rox has quit ["Rebooting .... gonna try out BeOS"]
giedi has quit [Read error: 104 (Connection reset by peer)]
giedi has joined #ocaml
__DL__ has quit [Remote closed the connection]
giedi has quit [Read error: 104 (Connection reset by peer)]
stefp has quit [Read error: 60 (Operation timed out)]
stefp has joined #ocaml
lament has joined #ocaml
giedi has joined #ocaml
stefp has quit [Read error: 60 (Operation timed out)]
stefp has joined #ocaml
lament has quit ["I AM NOT A MENSA MEMBER"]
stefp has quit [Read error: 60 (Operation timed out)]
stefp has joined #ocaml
jao has quit ["leaving"]
mattam has quit [Read error: 113 (No route to host)]
Kinners has joined #ocaml
platypus__ has joined #ocaml
mattam has joined #ocaml
lament has joined #ocaml
platypus_ has quit [Read error: 113 (No route to host)]
TachYon26 has joined #ocaml
TachYon76 has joined #ocaml
TachYon76 has quit [Read error: 54 (Connection reset by peer)]
Kinners has left #ocaml []
platypus__ has quit [calvino.freenode.net irc.freenode.net]
platypus__ has joined #ocaml
lament has quit ["I WILL NOT DO MATH IN CLASS"]
reltuk has joined #ocaml
two-face has joined #ocaml
skylan has joined #ocaml
TachYon26 has quit ["bez ki³y nie ma zaliczenia (z prawd studentek AM)"]
phubuh has quit ["leaving"]
two-face has quit ["Client exiting"]
Vincenz has joined #ocaml
<Vincenz>
Riastradh?
<Riastradh>
Yes?
<Vincenz>
I know why it didn't compile
lus|wazze has joined #ocaml
<Vincenz>
urgh
<Vincenz>
sml seems to allow reuse of labels in different records, Ocaml doesn't :/
<lus|wazze>
yup
<Vincenz>
one wonders why
<lus|wazze>
but on the other hand, you have to explicitly type records in sml
<Vincenz>
hmm?
<Vincenz>
in ocaml too, no?
<lus|wazze>
no
two-face has joined #ocaml
* Vincenz
is converting some code from sml to ocaml
* Vincenz
doesn't know sml and just a bit of ocaml
<Vincenz>
I have this sumtype
<Vincenz>
exp
<Vincenz>
which is built up of many records
<Vincenz>
and many of these records have a field "pos"
<Vincenz>
but now I have to differentiate them :/
<Vincenz>
yicky
<lus|wazze>
:|
<Riastradh>
In a type 'rt' defined as a record type, I always prefix the fields with 'rt_' -- I don't think you lose readability or anything, and you don't have to worry about name clashes.
<Vincenz>
I know
<Vincenz>
that's what I'm doing right now
<Vincenz>
but it was handy to just refer to pos
<Vincenz>
instead of let_pos
<Vincenz>
or while_pos
<Vincenz>
besides
<Riastradh>
What code are you translating?
<Vincenz>
assignexp becomes ass_pos
<Riastradh>
Or rather, what does it do?
<Vincenz>
AST
<Vincenz>
well the definition of the types
Zadeh has quit [Read error: 60 (Operation timed out)]
<Vincenz>
hmm
<Vincenz>
let's say I have module A
<Vincenz>
which defines some type =... | Plop of plop
<Vincenz>
and plop = {a : int }
<Vincenz>
in another module I do A.Plop {a = 2}
<Vincenz>
and it complains it can't bind the record field a
<Vincenz>
doh!
<Vincenz>
I have to use {A.a = 2}
* Vincenz
sees his code bloat and bloat
* Vincenz
notes more idiosyncrasies : parser says : %token <string*int> STRING ......then STRING -> {StringExp $1} (because StringExp expects string*int)....that doesn't work so I have to do StringExp (fst $1, snd $1)
<Vincenz>
| FUNCTION ID LPAREN tyfields RPAREN EQ exp {()}
<Vincenz>
you see?
<Vincenz>
tyfields
<taw>
valuesopt varsopt procs are opional elements
<Vincenz>
I can't make tyfields empty
<taw>
no ?
<Vincenz>
so I need two cases
<taw>
why not ?
<Riastradh>
Hmm...
<Riastradh>
ocamldoc is complaining that it can't find two constructors that are referenced, but those constructors are defined in one of the files it's processing.