#foswiki 2012-02-28,Tue

↑back Search ←Prev date Next date→ Show only urls(Click on time to select a line by its url)

WhoWhatWhen
gac410Even if Mapping is set to BaseUserMapping, the function groupAllowsChange still returns true for any group. [00:10]
.... (idle for 16mn)
I'm not sure what to expect, but it appears, if mapping manager is set to 'none'', it's changed to TopicUserMapping. And If I set the mapping to BaseUserMapping, it returns true for groupAllowsChange for any group except for the base groups. [00:26]
SvenDowideitoooo, complex [00:26]
gac410There seems to be no way to tell the groups API that the groups cannot be changed. [00:27]
SvenDowideitiirc if you have basemapper return false, it prevents the topic mapper from adding users to the admin group
but if you have none
[00:27]
gac410right. users.pm ANDs the results from the Base and the Impl mappers. [00:27]
SvenDowideitthen you should not be able to change any rgoups that are _not_ int he base [00:28]
gac410And Users.pm says. Mapper = TopicUserMapping if mapper = 'none' [00:28]
SvenDowideitargh
you wa?
[00:28]
gac410yup [00:28]
SvenDowideitthat sounds quite wrong [00:28]
gac410In Routine new() $implUserMappingManager = 'Foswiki::Users::TopicUserMapping' if ( $implUserMappingManager eq 'none' ); [00:28]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::Users::TopicUserMapping [00:29]
SvenDowideitoh gads, thats probly because i stopped work on what i'd actually intended
which was n-mappers
not 2-mappers
[00:29]
gac410gac410 is rapidly exceeding his complexity level [00:29]
SvenDowideitits amazing how much more convoluted code is if you use a constant :/ [00:30]
gac410but he's convinced it's broken and is going to open a blocker :-( [00:30]
SvenDowideitfor 1.1.5 i presume [00:30]
gac410yeah. [00:30]
SvenDowideitSvenDowideit screams at self for being busy [00:31]
gac410%GROUPINFO format=$allowschange should return 0 for the BaseUserMapper or 'none' mapper.
well before I panic and open a blocker.... How long has this been broken.
[00:31]
SvenDowideitforever
i would expect it was original dev
and is (i'm guessing) because users.pm has code everywhere to basemapper-> then othermapper->
[00:32]
gac410yes. CDot from 2008, [00:32]
SvenDowideitbut tbh, i'm starting to doubt
mmm, oh dear, is there a relevant comment/task?
oh, sorry,
you mean foswiki orig, not original dev which should be ~2006
[00:32]
gac410well - who knows. Blame show's it was a tidy against lib/TWiki/Users.pm in 2008
so probably original development.
[00:35]
SvenDowideittbh, fixing that will be out of safe scope for 1.2
er, 1.1.5
[00:36]
gac410Okay - so going back that far. doesn't make sense to do anything for 1.1.5. i was just trying to prove that GROUPINFO could reliably be used to active or inactivate the addUserToGroup API. [00:36]
SvenDowideitit means changing every users.pm func
if you add failing tests to > 1.1, then task block for me on 1.2
[00:36]
gac410yup - not going near it with a 3.048 foot pole. [00:37]
SvenDowideitatm, my schedule for next month is clear enough
argh :(
why does ssh-copy-keys not work with a port
[00:37]
gac410gac410 doesn't do why questions [00:38]
SvenDowideitargh!
so ssh -p, but scp -P
why can't we standardise :)
[00:40]
gac410Okay Item11578 blocker for 1.2. [00:44]
FoswikiBothttp://foswiki.org/Tasks/Item11578 [ Item11578: =groupAllowsChange= returns true when 'none' is set as the mapping manager ] [00:44]
gac410Hm. It also returns true if no user is logged in. Where as allowschange is documented as returning true if current user is allowed to change the group. [00:47]
.... (idle for 17mn)
gac410 is very confused. GROUPINFO opens an iterator on the group members, and checks groupAllowsChange for each member. The only member of AdminGroup is BaseUserMapping_333 ... and TopicUserMapper returns true - allows change.
So ... How do I differentiate between a *Group* itself allowing change, and if members of the group allow change. I think we are mixing our metaphors here somehow.
The Users.pm seems to suggest groupAllowsChange is a group-level function. Can the group be changed. But GROUPINFO is checking members of the group.
gac410 is hopelessly lost.
[01:04]
.................... (idle for 1h38mn)
pharveyMichaelDaum set Item779 WaitingFor me, I wonder why (did I just volunteer to fix that task? :) [02:44]
FoswikiBothttp://foswiki.org/Tasks/Item779 [ Item779: {AuthScripts} setting needs a * option, and a dynamically created list of cgi's that are installed. ] [02:45]
SvenDowideitbecause you did volunteer for it!!! [02:45]
gac410pharvey - I'm going to build an early beta 1.1.5 tomorrow evening to install on f.o, so I can see if the Logger rollover bug is really fixed. Any thoughts on the wysiwyg errors? [02:47]
pharveyDid I? I thought I just said I'd looked at the code
gac410: sorry, that work is progressing slowly and carefully. I spent most of Sunday on WysiwygPlugin, but nothing to commit yet.
[02:48]
gac410okay - no prob. just figured it was worth checking. [02:49]
pharveyI know. I'd love to get some new fixes in. I guess there's nothing stopping a 1.1.6. And WYSIWYG/TinyMCE can be upgraded separately. [02:50]
gac410yup.
still talking another month for the final 1.1.5, I mostly want to make sure the logs roll correctly on a production system.
[02:50]
...... (idle for 25mn)
GithubBot[foswiki] foswiki pushed 1 new commit to master: http://git.io/n-1V5g
[foswiki/master] Item11579: Initial commit of AdminToggle. - GeorgeClark
[03:17]
***GithubBot has left [03:17]
FoswikiBothttp://foswiki.org/Tasks/Item11579 [ Item11579: Implement the AdminToggle button ] [03:17]
..................... (idle for 1h42mn)
pharveygac410: btw Item11518 - massively cool :) [04:59]
FoswikiBothttp://foswiki.org/Tasks/Item11518 [ Item11518: Add BCrypt encryption to =HtPasswdUser.pm= ] [04:59]
gac410great! Thanks pharvey. Really not much to it though. It's just pluggin in another cpan module. :-)
The hard work will be getting agreement and then implementing something that doesn't pass the passwords in the clear.
[04:59]
pharveyoh! I thought it included the client-side stuff :P [05:00]
gac410No! I hope you are not serious. I thought I was really clear on my checkin. [05:01]
pharveyah, I only briefly scan the messages
never mind, I just can't read
[05:01]
gac410Yeah the task says it outright, not just the commit. "It doesn't do anything to improve the security of template auth. It does provide more robust encoding of the .htpasswd file in the event that it is obtained. "
Okay... so I only deserve "a little cool" definitely not "massively cool" :-D
[05:01]
SvenDowideithum. is there a way to reset just one subdirs files using git? [05:17]
gac410what do you mean by reset?
checkout reset HEAD or ??
[05:18]
SvenDowideitundo all uncommited changes
y, but only for the data/Main dir
SvenDowideit would try, but i don't want to risk losing, everything :)
[05:19]
gac410Reset HEAD takes a path - can you do a Find -exec
or git checkout?
[05:20]
SvenDowideitSvenDowideit will just do something else
as i really don't want to lose all these uncommted changes
[05:21]
gac410find data/Main -type f -exec git checkout {} \; or something like that?
oh - checkout would loose them.
reset HEAD --soft would leave them modified but undo the commits.
[05:21]
SvenDowideithuh?
i clearly don't grok whats going on
i thought reset would bring my sandbox back to whats commited
[05:22]
gac410What do you mean by loose uncommitted changes. does git status show them as modified? [05:23]
SvenDowideity
what worries me
is that i typed
git reset HEAD .
in core/data/Main
and it tells me about BuildContrib files
even though
the docco implies that it'll onlywork on the current dir
fact that git status is repo-wide makes the ui not-confidence-inspiring :/
[05:23]
gac410so reset HEAD default to --soft I beleve. So that should leave the files modified. git stash save YourMess will save all the outstanding changes. Then you can git stash apply them to restore the changes but keep a copy saved. [05:25]
SvenDowideitworse.
i ran git reset HEAD . in dataMain
is it didn't reset anything
[05:26]
gac410HEAD is the top of the index. So it doesn't change much. HEAD^ would move you back one commit.
reset HEAD --hard however will remove all modifications as well.
[05:27]
SvenDowideiti seriously can't see that git reset did anything at all
this is discombobulatin
sven@quiet:~/src/foswiki/core/data/Main$ git reset --hard .
fatal: Cannot do hard reset with paths.
[05:28]
gac410--hard reset will kill everything danger danger. [05:29]
SvenDowideitso, i want to reset a dir, and the only way to do it, is to checkout that dir [05:29]
gac410If you want to save all your local modifications, git stash save NameOfStash will save all your local changes [05:29]
SvenDowideitand clearly, i have to delete the files there and then checkout [05:29]
gac410no - you can just checkout a file. Don't have to delet it. [05:30]
SvenDowideitcept i want to delete any new files too
recursively
so its probly easier to rm -r data/Main ; git checkout data/Main
whch is what i expected get reset data/Main to do, but it doesn't
[05:30]
gac410I have a shell script that does a backup of any files listed in git status. [05:31]
SvenDowideity, thats just admitting defeat :)
i'm using this dumb little argh to force myself to explore git a little more
[05:31]
gac410Well - tis still a good fallback.
I've found I use stash alot more as it can put stuff back.
[05:32]
SvenDowideitmmm, and yet
sven@quiet:~/src/foswiki/core/pub$ git checkout Main
error: pathspec 'core/pub/Main' did not match any file(s) known to git.
[05:32]
gac410pharvey and babar probably would be better to help than me. ;( [05:33]
SvenDowideitmeh :)
we all learn
[05:33]
gac410hm. pub/Main doesn't contain any files, so git won't do anything with a checkout anyway.
works fine if the path contains files.
[05:39]
SvenDowideitinteresting really - i didn't realise we didn't have anyting in Main til then [05:40]
gac410I didn't either. I was trying to recreate your path error but it worked - but on a diff directory. Then tried Main and same error. [05:41]
pharveySvenDowideit, gac410 it's as easy as git checkout <branchname> <dir> [05:47]
gac410unless the dir doesn't contain any files :-) [05:47]
pharveyif you checkout a dir it will give you everything under it [05:48]
gac410git checkout -- Main
error: pathspec 'core/pub/Main' did not match any file(s) known to git.
[05:48]
SvenDowideitpharvey, what is bizzare to me, is that git reset --hard would make sense
except it does not take a pathspec
[05:48]
pharveygac410: do we ship a core/pub/Main ? [05:49]
gac410Yes, but no files. And I didn't think git tracked empty directories [05:49]
pharveypharvey is confused: what is the question? [05:49]
SvenDowideitno question (anymore :))
i'm testing 'import old users from twiki to Main dir' on my git checkout
[05:49]
gac410gac410 has reached that point too. Sven wanted to reset pub/Main but preserve local changes. [05:50]
pharveyI read that you wanted to restore a directory... git checkout is how I would do it
and you had 'git add'd stuff under pub/Main?
[05:50]
SvenDowideitnope [05:51]
gac410For files that git does NOT track, that's where my shell script comes in handy - to tar up anything identified by git status as untracked [05:51]
SvenDowideitmy issue was basically 2fold
git mixes contexts, some operations are doccoed to be local to pathspec, but return results outside - like get reset .
and that git reset doesn't actually do much
[05:51]
gac410well git reset will do a LOT esp if you give it the wrong arguments. :-) [05:52]
pharveygit reset --hard is a commit-level thing, not path-level
if I'm trying to recover mangled files, I just use git checkout... seems easier... but you wanted to 'reset' untracked files? How does that work :)
[05:52]
SvenDowideitpharvey, y, which doesn't match well with the other get reset options
simple thing i really wanted to do
call git reset --hard for one dir only
instead, i
[05:53]
pharveyI think we use git reset --hard too often, it's a very blunt instrument... we should probably be doing git checkout <foo> instead [05:54]
SvenDowideitchmod -R 777 data/Main ; rm -r data/Main ; git checkout data/Main ; ./pseudo-install developer [05:54]
pharveygit reset --hard will also *delete* any files you accidentally add'd from new, but haven't committed yet [05:54]
SvenDowideitbeats me, i want to know that the dir i'm targeting is identical to a fresh checkout
so yes, thats exactly what i want
[05:54]
gac410git clean [05:54]
pharveygit clean -fdx pub/Main [05:55]
gac410git clean will clean up all untracked. Right. [05:55]
SvenDowideitdon't make me use options :)
i want to clean untracted and reset any modified
ie, git checkout works?
[05:55]
pharveygit checkout <what/ref/branch/commit-id> <where/pathspec> [05:55]
SvenDowideitgit reset --hard still sounds like what i should use when i want to have a totally clean sandbox [05:56]
pharveybut checkout wont' clean untracked [05:56]
SvenDowideitcheckout doesn't give me what i want [05:56]
pharveyas long as you remember that --hard is a checkout-wide thing [05:56]
SvenDowideitso rather than providing get reset --hard data/Main - I need to git clean and git checkout :( [05:56]
pharveyI fail to see how git reset --hard is any different, except that reset --hard nukes your whole checkout, instead of a part of it
reset --hard doesn't clean either
[05:56]
SvenDowideitthats exactly the incinsistency i'm talking about
huh?
SvenDowideit mumbles :()
i thought it did, but its been a while since i was able to reset --hard
too many things cooking at once
[05:57]
pharveyreset --hard will delete newly-added files (I have lost *hours* of work on a few occasions this way)
but will leave untracked stuff alone
even git clean will leave untracked stuff alone
unless you do -f (force)
d removes untracked/empty dirs, x removes ignored files
[05:58]
SvenDowideitoh well, so much for dwim
i guess i'm going to have to write some scripts
laters :)
[05:58]
***gac410 has left [05:59]
pharveyI suspect we've tried to use git reset more often than it's meant to
as in fact, git reset can also do nasty things to which commit your branch head points to
it's a multi-fanged thing, reset is
after I got bit a few times, I used checkout instead
[05:59]
SvenDowideitdarnit :) [06:01]
pharveygit reset --hard && git clean -fdx is a sure-fire way to nucke the checkout though
nuke*
(nuke it back to a clean/good state)
[06:01]
SvenDowideitSvenDowideit wants a knuckle [06:01]
pharveyhaha
speaking of inconsistencies, it's marvelous how listing and renaming 'things' (branches, remotes) are so needlessly different
btw, did you replace your rsync job with a git submodule update --init yet?
pharvey is under the impression mercurial is almost as awesome, and much more user-friendly, but meh
[06:02]
SvenDowideitgod no, i've not even installed the modules Babar wants me to for the nightly builds
oh crap, and i have to go to town - i should have left already
SvenDowideit bumbles off
[06:04]
.................. (idle for 1h29mn)
GithubBot[foswiki] foswiki pushed 1 new commit to master: http://git.io/RLW8IQ
[foswiki/master] Item11579: some cleanup - MichaelDaum
[07:33]
***GithubBot has left [07:33]
FoswikiBothttp://foswiki.org/Tasks/Item11579 [ Item11579: Implement the AdminToggle button ] [07:33]
.... (idle for 16mn)
***ChanServ sets mode: +o MichaelDaum [07:49]
.... (idle for 17mn)
GithubBot[foswiki] foswiki pushed 1 new commit to master: http://git.io/7YSLcg
[foswiki/master] Item11549: keep refresh param out of the toc - MichaelDaum
[08:06]
***GithubBot has left [08:06]
FoswikiBothttp://foswiki.org/Tasks/Item11549 [ Item11549: rewrite of page cache ] [08:06]
AlexanderStHi there. i need to extract a html file form a given topic. Is there another possibility than PublishPlugin? [08:17]
MichaelDaumAlexanderSt, ./view topic=Some.Topic > Some.Topic.html [08:19]
AlexanderStthats easy :-)
but i have to do that in console?
[08:20]
MichaelDaumerm no
wget is your friend?
or whatever
[08:21]
AlexanderStwhat i need is a button "create html" and the user get´s his html file [08:22]
MichaelDaum<a href="%SCRIPTURLPATH{"view"}%/%WEB%/%TOPIC%?contenttype=text/plain">HTML</a> [08:24]
AlexanderSti give it a try
hmm, but i still have to get the content into a file :-)
e.g. called topic.html
[08:27]
MichaelDaumI guess for anything more sophisticated you should use PublishPlugin. Why not using it?
or ask andreli. he built his own publish framework that he never contributed back to the community ;)
[08:33]
AlexanderSti would like to use it, but can i remove the history log in an easy way? [08:35]
MichaelDaumdunno. never used it seriously. [08:35]
AlexanderStok
Michael, you have a lot of experience with ajax rquests... i have a cross domain ajax get request to get json data. if i use the datatype json, it´s not working. i don´t get any response. if i use jsonp i get the json object but a parse error because the server does not supprt jsonp at the moment
i would like to understand that. why does json not work at all??
datatype = json
[08:35]
flexibeastMichaelDaum: Should putting a SETVAR macro in a page add "Set"s in the wiki text of a topic upon normal save? [08:48]
MichaelDaumflexibeast, depends on how complex your requirements will get.
either you put SETVAR into the topic text itself, or better get it out of sight within a view_template
so users wont mess around accidentally
[08:49]
flexibeastWell, i've got a topic with a data form attached.
And basically i want to set access vars depending on the value of one of the form's fields.
[08:50]
MichaelDaumAlexanderSt, javascripts and thus json not working x-domain is a limitation for security reasons build into browsers
flexibeast, right
[08:51]
flexibeastSo at the moment, in the topic, i've got: "%SETVAR{
"DENYTOPICVIEW"
value=" "
field="Visibility"
regex="Public"
}%"
So when the Visibility field is set to 'Public', shouldn't that then add 'Set' statements to the topic?
[08:51]
AlexanderStMichaelDaum: But why is jsonp working?
it´s still the same function $.ajax() i´m using
with the same param crossDomain = true
[08:57]
.... (idle for 16mn)
CDotAlexanderSt: http://en.wikipedia.org/wiki/JSONP explains why jsonp works cross-domain. [09:14]
AlexanderStCDot...i´m just reading it...thanks ;-) [09:15]
MichaelDaumflexibeast, found an error [09:17]
flexibeastMichaelDaum: In my usage, or in the plugin? [09:17]
MichaelDaumin the plugin [09:17]
flexibeastMichaelDaum: Ah. :-) [09:18]
PeterSvendsenHi everybody, have anybody good experienct with creating a dashboard that take input from different servers e.g. build status (from the build server), static analysis tool etc.? [09:19]
flexibeastMichaelDaum: Is it simple to fix? [09:20]
MichaelDaumam working on it [09:20]
flexibeastMichaelDaum: Oh okay, thanks. :-) [09:20]
MichaelDaumflexibeast, reported in http://foswiki.org/Tasks/Item11581
flexibeast, new release available http://foswiki.org/Extensions/SetVariablePlugin
try your example again
[09:30]
flexibeastMichaelDaum: Thank you for working on this! Upgraded plugin, but example still isn't working.
(Nb. There is a "Set VIEW_TEMPLATE" in the topic.)
[09:41]
MichaelDaumadd an %UNSETVAR rule as well like in...
%SETVAR{"ALLOWTOPICVIEW" field="State" value="foobar" matches="Asked"}% %UNSETVAR{"ALLOWTOPICVIEW" field="State" matches="Answered"}%
[09:42]
flexibeastWait, i've been using a 'regex' parameter rather than 'matches' - is 'matches' what i should be using? [09:44]
MichaelDaumthats an alias [09:45]
flexibeastOh okay. [09:45]
MichaelDaumpoint is that you should also specify what will happen to your variable when the formfield has got another value [09:45]
flexibeastWell at the moment i have two SETVARs per preference.
i'll pastebin what i've got.
http://pastebin.com/b4zDmqdN
[09:45]
GithubBot[foswiki] foswiki pushed 1 new commit to master: http://git.io/JnRMxQ
[foswiki/master] Item11580:Item11581: - MichaelDaum
[09:47]
***GithubBot has left [09:47]
MichaelDaumflexibeast, have a look at NatEditPlugin. it does exactly what you are trying to do in a separate "Permissions" tab in the editor [09:47]
FoswikiBothttp://foswiki.org/Tasks/Item11580 [ Item11580: add a way to read global variables ] http://foswiki.org/Tasks/Item11581 [ Item11581: rules not parsed when there is no view template ] [09:48]
flexibeastMichaelDaum: So when a specific field value changes, it can change topic permissions? [09:48]
MichaelDaumno. but it adds a ui to set permissions similar to the formfield that you modelled. [09:49]
flexibeastWell, i want to change permissions automatically based on the form field value.
This is a task management WikiApp.
[09:50]
MichaelDaumso what happens that you say it does not work? [09:51]
flexibeastSome tasks, marked as 'public' through via a FlexForm-based UI, should be available to the general public - i.e. without a login to our Foswiki install - for viewing and editing.
But if that task is then marked 'private', such access needs to be revoked.
Well, that topic's permissions aren't getting changed.
[09:51]
MichaelDaumfirst, put these %SETVAR settings into the TaskViewTemplate
otherwise you end up in hell ;)
[09:53]
flexibeastHeh, okay. *goes to do that* [09:53]
MichaelDaumhere's how that looks like for BlogPlugin: http://svn.foswiki.org/trunk/BlogPlugin/data/Applications/BlogApp/BlogEntryViewTemplate.txt
the PERMSET_* things are only needed for compatibility with NatEdit's permission ui
[09:54]
MichaelDaum off to the supermarket buying coffee n milk [09:59]
AlexanderStflexibeast: we changed the SetVariable plugin to set topic rights with a given formfield...
you have the possibility to put wikinames into a formfield and these values are taken into the allowtopicview setting for example
[10:00]
flexibeastAlexanderSt: Ah, okay. Well, that's not quite the same thing as what i'm trying to do here. [10:07]
AlexanderStok, i thought you had a group of users in a formfield you want to use with setvar... [10:09]
flexibeastNo, the formfield simply has the options "Public" or "Private".
If that formfield is set to 'Public', then the topic that dataform is attached to has to have both DENYTOPICVIEW and DENYTOPICCHANGE set to " ".
But if it's set to 'Private', then the topic has to have those preferences set to "WikiGuest", so that the topic isn't viewable/editable by the general public.
MichaelDaum: The problem i'm having is that if i use SETVAR in the view template i've set up for individual public tasks, that still doesn't make those individual tasks available in the public task list.
[10:09]
AlexanderStah i see, that should be possible without any hacks :-) [10:12]
flexibeastAlexanderSt: i certainly hope so! [10:13]
AlexanderStMichaelDaum: wasn´t there a jquery plugin checking mandtory formfields before save?
...this has nothing to do with setVariable plugin of course :-)
[10:15]
flexibeastOh wait, think i might have got all this working .... [10:23]
Okay!
MichaelDaum: The problem was that SETVAR /requires/ the type param to be passed to it.
MichaelDaum_: Once i added 'type="Set"' to each instance of SETVAR, things worked fine. :-)
[10:31]
..... (idle for 24mn)
MichaelDaumflexibeast, what type?
it defaults to Local
I never use it
[10:56]
flexibeast'Set' [10:57]
MichaelDaumto %SETVAR? [10:57]
flexibeast*nod* [10:57]
MichaelDaumthats an optional parameter, defaults to Local
ah I see.
Local DENYTOPICVIEW = ... does not work
[10:57]
flexibeastThat certainly makes sense to me, once i apply a moment's thought. :-)
However, i wonder if it might be useful to add a note to the plugin docs noting that access control vars will need to be modified using the 'Set' type, since making them Local won't have the desired affect?
*effect
[11:00]
................. (idle for 1h24mn)
PeterSvendsenHi everybody, have anybody good experienct with creating a dashboard that take input from different servers e.g. build status (from the build server), static analysis tool etc.? [12:26]
***jurov has left [12:35]
..... (idle for 23mn)
PeterSvendsenis it possible to split the screen so I could have a left and right screen and indicate the split with a line? [12:58]
............................ (idle for 2h18mn)
AlexanderStI have a problem with a running wiki: i got a setArchivist Storable error. I cleaned the cache in the working_area but it´s still not running. do you have an idea? [15:16]
foswiki_irc5Is there anyone here who might be able to help with using the JHotDrawPlugin using IcedTea-Web? We are having trouble using SL6 (based on RHEL6), which only provides OpenJDK instead of Oracle's JDK.
SL5 and SL4 (which use Oracle's JDK work fine.
[15:20]
***ChanServ sets mode: +o Babar [15:35]
gac410AlexanderSt: I finally found what calls setArchivist. DBCacheContrib. Unfortunately I am completely unfamiliar with that extension. [15:37]
AlexanderStgac410: thanks i´m just navigation through dbcachcontrib issues... now i don´t have the setArchivist anymore but: Can't call method "fastget" on an undefined value [15:40]
MichaelDaumAlexanderSt, rebuild the dbcache using http...?refresh=dbcache [15:42]
AlexanderStnow i have a internal server error :-) [15:44]
MichaelDaum: Can i disable the DBCacheContrib? [15:55]
MichaelDaumrequired by DBCachePlugin and ClassificationPlugin
you'll loose the TopicTitle feature as well
also required by WikiWorkbenchContrib
[15:56]
AlexanderStyes i know, but this wiki doesn´t have wikiworkbench, classifaction, etc. features [15:58]
***foswiki_irc5 has left [15:58]
MichaelDaumso why did you install it?
and why does it error out?
[15:59]
gac410MichaelDaum: The foswiki doc changes - do you want to get any of that stuff into a 1.1.5 early beta? I'm working on building to install on f.o - so we get a "live test" of the log rotation changes on the 1st. [16:11]
MichaelDaumMichaelDaum just checked in the same stuff to 1.1.x [16:11]
gac410never mind - just got the email :-) [16:11]
GithubBot[foswiki] foswiki pushed 1 new commit to Release01x01: http://git.io/nPzW0g
[foswiki/Release01x01] Item11582: merge from trunk - MichaelDaum
[16:18]
***GithubBot has left [16:18]
FoswikiBothttp://foswiki.org/Tasks/Item11582 [ Item11582: use style attributes instead of bgcolor ] [16:18]
GithubBot[foswiki] foswiki pushed 2 new commits to master: http://git.io/hJ-RkQ
[foswiki/master] Item11582: - MichaelDaum
[foswiki/master] Item11582: removing hard-corded colors from docu that override the skin's styles - MichaelDaum
[16:19]
***GithubBot has left [16:19]
GithubBot[foswiki] foswiki pushed 1 new commit to master: http://git.io/qX3UQA
[foswiki/master] Item11582: cleaning up - MichaelDaum
[16:32]
***GithubBot has left [16:32]
GithubBot[foswiki] foswiki pushed 1 new commit to Release01x01: http://git.io/bCFhfQ
[foswiki/Release01x01] Item11582: cleaning up - MichaelDaum
[16:45]
***GithubBot has left [16:45]
.... (idle for 17mn)
GithubBot[foswiki] foswiki pushed 1 new commit to master: http://git.io/4GEVtQ
[foswiki/master] Item11582: tidy - MichaelDaum
[17:02]
***GithubBot has left [17:02]
FoswikiBothttp://foswiki.org/Tasks/Item11582 [ Item11582: use style attributes instead of bgcolor ] [17:02]
.... (idle for 15mn)
GithubBot[foswiki] foswiki pushed 2 new commits to Release01x01: http://git.io/GVp2dg
[foswiki/Release01x01] Item11582: html tidy - MichaelDaum
[foswiki/Release01x01] Item9526: gettext run for 1.1.5 beta - GeorgeClark
[17:17]
***GithubBot has left [17:17]
FoswikiBothttp://foswiki.org/Tasks/Item9526 [ Item9526: Translation work (needed for Pootle auto-checkin - please don´t close) ] [17:17]
gac410Does Item11454 need to block 1.1.5? [17:29]
FoswikiBothttp://foswiki.org/Tasks/Item11454 [ Item11454: 1.1.4 is slower and shows =info.author= as 'UnknownUser' when =.txt= is mauled by an external process ] [17:29]
gac410MichaelDaum: Item11480 is yours - is that a 1.1.5 blocker? [17:31]
FoswikiBothttp://foswiki.org/Tasks/Item11480 [ Item11480: exclamation (!) marker not protected in input values ] [17:31]
MichaelDaumgac410, yea and I don't know how to fix it [17:31]
gac410:-( [17:32]
CDotgac410: I was just looking at Item11454, and wondering why it is a release blocker
it's a rewrite of RcsLite, which could have been done at any time in the last 10 years....
I assume the rationale for it suddenly becoming urgent is the "touched .txt" problem
IMHO that problem is well controlled.
[17:32]
gac410right. and the UnknownUser is working as designed. Yes. The change in not trusting topicinfo resulted in big perf. issues for some.
agreeed though it does seem reasonably controlled.
[17:33]
CDotOn Item11480 - again I would say not a release blocker (though I would be very happy if we deprecated the exclamation point altogether; I never liked it) [17:35]
gac410That would be tough to do - ! is used pretty extensively, and is a lot more friendly than making people enter <nop> [17:36]
MichaelDaum: Item11492 - Is that more of an enhancement that rename should use a pluggable search for backlinks? I'll defer to 1.2 unless you see that as an easy fix. Rename is a bit of a beast. [17:45]
FoswikiBothttp://foswiki.org/Tasks/Item11492 [ Item11492: rename computes the list of referring topics even though there's no LOCAL_SEARCH or GLOBAL_SEARCH on the rename template ] [17:45]
MichaelDaumgac410, this fix is actually straight forward: a perl cleanup for performance reasons.
didnt I check it in?
sec
[17:47]
gac410no commits
and every time I touch rename, the world seems to become a much more difficult place. :-)
This is the list I'm reviewing before building the beta http://foswiki.org/Tasks/TasksByRelease?release=1.1.5;sortcol=4;table=1;up=0#sorted_table
[17:47]
MichaelDaumchecked it in [17:48]
gac410no freezing yet - just trying to be tidy.
okay thx.
[17:48]
MichaelDaumon trunk ... diffing to stable branch
and marked it WaitingForRelease in 115
[17:49]
gac410ty [17:53]
MichaelDaumy
w
fixing / rewriting RcsLite would be gr8 on the way to make it the default store impl instead of rcswrap
[17:54]
GithubBot[foswiki] foswiki pushed 2 new commits to Release01x01: http://git.io/IcoeCg
[foswiki/Release01x01] Item11492: merge from trunk - MichaelDaum
[foswiki/Release01x01] Item11492: perltidy - MichaelDaum
[18:03]
***GithubBot has left [18:03]
GithubBot[foswiki] foswiki pushed 2 new commits to master: http://git.io/RY9fzQ
[foswiki/master] Item11492: only compute GLOBAL_SEARCH and LOCAL_SEARCH when they are in the template ... gawd yet another action-specific bunch of "macros". - MichaelDaum
[foswiki/master] Item11492: perltidy - MichaelDaum
[18:04]
***GithubBot has left [18:04]
gac410MichaelDaum: Item11423 is listed as release in 1.1.5, but only has trunk checkins. [18:14]
FoswikiBothttp://foswiki.org/Tasks/Item11423 [ Item11423: can't disable default tableplugin settings ] [18:14]
MichaelDaum... a bogus one
MichaelDaum trying to repro
[18:15]
works fine now
http://trac.foswiki.org/changeset/13588
http://trac.foswiki.org/changeset/13589
they aren't listed in the task item
[18:22]
gac410okay great. ty I'll edit them into the task. [18:25]
MichaelDaumMichaelDaum made some comments already [18:26]
.... (idle for 15mn)
CDotI just merged the changes from trunk onto Release01x01, and find the WyswiygPlugin tests flaming out all over the show.... WTF? [18:41]
gac410The TMCE and Wysiwyg changes? I don't think they are close to ready. pharvey was working on them. [18:42]
wryfiare there any faq extensions or macros for foswiki?
(tools for making faq's, that is)
[18:43]
CDotFFS - looks like the code has been reformatted on one branch but not the other. They should be *identical*
CDot is now confused as to which branch has been taken as "master"
[18:44]
GithubBot[foswiki] foswiki pushed 1 new commit to master: http://git.io/CG3Diw
[foswiki/master] Item11378: support pass-through of DEL and INS tags - CrawfordCurrie
[18:47]
***GithubBot has left [18:47]
FoswikiBothttp://foswiki.org/Tasks/Item11378 [ Item11378: Strike-out =<del>=, =<s>= tags are lost in WYSIWYG transition ] [18:47]
CDotCDot decides to leave a merge to Release01x01 until after he has spoken to pharvey [18:49]
CDot wandered around the AppleStore while waiting for his car to be serviced, navigating all their computers onto Foswiki.org (as you do). In the process, I noticed how nasty the tagline on the main page is; so I changed it. Any complaints, talk to the hand. [18:59]
GithubBot[foswiki] foswiki pushed 1 new commit to master: http://git.io/mADPGA
[foswiki/master] Item11378: perltidy, and re-enable tests - CrawfordCurrie
[19:00]
***GithubBot has left
wryfi has left "Konversation terminated!"
[19:00]
MichaelDaumCDot, death by git branch [19:02]
CDotmaybe, maybe [19:02]
............... (idle for 1h11mn)
tartanHello foswiki gurus!
Could you please explain a couple of things about Debian installation?
I run Debian Squeeze where I want to install foswiki. I use the _fosiki_ repository, and I'm confused with a question about apache user. Should I type www-data there, which is the user Apache rund as?
[20:13]
OK, for future users: yes, and no password
And disable the short URL option in config generator
[20:23]
........... (idle for 54mn)
***card.freenode.net sets mode: +oooo Babar Lavr Lynnwood SvenDowideit
card.freenode.net sets mode: +oovo mrintegr1ty uebera|| FoswikiBot gmc
[21:18]
btiffinHow do I turn off, site-wide, WYSIWYG editting? [21:20]
gac410set NOWYSIWYG=1 in SitePreferences ... (I think) [21:21]
btiffinThanks, trying. [21:21]
gac410You can also * Set TINYMCEPLUGIN_DISABLE = 1 [21:23]
btiffingac140, NOWYISWYG for the win. [21:26]
................. (idle for 1h24mn)
***flexibeast has left "PING 1330469406" [22:50]
.... (idle for 18mn)
SvenDowideitpharvey, i noticed that if someone has <script> and code in a topic that wysiwyg still gets used
is this a sensible thing?
or should i/we add script (and style?) to the disable wysiwyg regex
[23:08]
pharveySvenDowideit: WYSIWYG doesn't try to do anythign with <script>, and yes it should be protected, raise-a-task [23:09]
SvenDowideitwilko :)
crap. i meant to bid, but was taking girls to day care
http://www.ebay.com.au/itm/120864451188?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649#ht_1158wt_745
its crazy , but clearly they're coming off lease now
[23:09]
btiffinSvenDowideit, pharvey; Yayy. ;) Not that I've tripped on that WYSIWYG issue yet, I'd prefer not to in future. WYSIWYG blows for technical markup, and in my humble opinion, 100% of all cool Foswiki pages are technical. Thanks for the efforts. [23:13]
SvenDowideitbtiffin, hehe
i advocate putting all the techy bits in a separate web
and turning wysiwgy off for it
but ssssshhhhhhh
i should put that in my 'book'
working title? Foswiki: from small beginnings to big data
[23:13]
btiffinOh, I've got a lot of Foswiki-Fu to learn, but now the explorations (with simple text edits) go faster.
Nice title.
[23:15]
FoswikiBothttp://foswiki.org/ from [ WebHome ] [23:15]
SvenDowideitactually - for explorations, i need more - one of those sp(l)it screen textarea+js submit
course, i have to convince pharvey to write a chapter >grin<
[23:15]
.... (idle for 19mn)
mmmm, why do i not know how to set my console font :/ [23:35]
pharveybtiffin: I've got some scratchings of a TML mode for http://codemirror.net/ - apart from enabling codemirror in raw-edit-mode, it'd be cool (I think?) to have it inside WYSIWYG's verbatim blocks
SvenDowideit, that's a damn cheap tablet
[23:48]
SvenDowideitya
and non tablet x61's are now $100ish
and i still can't see a reason to go to the x200's
i have 8GB ram, and thats the biggest limitation for all my use cases
[23:49]
btiffinpharvey: Thanks for the link, checking. And with that, have good everyone, nothing but good. [23:52]
***btiffin has left [23:52]
SvenDowideitpharvey, that tml mode scratching - um, can we have it?
and, can you blog about it - in future tense on foswiki.org/Blogs ?
SvenDowideit has nothing prepared for this week
[23:58]
BabarSvenDowideit: *cough* modules *cough* [23:58]
SvenDowideitBabar, ta for the email, that has put it in the que [23:58]
pharveySvenDowideit: MichaelDaum isn't perfectly happy with codemirror, the kool kids use ACE, but that's IE9+ only and has infinitely crappier doc [23:59]
Babarright. As did the previous poking... [23:59]
SvenDowideitthough really the problem is that i have zero idea where the builds are taking place :( [23:59]

↑back Search ←Prev date Next date→ Show only urls(Click on time to select a line by its url)