schemaxisCalc operation reference
Generated from the current declarations. Do not edit this file by hand.
Types, limits and defaults below describe resolved values. Runtime expressions must produce these types. Unknown options are rejected; defaults apply only when omitted. Constructor options must be constant after document composition.
Index#
fromRows#
Builds one LibreOffice Calc workbook from a numeric table typed directly into the document: a Data sheet with your column labels and rows, and a Summary sheet with count, sum, mean, minimum, maximum and sample standard deviation per column. It publishes analysis.xlsx, analysis.ods (reopened by a second Calc process) and data.csv, each checked against independent arithmetic. Choose it for small reproducible tables (2 to 10,000 rows, 1 to 64 columns); use readCSV when the table already exists as a file or comes from another workflow step.
Common mistakes and assumptions:
- Every cell must be a finite number within ±1e100; text, booleans, blanks and formulas are rejected (INVALID_TABLE).
- Column labels are only labels: a label such as "Force (N)" does not convert units or affect any calculation.
- Workbook formatting, custom formulas, macros and charts are not produced; only the fixed Data and Summary sheets exist.
Native binding: LibreOffice Calc. Output kind: ["spreadsheet"].
Inputs: none.
| Option | Meaning | Type / constraints | Default |
|---|---|---|---|
columns |
Give 1 to 64 unique literal column labels in data order. Labels are at most 128 characters and can describe units; they do not cause unit conversion. | {"type":"array","items":{"type":"string","minLength":1,"maxLength":128},"minItems":1,"maxItems":64} |
required |
rows |
Provide 2 to 10,000 numeric records, each matching the column count. Row order is preserved in Data and the CSV round trip. | {"type":"array","items":{"type":"array","items":{"type":"number","minimum":-1e+100,"maximum":1e+100},"minItems":1,"maxItems":64},"minItems":2,"maxItems":10000} |
required |
readCSV#
Builds the same Data/Summary workbook as fromRows, but reads the table from an existing comma-separated file: the first line holds column labels and every later line holds numbers. The file bytes are copied into the run so later verification does not depend on the original file. Choose it for measured data already on disk or for a numeric table produced by another workflow step (for example an R or ngspice results table).
Common mistakes and assumptions:
- The path must be absolute; relative paths are rejected during planning.
- The CSV must be UTF-8, comma-delimited, at most 8 MiB, with at least two data rows and no empty, text, NaN or infinite cells.
- Rows must all have the same number of values as the header; there is no padding or column type guessing.
Native binding: LibreOffice Calc. Output kind: ["spreadsheet"].
Inputs: none.
| Option | Meaning | Type / constraints | Default |
|---|---|---|---|
file |
Absolute path of a UTF-8, comma-separated file whose first line holds 1 to 64 distinct column labels and whose remaining 2 to 10,000 lines hold one finite number per column. The file must be a regular file of at most 8 MiB; it is copied into the run directory and read as plain text, never opened as a workbook. | {"type":"string","minLength":1,"maxLength":4096} |
required |