[erlang-questions] page counter - thx for everyones help

[Wes James]

On Fri, Apr 30, 2010 at 4:17 PM, Colm Dougan <colm.dougan@REDACTED> wrote:
> Wes,
>
> It would be typical to move the "extract page count" code into a
> function.  Depending on how happy you are depend on the format of the
> page you are scraping you could use something like this (see below).
> Note: I've used the default "http" library here rather than "ibrowse"
> as I don't have ibrowse installed.  If you decide to use "re" instead
> then you only have to change the extract_page_count function.
>
> Typically lists:nth isn't very erlangy.  It would probably be more
> typical (and more tolerant to the page changing) to iterate over Lines
> looking for a given pattern.
>
> Colm
>
> -define(MAGIC_LINE_NUMBER, 424).
> -define(URL, ""http://192.168.1.1/printer/configu.html?autoref=0&weblang=0").
>
> getc() ->
>        inets:start(),
>        {ok, {_StatusLine, _Headers, Body}} = http:request(?URL),
>        Lines = string:tokens(Body, "\r\n"),
>        PageCount = extract_page_count(lists:nth(?MAGIC_LINE_NUMBER, Lines)),
>        io:format("COUNT [~p]~n", [PageCount]),
>        ok.
>
> extract_page_count("<B>Page Counter</B></TD><TD>" ++ Rest) ->
>        {Count, _} = string:to_integer(Rest),
>        Count.

Colm,

Thx! I like this a lot better without having to use ibrowse.

I appreciate this info!

-wes