 |
I can NOT reproduce this issue using build 130213. You can test here: http://www.limesurvey-templates.com/20/index.php/survey/index/sid/129549/newtest/Y/lang/en Nevertheless I am pretty sure that this bug exists since it was reported twice. My system details:
Using an older build I can also NOT reproduce the problem. System details are:
|
|
|
Here is a link to my test in situ - the issue is pretty clear. http://www.nrgstrategy.com/total-energy-survey/index.php/129549/lang-en |
 |
I can't reproduce the problem with the survey or at the nrgstrategy website. If I enter a number with a decimal, it is fine. However, if I try to add a second decimal (like 12.34.56), the whole number disappears as soon as I type the second decimal. That is expected, since each time you type a character, JavaScript removes non-numeric keystrokes and checks whether the value is a valid number. The function that does this is fixnum_checkconditions() |
|
|
Thanks for all your time. Marcel's version also works fine for me, but my version I cannot get to work on any of three machines and the major PC browsers. Here is a short screen video to show you the issue I am facing. http://www.screencast.com/t/J017hXBim Just so you know, the main survey I am building requires measurement units to be entered in either cms or inches. some people will definitely need/ want to use decimals, but it cannot be forced - and resorting to just integers will make it a very blunt instrument indeed! Just to clarify the issue is with a single decimal entered in a cell without any validation string, and then also with a validation string as that also does not work for me (and perhaps the two are connected). |
 |
camills, I cannot reproduce it here locally, either. However I can see the problem on the nrgstrategy website. To rule out that you for some reasons are using old LimeSurvey files please update manually (= NOT using ComfortUpdate) to the latest build (do it even if it is the one you are already running). Before trying again please clear your browser cache, too. It is not a issue of the PHP version, I checked that one. |
|
|
I have now manually reinstalled the whole of limesurvey on my server. Same result. Please check the link at your end to see if there is any change just in case. I guess my only hope unless you can help is to move my survey to a different server. That will have to wait until next week. |
|
|
Trying on a different server is your best bet. |
|
|
Feedback please? |
|
|
Thanks for the offer - I have just now moved the survey to a new host (infrenion) from the previous one (bluehost). The decimal issue appears to have resolved itself and now works fine. There are other issues with some scripts (Mazi is looking into this, as they are his)but it does look as though the issue I reported was caused by something in the bluehost setup. If you want to look into this further I am happy to supply any further details or to give you access to the bluehost ftp. For my part - as i know time is tight for all of us - I will just continue with my new host set up and not set up any further surveys on bluehost! |
|
|
In general I would still like to know the cause of this so if you can provide the access data , that would be great. |
|
|
Pretty please? |
|
|
camills? |
|
|
Colin will provide you with the required information for debugging this weird issue the next days. |
|
|
The issue is caused by a buggy version of the GMP library(4.3.2) used by LimeSurvey to be able to handle large decimal numbers with high precision. |
|
|
Fix committed to master branch: http://bugs.limesurvey.org/plugin.php?page=Source/view&id=11790 |
|
|
Good catch! |
|
|
Fix committed to 2.05 branch: http://bugs.limesurvey.org/plugin.php?page=Source/view&id=11800 |
|
|
2.00+ build 130311 released |
- Anonymous
