Follow-up from PR #4 review (Standards axis).
at-token-re is duplicated with a subtle divergence:
src/eca_cli/chat.clj:108 — #"(?:^|\s)@(\S+)" (non-capturing lead group)src/eca_cli/view/blocks.clj:23 — #"(^|\s)@(\S+)" (capturing lead group)
Both parse the same @path token grammar (send-time context matching vs render-time styling), but the group shapes differ, so the meaning of each capture group is not the same across the two sites. Not a bug today, but the copies will drift when a third use site lands.
Per Clojure Craft "wait for the third occurrence before extracting" this was deliberately left duplicated. When the third site appears, extract one shared @-token regex with a single documented capture-group contract that both namespaces consume.
Follow-up from PR #4 review (Standards axis).
at-token-reis duplicated with a subtle divergence:src/eca_cli/chat.clj:108—#"(?:^|\s)@(\S+)"(non-capturing lead group)src/eca_cli/view/blocks.clj:23—#"(^|\s)@(\S+)"(capturing lead group)Both parse the same
@pathtoken grammar (send-time context matching vs render-time styling), but the group shapes differ, so the meaning of each capture group is not the same across the two sites. Not a bug today, but the copies will drift when a third use site lands.Per Clojure Craft "wait for the third occurrence before extracting" this was deliberately left duplicated. When the third site appears, extract one shared
@-token regex with a single documented capture-group contract that both namespaces consume.