@samth https://gist.github.com/cderici/f491054617802ac5b88a6d01a5099bb7

@cadr stack traces shouldn’t print for exn:fail:user

but also: [samth@huor:~/sw/pycket/pycket (master) plt] pk -l compiler/commands/make -- --help racket [ <option> ... ] <file> [<another-file>] ... where <option> is one of -j <n> : Compile with up to <n> tasks in parallel --disable-inline : Disable procedure inlining during compilation --disable-constant : Disable enforcement of module constants --no-deps : Compile immediate files without updating dependencies -p <file>, --prefix <file> : Add elaboration-time prefix file for --no-deps --no-prim : Do not assume `scheme' bindings at top level for --no-deps -v : Verbose mode --vv : Very verbose mode --help, -h : Show this help -- : Do not treat any remaining argument as a switch (at this level) Multiple single-letter switches can be combined after one `-'; for example: `-h-' is the same as `-h --'

[samth@huor:~/sw/pycket (master) plt] pk -l compiler/commands/make -- -v compile-file-pycket.rkt
"compile-file-pycket.rkt":
  making #<path:/home/samth/sw/pycket/compile-file-pycket.rkt>
module-prefetch: (racket/path racket/list racket/file racket/cmdline syntax/modread) in: #<path:/home/samth/sw/pycket/>
 [output to "#<path:compiled/pycket/compile-file-pycket_rkt.zo>"]

@samth yeah the stack trace printing is currently not conditional to anything, error display handler will always print it

[samth@huor:~/sw/pycket (master) plt] time PLT_LINKLET_TIMES=1  pk -l racket
;; boot                                      7 [0.000000] ms
;; make-instance                             0 [0.000000] ms ; 799 times
;; instantiate-linklet                    1257 [74.000000] ms ; 1639 times
;; startup                                 905 [376.000000] ms ; 1 times
;;   expander-linklet                        0 [0.000000] ms ; 1 times
;;   json-load                             293 [197.000000] ms ; 2 times
;;   json-to-ast                           609 [179.000000] ms ; 2 times
;;   fasl-linklet                            0 [0.000000] ms ; 1 times
;;   set-params                              2 [0.000000] ms ; 1 times
;; read                                  15030 [8868.000000] ms
;;   fasl->s-exp                          1155 [162.000000] ms ; 486 times
;;   s-exp->ast                          13361 [8613.000000] ms ; 486 times
;;   assign-convert-deserialize            513 [93.000000] ms ; 11516 times
;; total                                 17201 [9318.000000] ms

real    0m20.279s
user    0m19.824s
sys     0m0.378s

@samth wait are we able to run #lang racket now?

yup

and that was raco make that I pasted above

@samth horrayy! Yeah I wasn’t sure if I was understanding correctly :slightly_smiling_face:

@samth Awesome!

the -N flag seems to crash

@samth Also, I was trying to get some traces for the fasl functions, but it doesn’t put anything in the logfile, my guess is that there’s not enough computation for it to generate traces, as it fills the logfile when I run it with loading the expander

unfortunately I’m currently unable to use string_to_sexp to generate big s-expressions for fasl to work on, rpython keeps erroring on the re library

Just write a racket program that generates a big s-exp and compile and write it

@samth you mean load it from another json?

no

how am I gonna read that without loading the expander?

ah you’re trying what I suggested about not loading the expander

yes

don’t read the sexp, generate it

write a for loop

yeah that was what I was about to do

successfully compiled: racket, net/url, racket/fasl, typed-racket/utils/utils, compiler/compiler, raco/main, most of the db collection, file/unzip, and many others