2ec097a955
Because of a limitation in our GC tech (and because I have not sorted out how to program around it (which should be possible)) when the "is it safe to rename this variable" check may silently just say "yes" when it really doesn't know the correct answer. It turns out that this was happening on every attempt to rename after the first one (in a given online check syntax run) due to a bug, now fixed in this commit (the "loop" didn't actually loop...) Also, moved the creation of the thread to its own function to make the things it closes over more syntactically apparent.
97 lines
3.9 KiB
Racket
97 lines
3.9 KiB
Racket
#lang racket/base
|
|
(require racket/class
|
|
racket/place
|
|
(for-syntax racket/base)
|
|
"../../private/eval-helpers.rkt"
|
|
"traversals.rkt"
|
|
"local-member-names.rkt"
|
|
"intf.rkt"
|
|
"xref.rkt")
|
|
|
|
(provide go)
|
|
|
|
(define obj%
|
|
(class (annotations-mixin object%)
|
|
(init-field src orig-cust)
|
|
(define trace '())
|
|
|
|
(define-values (remote-chan local-chan) (place-channel))
|
|
(define table (make-hash))
|
|
|
|
(create-rename-answerer-thread orig-cust local-chan table)
|
|
|
|
(define/override (syncheck:find-source-object stx)
|
|
(and (equal? src (syntax-source stx))
|
|
src))
|
|
|
|
;; send over the non _ variables in the message to the main drracket place
|
|
(define-syntax (log stx)
|
|
(syntax-case stx ()
|
|
[(_ name args ...)
|
|
(with-syntax ([(wanted-args ...)
|
|
(filter (λ (x) (not (regexp-match #rx"^_" (symbol->string (syntax-e x)))))
|
|
(syntax->list #'(args ...)))])
|
|
#'(define/override (name args ...)
|
|
(add-to-trace (vector 'name wanted-args ...))))]))
|
|
|
|
(log syncheck:add-arrow
|
|
_start-text start-pos-left start-pos-right
|
|
_end-text end-pos-left end-pos-right
|
|
actual? level)
|
|
(log syncheck:add-tail-arrow _from-text from-pos _to-text to-pos)
|
|
(log syncheck:add-mouse-over-status _text pos-left pos-right str)
|
|
(log syncheck:add-background-color _text color start fin)
|
|
(log syncheck:add-jump-to-definition _text start end id filename submods)
|
|
(log syncheck:add-definition-target _text start-pos end-pos id mods)
|
|
(log syncheck:add-require-open-menu _text start-pos end-pos file)
|
|
(log syncheck:add-docs-menu _text start-pos end-pos key the-label path definition-tag tag)
|
|
(define/override (syncheck:add-id-set to-be-renamed/poss dup-name?)
|
|
(define id (hash-count table))
|
|
(hash-set! table id dup-name?)
|
|
(add-to-trace (vector 'syncheck:add-id-set (map cdr to-be-renamed/poss) remote-chan id)))
|
|
|
|
(define/public (get-trace) (reverse trace))
|
|
(define/private (add-to-trace thing)
|
|
(set! trace (cons thing trace)))
|
|
(super-new)))
|
|
|
|
(define (create-rename-answerer-thread orig-cust local-chan table)
|
|
;; the hope is that changing the custodian like this
|
|
;; shouldn't leak these threads, but it does seem to
|
|
;; so for now we don't use it
|
|
(parameterize (#;[current-custodian orig-cust])
|
|
(thread
|
|
(λ ()
|
|
(with-handlers ((exn:fail? (λ (x) (eprintf "online-comp.rkt: thread failed ~a\n" (exn-message x)))))
|
|
(let loop ()
|
|
(define id/name (place-channel-get local-chan))
|
|
(define id (list-ref id/name 0))
|
|
(define name (list-ref id/name 1))
|
|
(define res ((hash-ref table id) name))
|
|
(place-channel-put local-chan res)
|
|
(loop))))))
|
|
(void))
|
|
|
|
(define (go expanded path the-source orig-cust)
|
|
(parameterize ([current-max-to-send-at-once 50])
|
|
(with-handlers ((exn:fail? (λ (x)
|
|
(printf "~a\n" (exn-message x))
|
|
(printf "---\n")
|
|
(for ([x (in-list
|
|
(continuation-mark-set->context
|
|
(exn-continuation-marks
|
|
x)))])
|
|
(printf " ~s\n" x))
|
|
(printf "===\n")
|
|
(raise x))))
|
|
(define obj (new obj%
|
|
[src the-source]
|
|
[orig-cust orig-cust]))
|
|
(define-values (expanded-expression expansion-completed)
|
|
(make-traversal (current-namespace)
|
|
(get-init-dir path)))
|
|
(parameterize ([current-annotations obj])
|
|
(expanded-expression expanded)
|
|
(expansion-completed))
|
|
(send obj get-trace))))
|