Address "Click" option match number

[elescamilla]

When using the "click" option, the user may be presented with "CSSSelector, 9 matches". However, the trace will only capture one of the matches. That is not reflected to the user when creating the trace in the UI.

For this issue:

• Which of the 9 matches is captured by the Trace? The first one or the one that was selected?
• Modify what is conveyed to the user to clarify that only 1 item will be Traced

[luda171]

It is not true. If the user selects click once and sees that CSSSelector has 9 matches, he should select XPATH.

[luda171]

I did put a workaround for "click until" to click on the first element of CSS selector. We can highlight red the first element in this case.

[martinklein0815]

@elescamilla maybe test and see what needs to be done here?

[talyacooper]

I think the difference between "click all links in an area" and "click element" (which CAN result in multiple elements being clicked by the crawler) is somewhat confusing and hard to document. Here's an example:

For example, here I've captured 3 attempts at capturing all the links on one page of issues: Attempt 1: I do "click once," click on the title of an issue and then select the css option that lists 12 matches. This willcrawl 12 matches

Attempt 2: I do "click all links in an area" and click on the title of an issue (the same place where I clicked above). It doesn't list any matches, so there is nothing to crawl.

Attempt 3: I do "click all links in an area" and click in the wider box around the issue title. That will select 12 issues links, as well as all the tags and authors around them.

Basically, either "click" or "click all links in an area" can be used to select a group of links in a given area. How do we define the difference and guide a user to which is more useful?

FWIW Luda's changes to "click all in area" may have made my initial observation to Emily no longer relevant, but I still feel puzzled over the larger click vs click all functionality.

[luda171]

I would say “click all links in area “ is special simplified not generic macro - we can call it “download all links in area”? It collects urls from href attributes. It is also not disrupting navigation, since download happens later. The “click” action is more generic , and assumes that crawler actually perform click action on the go. It can change state of the page. The elements to click can be anything “clickable” including link. So user need to think more what will happen if he select multiple elements to click. It is also can have sub-events or children events. Yes, in some cases you can produce same results using different action items.

On Mon, Sep 12, 2022 at 2:56 PM talyacooper @.***> wrote:

I think the difference between "click all links in an area" and "click element" (which CAN result in multiple elements being clicked by the crawler) is somewhat confusing and hard to document. Here's an example:

For example, here I've captured 3 attempts at capturing all the links on one page of issues: Attempt 1: I do "click once," click on the title of an issue and then select the css option that lists 12 matches. This willcrawl 12 matches

[image: Screen Shot 2022-09-07 at 12 24 29 PM] https://user-images.githubusercontent.com/54854192/189753248-da1eecac-f80d-4ea6-8d25-07cf9c4a1624.png

Attempt 2: I do "click all links in an area" and click on the title of an issue (the same place where I clicked above). It doesn't list any matches, so there is nothing to crawl. [image: Screen Shot 2022-09-07 at 12 28 06 PM] https://user-images.githubusercontent.com/54854192/189753448-beb68fbd-5ed3-4968-af43-437be7ff6b5c.png

Attempt 3: I do "click all links in an area" and click in the wider box around the issue title. That will select 12 issues links, as well as all the tags and authors around them. [image: Screen Shot 2022-09-07 at 12 27 14 PM] https://user-images.githubusercontent.com/54854192/189754024-3e5039db-8a0a-41ce-a651-c77b154c35fa.png

Basically, either "click" or "click all links in an area" can be used to select a group of links in a given area. How do we define the difference and guide a user to which is more useful?

FWIW Luda's changes to "click all in area" may have made my initial observation to Emily no longer relevant, but I still feel puzzled over the larger click vs click all functionality.

— Reply to this email directly, view it on GitHub https://github.com/lanl/Memento-Tracer-Extention/issues/16#issuecomment-1244458821, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB32EG3HVLU56F27RPHLDHDV56KJRANCNFSM6AAAAAAQATTXXQ . You are receiving this because you commented.Message ID: @.***>

[talyacooper]

This is still very confusing but I think will be best addressed with documentation for the time being.