- 29 Aug, 2016 1 commit
-
-
anonym authored
-
- 13 Feb, 2016 1 commit
-
-
intrigeri authored
-
- 11 Feb, 2016 1 commit
-
-
intrigeri authored
-
- 10 Jan, 2016 1 commit
-
- 22 Dec, 2015 1 commit
-
-
kytv authored
-
- 10 Dec, 2015 5 commits
-
-
kytv authored
Instead of seeing the site we're looking for we may have an "eepsite unreachable" error page returned in the browser well before the timeout is reached.
-
kytv authored
-
kytv authored
Previously I tried to check whether I2P is ready by looking for the green star next to the Shared Client Tunnel entry. Unforunately, however, Sikuli couldn't tell the difference between a green star and a yellow one. With recent versions of I2P, there are two shared client tunnel entries, one for the old DSA encryption and one for modern edDSA. When *both* entries are visible, I2P should be ready to use. This should be much more reliable than checking whether port 4444 is open, especially since the static sleep added to work around the upstream bug does not work when the function is called from the shell library.
-
kytv authored
-
kytv authored
After lots of testing I've found that I2P will generally fail within 80 seconds if it's going to fail. Recently I discovered that sometimes, however, it takes a bit longer. The page is still trying to load and we press ESC, the browser may be at a stage in which the URL we were trying to load will not be in the address bar. Pressing reload in this case will be for naught. After a fair amount of testing I have not seen it take longer than 120 seconds. This means that we should always be at a state in which pressing reload will actually be able to reload the page.
-
- 09 Dec, 2015 2 commits
- 11 Nov, 2015 1 commit
-
-
anonym authored
-
- 30 Oct, 2015 1 commit
-
-
kytv authored
-
- 26 Oct, 2015 3 commits
- 04 Oct, 2015 1 commit
-
-
kytv authored
-
- 21 Sep, 2015 2 commits
- 12 Sep, 2015 11 commits
-
-
kytv authored
-
kytv authored
…so they can also be used for the I2P Browser.
-
kytv authored
-
kytv authored
-
kytv authored
Note: Retrying will have to come in a future commit.
-
kytv authored
At the same time, rename the image file to match current TestSuite-standards.
-
kytv authored
-
Tails developers authored
-
Tails developers authored
-
Tails developers authored
- 07 Sep, 2015 1 commit
-
-
anonym authored
-
- 15 May, 2015 1 commit
-
-
anonym authored
-
- 10 Apr, 2015 1 commit
-
-
anonym authored
"Persistent" in the sense that it survives across scenarios. This is a first step towards using a single Libvirt::Domain instance across the full test suite run, which in turn is a step towards being able to use Libvirt snapshots more easily (or at all, I think). This was achieved with the following command: sed -i 's/@vm\>/$vm/g' -- features/step_definitions/*.rb \ features/support/*.rb \ features/support/helpers/*.rb
-
- 10 Feb, 2015 1 commit
-
-
Tails developers authored
-
- 09 Feb, 2015 1 commit
-
-
Tails developers authored
-
- 03 Feb, 2015 1 commit
-
-
Tails developers authored
-
- 17 Oct, 2014 2 commits
-
-
Tails developers authored
-
Tails developers authored
Just hovering a sub-menu (e.g. "Accessories") and waiting a bit is enough for the submenu to open. But we did click on it, and if the sub-menu was already open by hovering, then clicking closed it, of course resulting in FindFailed exceptions to be thrown.
-
- 08 Oct, 2014 1 commit
-
-
Tails developers authored
-