Revisions on the Revisions of the Data Provenance branch #53
Conversation
None of the stuff it describes is needed anymore. Note that it is now possible for both Simplification.IN_PROGRESS and Simplification.FORBIDDEN to exist in the same Hints object. This should be fine
It is only used in one place, and it is only one line.
This removes "an op" and "{@link Op}" from the codebase
This required we delete the PluginBasedClassOpInfoGenerator, since the Op interface no longer extended SciJavaPlugin. But that's fine, since we were going to delete that anyways!
Also, changed the line endings in OpInstance from Windows to Linux.
We were conflating too many things with the name op()
Don't confuse people with the flavoring
ctrueden
force-pushed
the
scijava/scijava-ops-engine/data-provenance-revisions-revisions
branch
from
98850ae to
d2f2aba
Compare
ctrueden
left a comment
ctrueden
left a comment
There was a problem hiding this comment.
Looks good. A couple of small comments, and some commit messages do not use imperative tense as is standard, but this is good to go!
|
|
||
| static final String IMPL_DECLARATION = "|Vanilla:"; | ||
| /** Identifier for an unaltered OpInfo in an Op signature **/ | ||
| String IMPL_DECLARATION = "|Info:"; |
There was a problem hiding this comment.
Some reason this shouldn't be |Op: instead?
| @@ -1,53 +1,97 @@ | |||
|
|
|||
There was a problem hiding this comment.
A bunch of files in the repo have been Windows line endings up till this point... but now they are getting converted over one by one. This is much better done as its own commit fixing all of them at once, when this type of problem is noticed. Otherwise, the diff here is obfuscated by the line ending shift.
| * Creates an Op instance from an {@link OpInfo} with the provided | ||
| * {@link MatchingConditions} as the guidelines for {@link OpInfo} selection. | ||
| * This Op instance is put into the {@code opCache}, and is retrievable via | ||
| * This Op instance is put into the {@link #opCache}, and is retrievable via |
There was a problem hiding this comment.
By default when you generate javadoc, private members are not included in the generated output. So then this sort of link, which goes to a private field, will not actually be functional. I am actually not sure how this ends up rendering in the default case, but we should check. If it looks bad or confusing, we can just remove the @link in favor writing "op cache" in normal language. But if it looks OK, I favor leaving it so that in IDEs the linkage is still there.
ctrueden
deleted the
scijava/scijava-ops-engine/data-provenance-revisions-revisions
branch
Fixes to #46.