最新的docker镜像版本为1.6.1，能否提供2.0以上版本的dev和runtime镜像呢

[zjd1988]

在提交BUG之前，请阅读帮助文档的FAQ，或在现有issue中搜索是否有类似问题。
Before submitting a bug, please read the FAQ of the help documentation, or search for similar issues in existing issues.

运行环境信息 | System information (请提供足够详细的信息 | Please provide as much relevant information as possible)

• 操作系统信息 | Operation System information
• 设备信息 | Device information
• ModelBox版本 | ModelBox version
• GPU或加速卡信息 | GPU or NPU information
• 使用的推理引擎 | Inference engine information
• 编程语言 | Programming language （C++, Python, Java）

描述问题 | Describe the current behavior:

期望的行为 | Describe the expected behavior:

重现步骤描述 | Standalone code to reproduce the issue:

提供具体重现问题的步骤，如果可能，提供相关的截图信息，日志信息。
Provide a reproducible test case that is the bare minimum necessary to replicate the problem.

日志信息 | Logs

收集ModelBox的运行日志，路径为/var/log/modelbox Please Provide modleobx logs, log path /var/log/modelbox

其他信息 | Other Info.

[lzehl]

@jagru20 and @apdavison can you explain more what the differences are?

I don't understand the convention based on the examples we already have in. From those I understood the convention the following way, but maybe I'm wrong:

• a.b is be used when a has the format subtype b and b is not based on any other format
• a+b is used when a has the format subtype b which is a standalone base format (less structured than a+b; same extensions as b)
• a.b+c is used when a has the format subtype b which is based on a standalone base format c (different extensions as c)

Which also means I messed up a bit the instances above... If I follow this idea it should have been vnd.espina+txt vnd.espina+ms-excel and (from another PR where I did it correctly) vnd.espina.seg+zip.

@MaaikevS & @UlrikeS91 FYI (related question to your PRs #311 #309 )

[MaaikevS]

@lzehl Thank you for pointing this out. This is also seems related to the already merged PR #302 I will change all of these content types if the names used are indeed not in line with the naming convention!

[jagru20]

Sorry if I am causing confusion here. The naming convention (link) proposed by Andrew states that "Where the format uses a standardised container format, such as XML, JSON, HDF5, we add +{ container_format }, e.g. application/vnd.nwb.nwbn+hdf" Whereas another .{sub_sub_type} shows that this is another subtype of the same supertype. This means, a + Suffix does not sign a subtype, only a container format that applies to this contenttype. In your example above, @lzehl I agree that there should be three types for espina, but not named as vnd.espina+txt and vnd.espina+ms-excel but rather vnd.espina.ms-excel and vnd.espina.ms-excel (better vnd.espina.xls) as these seems to be subtypes of espina, but not container types.

[apdavison]

What is the difference between the two formats? Are they interchangeable (the same data can be stored in either), or do they have different roles (contain different types of data)?

[lzehl]

@apdavison as far as I understand, the data in these formats hold different data information. They exist in parallel in datasets that are using Espina and even if (xls and txt do not have the same structure, but I can double check) are differently structured

[apdavison]

ok, then I would name them application/vnd.espina.purpose1 and application/vnd.espina.purpose2 where "purpose1" and "purpose2" are the kind/type of data that they contain or what those data are used for. If there is value in also being able to open the file in Excel then you could add +ms-excel to the one that is an Excel file (almost anything can be opened in a text editor, so I'm not sure adding "+txt" is valuable).

[lzehl]

@apdavison thanks a lot for the feedback. I will try later also to document this for contributions to content types (I will pin you and @jagru20 on this for corrections).

@jagru20 I will revise today the PR and pin you when I'm done with the corrections.

[UlrikeS91]

I'm closing this PR. It is replaced by the new PR in the instances-repo.