Re: [android-developers] Compare MonkeyImage

2011-03-10 Thread aniruddha dhamal
Thanks Bill, That really helps a lot. On Mon, Mar 7, 2011 at 4:09 PM, Bill Napier nap...@google.com wrote: Yes. Our tests are written on top of pyunit (which runs inside monkeyrunner). We use sameAs and an asset. For the test cases we are developing, sameAs (with a percentage) is

Re: [android-developers] Compare MonkeyImage

2011-03-07 Thread aniruddha dhamal
@bill, You mentioned that you compare the subsequent snapshots. I have questions: Do you use *sameAs* method as an assertion method (like Assert in JUnit) to verify pass or failure of the testcase? How do you say that this percentage have changed with the next screenshot so that testcase is

Re: [android-developers] Compare MonkeyImage

2011-03-07 Thread Bill Napier
Yes. Our tests are written on top of pyunit (which runs inside monkeyrunner). We use sameAs and an asset. For the test cases we are developing, sameAs (with a percentage) is sufficient. This allows things like the time in the status bar and any other possible notifications to get ignored. If

[android-developers] Compare MonkeyImage

2011-02-15 Thread Christopher M. Judd
According to the monkeyrunner documentation, you can compare screenshot output to known screenshot for regression testing. How do you do this? I see the method sameAs on MonkeyImage but that takes a MonkeyImage presumably one saved on disk for regression testing but there appears to be no way to

Re: [android-developers] Compare MonkeyImage

2011-02-15 Thread Bill Napier
Hi Christopher, Being able to load an image into MonkeyRunner from the filesystem makes a lot of sense, I don't see how I had missed that. All the test's we've been using MonkeyRunner before only ever wanted to compare to subsequent snapshots (to see what has changed). This looks pretty