Split centralized manifest.json into per-spec manifest.json files#180
Conversation
Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Signed-off-by: Andrew Helwer <ahelwer@pm.me>
…anifests Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Signed-off-by: Andrew Helwer <ahelwer@pm.me>
| They are not covered by CI testing so it is possible they contain errors, the reported details are incorrect, or they are no longer available. | ||
| ## Other Examples | ||
| Here is a list of specs stored in locations outside this repository or not validated by the CI, such as submodules. | ||
| Since these specs are not covered by CI testing it is possible they contain errors, the reported details are incorrect, or they are no longer available. |
There was a problem hiding this comment.
This is factually incorrect. These specifications may not be covered by the CI in this repository, but they could be managed by their own dedicated CI pipelines. For example, the BlockingQueue specs have their own CI.
There was a problem hiding this comment.
If the BlockingQueue repository is moved or deleted then it will no longer be available, as stated. Although this itself will probably be detected in the next CI run, when git tries to clone a submodule that no longer exists.
|
|
||
| ## Examples Included Here | ||
| Here is a list of specs included in this repository, with links to the relevant directory and flags for various features: | ||
| ## Validated Examples Included Here |
There was a problem hiding this comment.
Again, why not have a single table that has an additional column that indicates if a spec has CI?
There was a problem hiding this comment.
We can certainly look at doing that in the future, but that's beyond the scope of this PR.
lemmy
left a comment
There was a problem hiding this comment.
I don't know enough about the python tooling to review it. The markdown files and the CI.yml look ready to merge (modulo my comments below).
Reflect changes in tlaplus/Examples#180 [CI] Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Reflect changes in tlaplus/Examples#180 Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Reflect changes in tlaplus/Examples#180 [CI] Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Reflect changes in tlaplus/Examples#180 Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Reflect changes in tlaplus/Examples#180 Signed-off-by: Andrew Helwer <ahelwer@pm.me>
Reflect changes in tlaplus/Examples#180 [CI] Signed-off-by: Andrew Helwer <ahelwer@pm.me>
In furtherance of #171. These changes:
generate_manifest.pyto split the existing manifest into per-directory manifestsgenerate_manifest.pyscript, which is a large number of small manifestsmanifest-schema.jsonto validate the split manifestsmanifest.jsonfileNow that we have per-spec manifests, we can incrementally devolve the various components of the manifests into either the README table or the specs themselves. Ideally we will reach a world where no manifests exist at all, but some features like recording the model state count will make that difficult. Putting those details as comments in the model is one possibility.