View Issue Details

This issue affects 1 person(s).
 6
IDProjectCategoryView StatusLast Update
06400Bug reportsSurvey editingpublic2012-07-30 14:36
Reporterabita1 Assigned Toc_schmitz  
PrioritynormalSeveritypartial_block 
Status closedResolutionwon't fix 
Product Version1.92+ 
Summary06400: 1) Short Free Text Question breaks if Map is set, and 'Display Rows' is non-blank 2)...
Description

1) Short Free Text Question breaks if Google Map is set, and 'Display Rows' is non-blank

I assume the intent of the question type is to allow a SFT question, or a Map, or both. I didn't notice, in the documentation, a statement to that effect, or to the contrary.

If SFT-MAP is not broken [ie, at least in the fashion it does work: the display row is blank, and there is a Map Set], but the question is set mandatory, the Q expects an answer to the text box, which does not appear.

Steps To Reproduce

Create a short free text question, and add a google map [yes]
Leave the display row line blank.
Map shows up, without a blank space to enter text. As long as the question is not mandatory, this should be no big deal.

Change the display row line to a number: 1 or 0. In either case, the map disappears on survey execution. It does not appear in the generated code

Additional Information

Originally reported as forum inquiry
See: http://www.limesurvey.org/en/forum/can-i-do-this-with-limesurvey/83945-google-map-suddenly-stops-working-on-short-free-text-question#83991

Error occurs within 120721, I believe, but latest testing was with the 1.92+ variant that was last generated on 7-26 by Tom White and stored on his Git [nominated 0720 for some reason...]

Same error occurs on 2.00RC7 branch

TagsNo tags attached.
Bug heat6
Complete LimeSurvey version number (& build)120721
I will donate to the project if issue is resolvedNo
BrowserFireFox 14
Database type & versionMySQL
Server OS (if known)Linux / Apache
Webserver software & version (if known)1.3.3
PHP Version5.2.17
Git branch / Pull Request URL

Users monitoring this issue

There are no users monitoring this issue.

Activities

tpartner

tpartner

2012-07-29 16:12

partner   ~20059

I don't think this is a bug.

Entering a value in the display rows input converts the answer element from an input[type="text"] to a textarea (to allow for rows) and disables the map feature - this is expected behaviour.

The map feature will only work with an input[type="text"] and when enabled the input[type="text"] is used to record the map coordinates only, not other text.

abita1

abita1

2012-07-30 02:15

reporter   ~20061

Ah...
That is interesting,

Setting aside the utility of allowing both, if the user asks for it -- If your description is expected:

1) Then, one part of the bug is both a documentation hole, and the lack of a cross check for the behavior I accidentally landed on.

2) Then the second part of a bug, remains -- i.e., that the selection of mandatory still expects an entry in the non-displayed text box ? And won't proceed to subsequent questions [which, to me implies there is a separate sQ code; see below]. I would need to look back for the text input in my generated code, but I think this is what happens.

3) There probably should be some consideration to the consequences of 0, 1 or blank/default, as to what form of box is created -- documentation or some note on the entry fields?
3a) In any case, as I note, below, why eliminate the map if called for. Seems like that would craft a slew of code logic that isn't really necessary; the user gets what he asks for in the HTML generated code: A: [Yes/No on the map] and B:[Text vs TextArea on the textInput, dependant on the DisplayRows, with some documented for default/blank, 0, 1, n]


There is some certain utility to having both a text input and a map


Why have an editable long/lat, if it is a graphic/map entry system? But, I'm ok with that being there.


If your expected behavior is the case . . . why not just segregate the two question behaviors into...well...a SFT, and a MAP -- two QTypes. That way, at least the logic mess I discovered can't happen. Again, while I like that there is the possibility of a text answer, along with a map answer to the question....that implies TWO subq's and appropriate subQCodes, and more, to make all the evaluation etc codes work, a certain tangle of code improvement and documentation. None of the codes are detailed in the documentation; though I am suspecting it already exists.


I do notice that there are some other returnable variables [city, etc..] though there is nothing about those in the documentation. Perhaps it is already plausible that a slew of subcodes already exist for the map data, separate from any Text data. In that case, why not just eliminate the restriction you identified, so that the HTML generated code puts in a map if the the designer/user so states.

tpartner

tpartner

2012-07-30 11:59

partner   ~20070

abita1, please feel free to update the documentation as required - it's a wiki.

Carsten, your thoughts on this? I can't reproduce the problem of missing input fields and the question behaves as I would expect.

c_schmitz

c_schmitz

2012-07-30 14:36

administrator   ~20071

I agree to Tony. It surely can be optimized but it is not a bug per se. Nothing to see here, please move along.

Issue History

Date Modified Username Field Change
2012-07-28 18:36 abita1 New Issue
2012-07-29 16:12 tpartner Note Added: 20059
2012-07-29 16:12 tpartner Assigned To => tpartner
2012-07-29 16:12 tpartner Status new => feedback
2012-07-30 02:15 abita1 Note Added: 20061
2012-07-30 02:15 abita1 Status feedback => assigned
2012-07-30 11:58 tpartner Assigned To tpartner => c_schmitz
2012-07-30 11:59 tpartner Note Added: 20070
2012-07-30 14:36 c_schmitz Note Added: 20071
2012-07-30 14:36 c_schmitz Status assigned => closed
2012-07-30 14:36 c_schmitz Resolution open => won't fix
2019-11-01 17:25 c_schmitz Category Survey design => Survey editing