studyzy/imewlconverter · error · InvalidDataException

有异常的词库，解析失败

Error message

有异常的词库，解析失败

What it means

BaiduBdictImporter.ReadOneWord reads a 4-byte little-endian length (number of pinyin/character units in one entry). A genuine bdict entry is short; if len exceeds 1000 the parser assumes the stream is misaligned or the file is corrupt/not-a-real-bdict and aborts with InvalidDataException.

Source

Thrown at src/ImeWlConverter.Formats/BaiduBdict/BaiduBdictImporter.cs:58

        while (input.Position < endPosition)
        {
            ct.ThrowIfCancellationRequested();

            var entry = ReadOneWord(reader, input);
            if (entry == null)
                break;
            if (entry.Word.Length > 0 && entry.Code != null)
                results.Add(entry);
        }

        return results;
    }

    private static WordEntry? ReadOneWord(BinaryReader reader, Stream stream)
    {
        var len = reader.ReadInt32();
        if (len > 1000)
            throw new InvalidDataException("有异常的词库，解析失败");
        if (len == 0)
            return null;

        var pinyinList = new string[len];
        for (var i = 0; i < len; i++)
        {
            var smIndex = reader.ReadByte();
            var ymIndex = reader.ReadByte();
            if (smIndex < Shengmu.Length && ymIndex < Yunmu.Length)
                pinyinList[i] = Shengmu[smIndex] + Yunmu[ymIndex];
            else
                pinyinList[i] = "";
        }

        var wordBytes = reader.ReadBytes(2 * len);
        var word = Encoding.Unicode.GetString(wordBytes);

        var hasValidPinyin = false;

View on GitHub (pinned to 16744a12ed)

Solutions

1. Confirm the file is genuinely a Baidu bdict and re-select the correct importer for its real source app.
2. Re-download or re-export the bdict to rule out truncation/corruption.
3. If it is a new bdict revision, the importer's fixed offsets (0x60 end-position, 0x350 word start) likely need updating to match the new layout.

// best-effort header sanity before full parse:
if (input.Length < 0x350) throw new InvalidDataException("too small to be a bdict");

try { var entries = importer.ParseBinary(stream, ct); }
catch (InvalidDataException ex) { ReportError($"unsupported/corrupt bdict: {ex.Message}"); }

When it happens

Trigger: Feeding a non-bdict file (e.g. an SCEL, QQPY, or plain text) to the bdict importer; a truncated/corrupted bdict where the read position drifted; an unsupported bdict revision whose different binary layout causes the length field to read as garbage.

Common situations: Wrong format selected for the input file; a partially downloaded or byte-corrupted bdict; a bdict exported by a newer Baidu IME version with a changed header/offset structure (the importer assumes header end-position at 0x60 and words at 0x350).

Related errors

• 词条数量异常: {cnt},可能是文件格式不兼容
• 词库文件格式不正确,文件大小至少需要{UserWordBase}字节,当前为{fileSize}字节
• 未能成功解析任何词条,可能是文件格式不兼容或文件已损坏

AI-assisted analysis of studyzy/imewlconverter@16744a12ed (2026-08-13). Data as JSON: /api/errors/7f4e118dd37d5bc4. Report an issue: GitHub.