Closed tbranyen closed 9 years ago
is "host" (I presume dns host) the right level of granularity, or should it be author/provider (there may be more than one on a site, and more than one site to an author)?
On 12/14/2014 4:20 PM, Tim Branyen wrote:
I added [data-host] to the rows to make it easier to select a particular host with the DOM. This made the selectors much simplier in the tests.
I changed the moment humanizer formatter to show more accuracy. Before it was 'a few seconds', now it's '5 seconds'.
This also adds in a test asserting basic tracking. I adjusted the watcher code to always start tracking a new tab in the onUpdated event.
You can merge this Pull Request by running
git pull https://github.com/tbranyen/tipsy adjust-idle-calculation
Or view, comment on, or merge it at:
https://github.com/haystack/tipsy/pull/42
Commit Summary
Ensure that pages are tracked with more precision
File Changes
- M shared/scripts/lib/components/log-table/log-table.html https://github.com/haystack/tipsy/pull/42/files#diff-0 (4)
- M shared/scripts/lib/index.js https://github.com/haystack/tipsy/pull/42/files#diff-1 (24)
- M shared/scripts/lib/watcher.js https://github.com/haystack/tipsy/pull/42/files#diff-2 (4)
M test/integration/tests/watcher.js https://github.com/haystack/tipsy/pull/42/files#diff-3 (51)
Patch Links:
- https://github.com/haystack/tipsy/pull/42.patch
- https://github.com/haystack/tipsy/pull/42.diff
— Reply to this email directly or view it on GitHub https://github.com/haystack/tipsy/pull/42.
I'll definitely need to identify more specific targets as well, I'll probably be doing this next so I can assert times are correct for two different authors on a given site.
For now though this ensures correct behavior for simple tracking in both Chrome and Firefox.
These tests are a huge pain to write, but I'm going to go through every combination I can think of and then migrate the tests and library into a separate repository.
I added [data-host] to the rows to make it easier to select a particular host with the DOM. This made the selectors much simplier in the tests.
I changed the moment humanizer formatter to show more accuracy. Before it was 'a few seconds', now it's '5 seconds'.
This also adds in a test asserting basic tracking. I adjusted the watcher code to always start tracking a new tab in the onUpdated event.