WebWatcherScan or pick
This page is for anyone at step 2 of the Which element? assistant who is deciding between the automatic scan list and Pick in Safari. Both end with the same result, a selector and a watch type, so the choice is only about which is quicker for your page. The full walkthrough of the assistant is in Finding the element.
#Compare the two
| Helper | How it works | Choose it when |
|---|---|---|
| Scan | WebWatcher reads the open tab when the page step finishes and lists what looks watchable in three groups. Rescan runs it again. | You are not sure what the page offers, or the element is a normal badge or counter. |
| Pick in Safari | You click the element in the page and move the outline with the arrow keys. | The scan does not list your element, it lists the wrong thing first, or you want one item in a list. |
#Scan
The scan needs no input from you. It groups candidates under "Showing a number now", "Could get a badge later" and "Other", and each row has a Use button and an eye button that highlights the element in Safari.
Scan looks for notification-style counts. When the page has none, the list is replaced by a note that points to Pick in Safari. For a price, a block of text or an element that may appear, use Pick in Safari, or look in the Other group.
#Pick in Safari
Pick in Safari puts you in control of the exact element. You click it, adjust the outline with ↑ ↓ ← →, then press ⏎ or click Use this. Choose it for a status area, one row of a list, or a value the scan would not call watchable. It also lets you point at an icon that shows no number yet, and then choose "Anything changes inside it".
#Where both stop
Neither helper can reach an element in a cross-origin iframe or inside a closed shadow root. Pick in Safari tells you about an iframe with "That element is inside an embedded frame WebWatcher can't reach — pick something outside it." Closed shadow roots give no message: the element is never listed or outlined. The fix for each is in the "If it does not work" table of Finding the element.