I upgraded my tiki from 15.4 to 16.2. Everything runs fine. The problem is the site registration does not work. I have a tracker to extend my user information. In the tracker, I created a field called "Network" that allows the users to choose a group that them want to be in. Hence, they won't see each other content due to different group permission. This registration broke. The error message showed:
Cannot add nonexistent user xxx to group students
I found out the problem was that there is a change in trackerlib.php line 3671. Instead of returning "null", it now returns an empty array().
else {
return array();
}
In GroupSelector.php, line 124, the condition check
if (empty ($creators))) $creators=array($user);
This if will never be executed since the function now return an array not null. Hence causing the problem that no one can be registered if they choose a group in the registration form.
Propose solution, change the if condition in line 124 to:
if (empty($creators[0]))
My fix has now been improved by our colleague kroky6 (thanks!) and is now in the 16.x branch, so you can test it by using the 16.x nightly tarball from here tomorrow and will be in 16.3 (and 17.0 onwards) coming soon...
Did you check the actual 16.x version (dev branch) of this file (lib/core/Tracker/Field/GroupSelector.php) ?
https://sourceforge.net/p/tikiwiki/code/HEAD/tree/branches/16.x/lib/core/Tracker/Field/GroupSelector.php
Chances are, this is fixed already. :)
Bernard
Related
I'm seeing some very strange behavior with FileMaker 14. I'm using LayoutObjectNames for some required functionality. On the development system it's working fine. It returns the list of named objects on the layout.
I close the file, zip it up and send it to the client, and that required functionality isn't working. He sends the file back and I open it and get a data viewer up. The function returns nothing. I go into layout mode and confirm that there are named objects on the layout.
The first time this happened and I tried recovering the file. In the recovered file it worked, so I assumed some corruption had happened on his end. I told him to trash the file I had given him and work with a new version I supplied. The problem came up again.
This morning he sent me the oldest version that the problem manifested in. I confirmed the problem, tried recovering it again, but this time it didn't fix the problem.
I'm at a loss. It works in the version I send him, doesn't on his system. We're both using FileMaker 14, although I'm using Advanced. My next step will be to work from a served file instead of a local one, but I have never seen this type of behavior in FileMaker. Has anyone seen anything similar? Any ideas on a fix? I'm almost ready to just scrap the file and build it again from scratch since we're not too far into the project.
Thanks, Chuck
There is a known issue with the Get (FileName) function when the file name contains dots (other that the one before the extension). I will amend my answer later with more details and a possible solution (I have to look it up).
Here's a quote from 2008:
This is a known issue. It affects not only the ValueListItems()
function, but any function that requires the file name. The solution
is to include the file extension explicitly in the file name. This
works even if you use Get (FileName) to return the file name
dynamically:
ValueListItems ( Get ( FileName ) & ".fp7" ; "MyValueList" )
Of course, this is not required if you take care not to use period
when naming your files.
http://fmforums.com/forums/topic/60368-fm-bug-with-valuelistitems-function/?do=findComment&comment=285448
Apparently the issue is still with us - I wonder if the solution is still the same (I cannot test this at the moment).
Is there a REST API endpoint to get a collection of changes that are pending for a build in TeamCity?
We have the build set to manual and it is triggered outside TeamCity and would like to show a bullet point list of commits that'd be in that build.
In the user interface you can see this under the "Pending Changes (X)" tab.
I can't find any examples of doing this and the closest I've found is:
http://<server>/httpAuth/app/rest/changes/buildType:<build type id>
This seems to return the last change though.
Anyone done this before?
I just found a working solution thanks to this question. I'll show it here in case other people are looking for a full solution :
You need to know the buildTypeId of the build on which you want to get the pending changes. In this case lets say buildTypeId=bt85
1
http://<server>/httpAuth/app/rest/buildTypes/id:bt85/builds/
// Get the last build from the XML returned.
// Lets say last build id = 14000
2
http://<server>/httpAuth/app/rest/changes?build=id:14000
// The newest change returned is the one you need.
// Lets say newest change id = 15000
3
http://<server>/httpAuth/app/rest/changes?buildType=id:bt85&sinceChange=15000
// You're now looking at the pending changes list of the buildType bt85
My eventual solution in a work around kind of way is to:
Find the latest change ID from my database of builds outside of TeamCity (I guess you could query the TeamCity API to find the last successful build and pull it from there)
Then call:
http://<server>/httpAuth/app/rest/changes?buildId=id:<build id>&sinceChange=id:<last change id>
Then fetch each individual change from that list.
A bit of a workaround but I couldn't see anyway otherwise to get the list of pending changes.
This is my 1st question on this forum... So, please, be indulgent !
I'm using TYPO3 4.7.11 (PHP 5.3.3) with extension direct_mail 3.1.1 for the intranet site of a non-profit firm.
My problem (maybe connected to Bug #51583 : http://forge.typo3.org/issues/51583) is that, after numerous tests and attempts, it seems impossible to have an updated version of a page saved as draft for newsletter in an automatic scheduler driven way : the same newsletter is produced with the same informations that were already there on the day it was first created and saved.
The specific page used for newsletter includes a content element 'Menu/Sitemap' with 'Recently updated pages' as 'Menu type'. It has been saved as 'draft (for recurring sendings)' in Direct Mail.
The scheduler contains these 2 tasks with recurring type :
- Direct Mail: Create Mail from Draft (direct_mail)
- Direct Mail: Mailing Queue (direct_mail)
Note : the manual way is fully functional and the newsletter produced is really updated. Same with option "Testmail - Simple" !
So, my problem seems to be linked to the automatic scheduled mailing ! It looks as if the newsletter draft has turned into a freezed snapshot of a specific moment and that Typo3 is unable to update/recalculate this page when invoked in scheduler mode.
On the web, I saw reported problems that could be related like "When mails get sent via the scheduler the same subject is used for all sendings ( https://review.typo3.org/21313 )" and "Adding hooks when sending direct mails via scheduler ( forge.typo3.org/issues/48994 )", but these issues seem to be fixed with direct_mail 3.1.1 version.
I made these observations and, in my opinion, there is some relevancy :
1.There is no domain proposed in the 'Domain of internal links' drop-down list in 'Set default values for mail content fetching options' in Direct Mailer, and yet I have a single record in sys_domain table with a domain name (with no protocol and no final slash). Is there a reason why this record is not considered good, or isn't it the right table ? (uid=3, pid, tstamp, crdate, cruser_id, hidden, sorting, prepend_params and forced=0, redirectHttpStatusCode=301, domain_name=site.subdomain.domain, redirectTo=)
2.In the Typo 3 Log, I get this systematic error message for user _cli_scheduler#LIVE :
Core: Error handler (BE): PHP Warning: Invalid argument supplied for
foreach() in
...typo3conf/ext/direct_mail/Classes/Scheduler/MailFromDraft.php line
125.
The concerned part of MailFromDraft.php is this function : initializeHookObjects
...
/*
* Initializes hook objects for this class
*
* #return void
*/
function initializeHookObjects() {
foreach ($GLOBALS['TYPO3_CONF_VARS']['SC_OPTIONS']['direct_mail']['mailFromDraft'] as $hookObj) {
$hookObjectInstance = t3lib_div::getUserObj($hookObj);
if (is_object($hookObjectInstance) && ($hookObjectInstance instanceof x_directmail_Scheduler_MailFromDraftHook)) {
$this->hookObjects[] = $hookObjectInstance;
}
}
}
...
I'm not sure of understanding very clearly the origin and the use of the hook Object... (in spite of this interesting article by Robert Lemke : typo3.org/documentation/article/how-to-use-existing-hooks-in-your-own-extension/ )
3.Nothing like the apparently requested GLOBALS['TYPO3_CONF_VARS']['SC_OPTIONS']['direct_mail']['mailFromDraft'] seems to exist in TYPO3_CONF_VARS (Global configuration).
Can anybody give me an advice or a clue about what's going on and why I can't get a weekly updated newsletter with the scheduler ? I feel a bit confused !
Thanks in advance for any suggestion or solution (if a miracle is possible).
Greetings.
P-H SILLIAU
I've read about this issue before, but couldn't remember where.
Googling "direct_mail draft (for recurring sendings)" helped.
Try this bug: http://forge.typo3.org/issues/4421
User Markus says:
Things work fine, when you set a domain-record in your system and
select it in the direct_mail settings!
If you don't have a domain-record and specify it in the direct_mail
setup you're able to send normal newsletters, but if you try the
draft-functionality it won't work because the getUrlBase function in
class.tx_directmail_static.php returns an unsuable URL to the System
so it can't use the fetchHTML($file) and quits - therefore not
replacing the old draft contents created when starting the first time.
I don't really get why this works the first time you set up the draft
though....
So setting up the domain-record is a work-around that works.
I hope it does!
Probably, you will find more related topics.
Else, workarounds would be
re-considering the task. As it's a NPO intranet, maybe requirements are not that required suddenly, if asked again :-)
setting up a custom notification tool that only does that precise job.
To put an end to this problem, after many attempts (and informations gleaned from the internet), here is the solution we finally used in our specific case to make the newsletter work :
1st. We created a record in table sys_domain. This was a recurrent instruction in the manual and the forums, and it was IMHO legitimate.
Important : note that the redirectTo field had to remain empty, due to global malfunction of the site if filled (whatever we put in it like /, /var/www/sitename, ...)
2nd. All images, CSS, JS included in template had to be hardcoded (i.e. http ://site/fileamin/images/xxx.png for instance). If we didn't do that, the result would have been an abort in the production of the newletter : Not Found... Maybe, by digging a bit deeper, should we be able to find a parameter we forgot or neglected in some way to solve this issue...
3rd. In the newsletter template TS setup, we added these 2 parameters :
mod.web_modules.dmail.use_domain=[uid of sys-domain]
config.absRefPrefix = / (in order to get rid of PHP DOCUMENT_ROOT (or TYPO3_DOCUMENT_ROOT ?) otherwise wrongly present in all generated links.)
The result is now a well dynamically-generated newsletter, the date is OK, all links are correct and realUrl-compliant (no ../index.php?id=nnn) .... and you know what ?... We're happy ! :-)
Hope it will help !
Many thanks to everybody who answered (Markus, Urs...) or even thought of a possible solution...
P-H Silliau
We have had a PowerShell script scheduled and executing successfully for the past 3-4 months (In both Test and Prod). The purpose of the script is to update document properties in SharePoint when certain triggers are fired from external systems. Without getting into too much detail, below is the code that has been used to update item properties for a document that has been declared a record:
$recordsmanagement=[Microsoft.Office.RecordsManagement.RecordsRepository.Records]
$recordsmanagement::UndeclareItemAsRecord($item)
$item = $list.GetItemById($item.id)
$item.File.CheckOut()
$item[$sSpFieldName]=$sDbValue
$item.Update()
$item = $list.GetItemById($item.id)
$item.File.CheckIn("")
$recordsmanagement::DeclareItemAsRecord($item)
This code has worked hundreds of times without a problem. For some reason, this code started bombing a week ago on the last line (when re-declaring as a record):
System.Management.Automation.MethodInvocationException: Exception calling "DeclareItemAsRecord" with "1" argument(s): "The file /lib/folder/file.pdf has been modified by SHAREPOINT\system on 10 Oct 2012 00:00:47 -0500."
The other weird part is that this is only happening in Prod. The Test environment seems to execute just fine. I haven't tried a fix for production yet, but I'm pretty sure I can just get the $item object again using GetItemById (after the CheckIn). I'm a little hesitant to do this just yet as I wanted to get some other people's perspective first.
Does anyone have any input on this? Thanks in advance.
I think the best bet is to get the item again after the checkin as you say.
The error message indicates exactly that. You are trying to perform an operation on a SPListItem that has been modified. So pull it again using GetItemById before you declare it as a record.
Why it only happens on some records and on some environment I am not sure. I guess Sharepoint is a bit temperamental.
Thing I would try as well:
Check if there is any workflow doing some work on that item when you check it in
Try to use SystemUpdate() instead of Update() if that suits your requirements
Good luck
It seems kinda silly that this feature wouldn't be there. I feel like I am overlooking something obvious, but I haven't been able to Google any answer successfully.
#LyndenShields's answer doesn't seem to work anymore.
At this date (March 16, 2015), the way to search for issues that have no assignee is to use the no: qualifier in the issues Filter field:
no:assignee
Alternatively, you could use the following query string to achieve the same:
https://github.com/<username>/<repo>/issues?q=no%3Aassignee
EDIT: Note that I could not make it work if I only had one qualifier specified (ie. you may need to add another one and have it like no:assignee is:open).
EDIT #2 (March 16, 2016): the above edit is not accurate anymore (hence the strikethrough); GitHub now accepts search queries with one qualifier only.
EDIT #3 (July 26, 2017): removed the %01 from the end of the URL - it basically is an unused character, but I believe GitHub interpreted it as a space. Indeed, when you use it this way, GitHub will automatically add a space, I guess to make sure other qualifiers can easily be added. Strangely, though, it will add a %20 to the end of the URL today, but again, you can simply omit that part. The %3A between the words no and assignee is the character code for Colon. See a full list here.
EDIT #4 (June 30, 2020): this is now possible via the UI: simply go to your issues list (https://github.com/<username>/<repo>/issues) then choose Assigned to nobody from the Assignee dropdown:
Go to the issues list and show issues assigned to yourself. Look in the URL bar and change where it has your username to none just after 'assigned/' and press enter. The UI will look the same but the list will have only unassigned issues.
e.g.:
https://github.com/<username>/<repo>/issues/assigned/none
One workaround would be to compare:
the list of issues
with the list of issue assigned to anyone
(the difference being the list of issues not yet assigned)
See GitHub V3 Api "list-issues-for-a-repository" section:
assignee
String User login
none: for Issues with no assigned User.
*: for Issues with any assigned User.
2021 Update.
In the search bar you must type no:assignee and this will show all Github issues without an individual assigned to it.