DocsWorkflowsComponentsDatadata:shape:read

data:shape:read

Reports the structure of a document a run is holding.

Options

Name	Type	Description
`content`	`string`	The document to report the structure of
`name`	`string`	The file name the document came under, read as a hint where its bytes match no known kind
`format`	`string`	Overrides the detection, one of xml, csv, json, pdf, html or zip
`delimiter`	`string`	The field separator of a delimited document, worked out from the first rows when left out
`max_nodes`	`int`	How many distinct paths the report names
`max_rows`	`int`	How many rows of a delimited document the report samples

Outputs

Name	Type	Description
`format`	`string`	What the document turned out to be
`readable`	`bool`	Whether the structure below is a reading of the document
`complete`	`bool`	Whether nothing was left out of the report
`clipped`	`[]string`	Each bound the reading reached
`notes`	`[]string`	What to know before trusting the report
`nodes`	`[]Node`	The tree, for a document read as an element tree or as nested keys
↳ `path`	`string`	Locates the node from the document root. An attribute is written with a leading at sign on its last segment and an array's elements are written with an empty pair of brackets.
↳ `kind`	`string`	Is what the node is, one of element, attribute, object, array, string, number, boolean or null.
↳ `count`	`int`	Is how many times the path occurred in what was read.
↳ `max_per_parent`	`int`	Is the most occurrences under a single parent, which is what separates a repeating node that becomes rows from a single one that becomes a field.
↳ `types`	`[]string`	Names every scalar type seen at this path where more than one was, which is how a field that is a number in most records and text in a few shows itself before somebody types it as a number.
↳ `sample`	`string`	Is one value from the node, cut to the reading's sample length.
`table`	`Table`	The reading of a delimited document
↳ `delimiter`	`string`	Is the field separator that was used, as one character.
↳ `delimiter_detected`	`bool`	Says the separator was worked out from the first lines rather than stated by the caller.
↳ `header`	`[]string`	Is the first row, which is taken as the column names.
↳ `rows`	`int`	Is how many rows followed the header in what was read.
↳ `ragged`	`int`	Is how many of those rows carried a different number of fields than the header did.
↳ `sample`	`[]string`	Is the first rows after the header, each value cut to the reading's sample length.
`pages`	`Pages`	What a page description document says about itself
↳ `version`	`string`	Is the format version from the file header.
↳ `encrypted`	`bool`	Says the file declares encryption, which is why no content could be read even where a reader for it existed.
↳ `page_objects_seen`	`int`	Is how many page objects were found in the parts of the file that are not compressed, which is a floor rather than the page count. It is zero for a file that keeps its object tree compressed.
`head`	`string`	A short printable sample of the start of a document whose structure was not read

Example

name: Work out what an exported file holds
actions:
  - name: exported
    component: storage:local:file:read
    vars:
      path: export.xml

  - name: structure
    component: data:shape:read
    vars:
      content: "{{ exported.content }}"
      name: export.xml

outputs:
  format: "{{ structure.format }}"
  complete: "{{ structure.complete }}"
  nodes: "{{ structure.nodes }}"

Edit or read the source on GitHub