Re: Fair Criticism, etc
Speaking of Lua, has anyone compared the Eu interpreters coded in Eu to
the Lua ability to exec commands stored in strings? Are they as fast? Less
able? What about nested procedures and functions, and lengths of the
strings? If Eu has a line length limit, is a string that exceeds this limit non-
executeable in the interpreters? So i cannot pass a 1/2 megabyte file to the
doubley interpreted Eu, right? I accept that the interpreted languages are
slower than compiled languages, but is the doubly-interpreted Eu still as fast
as the Lua in a dostring()?
Re Igor's method of storing the Eu interpreter in a ramdrive, i had mentioned
that ages ago, and i used that to store a copy of command.com to do dos
shells with pascal. Sharing can be done with Eu this way because one can
always shuffle the data back and forth between instances using files in the
same ramdrive, or dde, or socks (be careful with socks if you are online and
not behind a firewall), or win_msgs, or fixed reserved memory elsewhere, or
have it print out on paper and the paper fall on the scanner which then scans
it back in to an OCR app, or etc etc....
About the "goto" from Igor: that doesn't clean up the example from another
email which i have already deleted, where someone had a *huge* pile of
nested-nested-nested if-then-elsif. Never mind that i would have used a
nested sequence to hold the data, and it was a windoxe app which i have no
talent in, that whole procedure could have been made infinitely clearer with 4
simple logical tests and a "goto eoprocedurename". The same way a case
statement would look cleaner and clearer. But in terms of the work to add
new keywords to the language, "goto" would give us enough of the "case"
command that the "case" and "exit(lots_of_parms)" wouldn't be needed.
Besides, "goto" would give us "repeat" and a really loaded way to do "while",
simply.
Re:
> I go away for 2 days and comeback to 102 emails??? I think you've beat
'fair
> critisizm' to death. The only thing that matters is am I having fun
programming
> with EU and is Rob having fun writing it... I am and I think He does... are
> you?? if so nothing else matters!!
Well, i think a real world application that gets work done is more important
than mental mas,, err,, nevermind. But it's not all fun.
Re:
> > So structures would
> > not be a mere tweak of sequences. They would
> > have to be treated as a distinct data type.
>
> If you want typechecking of the elements, yes.
> Otherwise, nothing you have mentioned so far requires anything
> more than a little syntactic sugar.
One of the huge draws of an interpreter, that places an interpreter heads and
tails above a compiled language, is flexability. Exception processing. The
good ole "if-then"-else go on and do it like we have always done it. This
would almost be trivial to add, to the interpreter at least, altho i don't know
about the translator, cause it's not on my mind at the moment. Internally, the
interpreter simply refers to the table as a nested *normal* sequence. If
additional restrictions need be imposed, that's extra code and processing
time, but do-able. I vote for no restriction added for the keyword "table" in
native code that Rob writes, but expanded typecheck abilities, code that we
can write. Added abilities includes altering the contents of the variable in
question, and re-typing it. Then anywhere the table is used, you can check it
for length, contents, etc, as the coder wishes, the same as we can do now.
This is a little like the "with-do" command. It's not really needed, if you can
do this:
global procedure getnetstats(atom netnum)
sequence ircnet
ircnet = serverlist[netnum]
-- get the global vars for this net
sock = ircnet[1]
etc etc
I do that for *reading* from a huge nested-nested sequence (serverlist), but if
we could also allowed to do the same to *write* to it, the new code would not
break anything, and not need an added keyword,, just transparently added
code internal to the interpreter.
Kat
|
Not Categorized, Please Help
|
|
