Our Add-in for MS Word failed validation again. They cited that it didn’t work in the Word 2016 Client on Windows 10. But I developed and tested it in W10/O16 and it works for me. I've had several other people test using W10/O16 and it works for them. What could be so different about MS's testing setup?
They informed me they are testing using Azure VM's and provided a snapshot of the console errors they are seeing
POST https://word-edit-officeapps.live.com/we/RemoteUls.ashx?build=16.0.12231.32656&waccluster=IE3 500 MicrosoftAjax.js: 5
I've never seen this error before. They also showed another error I've never seen in any of my tests...
Uncaught TypeError: Converting circular structure to JSON -->
starting at object with constructor 'a' | Property
'bindings' -> object with constructor 'OSF.DDA.BindingFacade'
--- property 'document' closes the cricle
at JSON.stringify (<anonymous>)
at HTMLDocument.<anonymous> (Home.js:229)
at blah blah blah
at blah blah blah
at office.js:41
I look at line 229 in Home.js and it's
228: Office.onReady(function (info) {
229: log("Office.onReady info: " + JSON.stringify(info));
230: isJqready(); // run other functions
231: });
232:
233: Office.initialize = function () {}; // empty initialize function
I was wondering if it is possible that their environment doesn't support ES6? I've coded Home.js in vanilla JavaScript ES6, not TypeScript.
As a test, I've converted my ES6 code to ES5. But I'm hesitant to resubmit not knowing exactly why it failed.
Does anyone have any pointers?
I'm actually working on something similar. I can tell you the POST error is happening on OfficeOnline, not the desktop client so apparently they misreported that. Try running it on Office Online in various browsers.
It definitely needs to be either transpiled to ES5 or written natively in it because earlier versions of Word 2016 (before ~May this year) and in fact Word 2013 in general will need to run it in IE11 which does not support ES6.
Related
I have this very simple 4 lines of code to retrieve a duration value in google sheet from Google Apps Script. The sheet is blank and has this only value for this test.
function myFunction() {
var my_ss = SpreadsheetApp.getActiveSpreadsheet();
var my_sh = my_ss.getSheetByName("test");
var my_value = my_sh.getRange("A1").getValue();
console.log(my_value);
}
When I run this in the new runtime V8, I got the good duration of 5 minutes (with bad GMT but I don't use this). But with the current bug on V8 where we can't use the console to develop arrays and object, I downgraded the script to Apps script Legacy, and when I execute the exact same code (I mean I don't do anything but downgrading the environment with the Run menu), then I got the good GMT but the wrong duration! I got 55 minutes and 39 seconds instead of 5 minutes!
I checked the time zone in apps script V8, apps script legacy and google sheet, it's the same every where, "GMT+01:00 Paris".
Even stranger: this test is actualy performed from a copy of an original file created on august 2019. In the original file I got the opposite behavior. I got the good value with apps script legacy, but with runtime V8 I got a different value (I got 23:14:21 instead of 00:05:00).
Does anyone else encounter this behaviors? Do you know how to run apps script legacy on new files without these problems? Thanks.
As a workaround you can use getDisplayValue() instead :
var my_value = my_sh.getRange("A1").getDisplayValue();
and then, based on this value, you can construct a date object of your preference.
So, I noticed that we had this issue only in the case we upgrade to V8 on a file that has was created before V8 was published (and vice versa). So to avoid this, if I have an old file where I want to upgrade apps scripts to V8, I copy the file first, and in the new file (so a recent one when V8 already exists), then I don't encounter the problem...
I have a problem when I tried to integrate tiki with my LDAP server. In the test_ldap.php I wrote the code to debug according to this website (https://doc.tiki.org/LDAP%20authentication). This code returned a success. So I know that my LDAP is working fine. I have a problem at
$entry = Net_LDAP2_Entry::createConnected($this->_ldap, $this->_entry);
In the shiftEntry function in Search.php. When I return debug in here, it goes to ClassLoader.php and it go to
register_shutdown_function(function () {
TikiLib::events()->trigger('tiki.process.shutdown', []);});
In my opinion, maybe the problem is the
spl_autoload_call
The spl_autoload_call function called Net_LDAP2 which extends PEAR somewhere before this line in Search.php. This would lead both Net_LDAP2 and PEAR and PEAR_ERROR... Then when it comes to Net_LDAP2_Entry class, it would also load PEAR.... Would this create a fatal error?
I keep having blank screen in my tiki. Tried to cut the code from Search.php to the test_ldap.php to test. It has the same problem.
I am using:
Version 16.2
OS: Clear OS
Thanks for your help. I am blocked now.
I think this may be a good start: https://dev.tiki.org/item6283. If you look in the comments by albertgi he states changing some function names in PEAR.php. I was having LDAP integration problems and this was one of the key problems.
Hi I'm using a Confluence macro called 'PockketQuery'(PQ). I have connected to a server located at my client's base through PostgreSQL. I run PQ to fetch results from the db into my confluence page. However, it's fecthing an extra unwanted word "Hallo" along with every result. I m unable to figure out where this string maybe coming from and getting attached to my results like this. Please help me get rid of it.
For example I run a PQ on the db which is supposed to fetch me result "Jack London", so the result that I see is "hallo Jack London".
Note: I use VPN to connect to my client's server and Confluence.
Are you using the latest version from the Marketplace 1.14.6? This issue shouldn't exist in the latest version.
I got an upgrade to version 1.14.6 of Confluence's PocketQuery macro. The issue that I had is resolved, the unwanted string in the result is there no more. The bad part is they don't mention it anywhere on the macro's bug fixes. There are no release notes attached to this fix.Thank you Felix for your help.
I have the following DB Model
At home I work with Visual Studio and SQL-Server on my PC. Now I took the PC at work and tried to create an EFModel using a SQL-Server in our network.
Creating the Model at work I get 44 Errors.
First curious thing (for me) is i get a msgbox with "Running this text template can potentially harm your computer.Do not run it , if u obtained it from an untrusted source..." i confirmed with ok (two times for 2 pop ups). Afterwards I see 44 Errors. After a few minutes I get the same msgbox as above again. After pressing OK a few errors are deleted (now I have 32 Errors).
Since I have the german express version I try to translate the remaining errors by structure:
"Public Property WindesName as String" has many definitions with identical signatures.
variable "_WindowsName" creates a conflict with a property "WindowsName" that is an implicitly declared member of class "Name"
I have this kind of errors for every field.
What I tried: I actually wanted to create the Model into a program, that doesnt work. Then i used a blank application, doesent work too. I reinstalled EF from nuget with no effect.
Maybe it has something to do with the model or with the network. Any help is welcome.
Ok, I was able to fix it. I changed Code Generation Strategy to Standard, deleted the two .tt files and rebuild the project. Now it is working.
I have a simple test setup:
component displayname="test1" extends="test2"
{
remote Any function ping1() {}
}
component displayname="test2"
{
remote Any function ping2() {}
}
When I cal the cfc using test1.cfc?wsdl something strange happens.
When I tell my application to use WSDL version 1 using this.wssettings.version everything looks fine, pretty much the same as in CF9. But when I tell it to use version 2 the function "ping2" does not show up in my definition xml.
Why doesn't WSDL 2 show my extended function "test2" when I call test1.cfc?wsdl ??
I tried this on ColdFusion 10 and both WSDL and WSDL2 do not show the inherited methods. I only see the ping1() function of the test1.cfc, I do not see ping2() inherited from test2.cfc. If you change the "Select web service version" value in the ColdFusion Administrator from 2 (the default) to 1, then ColdFusion 10 performs as ColdFusion 9 does and shows both methods.
On ColdFusion 9, the WSDL shows both methods.
In my opinion, this is a bug in ColdFusion 10, you should log it with Adobe (https://bugbase.adobe.com/), and post the bug number here so others can vote for it. This could break existing functionality if people upgrade without changing the default web services version.
I tried this exact code on CF 9.01 and did not get the same result. My WSDL shows both methods. Something is unique about your CF install - perhaps the wrong version of Axis or something? I'm sorry I can't be more helpful - but I'm struggling to figure out a way to duplicate your issue.