Python实现字符串拆分并将Discord表情替换为[name]格式

需求说明 #

• 核心处理目标：将Discord入站消息中的所有表情统一替换为[表情名称]格式
• 输入参考：<a:GG:123456789> <:1Copy:12345678><:14:1256678>:eyes:Hello friend!:eyes:
• 预期输出：[GG] [1Copy][14][eyes]Hello friend![eyes]

初始实现代码 #

最初编写的处理逻辑仅能部分满足转换需求，针对上述参考输入的实际输出为[GG] [1Copy] [14] [eyes]，原始代码如下：

def shorten_emojis(content):
    seperators = ("<a:", "<:")

    output = []

    for chunk in content.split():
        if any(match in chunk for match in seperators):
            parsed_chunk = []

            new_chunk = chunk.replace("<", ";<").replace(">", ">;")

            for emo in new_chunk.split(";"):
                if emo.startswith(seperators):
                    emo = f"<{splits[1]}>" if len(splits := emo.split(":")) == 3 else emo

                parsed_chunk.append(emo)

            chunk = "".join(parsed_chunk)

        output.append(chunk)

    output = " ".join(output)

    for e in re.findall(":.+?:", content):
        output = output.replace(e, f"<{e.replace(':', '')}>")

    return output

实际运行偏差 #

测试用例1 #

• 输入：<a:GG:123456789> <:1Copy:12345678><:14:1256678>:eyes:Hello friend!:eyes:
• 实际输出：[GG] [1Copy] [14] :eyes:Hello friend!:eyes:
• 期望输出：[GG] [1Copy][14][eyes]Hello friend![eyes]

测试用例2 #

• 输入：<a:cryLaptop:738450655395446814><:1Copy:817543814481707030><:14:817543815401439232> <:thoonk:621279654711656448><:coolbutdepressed:621279653675532290><:KL1Heart:585547199480332318>Nice<:dogwonder:621251869058269185> OK:eyes:
• 实际输出：[cryLaptop] [1Copy] [14] [thoonk] [coolbutdepressed] [KL1Heart] Nice [dogwonder] OK:eyes:
• 期望输出：[cryLaptop] [GG] [1Copy] [14] [thoonk] [coolbutdepressed] [KL1Heart] Nice [dogwonder] OK[eyes]

状态说明 #

目前已针对上述代码的逻辑缺陷完成调整，调整后的输出结果可完全匹配预期要求。

内容的提问来源于stack exchange，提问作者Joshua Nixon