Re: Do all programming languages use files?


In article <1125311420.471389.249420@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>, "Past" <futureofai@xxxxxxxxx> writes:
> I was doing a research to see if all the programming languages ever
> written have atleast some small level capability of successfully
> writing and reading from files?

No.

Few assembly languages, for example, include file I/O operations,
since few CPUs provide file I/O operations. Of course, assembly
programs for systems with filesystems (a minority of all computers
in use today, but it's a highly-visible minority) are able to
perform file I/O, but not through any facility inherent in the
language itself; they have to invoke system-specific mechanisms.

There are many experimental, toy, and esoteric languages which do
not have file I/O capabilities. The Unlambda and ETA languages
do not have them, for example. Web sites like "The Turing Tarpit"
will provide you with many examples of these.[1]

For that matter, the C language does not have any capability for
"successfully writing and reading from files". The C standard
requires that hosted implementations provide a series of functions
for file I/O, but these are permitted to fail for any reason
whatsoever, including being written by the implementor to always
fail. *Successful* file I/O is a quality-of-implementation and
environmental issue as far as C is concerned.

Some languages are loaded with features, including things like
file I/O, as part of the standard package; others go for a more
minimal approach. And different languages that do provide for
file I/O approach it in some very different ways. You might want
to compare, for example, the file I/O support in Pascal (intended
to be easy for students), C (intended to be flexible), and COBOL
(intended to facilitate implementing business rules).


1. http://www.geocities.com/ResearchTriangle/Station/2266/tarpit/tarpit.html

--
Michael Wojcik michael.wojcik@xxxxxxxxxxxxxx

Maybe, but it can't compete with _SNA Formats_ for intricate plot
twists. "This format is used only when byte 5, bit 1 is set to 1
(i.e., when generalized PIU trace data is included)" - brilliant!
.

Relevant Pages

  • Re: Settling an Argument - Assembly *IS* a Language, Right?
    ... The limited range of expression is what's rude about early Sumerian ... the oldest Sumerian cuneiform writing could not render ... the silliness of expecting computer languages to have anything to do ... And that's not the meaning of "trolling." ...
    (sci.lang)
  • Re: Settling an Argument - Assembly *IS* a Language, Right?
    ... Peter T. Daniels wrote: ... The limited range of expression is what's rude about early Sumerian ... the oldest Sumerian cuneiform writing could not render ... the silliness of expecting computer languages to have anything to do ...
    (sci.lang)
  • Re: Writting Sign Languages
    ... have worked with systems of writing down signed languages. ... There are a bunch of systems kicking around for writing sign ... SignFont materials fwere hard to come by in 1991 when I first went ...
    (sci.lang)
  • Re: Writing and the Tower of Babel story
    ... after the Tower fell. ... but isn't it a but odd that most or all writing seems to have begun ... some of whom speak different languages. ... be found after the formation of a city, ...
    (talk.origins)
  • Re: Reading depends on writing, in Chinese
    ... >>> listening, speaking, reading, and writing. ... > the local cinema films in one of the two languages were ... file .signature infected by signature virus. ...
    (sci.lang)