html into R

Hi Rui again sorry to have to ask this but although your code prints out a
tibble I can't seem to be able to identify it ie find its name  I assumed
that it's "y" but outside of your code R tells me that y is not found.
I've tried various things but nothing gives me the tibble as an object with
a name that I can use...?   Thanks Nick

On Fri, 26 Aug 2022 at 13:37, Nick Wray <nickmwray at gmail.com> wrote:

Hi Rui That is brilliant   Thanks v much - what is even better is that I
have loads of data from different years, rivers and stations to download,
each of which entails a different set of numerical inputs and was thinking
about how I could loop through the URL with different inputs to that - but
by using paste I can create all the links I need   Thanks again Nick

On Fri, 26 Aug 2022 at 11:57, Rui Barradas <ruipbarradas at sapo.pt> wrote:

Sorry, there's simpler code. I used html_elements (plural) and the
result is a list. Use html_element (singular) and the output is a tibble.


page |>
    html_element("table") |>
    html_table(header = TRUE) |>
    (\(x) {
      hdr <- unlist(x[3, ])
      y <- x[-(1:3), ]
      names(y) <- hdr
      y
    })()


Hope this helps,

Rui Barradas

?s 11:53 de 26/08/2022, Rui Barradas escreveu:

Hello,

You can try the following. It worked with me.
Read from the link and post-process the html data extracting the

element

"table" and then the table itself.

This table has 3 rows before the actual table so the lapply below will
get the table and its header.


library(httr)
library(rvest)


link <-
"

https://timeseries.sepa.org.uk/KiWIS/KiWIS?service=kisters&type=queryServices&datasource=0&request=getTimeseriesValues&ts_path=1/14972/Q/15m.Cmd&from=2020-01-01&to=2020-01-07&returnfields=Timestamp,Value,Quality%20Code"

page <- read_html(link)
page |>
    html_elements("table") |>
    html_table(header = TRUE) |>
    lapply(\(x) {
      hdr <- unlist(x[3, ])
      y <- x[-(1:3), ]
      names(y) <- hdr
      y
    })


Hope this helps,

Rui Barradas

?s 09:43 de 26/08/2022, Nick Wray escreveu:

hello - I need to download flow data for Scottish river catchments.

The

data is available from the Scottish Environmental protection Agency

body

and that doesn't present a problem.  For example the API beneath will
access the 96 flow recordings on the River Tweed on Jan 1st 2020 at one
station:

https://timeseries.sepa.org.uk/KiWIS/KiWIS?service=kisters&type=queryServices&datasource=0&request=getTimeseriesValues&ts_path=1/14972/Q/15m.Cmd&from=2020-01-01&to=2020-01-07&returnfields=Timestamp,Value,Quality%20Code

But this data comes as HTML.  I can copy and paste it into a text doc
which
can then be read into R but that's slow and time-consuming.  I have

tried

using the package "rvest" to import the HTML into R but I have got
nowhere.

Can anyone give me any pointers as to how to do this?


Thanks Nick Wray

     [[alternative HTML version deleted]]