Repository navigation
EMI: feat: say what each insights number means and drop the padlock from what comes back - #369
Merged
Merged
Conversation
…rom what comes back
…-data' into emi-feat-insights-number-lines
Base automatically changed from
emi-fix-russian-insights-history-data
to
main
October 20, 2026 04:28
# Conflicts: # brand/screens/home-screen-patterns.html # brand/screens/home-screen-patterns.png # brand/screens/home-screen-trend.html # brand/screens/home-screen-trend.png # brand/screens/home-screen.html # brand/screens/home-screen.png
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What. Each number on Insights now has one line under it that says what it is counted from and what it changes for her. The forecast line drops the confidence word, and the "What comes back" card shows the chart mark instead of a padlock.
Why. A beta tester read the numbers and did not know what they said about her, and she read the padlock as a paid feature.
Stack. #363 then this pull request. Merge 363 first.
Forecast, English. Before, after.
Your numbers, English. Before, after.
Chart and What comes back, English. Before, after.
Forecast, Russian. Before, after.
Your numbers, Russian. Before, after.
Chart and What comes back, Russian. Before, after.
The pictures are observed on the iPhone SE (3rd generation) simulator with Metro on this checkout. Before is the base branch at 2889b92. After is this branch before the merge of the moved base: English at a1e1374, Russian at a306441. The phone holds "Maria" with four recorded cycles. The redesign canvas has no board for Insights, so the lines use the existing tokens of the row (body small, secondary ink).
Issue. Part of #324, #324. Step 32 does the My data and export rows.
Draft text for review. Each key in English, Russian and Spanish, old and new.
forecast.confidence.sentencehome.numbers.hershome.numbers.publishedhome.numbers.cycleLength.meanshome.numbers.periodDuration.meanshome.numbers.cycleLengthVariation.meanshome.numbers.linehome.trend.captionhome.trend.allInsidehome.trend.outsidehistory.patternsNeedhistory.patternsCompletehistory.nothingRepeatsforecast.confidence.highforecast.confidence.mediumforecast.confidence.lowDecisions.
lengthDaysis the next start minus this start, and the forecast takes the median of the last six of these lengths (packages/cycle/src/forecast.ts). Last period:periodDayscounts only the period days she logged, and a day with no record neither adds nor ends a period (cyclesFrominpackages/cycle/src/cycles.ts). Variation: one standard deviation of the same lengths, and the range is that many days either side of the middle, at least one (halfWidth).confidenceWordsand the threeforecast.confidence.high/medium/lowkeys are removed.packages/cycle/src/patterns.tsneeds the symptom in 3 cycles within one day of the same point. The Russian line from step 25 gets the same words.Checks.
make checkafter merging the moved base: pictures 124 of 124, workspace 1873 of 1873, application 3660 of 3662. Test 10.7 fails locally because ~/.npmrc prints "always-auth" and "email" warnings. Test 6.2 is the timing ratio, and it passed alone (12 of 12).npm run test:behaviour: 98 of 98.grep -c itvrepos package-lock.jsonprints 0.