#foswiki 2015-07-20,Mon

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

WhoWhatWhen
***wdenk has quit IRC (Ping timeout: 264 seconds) [00:04]
....... (idle for 32mn)
GithubBot[distro] gac410 pushed 1 new commit to master: http://git.io/vmQwy
distro/master 9376d4a George Clark: Item13450: Add template to MANIFEST
[00:36]
***GithubBot has left [00:36]
FoswikiBothttp://foswiki.org/Tasks/Item13450 [ Item13450: SubscribePlugin with famfamfam skin breaks styling when button switches state. ] [00:36]
........................................ (idle for 3h15mn)
GithubBot[UpdateAttachmentsPlugin] gac410 tagged 3.11 at master: http://git.io/vm7f4 [03:51]
***GithubBot has left [03:51]
....... (idle for 34mn)
gac410 has left [04:25]
..................... (idle for 1h43mn)
ChanServ sets mode: +o CDot [06:08]
...... (idle for 26mn)
ChanServ sets mode: +o MichaelDaum [06:34]
............................... (idle for 2h34mn)
GithubBot[distro] MichaelDaum pushed 1 new commit to Item13525: http://git.io/vm5Ox
distro/Item13525 b825ac4 MichaelDaum: Merge branch 'master' into Item13525
[09:08]
***GithubBot has left [09:08]
FoswikiBothttp://foswiki.org/Tasks/Item13525 [ Item13525: Store reads the same file repeatedly from disk ] [09:08]
......... (idle for 40mn)
jastlooks like sourceforge has less than minor issues. perhaps switching to a different provider will need thinking about [09:48]
GzyniuHello, is there someone from Foswiki Developers or anyone I could ask about Single Sign-On in foswiki? [10:02]
jastsure. no promises, though. :) [10:13]
GzyniuI have a website in the same domain as Foswiki, and when I log in to my website, a cookie is created, and I want to use this cookie in Foswiki so I would not have to log in again in Foswiki. In my website I use LDAP, and the same login and passwords are used in both websites (Foswiki and my website) I am using also Webservice, and I see the cookie value and information about session expire from my Website in Foswiki, but I do not kn
Where should I use it? or maybe You have some other ideas how to do this as simple as possible?
[10:21]
***ChanServ sets mode: +o MichaelDaum [10:23]
Gzyniujast, You think You could help? [10:29]
JulianLevensGzyniu: what's you set-up? Which OS and version etc [10:31]
Gzyniulinux and foswiki version is 1.1.6
did you saw my question? I can paste it again
[10:31]
jastwhat kind of single-sign-on is that? standard protocol like CAS or SAML? [10:34]
Gzyniuwell I wanted to try write my own SSO using cookie [10:36]
jastokay
what's in the cookie? something like a session ID?
[10:36]
Gzyniuyes, session id, expire time etc [10:36]
jastif it's your own SSO, you're going to have to write your own login manager for Foswiki, so you better know Perl :) [10:36]
GzyniuI know perl, it's not a problem but is there a simple way to use a cookie to login in foswiki? [10:38]
jastin any case, as a prerequisite, LdapContrib must be set up. I wouldn't start setting up SSO before auth with LdapContrib works
sure. you just have to write your own login manager, more or less like ApacheLogin (shipped with Foswiki)
only instead of looking at the authorization header, you look at the cookie
[10:38]
***ChanServ sets mode: +o SvenDowideit [10:50]
................. (idle for 1h21mn)
ChanServ sets mode: +o gac410 [12:11]
Gzyniuis here someone who can help me with SSO - single sign-on in foswiki and my browser using cookies?
my website*
[12:12]
andreliGzyniu: I read your texts earlier. We go the other way. We use the Foswiki Session to authenticate in some non Foswiki scripts. [12:18]
gac410CDot, Any thoughts about anonymous commenting on foswiki.org. Seems like it would draw spam and we should require authentication. [12:31]
..... (idle for 22mn)
***ChanServ sets mode: +o MichaelDaum [12:53]
CDotI'd say "no" [12:53]
gac410okay I'll disable it.
I probably overlooked a default
I had noticed one of my comments posted as wikiguest ;)
MichaelDaum: CDot anyone else? Any thoughts on whether we are ready to build Foswiki 2.0.1?
[12:54]
CDotas long as you're happy with bulk_copy, go for it. [12:56]
gac410I'd be really happy if someone else could also test it. My test data was broken [12:57]
MichaelDaumgac410, I've got some JQueryPlugin fixes [12:59]
gac410okay
Any thoughts on the cache issues on f.o ... anything worth investigating there for 2.0.1?
[12:59]
MichaelDaumimproved browser detection on logger
gac410, seen my comments in the task item?
MichaelDaum spotted one last reference to tmwiki.org in Request.pm
[13:00]
gac410Browser detection in logger? That probably needs a better CPAN module. I resisted that when I was in the logger ages ago. [13:02]
MichaelDaumMichaelDaum assembled his new desktop mac-hine ... sdd ... lots of ram ... [13:02]
gac410It's pretty complex, not anything I'd want to fix in a patch. [13:02]
MichaelDauma simple regex on the agent string
I know it is more complex i.e. starting with i.e.11
[13:03]
gac410It's almost useless. So many bots use bogus agent strings, why add the overhead
I'd rather log the raw string and let post processing deal with trying to make sense out of garbage.
[13:03]
MichaelDaumokay [13:04]
gac410I thnk we do process it a bit though. I don't recall now. [13:05]
MichaelDaumif at all distinguishing the IEs is probably of some use
CDot, do you need help setting up a profiling script to run foswiki under nytp?
MichaelDaum got one generic enuf ... including virtualhosting
[13:07]
CDotMichaelDaum: attach it to your report, I might get to it later in the week [13:09]
***Gzyniu has left [13:10]
MichaelDaumyou'll need to adjust it to your paths as well as take out virtualhosting ... but it is almost too simple y
ll see
gac410, the X on the broadcastmessage ... outch
could you add some css to make it float to the right please
[13:11]
gac410I could not get your <span...> to work. [13:12]
MichaelDaumall it needs is a float:right to bring it inside the msg div [13:12]
gac410I ended up using JQICON [13:12]
MichaelDaumy [13:13]
gac410It's in a temporary skin override. foswiki.fo.tmpl
I locally tried JQICON ... style="float:right" ... That seems to do it. Okay?
[13:13]
MichaelDaumCDot, okay uploaded it...lotsa room for improvements of course
gac410, anything that helps :)
[13:17]
gac410Sourcforge is a mess. [13:18]
MichaelDaum"error Warning: Updates found for 0 extension(s). tick Upgrade cross Ignore for 7 days" [13:18]
gac410yeah I saw that as well. /me really doesn't like that extension
Oh ... and there are 4 updates pending, but I've not updated because of the configuration overlay issue.
[13:18]
MichaelDaumwe should fix http://foswiki.org/Tasks/Item13524 [13:19]
gac410So not only is it silly wording, it's wrong.
Nothing I've worked on. I looked at the js and gave up.
[13:19]
MichaelDaumin principle the extension is right ... though it probably better would fit as part of configure [13:20]
gac410yeah I'd rather only see it in configure. But who am I to argue with the foswikicamp :) [13:21]
MichaelDaum... or the configure masters [13:22]
CDotNothing to do with configure, is it?
it's the UpdatesPlugin, AFAICT
[13:23]
gac410correct
Just saying given that AdminGroup can't necessarily do anything about extensions, it's rather noise / clutter and really IMO an unwanted "phone home" feature
[13:24]
MichaelDaumCDot, but having it in configure might be better than on the user-facing side
e.g. WordPress only displays any update info on the admin interface
[13:25]
gac410MichaelDaum: I added the style to the message [13:28]
***ChanServ sets mode: +o Lynnwood [13:35]
gac410CDot: ... See feedback posted to Item13516 [13:37]
FoswikiBothttp://foswiki.org/Tasks/Item13516 [ Item13516: bulk_copy.pl has fails to copy some webs, crashes copyng attachment revisions. Unusable. ] [13:37]
gac410Looks like JulianLevens found another issue. [13:37]
CDotconfigure runs as a plugin these days, but still isn't "user facing". No reason UpdatesPlugin shouldn't do the same.... [13:37]
gac410CDot ... the UpdatedPlugin makes the assumption that AdminGroup members can do something about extensions. [13:38]
CDotthe feedback is wrong.....
my tests run fine. I suspect Julian has some mixed-up code.
and ConfigurePlugin only runs for AdminGroup members too.
[13:38]
gac410Actually it only checks admin group if configure AuthorizedUsers is not set.
The problem though is more that 1) it warns about 0 updates, and 2) it reports 0 updates, even when updates are possible.
I suspect it's confused about the extensions report changes.
Actually MichaelDaum_ wrote UpdatesPlugin if git log is to be believed.
Although he might have just been the scribe for the FoswikiCamp
[13:40]
CDotNo idea, I don't think I was involved [13:44]
gac410gac410 was on the wrong side of the ocean ... and in the wrong side of daylight during it's creation :) [13:45]
darn. Configure::Auth is written to throw an error if access not allowed. Not really usable in its current form from UpdatesPlugin::initPlugin
Seems rather high overhead to try / catch for EVERY transaction for non-authorized users.
[13:54]
.... (idle for 19mn)
JulianLevensJulianLevens Checking for mixed up code
Didn't find any, not that that proves anything
[14:13]
gac410, CDot: I added the Dump after my test failed to get more info
My original test was with latest commit: 9376d4a9e
[14:20]
CDotJulianLevens: the reference to the scalar is correct. I use references to scalars to protect data from encoding when passing over the pipe. The function that encodes the message also flattens out the scalar
sub convert, line 99
[14:25]
JulianLevensy, I saw your comment: do you mean utf encoding? [14:26]
CDotyes
things like file names are byte strings already, so don't need encoding
[14:26]
JulianLevensThe failure on L136 is the json->encode [14:27]
CDotthe content of attachments also doesn't need encoding [14:27]
JulianLevensthe temp file name used $tfn is passed as a ref to sub and json->encode is complaining
I have JSON 2.61
[14:28]
GithubBot[distro] gac410 pushed 1 new commit to master: http://git.io/vmFOR
distro/master 75c9bf5 George Clark: Item13504: Add usage notes to UpgradeGuide
[14:33]
***GithubBot has left [14:33]
FoswikiBothttp://foswiki.org/Tasks/Item13504 [ Item13504: Documentation changes Foswiki 2.0.1 / 2.1.0 ] [14:33]
gac410gac410 building a beta 2.0.1 locally, so I can test this from scratch [14:35]
Doesn't work for me. Wide character in print at tools/bulk_copy.pl line 70. [14:41]
***mephinet- has quit IRC (Ping timeout: 248 seconds)
donbarry___ has quit IRC (Ping timeout: 248 seconds)
[14:42]
gac410With no clue as to what's actually wrong.
And that's on a --check run.
with --iweb TestCases
[14:42]
JulianLevensI'm bulk_copying from 2.0 to 2.0 hence $Foswiki::UNICODE is always 1, therefore convert is never called [14:44]
FoswikiBothttp://trunk.foswiki.org/System/PerlDoc?module=Foswiki::UNICODE [14:44]
gac410I'm copying from 2.0 to 2.0 as well. [14:44]
JulianLevensy, 1.1.9 to 2.0 is probably OK
I think convert should always be called and $Fosswiki::UNICODE checks moved inside sub convert
[14:44]
gac410Okay .. I've recreated your issue JulianLevens I removed my test webs, so that it only sees TestCases and it fails
cannot encode reference to scalar at tools/bulk_copy.pl line 136. Processing the Volcano attachment
[14:46]
JulianLevensy, an eruption occurs :( [14:47]
gac410CDot ... something's definitely wrong. I've got 2 failures. One with my test data, can't even get to the --iweb I specified, and then it crashes with the issue Julian if I remove my test data [14:47]
JulianLevensthe $Foswiki::UNICODE test needs moving from L136 to L99 [14:47]
gac410and the wide char in print error I'm encountering needs some hint of what's wrong where. :( [14:49]
JulianLevensneed to be careful there are other conditional ($FW::UNICODE) calls to convert and maybe they all need the conditional removing
Hmm, it's starting to look more complicated; but there is definitely an issue
[14:49]
gac410Okay. The WideCharInPrint is caused because I have a unicode webname. Running with -CS gets the message printed: Skipping not-iweb Litterbox/Úňíčôďě
So that's also a legitimate bug and not invalid test data.
[14:52]
JulianLevensThe convert sub is also handling dereferencing of scalars (needed as a way to stop scalars being utf8 encoded). This is needed every time irrespective of whether $FW::UNICODE is relevant [14:55]
jastdang, Devel::NYTProf::Apache keeps segfaulting on me [14:59]
gac410CDot: Adding at the top: binmode STDOUT, ":utf8"; binmode STDERR, ":utf8"; Resolves the error messages anyway. Still crashes on volcano though.
Minimal testcase here to fail: tools/bulk_copy.pl --iweb TestCases --itopic TestCaseAmISane --check /var/www/foswiki/distro/core/bin ./bin
[15:01]
CDotgac410: "can't even get to the --iweb I specified" - are you sure the web passes the "this is a web" test in the source Foswiki install?
remember that the locale will pervert the current charset used for [:upper:] etc if you haven't enabled locales in LSC.
[15:07]
gac410Yes. The utf8 web was alphabetically lower than TestCases [15:09]
CDotJulianLevens: 2.0 to 2.0? I didn't think of that. Hmmmm. OK, that explains what you are seeing.
gac410: and what are you copying from/to?
[15:09]
gac410From 2.0 utf8 to 2.0 utf8 Skipping not-iweb Litterbox/Úňíčôďě
So it crashes in print before it hits TestCases web.
As I said, either running with perl -CS or adding binmode STDOUT, ":utf8" fixes that particular error.
[15:10]
JulianLevensI've added a $convert flag to sub convert and removed all the conditional calls to convert()
Hold on I'll create a branch and push
[15:11]
CDotJulianLevens: it's OK, I already fixed it [15:12]
gac410Glad we found this one. Probably would have encountered both issues if we ever move foswiki.org to plainfile store.
since it's already utf8 and converted to 2.0
[15:14]
CDotgac410: so have you found something else? Or is Julian's error the only extant one?
(16:10:38) gac410: So it crashes in print before it hits TestCases web.
binmode STDOUT, ":utf8" is always done in Foswiki.pm, but that isn't loaded until the fork so a print might predate it
that's one reason I use a directory path containing unicode when testing
either way, I need to know what "it crashes" means (where)
[15:19]
gac410I've posted the message twice now. 3rd time:
,... - Skipping not-iweb Litterbox/_default
Wide character in print at tools/bulk_copy.pl line 70.
[15:22]
CDotthird time lucky
unfortunately that's useless to me, it's a bit like saying "error in print"
need to know what it was doing when it died
[15:23]
gac410Gawd. It was trying to print: - Skipping not-iweb Litterbox/Úňíčôďě
I have a webname with unicode in it.
[15:24]
CDotso do I, and it works fine :-/
a wide character suggests the binmode was skipped somehow
CDot double-checks the binmode call in Foswiki.pm
aha; the binmode is only applied to STDERR
because STDOUT is used to print the CGI response, so it can't have an IO layer
ok, so, change sub announce to print STDERR instead of print
[15:25]
gac410Or add binmode to bulk_copy ... doubt it will ever run under cgi [15:27]
jastCGI can't use an I/O layer? [15:27]
CDotjast: yes, it can, but an IO layer would get mega-confusing
mainly for REST responses, which come pre-encoded
(JSON generates utf8 by default)
also, an IO layer would banjax binary data
(i.e. viewfile)
so I made the choice of encoded/unencoded explicit (use either $response->print for encoded, or $reponse->body for unencoded)
gac410: yes, I added the binmode.
CDot adds a unicode web name to the roster of tests in gen_bulk_copy_test.pl
[15:29]
.......... (idle for 45mn)
jastJSON doesn't generate utf8 "by default"
the encode_json function does that as an extra step. to_json passes everything through unchanged.
same for decode_json vs. from_json
if you use the OO interface, encoding defaults to off
but yeah, I see the point of not using an I/O layer on STDOUT
[16:20]
.......... (idle for 48mn)
CDotjast: yes, you are right, I should have said "the way it has been used", because most callers had used the UTF8 generation in keeping with the byte orientation of the 1.1.9 core. [17:10]
gac410: I'm done [17:20]
GithubBot[distro] cdot pushed 1 new commit to master: http://git.io/vmbwI
distro/master 45ca0df Crawford Currie: Item13516: fixed problems with bulk_copy between FW2 stores. Added difficult web/topic names to RCS test generator. Corrected doc in PFS (no functional changes)
[17:21]
***GithubBot has left [17:21]
FoswikiBothttp://foswiki.org/Tasks/Item13516 [ Item13516: bulk_copy.pl has fails to copy some webs, crashes copyng attachment revisions. Unusable. ] [17:21]
gac410Okay Thanks CDot ... did you see my changes to the UpgradeGuide ... I tried to document some of the bulk_copy limitations, no idea if I got it right. [17:21]
CDotno...
y, you got it pretty much right
though at some point the concepts of "pub" and "data" will disappear
but for now, it's fine
[17:22]
gac410I also tested the UpdateAttachmentsPlugin, It's working now.
hm no pub or data? That's not good. I'm aware of at least one company whose entire doc distribution strategy depends upon auto attach
Ah... I recall a question. When playing around with checking the store, a lot of the System topics don't have META for attachments - it's pretty hit or miss.
So CDot ... Should we be ensuring that every "distributed" attachment has corresponding META if it's in a position to be seen by Store (ie excluding the subdirectory stuff)
[17:23]
CDotyes, that is strongly preferred
if nothing else, the comment= should be populated
however it is not always realistic, if a topic has gazabillizillions of attachments
[17:29]
gac410hm We'll need to automate that. Maybe it should be part of the commit checker. Even the gazabillzillions are in good shape. iirc, famfam didn't kick out as a problem.
Maybe I'll look into a hacked UpdateAttachmentsPlugin rest handler to try to pick up the missing attachments. skipping files not in a MANIFEST
or just attach them all, and then remove the ones missing from manifest
[17:30]
...... (idle for 29mn)
hm CDot. bulk_copy and UpdateAttachmentsPlugin are not happy with each other. I ran UpdateAttachments against System.JQueryPlugin ... it autoattached everything. and then bulk_copy --check fails all attachments
ex: Attachment jquery-1.11.1.js has corrupt meta-data - cannot copy
Not sure what it doesn't like
[18:02]
CDotno version= in the META: probbaly [18:03]
gac410Whoops. Empty version in the Meta. hm...
I assume it really should insert version 1?
[18:03]
CDotit should insert whatever is at the head of the history.
so if there's only one version, it should be version 1
[18:04]
gac410Okay. So updateAttachmentsPlugin doesn't look for history. It's assuming that the files are new.
Hm. since it's reading outside of the store API, I suppose it could do a rlog if a ,v exists. But the primary use case is someone saving a file into pub from some external source,
So probably better to just ignore history. And maybe document that it is not intended to support attachment revisions.
[18:05]
....... (idle for 30mn)
MichaelDaum_: have you tested MetaCommentPlugin on 2.0? Issues reported in Foswiki:Support/Question1634 [18:36]
FoswikiBothttp://foswiki.org/Support/Question1634 [ MetaCommentPlugin is not working in Foswiki 2.0.0 ] [18:36]
gac410Jomo's test table says issues with substr, but not sure what that's all about [18:36]
disc0techHi - I've just installed foswiki on a ubuntu server under apache for the first time, can anybody help me out with an install issue? Been through the excellent docs / google. already
I can get to <mysite>/bin/view just fine, but all the next level of links are broken. E.g. I get a web server error NOT FOUND: The requested URL /System/WebHome was not found on this server.
[18:43]
gac410disc0tech: Are you using the debian packages (version 1.1.9) or the new Foswiki 2.0.0 tarball [18:43]
disc0techtarball
used apt-get only for the perl dependencies
[18:44]
GuilainCand what's the issues you encountered ? disc0tech ? [18:44]
gac410he just said :) [18:44]
disc0techNo errors in the install / configure process
just web broken links
[18:44]
GuilainCoops
GuilainC re-read
did you use the apacheconfigurator ?
do you run under apache ?
[18:44]
disc0techI do run under apache
I used the apache web config generator
[18:45]
gac410disc0tech: if bin/view/System/WebHome is working, but /System/WebHome (without bin/view) is failing, it seems like the short url configuration is messed up. [18:45]
disc0techand can confirm it is loaded at runtime (via httpd -S)
good point re short utls
[18:46]
gac410when you used ApacheConfigGenerator, did you specify to use Short URLs? [18:46]
disc0techyes I enabled it..
should I disable and test do you think?
[18:47]
gac410And when you first visited foswiki, did you visit a consistend view URL. ie. if you specified short URLs in ApachCfgGenerator, then your first visit should have simply been yoursite.com/Main/WebHome
or /System/WebHome ... either should work.
[18:47]
disc0techI just tried loading: http://mysite/bin/view/System/WebHome
and it worked fine
[18:48]
gac410So right now, if you enter into url http://yoursite.com/System/Webhome does that work or give a 404? [18:48]
disc0tech404 [18:49]
gac410okay. So that's saying something is wrong in the ApacheConfigGenerator.
results
If you omit the System/WebHome ie just enter http://mysite/ does that work?
These are related to the Alias and RewriteRules in the Apache configuration
[18:49]
disc0techYes it did take me to the first page, but subsequent links were broken. Am just trying to tweak to disable shorturls via the generator but I seem to have regressed a bit... [18:51]
gac410Alias / ... handles the case of the url without web/System
okay. So the alias is correct
hm That alias statement *should* allow the case of the .../System/Webhome to work. The Alias / ... gets apache to handle the / using bin/view
[18:51]
disc0techI've broken my alias somehow now though, redirecting to the wrong site...
:(
[18:54]
gac410Should be Alias / "/var/www/data/Foswiki-1.2.0/bin/view/" (replace the /var/www... with the full path to the view script including trailing /)
(cut/paste from an old config on my test system. I have not caught up with the renumbering of the release :) )
[18:55]
GuilainCto my point of view, the best disc0tech, is to reload the topic apacheConfigurator with your pref (you should get it at the top of the config file generated) and then check carefully all the different aspect
try to debug is well too, if you want to progress in your comprehension of apache :D
[18:56]
disc0techok thanks for the hints, I'll let you know [18:57]
gac410+1 just regenerate apache. And if the starting config was messed up, remove lib/LocalSite.cfg and redo the bootstrap as well.
Bootstrap really depends upon the URLs you use to get to the default foswiki landing page to be correct.
And clicking the link in the bootstrap banner passes along the detected url info to configure.
[18:57]
disc0techSo I regenerated the ApacheConfigurator with short urls disabled, regenerated LocalSite.cfg with ShortURLs=none
Now I can get to the welcome page that gives me lots of link options (We can't predict your webserver configuration etc), but none of the links work. Previously the /bin/view link worked.
[19:03]
GuilainCcould you copy paste the link to your apacheconfigurator ? [19:04]
disc0tech...and I bounced apache and now everything seems to be working! [19:04]
GuilainCah...
:)
[19:05]
gac410So ShortUrls should work just fine as well. Everything just needs to be consistent between Apache and Foswiki configs.
probably most use short urls if I could hazard a guess.
[19:05]
disc0techIs there a way to re-enable "New User Registration"? The admin credentials I set as part of the command line configuration don't see to work. [19:11]
gac410How did you set from command line tools/configure -save -set {Password}='yourpassword' ?
User registration should work
[19:12]
disc0techI ran the whole configure script from the command line [19:12]
gac410Ah. okay. Try setting the admin password again using what I just typed. [19:13]
disc0techWhen I go to System/UserRegistration, I see - New user registration is disabled on this site.
okay
[19:13]
gac410Ah. That's disabled if email is not enabled iirc [19:13]
disc0techyes I just saw that it is disbaled in v2 in the docs
I'm pretty sure I set the password correctly, I can still see it in the terminal buffer
So I guess the admin user is not enabled, and I've been changing browsers so I can't create a user
[19:14]
gac410try the explicit set rather than the command line prompting. [19:14]
disc0techOK will do [19:14]
gac410We did not do lots of commandline testing. Most focus on bootstrap has been the web UI. [19:15]
disc0techyay, works [19:15]
gac410okay. then bin/configure email tab, it should be able to autodetect and configure email Is your ubuntu server already working with email, for ex from cronjobs?
If so, then from email tab, put in admin email address and hit autoconfigure. It should find the underlying os sendmail if present
[19:15]
disc0techyep I can get into configure now
yes I have email on the server
[19:16]
gac410okay, so setup *should* be extremely simple. Just one field - enter email address and autocongif, then save
note that we have some serious bugs, especially in bulk_copy.pl, so migrating existing webs from an old setup is challenging.
we are probably pretty close to building 2.0.1 Still trying to test
[19:17]
disc0techtest email success :)
I'm not migrating - new install, evaluating for internal company wiki
[19:19]
gac410okay great. [19:20]
GuilainCdisc0tech, an advise totaly objectif
stop to evaluate
keep it and say it's wonderfull
and foswiki do all what we want
even the coffe
coffee
it will save your time
:D
more seriously, did you evaluate in comparison with other wikis ?
[19:23]
disc0techSo an intern did an evaluation, I got an installation request
just getting it up and running for them
[19:25]
GuilainCok ! [19:27]
disc0techthanks folks, all working, signing off [19:33]
.......... (idle for 48mn)
gac410MichaelDaum ... Should data/System/NatEditPluginJSUnitTests.txt be listed in the MANFIEST ?

(it's not) ;)
[20:21]
.............. (idle for 1h7mn)
***ChanServ sets mode: +o MichaelDaum
ChanServ sets mode: +o MichaelDaum
[21:29]
..................... (idle for 1h42mn)
GithubBot[distro] gac410 pushed 3 new commits to master: http://git.io/vmA7k
distro/master 0f280bd George Clark: Item13541: Don't return directories as attachments
distro/master 24f2b5f George Clark: Item13541: Don't try to write an empty attachment...
distro/master e365cf0 George Clark: Item13504: Fix up attachment data...
[23:13]
***GithubBot has left [23:13]
FoswikiBothttp://foswiki.org/Tasks/Item13541 [ Item13541: eachAttachment returns directory names with PlainFileStore, breaks bulk_copy.pl ]
http://foswiki.org/Tasks/Item13504 [ Item13504: Documentation changes Foswiki 2.0.1 / 2.1.0 ]
[23:13]

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