xpath

xpath本身是一套独立的技术，常用于web领域内。

此处uiautomator2也支持xpath，用于元素定位，可以实现复杂条件的元素的查找。

xpath实际用法举例

想要找到百度搜索结果的results下面的button，则对应语法是：

titleButtonSelectorList = d.xpath("//*[@resource-id='results']//android.widget.Button[@clickable='true']").all()

即可找到对应元素的列表

xpath_found_results_btn_list

注意：

分开写

resultsSelector = d.xpath("//*[@resource-id='results']")
titleButtonSelectorList = resultsSelector.xpath("//android.widget.Button[@clickable='true']").all()

却就是找不到元素。

具体原因，暂时未知。

xpath常见操作

定位节点和操作节点

tbsNodeList = self.driver.xpath("//com.tencent.tbs.core.webkit.WebView").all()

xpath的all()返回的是XMLElement而不是XPathSelector

self.driver.xpath("//com.tencent.tbs.core.webkit.WebView")返回的是uiautomator2.xpath.XPathSelector

而加上了all()后

self.driver.xpath("//com.tencent.tbs.core.webkit.WebView").all()

则返回的是：uiautomator2.xpath.XMLElement的list了

即，每一个都是uiautomator2.xpath.XMLElement，而不是uiautomator2.xpath.XPathSelector

其中每一个都是uiautomator2.xpath.XMLElement

然后就可以去操作属性了：

获取属性content-desc的值

eachXmlElem.attrib.get("content-desc", "")

给属性content-desc设置值

eachXmlElem.attrib["content-desc"] = "add something to avoid NAF"

删除一个属性

eachXmlElem.attrib.pop("NAF")

文档

关于Xpath的详细用法，见官网中的：

其内部用的lxml，具体功能和语法都可以参考：

The lxml.etree Tutorial

查找元素 相关函数

findAndClickNode：查找当前节点的父级符合条件的节点 并点击

def findAndClickNode(self, curNodeXpath):
    """
        寻找可以clickable=true的当前或父级元素，并点击

        注：主要用于当节点clickable=false，点击无效时，使用此方法
    """
    foundAndClicked = False
    matchDict = {"clickable": "true"}
    clickableParentNode = self.findParentNode(curNodeXpath=curNodeXpath, matchDict=matchDict)
    if clickableParentNode:
        foundNodeAttrib = clickableParentNode.attrib
        clickableParentNode.click()
        foundAndClicked = True
        logging.info("clicked element [%s] found by [xpath=%s, match=%s]", foundNodeAttrib, curNodeXpath, matchDict)
    else:
        logging.warning("Fail click %s for not found %s(parent) node", curNodeXpath, matchDict)

    return foundAndClicked

调用：

if curNodeXpath:
    foundAndClicked = self.findAndClickNode(curNodeXpath)

相关函数：

findParentNode：寻找父节点

def findParentNode(self, curNodeXpath, matchDict, maxUpLevel=3):
    """
        寻找符合特定条件的父级节点，最多向上找3级

        如果当前节点符合条件，则返回当前节点
    """
    matchNode = None

    try:
        curNode = self.driver.xpath(curNodeXpath).get()
        curNodeAttrib = curNode.attrib # .attrib contain 'clickable'
        # curNodeInfo = curNode.info # .info not contain 'clickable'
        isCurMatch = self.isMatchNode(curNodeAttrib, matchDict)
        if isCurMatch:
            # current is match
            matchNode = curNode
        else:
            # try parent nodes
            curUpLevel = 1
            curParentNodeXpath = curNodeXpath
            while(curUpLevel <= maxUpLevel):
                curParentNodeXpath += "/.."
                curParentNode = self.driver.xpath(curParentNodeXpath).get()
                curParentNodeAttrib = curParentNode.attrib
                isCurParentMatch = self.isMatchNode(curParentNodeAttrib, matchDict)
                if isCurParentMatch:
                    matchNode = curParentNode
                    break

                curUpLevel += 1

    except XPathElementNotFoundError as xpathNotFoundErr:
        logging.error("XPathElementNotFoundError: %s", xpathNotFoundErr)

    if not matchNode:
        logging.warning("Not found match parent for xpath=%s and match=%s", curNodeXpath, matchDict)

    return matchNode

isMatchNode：节点是否匹配

def isMatchNode(self, curNodeAttrib, toMathInfo):
    """判断当前节点属性是否满足条件"""
    isAllMatch = True
    for eachKey, eachToMatchValue in toMathInfo.items():
        if eachKey not in curNodeAttrib:
            isAllMatch = False
            break

        curValue = curNodeAttrib[eachKey]
        if curValue != eachToMatchValue:
            isAllMatch = False
            break

    return isAllMatch

findAndClickTextNode：寻找节点并点击

    def findAndClickTextNode(self, text):
        """
            对于text类型节点：android.widget.TextView, text=xxx
            寻找可以clickable=true的当前或父级元素，并点击


            注：主要用于当text=xxx的节点clickable=false，点击无效时，使用此方法
        """
        curTextNodeXpath = "//android.widget.TextView[@text='%s']" % text
        self.findAndClickNode(curTextNodeXpath)

xpathFindElement：用xpath查找元素

def xpathFindElement(self, curClass=None, curId=None, curBounds=None):
    """
        find element by xpath

        return value type
            is: u2.xpath.XMLElement
            not: u2.session.UiObject
    """
    foundElement = None
    curXpath = self.generateElementXpath(curClass=curClass, curId=curId, curBounds=curBounds)
    try:
        foundElement = self.driver.xpath(curXpath).get()
    except XPathElementNotFoundError as xpathNotFoundErr:
        logging.error("XPathElementNotFoundError: %s from %s", xpathNotFoundErr, curXpath)

    return foundElement

调用：

（1）

foundElement = self.xpathFindElement(curClass=locatorClass, curId=locatorId, curBounds=locatorBounds)

相关函数：

generateElementXpath：生成元素xpath

def generateElementXpath(self, curClass=None, curId=None, curBounds=None):
    """generate element xpath"""
    # nodeXpath = ""
    # if locatorClass:
    #     nodeXpath =  "//%s[@bounds='%s']" % (locatorClass, locatorBounds) # "//android.widget.TextView[@bounds='[191,2060][430,2135]']"
    # elif locatorId:
    #     nodeXpath =  "//*[@resource-id='%s' and @bounds='%s']" % (locatorId, locatorBounds)
    # else:
    #     nodeXpath = "//*[@bounds='%s'" % locatorBounds

    classRule = "*"
    if curClass:
        classRule = curClass # 'android.widget.ImageView'

    propertyRule = ""
    if curId:
        propertyRule += "@resource-id='%s'" % curId
        # "@resource-id='com.netease.newsreader.activity:id/hs'"

    if curBounds:
        if propertyRule:
            propertyRule += " and "

        propertyRule += "@bounds='%s'" % curBounds
        # "@resource-id='com.netease.newsreader.activity:id/hs' and @bounds='[75,2098][141,2134]'"

    # TODO: add other support: text, desc, instance, ...
    curXpath = "//%s[%s]" % (classRule, propertyRule)
    # "//android.widget.ImageView[@resource-id='com.netease.newsreader.activity:id/hs' and @bounds='[75,2098][141,2134]']"

    return curXpath

调用：

curClassname = None
curResId = None
curBoundsStr = None

# curAttrib = foundElement.attrib
# AttributeError: 'UiObject' object has no attribute 'attrib'
if hasattr(foundElement, "attrib"):
    curAttrib = foundElement.attrib
    # {'index': '0', 'text': '', 'resource-id': 'com.netease.newsreader.activity:id/hs', 'package': 'com.netease.newsreader.activity', 'content-desc': '', 'checkable': 'false', 'checked': 'false', 'clickable': 'false', 'enabled': 'true', 'focusable': 'false', 'focused': 'false', 'scrollable': 'false', 'long-clickable': 'false', 'password': 'false', 'selected': 'true', 'visible-to-user': 'true', 'bounds': '[75,2098][141,2134]'}
    curResId = curAttrib["resource-id"]
    curBoundsStr = curAttrib["bounds"]
else:
    # # for debug
    # self.debugPrintElement(foundElement, "no attrib")
    logging.debug("")

curInfo = foundElement.info
# {'bounds': {'bottom': 2134, 'left': 75, 'right': 141, 'top': 2098}, 'className': 'android.widget.ImageView', 'contentDescription': '', 'enabled': 'true', 'focusable': 'false', 'focused': 'false', 'longClickable': 'false', 'packageName': 'com.netease.newsrea....activity', 'scrollable': 'false', 'selected': 'true', 'text': ''}
if not curClassname:
    curClassname = curInfo["className"] # 'android.widget.ImageView'

if not curBoundsStr:
    boundsDict = curInfo["bounds"]
    x0 = boundsDict["left"]
    y0 = boundsDict["top"]
    x1 = boundsDict["right"]
    y1 = boundsDict["bottom"]
    curBoundsStr = "[%d,%d][%d,%d]" % (x0, y0, x1, y1)
    # '[75,2098][141,2134]'

if not curResId:
    if "resourceName" in curInfo:
        curResId = curInfo["resourceName"] # 'com.netease.newsreader.activity:id/bn5'

curNodeXpath = self.generateElementXpath(
    curClass=curClassname,
    curId=curResId,
    curBounds=curBoundsStr,
)

查找元素


    def find_element_Android(self, locator):
        """Android: find element"""
        foundElement = None
        locatorType = locator.get("type")
        locatorText = locator.get("text")
        locatorClass = locator.get("class")
        locatorDesc = locator.get("desc")
        locatorId = locator.get("id")
        locatorInstance = locator.get("instance")
        locatorBounds = locator.get("bounds")

        if locatorType:
            if locatorType == "text":
                foundElement = self.driver(text=locatorText)
            elif locatorType == "desc":
                foundElement = self.driver(description=locatorDesc)
            elif locatorType == "id":
                foundElement = self.driver(resourceId=locatorId)
            elif locatorType == "id+bounds":
                foundElement = self.xpathFindElement(curClass=locatorClass, curId=locatorId, curBounds=locatorBounds)
            elif locatorType == "class+bounds":
                foundElement = self.xpathFindElement(curClass=locatorClass, curBounds=locatorBounds)
            elif locatorType == "id+text":
                foundElement = self.driver(resourceId=locatorId,text=locatorText)
            elif locatorType == "id+desc":
                foundElement = self.driver(resourceId=locatorId,description=locatorDesc)
            elif locatorType == "class+instance":
                # foundElement = self.driver(className=locator["class"], instance=locatorInstance)
                instanceInt = int(locatorInstance)
                # foundElement = self.driver(className=locatorClass, instance=instanceInt)
                foundElementList = self.driver(className=locatorClass)
                if foundElementList:
                    curIdx = instanceInt
                    shouldMaxNumber = curIdx + 1
                    if foundElementList.count >= shouldMaxNumber:
                        foundElement = foundElementList[curIdx]

        if not foundElement:
            if locatorText:
                foundElement = self.driver(text=locatorText)
            elif locatorClass:
                foundElement = self.driver(className=locatorClass)
            elif locatorDesc:
                foundElement = self.driver(description=locatorDesc)
            elif locatorId:
                foundElement = self.driver(resourceId=locatorId)
            elif locatorInstance:
                foundElement = self.driver(instance=locatorInstance)
            elif locatorBounds:
                foundElement = self.xpathFindElement(curBounds=locatorBounds)

        return foundElement

调用：

EditTextClass = "android.widget.EditText"

curBounds = ""
if self.isAndroid:
    curLeft = eachEditTextLocation[0]
    curTop = eachEditTextLocation[1]
    curRight = curLeft + eachEditTextLocation[2]
    curBottom = curTop + eachEditTextLocation[3]
    curBounds = "[%s,%s][%s,%s]" % (curLeft, curTop, curRight, curBottom) # [191,2060][430,2135], '[939,423][1621,558]', '[939,569][1621,701]'
    # foundElement = self.xpathFindElement(curClass=EditTextClass, curBounds=curBounds)
    editTextLocator = {
        "type": "class+bounds",
        "class": EditTextClass, # 'android.widget.EditText'
        "bounds": curBounds, # '[939,423][1621,558]'
    }
    foundElement = self.find_element_Android(editTextLocator)
    # Note: Not use follow wait_element_setText, for it only support type=text
    # isSetOk = self.wait_element_setText(editTextLocator, curInputValue)
    if foundElement:
        foundElement.set_text(curInputValue)

results matching ""

    No results matching ""