Partial reading a webpage?

[Pured]

I am looking to create a script which refreshes/reads a webpage every few seconds. My goal is to see if the page has changed, then I will send myself a notification that the webpage has been updated.

 

However, rather than downloading the entire webpage every single time, is there a way to check when the webpage last updated?

 

If not, is there away to partially download/read html source until a specific tag is hit?

 

Goal: I would like to increase my poll rate and not excessively waste data.

[Nine]

There is multiple ways to achieve this.  Provide the URL of the page.  And show DOM of the specific part of the page you want to monitor.  What is it that you need to look for ?

[mLipok]

Do you want to use IE or WebDriver , Which browser ?

Edited April 6, 2021 by mLipok

[Pured]

@Nine Any site really... Whether it be a patch note for a game being posted (no rss) or a specific market listing being posted, e.g. https://www.mercari.com/jp/search/?keyword=トゥエニ

I'm not sure how to find a page's dom.

 

@mLipok IE is fine (what I usually use).

[Nine]

5 minutes ago, Pured said:

I'm not sure how to find a page's dom

Just right click on an element of the web page and select "Inspect" in the context menu.

If you are looking at any change in a page, it would be quite easy to do with a generic solution.  If you looking for a specific change in a specific tag (or part of the web page), I am afraid it is going to be very hard to find a generic solution.  You will need to adapt your code to each individual page.

[Pured]

@Nine Going through this forum, I know I can get the size of a page, but it seems like I have to download the entire page rather than just fetch the size.

 

It would be nice if there is a way to fetch the date modified or size of a webpage without getting anything else, since I don't care too much about what has changed (I hope...). 

 

Otherwise like you said, using doms (I have other scripts which traverse through html and to tag-specific jobs), so I can do that myself, but it would require a full load of the page, which I don't want.

[TheXman]

4 hours ago, Pured said:

It would be nice if there is a way to fetch the date modified or size of a webpage without getting anything else, since I don't care too much about what has changed

@Pured

If that is all you want, then just retrieve the web page's header.  That will only include information about the page and not the page itself.  The Content-Length header will give you the size of the content or if it exists, you can look for the Last-Modified header.  Depending on the framework used by the web page, looking at those headers may not be a very accurate way to tell if the content has changed.

There are several ways to get just the header.  If I were looking for the header, then an easy way would be to execute a simple cURL command or send a HTTP HEAD request using one of the numerous ways to do HTTP requests.

Example header:

HTTP/2 200
server: nginx
date: Tue, 06 Apr 2021 13:44:06 GMT
content-type: text/html;charset=UTF-8
content-length: 198416
vary: Accept-Encoding
x-powered-by: PHP/7.2.34
x-ips-loggedin: 0
vary: cookie,Accept-Encoding
x-xss-protection: 0
x-frame-options: sameorigin
expires: Tue, 06 Apr 2021 13:45:06 GMT
cache-control: max-age=60, public
pragma: public
set-cookie: ips4_IPSSessionFront=lfntg0ifj1a53lnh9s4vvvndpt; path=/; secure; HttpOnly
set-cookie: ips4_guestTime=1617716646; path=/; secure; HttpOnly
last-modified: Tue, 06 Apr 2021 13:44:06 GMT
x-powered-by: PleskLin

 

13 hours ago, Pured said:

Goal: I would like to increase my poll rate and not excessively waste data.

Be warned, continuously "banging" on a website, very frequently, will most likely get your IP address blocked if the host has any type of decent anti-hammering protection.  That type of behavior is usually frowned upon.  😉

 

Edited April 6, 2021 by TheXman

[samibb]

Dear 

i am new in Http API and i need your support 

i try this code but it did not gave me the head.

by the way this page need user ID and Password.

#include "WinHttp.au3"

Opt("MustDeclareVars", 1)

; Open needed handles
Global $hOpen = _WinHttpOpen()
Global $hConnect = _WinHttpConnect($hOpen, "https://intranet.airlineretailing.amadeus.com:55001")
; Specify the reguest:
MsgBox(0, " $hConnect",  $hConnect)    ;" works gave ID"
Global $hRequest = _WinHttpOpenRequest($hConnect, Default, "/1ASIHGAPSV/arp/index.html")
MsgBox(0, "$hRequest",  $hRequest)       ;" works gave ID"
; Send request
_WinHttpSendRequest($hRequest)

; Wait for the response
_WinHttpReceiveResponse($hRequest)

; Get full header
Global $sHeader = _WinHttpQueryHeaders($hRequest)

; Close handles
_WinHttpCloseHandle($hRequest)
_WinHttpCloseHandle($hConnect)
_WinHttpCloseHandle($hOpen)

; Display retrieved header
MsgBox(0, "Header", $sHeader)

 

Thank

[TheXman]

@samibb

You've been on this forum long enough to know better than to try to "hijack" a thread with your unrelated issues.  If you have questions about using the WinHTTP.au3 functions, then post your questions in the WinHTTP UDF topic or start a new one.

As it relates to the example that you posted, if you added @error checking then you would have seen that the _WinHttpReceiveResponse failed.  The most likely reason that it failed is because you tried to send an https request without using the $WINHTTP_FLAG_SECURE flag in your _WinHttpOpenRequest.

Look at the reply below for an example of a _WinHttpOpenRequest that is setting up a secure request using WinHTTP.  Note that you were a member of this topic.    There are numerous other examples of sending https requests using the WinHTTP UDF lib.

here's another one that gives a brief explanation of why you need the flag for secure connections:

If you have additional questions, then post them in a proper topic, not one that is asking about how to "partially read a webpage".

Edited April 23, 2021 by TheXman