Add download flag to click action, update dom logic (#310)

skyvern/forge/prompts/skyvern/extract-action.j2

@@ -16,6 +16,7 @@ Reply in JSON format with the following keys:
         "id": int, // The id of the element to take action on. The id has to be one from the elements list
         "text": str, // Text for INPUT_TEXT action only
         "file_url": str, // The url of the file to upload if applicable. This field must be present for UPLOAD_FILE but can also be present for CLICK only if the click is to upload the file. It should be null otherwise.
+        "download": bool, // Can only be true for CLICK actions. If true, the browser will trigger a download by clicking the element. If false, the browser will click the element without triggering a download.
         "option": {  // The option to select for SELECT_OPTION action only. null if not SELECT_OPTION action
             "label": str, // the label of the option if any. MAKE SURE YOU USE THIS LABEL TO SELECT THE OPTION. DO NOT PUT ANYTHING OTHER THAN A VALID OPTION LABEL HERE
             "index": int, // the id corresponding to the optionIndex under the the select element.

skyvern/webeye/actions/actions.py

@@ -52,9 +52,7 @@ class DecisiveAction(Action, abc.ABC):
 class ClickAction(WebAction):
     action_type: ActionType = ActionType.CLICK
     file_url: str | None = None
-
-    def __repr__(self) -> str:
-        return f"ClickAction(element_id={self.element_id}, file_url={self.file_url})"
+    download: bool = False
 
 
 class InputTextAction(WebAction):
@@ -162,7 +160,14 @@ def parse_actions(task: Task, json_response: List[Dict[str, Any]]) -> List[Actio
             actions.append(TerminateAction(reasoning=reasoning, errors=action["errors"] if "errors" in action else []))
         elif action_type == ActionType.CLICK:
             file_url = action["file_url"] if "file_url" in action else None
-            actions.append(ClickAction(element_id=element_id, reasoning=reasoning, file_url=file_url))
+            actions.append(
+                ClickAction(
+                    element_id=element_id,
+                    reasoning=reasoning,
+                    file_url=file_url,
+                    download=action.get("download", False),
+                )
+            )
         elif action_type == ActionType.INPUT_TEXT:
             actions.append(InputTextAction(element_id=element_id, text=action["text"], reasoning=reasoning))
         elif action_type == ActionType.UPLOAD_FILE:

skyvern/webeye/actions/handler.py

@@ -142,11 +142,30 @@ async def handle_click_action(
 ) -> list[ActionResult]:
     xpath = await validate_actions_in_dom(action, page, scraped_page)
     await asyncio.sleep(0.3)
+    if action.download:
+        return await handle_click_to_download_file_action(action, page, scraped_page)
     return await chain_click(
         task, page, action, xpath, timeout=SettingsManager.get_settings().BROWSER_ACTION_TIMEOUT_MS
     )
 
 
+async def handle_click_to_download_file_action(
+    action: actions.ClickAction,
+    page: Page,
+    scraped_page: ScrapedPage,
+) -> list[ActionResult]:
+    xpath = await validate_actions_in_dom(action, page, scraped_page)
+    try:
+        await page.click(
+            f"xpath={xpath}", timeout=SettingsManager.get_settings().BROWSER_ACTION_TIMEOUT_MS, modifiers=["Alt"]
+        )
+    except Exception as e:
+        LOG.exception("ClickAction with download failed", action=action, exc_info=True)
+        return [ActionFailure(e, download_triggered=False)]
+
+    return [ActionSuccess()]
+
+
 async def handle_input_text_action(
     action: actions.InputTextAction, page: Page, scraped_page: ScrapedPage, task: Task, step: Step
 ) -> list[ActionResult]:

skyvern/webeye/scraper/scraper.py

@@ -26,6 +26,7 @@ RESERVED_ATTRIBUTES = {
     "aria-role",
     "aria-selected",  # for option tag
     "checked",
+    "data-original-title",  # for bootstrap tooltip
     "data-ui",
     "for",
     "href",  # For a tags
@@ -373,7 +374,7 @@ def _trimmed_attributes(tag_name: str, attributes: dict) -> dict:
             new_attributes[key] = attributes[key]
         if key == "role" and attributes[key] in ["listbox", "option"]:
             new_attributes[key] = attributes[key]
-        if key in RESERVED_ATTRIBUTES:
+        if key in RESERVED_ATTRIBUTES and attributes[key]:
             new_attributes[key] = attributes[key]
     return new_attributes