考虑到您的格式保持不变。这里有一个灵活的解决方案，可配置：

class Separator():
    def __init__(self, data, title, sep, splitter):
        self.data = data # the data
        self.title = title # the starting in your case "<h5>"
        self.sep = sep # the point where you want to update res
        self.splitter = splitter # the separator between key | value
        self.res = [] # final res
        self.tempDict = {} # tempDict to append
    def clearString(self, string, *args):
        for arg in args:
            string = string.replace(arg, '') # replace every arg to ''
        return string.strip()
    def updateDict(self, val):
        if val == self.sep:
            self.res.append(self.tempDict) # update res
            self.tempDict = {} # renew tempDict to append
        else:
            try:
                if self.title in val: # check if it "<h5>" in your case
                    self.tempDict["category"] = [{"key": self.clearString(val, self.title, self.splitter), "value": self.clearString(self.data[self.data.index(val)+1],'<h6>', '|')}] # get the next value
                elif self.tempDict["category"][0]["value"] != self.clearString(val, '<h6>', '|'): # check if it is not the "value" of h6 in "category"
                    val = self.clearString(val,"<h6>").split("|")
                    if "codes" not in self.tempDict.keys(): self.tempDict["codes"] = [] # create key if not there
                    self.tempDict["codes"].append({"key": val[0], "value": val[1]})
            except: # avoid Exceptions
                pass
        return self.res
object = Separator(data, '<h5>', '', '|')
for val in data:
    res = object.updateDict(val)
print(res)

提供的示例输入的输出：

[
    {
        'category': [{'key': '1', 'value': 'Type of Care'}],
        'codes': [
            {'key': 'SA', 'value': 'Substance use treatment'},
            {'key': 'DT', 'value': 'Detoxification '},
            {
                'key': 'HH',
                'value': 'Transitional housing, halfway house, or sober home',
            },
            {
                'key': 'SUMH',
                'value': 'Treatment for co-occurring serious mental health ',
            },
        ],
    },
    {
        'category': [{'key': '2', 'value': 'Telemedicine'}],
        'codes': [
            {'key': 'TELE', 'value': 'TelemedicineTelemedicine/telehealth'},
        ],
    },
]