[
https://issues.apache.org/jira/browse/CAY-2980?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andrus Adamchik updated CAY-2980:
---------------------------------
Description:
Current model name generation is based on an old third-party "inflector"
library. Let's internalize these algorithms and get rid of the dependency and
try improve name generation intelligence.
✅ Get rid of the old inflector.
✅ Implement an internal pluralization based on a common two-form English
inflection algorithm. It is not fully backwards compatible. Hopefully we didn't
lose and pluralization rules, and looks like we fixed a few ("logos" instead of
"logoes", "equipment" instead of "equipments")
✅ Implement fk-based to-many relationship naming ("home_team_id" ->
"homeGames")
4. Check for Java reserved words
5. Strip common prefixes from relationships ("pp_table" -> "tables" instead of
"ppTables", if "pp_" is used by both source and target)
was:
Current model name generation is based on an old third-party "inflector"
library. Let's internalize these algorithms and get rid of the dependency and
try improve name generation intelligence.
✅ Get rid of the old inflector.
✅ Implement an internal pluralization based on a common two-form English
inflection algorithm. It is not fully backwards compatible. Hopefully we didn't
lose and pluralization rules, and looks like we fixed a few ("logos" instead of
"logoes", "equipment" instead of "equipments")
✅ Implement fk-based to-many relationship naming ("home_team_id" ->
"homeGames")
4. Check for Java reserved words
5. Strip common prefixes from relationships ("pp_table" -> "tables" instead of
"ppTables", if "pp_" is used for the entire model)
> Improve model name generation
> -----------------------------
>
> Key: CAY-2980
> URL: https://issues.apache.org/jira/browse/CAY-2980
> Project: Cayenne
> Issue Type: Task
> Reporter: Andrus Adamchik
> Assignee: Andrus Adamchik
> Priority: Minor
> Fix For: 5.0-M3
>
>
> Current model name generation is based on an old third-party "inflector"
> library. Let's internalize these algorithms and get rid of the dependency and
> try improve name generation intelligence.
> ✅ Get rid of the old inflector.
> ✅ Implement an internal pluralization based on a common two-form English
> inflection algorithm. It is not fully backwards compatible. Hopefully we
> didn't lose and pluralization rules, and looks like we fixed a few ("logos"
> instead of "logoes", "equipment" instead of "equipments")
> ✅ Implement fk-based to-many relationship naming ("home_team_id" ->
> "homeGames")
> 4. Check for Java reserved words
> 5. Strip common prefixes from relationships ("pp_table" -> "tables" instead
> of "ppTables", if "pp_" is used by both source and target)
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
