We're looking at Tarantula as a replacement to our current test tool. Is there the facility to import requirements, cases etc?
Help get this topic noticed by sharing it on Twitter, Facebook, or email.
Pentaho Kettle can be used for importing test cases:
Doors import could be used to import cases/requirements from .csv file.
Unfortunately it needs to be in specific Doors format, which is not currently documented.
If somebody is willing to test importing, i could write short instructions and provide example file.
Attached csv file example is suitable for importing cases.
Below is example file for importing cases:
Object Level,Object Identifier,Last Modified On,Object Heading,Priority
1,10,25.6.2008,Case 10 heading,
1,11,25.6.2008,Case 11 heading,
1,12,25.6.2008,Case 12 heading,
Object Level: This can be set to 1. This is used to describe hierarchy of objects (rows) in file. As we import only cases here, this can be set to level 1 for all objects.
Object Identifier: This is unique identifier of test cases. For case importing it is only required that is unique for each case. Row will be bypassed, if id is already imported.
Last Modified On: This has no effect for cases.
Object Heading: This will be set as case title.
Priority: This can be left empty.
To import this kind of file into cases, following options needs to be set on Tools - Doors Import screen.
Select Test Case as import target.
Set Level as 1-1 (This means that we will use rows with "Object Level=1" as test cases.
If you choose to simulate import, data is not really imported. This is useful for checking that all options have been set correctly.
Once everyting is all right, press import link.
Here is example of more complex file, for importing both requirements and cases:
"Object Level","Object Identifier","Last Modified On","Object Heading","Object Text","Description of the feature","Priority"
2,"2",25.6.2008,,"Doors Import","It should be possible to import data from doors","P1"
2,"3",25.6.2008,,"Flashing Lights","Front page should have big flashing lights","P5"
2,"5",25.6.2008,,"Robustness","Should be robust as hell","P1"
Here Level 1 rows would contain requirements and Level 2 test cases for these requirements. Import options would be set according to this (Import requirements from level 1 -1, and test cases from level 2 - 2).
This is a bit disappointing. I am currently evaluating test management tools and Tarantula was well ahead (including ahead of some commercial products) because of its ease of use and tagging features that allow tests to be grouped in many different ways. However the fact that I can't import the steps in test cases really kills it off as a candidate. Pity. Now I have to look at inferior candidates.
has someone managed to import test cases with steps by using a (more or less) simple CSV file in the meanwhile? I don't want to have our admins to have to run something on a command-line that "hacks" data into the DB. I need something that runs on user-level front-end.
I played around with exporting stuff and reimporting it. However it doesn't work as it seems to be able to only update existing test cases w/ steps... not to add ("insert") new ones - at least I didn't find the trick.
My approach was to use "export" > "cases" and then use the outputs' CSV structure to create an import file with 1 sample of a new testcase. The import then however complains that it couldn't find an entry with ID nnn (the one I entered).
Giving a null value for the ID doesn't work either, as it will complain as well that it couldn't find the case with id . To me this looks like the logic behind it is working with sql update statements only.
Any clues, pointers?
We are trialing Testia Tarantua.
Despite the little documentation and help available I have managed to get my head around it.
I am having problems with importing a csv file of requirements or cases though.
For this example:
I have added two requirements in Testia
Exported as a CSV and selected the 'requirements' check box.
Opened the exported file in Notepad
Added a new requirement following the same CSV convention identically
Imported the CSV file back into Testia.
Then it is not finding the new requirement ID. Yes I know, I would like you to add it.
Am I doing something wrong here? Any advice appreciated. If I can't Easily import requirements and cases from a spreadsheet or CSV notepad file at least then I can't see this working out for us.
Hi Thomas / Ander,
Thanks for your replies and apologies for not noticing your previous comments to the same effect earlier in this thread. I had read through prior to posting but not maximized those comments.
I have experimented with your suggestions, which make good sense but alas I am still unable to import correctly.
Now I'm basically getting "could not find case/requirement with ID new".
Just to clarify my steps (have tried with both cases and requirements):
1. Created one requirement/case in Tarantula
2. Exported without IDs
3. Opened CSV in Excel, added two new requirements with identical formatting
4. Saved file
5. Import file in Tarantula
6. Receive Fatal Import error.
I'm running the latest version of Testia Tarantula.
I'll post the images of the errors below. Any further help appreciated.
I have thought in use Tarantula in my big project to manage our tests. In this moment we are using test plans in excel. I would like to use this tool, because I have seen the tutorial and I like so much!
But I have a very important precondition , to import all my test cases in excel.
Is there any option to import xls directly from the menu ?
or Is the Kettle script the only way to import the test case, step ...?
thanks in advance!
Hi and sorry for not responding earlier!
We have had our share with imports from CSV too ;) and we discovered that most of the problems with importing are related to the fact that you sometimes need a null-parameter ("") and sometimes "less than null" by using the ;; notation. That depends on the field type.
I will need to check again with my co-worker who managed to solve all these little nasty problems. Will get back ASAP - but perhaps it is a good idea if you could share your Excel file (or the created CSV) so we can give it a try.
AFAIK, Tarantula can not import XLS directly. You will need to export your xls into a plaintext CSV file. The required format for CSV import can be determined by looking at an EXPORT file of Tarantula - basically they use the same structure.
Then it depends if you want to use the "simple" CSV or those processed by Kettle. I can't tell anyting about Kettle unfortunately, but you should be able to somehow create a CSV from your Excel that works with importing into Tarantula.
Geewhiz... I think the Tarantula community really really needs an Import/Export tutorial. We've been thinking about this for quite some time internally already just to save the time spent on repeating about pitfalls and howtos with our own people :) But because of workload I can'gt tell when we're able to pubölish this. Note, I'm not affiliated with Prove Oy...
If I understand correctly, you want to import more than 1 field for "expected result"? That won't work without changing the code and database schema. There is only 1 field for expected results. What is it that you want to achieve... can you give an example perhaps?
We are also trying to implement Tarantula as our test management tool and interested in importing test cases via CSV files and therefore wondering if there are any news about that?
With the Release 2012.36.2 we get pretty much the same results as described by Alex. If required I can provide the CSV file we are testing with.
first of all, I find tarantula great and quite fit for my purpose (mid-size software testing) and am trying to stay with it despite its being relatively immature here and there.
The issue is simply reproducible so I don't provide a minimum example:
In the release 2012.45.0 it is impossible to import test cases with tags into a project which doesn't yet contain these tags. The import stops with a fatal error immediately following the step "Creating a new Case.."
Both the export and import are done using tarantula's native csv-file.
Could you please comment or suggest a workaround? I've tried to create a dummy case with the required label in the target project just to get the label into it before starting the import but this has led to a less obvious fatal error saying "undefined method `project_id' for nil:NilClass" so I've given up.
Obviously lots of test cases are tagged so this issue affects the majority of import operations.
The easiest way to get the correct format for importing test cases is to:
1. Enter a case manually with steps
2. Export the case using the settings: Cases = selected, Recursion depth=1 (Case headers and steps), Export without IDs = selected.
This creates a file that can be reimported immediately. Your options for transforming it are many. My method is to:
3. Open my existing cases and the export file from step 2 in Excel
4. Use text-to-columns with a semi-colon delimiter on the export from step 2
4. Model my existing test case data identically to the export file from step 2
5. Save the test case Excel file as CSV
6. Perform string transformations (search and replace covers my needs) on the data to emulate the original export from step 2.
This is still a bit of a pain, but it makes it usable through the UI which is important for my team, having no Linux gurus.
Seems i get the same error while trying to upload the csv file in the demo site.
I have attached screenshot and other information below.
Screenshot : http://screencast.com/t/pZfyfaFzs
Case Id;Title;Date;Priority;Planned duration (minutes);Test Areas;Average Duration;Tags;Objective;Test Data;Preconditions & Assumptions
new;Testapple;2014-09-05;1;20;;600;;apple objective;apple test data;apple pre
"";new;Step 1a;one apple
"";new;step 2b;two bee
"";new;step 3c;three cat
I was able to Use the file in demo instance but not on my local instance. Can you please provide a test file for me to use on my local instance installed.
I get this issue:
2014-11-19 21:22:50 +0200 [INFO] Starting a new CSV import..
2014-11-19 21:22:50 +0200 [INFO] Found Case header.
2014-11-19 21:22:50 +0200 [INFO] Creating a new Case..
2014-11-19 21:22:50 +0200 [FATAL] undefined method `split' for nil:NilClass
/opt/tarantula/rails/lib/csv_exchange/model.rb:188:in `block in set_csv_attributes'
/opt/tarantula/rails/lib/csv_exchange/import.rb:26:in `block (2 levels) in process'
/opt/tarantula/rails/lib/csv_exchange/import.rb:25:in `block in process'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/actionpack-3.2.8/lib/abstract_controller/callbacks.rb:18:in `block in process_action'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/actionpack-3.2.8/lib/action_controller/metal/instrumentation.rb:30:in `block in process_action'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/activesupport-3.2.8/lib/active_support/notifications.rb:123:in `block in instrument'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/actionpack-3.2.8/lib/action_controller/metal.rb:246:in `block in action'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/journey-1.0.4/lib/journey/router.rb:68:in `block in call'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/actionpack-3.2.8/lib/action_dispatch/middleware/callbacks.rb:28:in `block in call'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/passenger-3.0.15/lib/phusion_passenger/rack/application_spawner.rb:171:in `block in handle_spawn_application'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/passenger-3.0.15/lib/phusion_passenger/spawn_manager.rb:253:in `block (2 levels) in spawn_rack_application'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/passenger-3.0.15/lib/phusion_passenger/spawn_manager.rb:246:in `block in spawn_rack_application'
/opt/tarantula/rails/vendor/bundle/ruby/1.9.1/gems/passenger-3.0.15/lib/phusion_passenger/abstract_server_collection.rb:82:in `block in synchronize'
@Ankush, I was getting the same errors you got. Don't use the file given that works on the demo site. You should follow @Bay's instruction of exporting your test cases and then edit that file in a text editor. Don't use excel. Then it should work in your local instance.