Web.Page
Reads an HTML web page and returns the tables it contains, so you can pick the one you need.
Web.Page(html)Returns: A table with a row for each table on the page. Its Data column holds each table's rows.
What it does
Web.Page reads an HTML page and finds the tables in it. It is useful for a public table that has no download, such as a published rate schedule or a list of holidays. The page is read as it is, so the query works until the page layout changes.
Syntax and arguments
html: the page content. In practice, pass the result of Web.Contents:Web.Page(Web.Contents("https://www.example.com/rates")).
Example
A published table of standard mileage rates appears on a web page. The query reads the page and takes the first table on it.
let
Source = Web.Page(Web.Contents("https://www.example.com/mileage-rates")),
Tables = Source{0}[Data],
WithHeaders = Table.PromoteHeaders(Tables, [PromoteAllScalars = true])
in
WithHeaders
Source has one row per table on the page, so {0}[Data] takes the first table's rows. The headers are promoted from the first row. The address is a placeholder.
Example: a second, different use
Choosing the table by its position in a list of tables after a quick check of the page. Look at the Source table in the editor to see which row holds the table you need.
Things to watch
- Web pages change layout without notice. Check the row count and the headers after each refresh.
- The table you want may not be the first one on the page. Count rows in the Source table before you choose.
- Check the site's terms before you copy its data into a report. Prefer a published file or an API when one exists.
- Values are text when they come from a page. Convert the types after the headers are right.
In business terms
Published rate tables, fee schedules and public reference data are sometimes only on a web page. Keep a dated copy of what you pulled, because a page can change after the figures were used.