QXmlReader for PyQt5 的读取循环不会返回预期数据

Question

我想做一个QAbstractItemModel，从一系列 Xml 文件中获取数据，这些文件都位于同一个目录中。由于 PyQt5 不再支持 QDomDocument(或者至少我找不到让它工作的方法(，我不得不求助于QXmlStreamReader。我正在将数据本身放入一个巨大的python字典中(嗯...按照计算机科学标准，它并不完全是巨大的(，它包含各种键下的其他字典，以创建一个树状结构。

这是我到目前为止的代码：

class DataModel(QtCore.QAbstractItemModel):
    def __init__(self, settingsDirectory, parent = None):
        super(DataModel, self).__init__(parent)
        settingsDirectory.setNameFilters(["*.xml"])
        files = settingsDirectory.entryList()
        print(files)
        self.data = {}
        for i in range(len(files)):
            filePath = str(files[i])
            file = QtCore.QFile(settingsDirectory.absolutePath() + "/" + str(filePath))
            fileOpens = file.open(file.ReadOnly | file.Text)
            if fileOpens:
                parser = QtCore.QXmlStreamReader(file)
                print("--------Beginning parsing----------")
                print("Reading file: "+str(filePath))
                while not parser.atEnd():
                    parser.readNext()
                    token = parser.tokenType()
                    print("Reading tag: " + str(parser.name()))
                    print("Tag type is: " + str(token))
                    if token == parser.StartDocument:
                        self.data["XML Version"] = str(parser.documentVersion())
                        self.data["XML Encoding"] = str(parser.documentEncoding())
                    if token == parser.StartElement:
                        tokenName = parser.name()
                    if parser.tokenType() == parser.Characters:
                        tokenText = parser.text()
                        print("This tag has a text value: " + str(tokenText))
                        print("current data: " + str(self.data))
                    if token == parser.EndElement:
                        if tokenText != None:
                            self.data[tokenName] = tokenText
                        else:
                            self.data[tokenName] = {}
                        tokenName = None
                        tokenText = None
            else:
                print(self.tr("xml file did not open properly"))
        print(self.data)

虽然这段代码不会崩溃或任何东西，但它确实有一些我不知道为什么会发生或如何解决的问题：

1.由于某种原因，tokenName永远不会从None改变 - 已解决

2.self.data字典的结构不会变成树状的，不知道为什么：|

示例数据：

<?xml version="1.0" encoding="UTF-8"?>
<tag>
    <description>This is a text</description>
    <types>
        <typesAllowed></typesAllowed>
        <typesEnabled></typesEnabled>
    </types>
</tag>

产生最终结果：

{'XML Encoding': 'UTF-8', 'XML Version': '1.0', 'typesAllowed': 'ntt', None: 'n', 'typesEnabled': 'ntt', 'description': 'This is a text'}

而不是通缉的：

{'XML Encoding': 'UTF-8', 'XML Version': '1.0', 'tag': {'description': 'this is a text', typesAllowed': 'ntt', 'typesEnabled': 'ntt'}}

我知道这些问题很可能是我对StreamReader如何工作原理了解不足的结果，因此欢迎任何和所有提示:)

编辑 1：

tokenName更改是一个愚蠢的定位错误，愚蠢的我。 代码反映了修复。

编辑2：

添加了示例和示例输出

Answer 1

这个问题现在已经解决了;我对这个问题采取了不同的方法。

我基本上取了一个列表，如果StartElement令牌具有属性parseAs == "element"，我将tuple的(name, {})附加到其中，并将一个计算的字符串(parseText函数(放入最后一个tuple的字典中。当它遇到一个EndElement令牌时，它会找到带有 name == 的tuple，tokenName (当前令牌的名称(将其作为带有键name的条目放入前一个tuple的字典中。

关于它是如何工作的还有一些细节，但如果我包括它们，我可能会过于复杂地解释(它如何知道何时向self.data提交currData等(。

class DataModel(QtCore.QAbstractItemModel):
    def __init__(self, settingsDirectory, parent = None):
        super(DataModel, self).__init__(parent)
        settingsDirectory.setNameFilters(["*.xml"])
        files = settingsDirectory.entryList()
        print(files)
        self.data = {}
        self.parsingLog = {}
        for i in range(len(files)):
            filePath = str(files[i])
            file = QtCore.QFile(settingsDirectory.absolutePath() + "/" + str(filePath))
            fileOpens = file.open(file.ReadOnly | file.Text)
            if fileOpens:
                parser = QtCore.QXmlStreamReader(file)
                currData = []
                haveStartToken = False
                print(self.tr("--------Beginning parsing--------"))
                print(self.tr("Reading file: "+str(filePath)))
                print(self.tr("---------------------------------"))
                while not parser.atEnd():
                    if not parser.hasError():
                        parser.readNext()
                        token = parser.tokenType()
                        print(self.tr("--------------------"))
                        print(self.tr("Token type: " + str(self.printTokenType(token))))
                        if token == parser.StartElement:
                            tokenName = parser.name()
                            attributes = parser.attributes()
                            parseAs = attributes.value("parseAs")
                            print(self.tr("Reading StartElement: " + str(tokenName)))
                            print(self.tr("parseAs: " + str(parseAs)))
                            if parseAs == "text":
                                textValue = self.parseText(parser.readElementText())
                                print(self.tr("Text Value: " + str(textValue)))
                                if len(currData) != 0:
                                    currData[len(currData)-1][1][tokenName] = textValue
                                else:
                                    print(self.tr("*******Terminating application*******"))
                                    print(self.tr("Reason: currData is empty"))
                                    print(self.tr("*******Terminating application*******"))
                                    sys.exit()
                            elif parseAs == "element":
                                currData.append((tokenName, {}))
                            else:
                                print(self.tr("******WARNING******"))
                                print(self.tr("parseAs attribute is not given correctly"))
                                print(self.tr("******WARNING******"))
                            print(self.tr("--------------------"))
                        elif token == parser.EndElement:
                            tokenName = parser.name()
                            print(self.tr("Reading EndElement: " + str(tokenName)))
                            print(self.tr("currData before: " + str(currData)))
                            if not haveStartToken:
                                startToken = currData[0][0]
                                haveStartToken = True
                            for i in currData:
                                if i[0] == tokenName:
                                    print(self.tr("Closing token: " + str(tokenName)))
                                    if i[0] != startToken:
                                        currData[len(currData)-2][1][tokenName] = currData[len(currData)-1][1]
                                        del currData[len(currData)-1]
                                        print(self.tr("currData after: " + str(currData)))
                                        print(self.tr("--------------------"))
                                    elif i[0] == startToken:
                                        print(self.tr("This is the final token, writing to self.data"), end = "")
                                        self.data[startToken] = currData[0][1]
                                        for i in range(5):
                                            time.sleep(0.25)
                                            print(self.tr("."), end = "")
                                        print(self.tr("done."))
                                        print(self.tr("--------------------"))
                        elif token == parser.Characters:
                            print(self.tr("Characters value: " + str(parser.text())))
                            print(self.tr("--------------------"))
                        elif token == parser.StartDocument:
                            self.parsingLog["File: "+str(filePath)] = {}
                            self.parsingLog["File: "+str(filePath)]["XML Version"] = str(parser.documentVersion())
                            self.parsingLog["File: "+str(filePath)]["XML Encoding"] = str(parser.documentEncoding())
                            print(self.tr("File Version: " + str(self.parsingLog["File: "+str(filePath)]["XML Version"])))
                            print(self.tr("File Encoding: " + str(self.parsingLog["File: "+str(filePath)]["XML Encoding"])))
                        elif token == parser.EndDocument:
                            print(self.tr("Cleaning up"), end = "")
                            for i in range(5):
                                time.sleep(0.25)
                                print(self.tr("."), end = "")
                            time.sleep(0.1)
                            print(self.tr("done."))
                            print(self.tr("self.data: " + str(self.data)))
                            print(self.tr("types of data: yesNo (should be str) - " +
                                          str(type(self.data["building"]["specialSlot"]["yesNo"])) +
                                          " - id - should be int - " + str(type(self.data["building"]["specialSlot"]["id"])) +
                                          " - isItFloat - should be float - " + str(type(self.data["building"]["specialSlot"]["isItFloat"]))))
                            print(self.tr("--------------------"))
                    else:
                        print(self.tr("XML file is not well-formatted"))

            else:
                print(self.tr("xml file did not open properly"))
    def parseText(self, text):
        if isinstance(text, str):
            if text == "":
                return str(text)
            for i in text:
                if i not in ("0123456789."):
                    return str(text)
            for j in text:
                if j not in ("0123456789"):
                    return float(text)
            return int(text)
        else:
            return ValueError
    def printTokenType(self, token):
        if token == QtCore.QXmlStreamReader.NoToken:
            return "NoToken"
        elif token == 1:
            return "Invalid"
        elif token == QtCore.QXmlStreamReader.StartDocument:
            return "StartDocument"
        elif token == QtCore.QXmlStreamReader.EndDocument:
            return "EndDocument"
        elif token == QtCore.QXmlStreamReader.StartElement:
            return "StartElement"
        elif token == QtCore.QXmlStreamReader.EndElement:
            return "EndElement"
        elif token == QtCore.QXmlStreamReader.Characters:
            return "Characters"
        elif token == QtCore.QXmlStreamReader.Comment:
            return "Comment"
        elif token == QtCore.QXmlStreamReader.DTD:
            return "DTD"
        elif token == QtCore.QXmlStreamReader.EntityReference:
            return "EntityReference"
        elif token == QtCore.QXmlStreamReader.ProcessingInstruction:
            return "ProcessingInstruction"