Chrome Driver - Traverse Iframes & Insert HTML

[Marcus Tettmar]

1. This code will not replace the ENTIRE HTML. It will only replace the innerHTML of the BODY. document.body is the body tag.

2. Javascript needs quotes to be escaped. Because of the way you are extracting the body I expect the JS is erroring because there are unescaped quotes. E.g. " needs to be come \"

[Grovkillen]

Again, if this is something people find useful we can turn this into a native function.

Ah, I would very much like that.

[Dorian (MJT support)]

Well, since we were working on this anyway and it was further ahead than I thought, I managed to reprioritise a few things and get an update pushed out today with added native ChromeSwitchFrame and EdgeSwitchFrame functions. Version 15.0.07. Available in the usual places.

Again, if this is something people find useful we can turn this into a native function.

Ah, I would very much like that.

It's already there, Grovkillen, see above.

[Grovkillen]

I thought I was referring to the "ExecuteJS". Instead of having a custom sub routine it would be ideal to have it taken care of with a command.

[Marcus Tettmar]

Code: Select all

Still struggling with this...
CODE: SELECT ALL
GoSub>ExecuteJS,session_id,document.body.innerHTML=\"<p>blah blah</p>\"
Works almost as expected, although instead of replacing the entire page, it actually just replaces everything inside the page body tag.

Correct. document.body.innerHTML = "set the html inside the BODY tag which belongs to the DOCUMENT".

This does not "replace entire document".

You could try setting document.innerHTML which I believe is a new HTML 5 feature supported by SOME browsers. I haven't tried it. It may not work. Give it a go. Another possibility is document.write(newDocument). Again I haven't yet tried it. Give it a go.

Also see:
https://stackoverflow.com/questions/675 ... e-document

But the long and short of it is that you have some javascript code which sets the inner html of the BODY element and seem surprised that only the BODY element is being updated. But that's what the code does.

[Grovkillen]

I would love to have a execute JavaScript command. Together with a " to /" flag. It could then be used almost like the vbscript and python implementation is working now.

[Marcus Tettmar]

I would love to have a execute JavaScript command. Together with a " to /" flag. It could then be used almost like the vbscript and python implementation is working now.

Not sure I follow. Can you explain?

[Grovkillen]

The ExecuteJS sub routine in your code example work just great. But couldn't that be made into a command?

ChromeExecuteJS>sessionID,JSCode

Instead of:

Code: Select all

Let>CHROMEDRIVER_EXE=C:\Program Files (x86)\Macro Scheduler 15\chromedriver.exe

ChromeStart>strSessionID
ChromeNavigate>strSessionID,url,https://www.google.com/
ChromeFindElements>strSessionID,css selector,body,strElementID
GoSub>ExecuteJS,strSessionID,document.body.innerHTML=\"THIS SHOULD BE REPLACING THE BODY INNER HTML?\";


SRT>ExecuteJS
  Let>HTTP_POSTJSON=1
  Let>body= { "script": "%ExecuteJS_Var_2%", "args": [] }
  Let>CHROMEDRIVER_PORT=9515
  HTTPRequest>http://localhost:%CHROMEDRIVER_PORT%/session/%ExecuteJS_Var_1%/execute,,POST,body,theResult
END>ExecuteJS

And what I meant by having a flag for the " is that it would be great to not have to care about adding the \ in front of the " before passing the code to the command.

[Marcus Tettmar]

Yes. It’s on the list already. The more these new functions get used the more ideas/requests/use cases emerge.

[Grovkillen]

Yes. It’s on the list already. The more these new functions get used the more ideas/requests/use cases emerge.

Ah, great!

[Grovkillen]

Is this being implemented Marcus? *fingers crossed*

[Dorian (MJT support)]

I'll ask at MJT Towers.

[Grovkillen]

Thanks!

[Marcus Tettmar]

Hi,

This has been added to v15.0.18:
https://www.mjtnet.com/manuals/v15/HTML ... ecute.html

[Grovkillen]

Thanks!! Will try.