published,
Yes, it should be the way you corrected it. Thanks.
Thanks for the patches!
** help wanted **
I've got this libical code, and I need to merge it back in to CVS, which
hasn't been touched in years. We have to figure out which files were
added, deleted, changed, etc.
There's also the beginnings of a CAP calendar server in there which we
can delete,
i'd prefer to start with merging the cvs into the uncensored svn
first.
afaik there is not much documentation of what our friend omar did to
libical, right?
so theres not much else todo than jump the big gap forward?
Might as well just jump it forward.
Bizarre bizarre bizarre.
I'm making an asynchronous HTTP call that populates the list of address
book rooms. That call has an onComplete callback set that is then
supposed to populate the list of contacts.
The problem is, for some reason, the second call is arriving at the
Wouldn't we have to figure in *every* distribution if we did that?
Feh. SVN doesn't like commit logs with exclamation points in them.
Already did.
Very nice. Slowly but surely we are getting on the map so to speak.
Just because Seth asked for two sets of buttons doesn't mean that it's a
design decision that is carved in stone. If you believe the UI would be
best with just one set of buttons, then go for it!
Aha. svn propset svn:keywords... was the magic incantation.
Thanks!
Apparently, if you have the latest MS Office installed, your computer will
send two additional HTTP requests to every single website you visit, in
order to determine whether the site supports some weird online
discussion plugin that's integrated into Office.
resize_msglist is the div that allows you to grab the bar between the
message list and the message viewing pane, and resize it. If I remember
correctly, overflow:none; is there to prevent a scroll bar from
appearing in that div. I'm not 100% certain that it needs to be there.
IG : Bug or feature ?
In Mail folder view, we can use Reply All and Forward buttons but not in
Bulletin Board view... I don't understand why
There is no such thing as reply all or forward in a bulletin board
conversation. A reply is posted to the same bulletin board.
hahA lolcode can has dokuwiki!! lol
We changed from text/x-vcard to text/vcard because that's what it's
supposed to be now that vCard is a standard; furthermore, that's what the
Outlook connector wants to use.
Citadel will *accept* either of those MIME types as a vCard, but we
changed what it generates by default.
Hey thierry -- I think there's a problem with the Who is online? list
in the sidebar. A lot of the time, when I first log in to a Citadel
system, instead of showing the list of users, it displays the Sorry! You
must have cookies enabled... message embedded in the sidebar.
Any idea
thierry: the bulletin board message view is broken in Internet Explorer
(I tried it in IE7, but the problem may exist in other versions as well).
The first message is perfect, but all others are showing without any
layout at all, and they're just all over the screen.
I'll see if I
Never mind, I fixed it ... there was a div tag that wasn't closed, so I
closed it and IE7 is displaying the bulletin board view properly now.
This is good ... in what I believe is a proof that our recommended backup
strategy actually *works* as advertised ... I fired up the replica of
Uncensored on my home server (it rsync's back there every night) and after
a quick check of the db, it came right up.
It even started trying to
Putting back a modified version of what you read in should be the right thing to do. I can't test it because I don't have a working build of ctdlphp right now.
Envisage You to update tinymce in the next version of webcit?
Yes, now that you mention it, this is probably a good idea.
I haven't had a chance to review GPLv3. Has it been finalized?
No. They seem to be getting towards the end of the draft process at this
time.
Envisage You to update tinymce in the next version of webcit?
TinyMCE has now been updated to the latest version in our svn tree.
The interesting thing is that they've split out the language packs. This
will make WebCit smaller because we only have to include the languages
that
solve the problem with the Who is online? list in the sidebar
It works now. I will test some more, to be sure.
TRANSLATORS: please get your .po files up to date; we will be doing a
release of WebCit soon.
There we go. As the IGlette likes to say, Happy to you. I think that
completely eliminates our use of setuid in the Citadel system.
I really like the project statistics, metrics, and graphs that Ohloh
generates. It does all of that automatically by fetching your entire SVN
repo and analyzing the code. Pretty cool.
Every time I try to attach my Ohloh account to citadel, it puts me on the
other Citadel (the one that was deleted). Blah. I'll try again later.
I will be on vacation this week, and only checking in occasionally.
Please be patient. :)
Yes, the wording is good. I only changed a couple of little things:
1. The response code for subject required is now numeric. There's no
need to output and then compare a string when the value will be found in
the same position every time.
2. Intead of automatically setting the
Thank you. I couldn't quite figure that out.
The first parameter it returns is the screen name you are using to post
the message. We don't look at that right now, but we should. It's not
terribly important though.
Ok, so the next thing you should do is, evaluate that return value, and
if it is nonzero, prompt for a subject. This
True, but technically this *is* a semi-public list. Imagine what would
happen if the vulnerability is serious and it fell into the hands of
someone who has an axe to grind.
StuartNCC: please email your report to ajc (at) citadel (dot) org, and I
will redistribute it to the appropriate
ok, it could be working that way. I think i made my way through it...
s.b. needs to check ;-) was some cut'n'paste, i hope i've hit the
right places...
It looks good! But I think the Internet mail recommends a subject
message is not necessary. Aside from that, the new code is
Ok folks, the GPL v3 is now official. We're going to move to it,
effective immediately, unless there are any objections. So if anyone
cares about this kind of thing, read the GPL v3 now.
You'd like to force the use of plain text?
I agree that this is something useful, but remember that WebCit *only*
generates HTML. We'd have to convert to plain text within the server.
html.c
Aha! I now understand the problem, but it'll take a while to locate it.
Follow TCP Stream in Wireshark showed a gzip-compressed HTTP output, to
which I said oh, we'll have to disable compression while we
troubleshoot. But then I noticed a /div at the end of the compressed
output. So
Sometimes you can hunt for a bug for hours and it just doesn't appear ...
it often takes a second person to look at it to spot the problem. Or,
sometimes you just have to walk away for a while and come back fresh, and
the bug appears quickly. The bug I fixed in r5263 was like that -- Willi
jamy chars ??
matt:
Both of your proposals seem legitimate. Please post patches.
Ok, there it is.
Now we can continue working on the short list of items we were going to
address. I don't mind doing another release very soon. We just had to
get this one out because there are security advisories being released
tomorrow, and this is our last opportunity to have the
Yes, but only if you're paranoid. The vulnerabilities are, in my opinion,
not major. Some of them I was not even able to confirm that they are
real.
In other news, I had a happy little discovery today:
http://www.citadel.org/doku.php/faq:troubleshooting:out_of_lock_entries
After upgrading Citadel, I couldn't get Uncensored to start back up
again. I got that same old error that I've seen in the past:
BDB: Lock
ASTERISK VOICEMAIL INTEGRATION, HERE WE COME! :)
Asterisk voicemail?!?
Asterisk 1.4 has a cool new feature that allows you to use an IMAP server
as the storage back end for voicemail boxes. The nice thing about doing
it this way, as opposed to simply having voicemail forwarded to your
email, is that there's only one copy of it.
there you go fleeb, ig needs to push it out to easyinstall now...
Ok, it's on the Easy Install site. Give it a try.
As they used to say in the Amiga demoscene: vgood!!
i've done some strlen removing...
in messages.c 36 out of 106 remain...
from 15 of them in loop heads only one remains.
Ok, most of these look good; the whole idea of calculating the string
length once before entering the loop is a good one.
I'm having trouble
If you perform a size-altering operation such as striplt() then you
definitely have to measure the string again.
Actually, I think the EVT_BEFORESAVE hook is the one you're interested in.
And yes, doing that way would be far more efficient than implementing a
new queue.
I did not write serv_funambol.c; it was contributed by the author of the
middleware with which it connects.
messages.c is allowed to have global variables because it's part of the
text client, which is not multithreaded. (Unless you're talking about
messages.c from webcit, in which case
I am completely in agreement that it would be a good idea to have the
build system scan a given directory for Citadel modules, and add them to
the build.
Unfortunately I don't know how to do that. :(
But I am willing to accept it as a patch if someone else codes it up.
I
Ok, I like that. For the unique names, we could do what the Java folks do
and use a namespace based on ownership of a domain, such as
org_citadel_foomodule_init()
Yes, we can do that.
I'm a bit concerned about citserver right now. I upgraded uncensored on
Sunday evening, and I've noticed that it's using a lot of CPU now. Didn't
someone else complain of excessive CPU utilization?
Do we have profiling tools available?
davew: that's probably it. Good catch. I think we have that data readily
available; I'll see about populating that field when I do some hacking
tonight. Thanks!
davew: the problem couldn't be in your patches, because I haven't applied
them yet. :)
I'm going to try to get that envelope_to problem fixed now. Once we've
got it stable, and you've got your tree updated, I'll ask you to generate
a new set of patches and we'll apply them then.
Ok, disregard previous message. I thought that RegisterRoomHook() was one
of your patches, but it isn't (unless dothebart applied them?) and you're
correct; it's in the wrong place. My bad.
As far as I can tell, vacation is doing the wrong thing in two situations:
1. When there is more than one envelope recipient
2. When the sender and recipient are both on the local Citadel system
On my test system, it doesn't crash, but it doesn't deliver the vacation
message
Ok, I've fired off a message to Aaron Stone asking about some libsieve
details. As far as I can tell, it *does* compare the body recipient(s) to
the envelope recipient in order to determine whether this message was
explicitly addressed to me. I'd love to be able to suppress that test.
WebCit is updated on Uncensored. Thierry - I like the menu button change you made to the 'who is online' button. It works flawlessly.This message is a reply to another in this room, so this will be a test to see if the subject bug is fixed.
Callbacks I can understand but the constant copying and duplicating
and transfering of data around the sieve context... eh.. what for?
I think it may have had something to do with the fact that libsieve did
not originate as a standalone library. It began as the Sieve code from
davew: I'm inclined to agree with you about the source of the crashes.
I've checked in some code that will force local style addresses to get
converted to Internet email style addresses for the purpose of the Sieve
GETENVELOPE call. For example, Friko Mumjiboolean would get converted
to
I notice that parsedate.c is in the svn. Should it be, after all it
is built from parsedate.y is it not?
That is intentional. By including parsedate.c we eliminate the
bison/yacc requirement for those building the code. This is especially
useful for reducing the number of sites
I notice that make distclean doesn't remove chkpw and
database_cleanup.sh is this an oversight?
Yes, that is an oversight.
The above commit is part of the solution to making local-to-local vacation
responses work. At the very least, we're not scanning the envelope
recipients anymore.
It isn't feasible to kludge CtdlMakeMessage() to fake an RFC822 style
recipient for local-to-local mail, because there are
If you're inside a module you're already privileged.
Yes, it's true ... some of the modules reference each other's symbols.
There are a couple of situations where I got a bit messy. Sorry :(
And there we go. A righteous hack indeed.
I like the CTDL_MODULE_INIT() framework. Nice and clean. One thing I
really didn't like about the conversion back to a monolithic server was
that we had to name all of the module init functions. This fixes that
very nicely.
I've just noticed that webcit shows the time as server time when
creating a message and in bbs view when reading a message instead of
local time.
If you have a reliable way of learning the client's time zone, getting it
back to the server, making strftime() aware of it, and doing
For a while, we were experimenting with a way to auto-reconnect clients
through a server restart, but it ended up being really unstable.
Restarting when everyone is idle might be a good idea.
Nope, just the client that ships with the server.
Everyone please go here:
http://tinyurl.com/ysbqfl
FSDaily is a site similar to Digg, but specific to free and open source
software. I've posted a link to our GPLv3 announcement, and would
appreciate it if everyone could vote for this story. It only takes
about a dozen votes to
davew: broken build alert!!
In file included from server_main.c:61:
modules_init.h:1: error: expected identifier or â(â before â-â token
modules_init.h:8: error: stray â#â in program
modules_init.h:9: error: stray â#â in program
modules_init.h:11: error: stray â#â in program
Still doesn't seem to work.
Oooh, the ability to easily create new maintenance threads would be a big
win.
Still broken. I checked out a completely new copy of the tree and attempted
to build it on Ubuntu 7.04 (32-bit Intel).
/mk_module_init.sh
Scanning extension modules for entry points.
This version of echo /bin/echo supports -e -E.
-e Found entry point in file serv_imap.c
-e Found entry point in
So far so good. It built cleanly on my laptop (Ubuntu 7.04), which is
the machine that barfed on it earlier. I normally don't do software
development on this machine, though, so I'll head back over to the usual
dev box and make sure it's still clean there.
To avoid having to
The Citadel build itself probably wouldn't be nearly as involved as making
sure you have all the correct tools installed. No big deal though,
because dothebart's excellent buildpackages script can build a
distribution tarball very quickly on a Linux machine, which can then be
brought over to
It means you're opening a unix domain socket.
Current web style
http://www.webdesignfromscratch.com/current-style.cfm
There are some good ideas in here. I'm going to try to use the examples
and guidelines as a reference.
As good as the gradient background image in webcit is - its way too
big - 132K.
Surely some compression won't hurt. Switching from a radial to
repeat-y gradient would be better bandwidth wise.
I don't know how to do that ... I just downloaded a canned background
from a
Oops, I spoke too soon. Thierry knows how to do it. Thanks.
The color you changed is the color of of last pixel on the bottom of
the gradient background image, which is not repeat vertically. So if
you have many messages the transition between the gradient image and
the bottom will be really ugly
While I agree with this in concept, the
so |foo|bar|bla
gimme number 3
will give you nada?
The tokenizer understands zero-length tokens. So in the example you
gave:
Token 0 is
Token 1 is foo
Token 2 is bar
Token 3 is bla
And if you supply it a string that doesn't contain the
201 - 300 of 1678 matches
Mail list logo
