SwiftHow we calculate
An answer should come with enough information to check it.
Inputs and assumptions
Each tool defines a model and validates its supported inputs. Academic tools distinguish credits, assessment weights, and possible points. Date tools identify endpoint conventions; time tools distinguish unzoned arithmetic from named-zone conversion. Validation catches entry mistakes but does not decide whether a model fits your situation.
Independent arithmetic checks
Calculation engines are separate from the interface. Tests cover known weighted averages, impossible grade targets, zero denominators, leap years, reversed ranges, holiday duplication, ISO week-year boundaries, and time-zone gaps and overlaps. English and Chinese forms call the same engines.
Examples and rounding
Example values are labeled as examples. Calculations keep intermediate numerical precision and round the display, generally to at most four decimal places. This is a display convention, not an institutional rounding policy. Browser date inputs may use the device locale; written results use unambiguous year-month-day notation.
Language and publishing
English and Simplified Chinese pages describe the same model. Switching language does not change the grading scale or select a national holiday calendar. Articles receive a Chinese route only when translated title, summary, and body are available. Published content and tool switches become public together after a versioned site build is deployed.
Corrections
A useful report includes the tool, non-sensitive example inputs, expected output, and a source or hand calculation explaining the difference. References explain specific assumptions and do not imply an endorsement of SwiftCalcKit.