From Newsgroup: comp.lang.awk

Someone has done a nice package for awk called rawk that adds some niceties—and a reasonably large “standard library”—to awk. It’s a single
(large) awk file and is expected to work in any posix awk (it doesn’t currently work with the plan9/p9p version, but I’m working with him on
that).

https://git.sr.ht/~eli_oat/rawk/

A blog post by the author introducing it:

https://eli.li/make-awk-rawk
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.awk

Your post appears to me to be completely misleading! According to the
examples it's NOT an Awk library at all. It's a new language with new
syntax (running with Awk just as its vehicle). So it's not so much a
"package FOR Awk", it's rather "Awk has been used FOR this language".
The bad thing with that - from an Awk users' perspective - is that to
use it it seems, to use these "library" features, you'd need to learn
a new language with an own syntax. For Awk users a library written in
the Awk language would be much more useful. YMMV.

Janis

On 07.08.2025 22:39, Anthony wrote:

Someone has done a nice package for awk called rawk that adds some niceties—and a reasonably large “standard library”—to awk. It’s a single
(large) awk file and is expected to work in any posix awk (it doesn’t currently work with the plan9/p9p version, but I’m working with him on that).

https://git.sr.ht/~eli_oat/rawk/

A blog post by the author introducing it:

https://eli.li/make-awk-rawk

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.awk

On 2025-08-08, Janis Papanagnou <janis_papanagnou+ng@hotmail.com> wrote:

Your post appears to me to be completely misleading! According to the examples it's NOT an Awk library at all. It's a new language with new
syntax (running with Awk just as its vehicle).

To be fair, there are languages in which such a thing is considered
a library, like Lisps!

Rawk has a translator which is written in Awk (directly, so if you know
nothing but Awk, you can understand the translator and work on it), and
the target language is Awk. You need an Awk program to run both the
translator and the resulting code.

The only thing that's missing is that you can't launch an interactive
Awk session where you load Rawk, and then write Rawk programs; it is
batch translated. But otherwise it is locked squarely into your Awk
ecosystem.

Moreover, the goal of this language is to implement Awk, but with
extensions, and not an entirely different language. Awk programs are
supposedly valid Rawk programs. It's kind of like Objective C to C,
... or something.

It has fewer non-Awk dependencies than that cppawk think I wrote
a few years ago, which has a shell script driver, and requires the GNU preprocessor cpp.
--
TXR Programming Language: http://nongnu.org/txr
Cygnal: Cygwin Native Application Library: http://kylheku.com/cygnal
Mastodon: @Kazinator@mstdn.ca
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.awk

Janis Papanagnou <janis_papanagnou+ng@hotmail.com> wrote:

Your post appears to me to be completely misleading! According to the examples it's NOT an Awk library at all. It's a new language with new
syntax (running with Awk just as its vehicle).

I don’t think that’s a fair description. Rawk is entirely additive; all existing awk is valid rawk. And any new bits you’d write using the new
syntax are nicely contained in the RAWK {} block.

So it's not so much a

"package FOR Awk", it's rather "Awk has been used FOR this language".
The bad thing with that - from an Awk users' perspective - is that to
use it it seems, to use these "library" features, you'd need to learn
a new language with an own syntax. For Awk users a library written in
the Awk language would be much more useful. YMMV.

Janis

On 07.08.2025 22:39, Anthony wrote:

Someone has done a nice package for awk called rawk that adds some
niceties—and a reasonably large “standard library”—to awk. It’s a single
(large) awk file and is expected to work in any posix awk (it doesn’t
currently work with the plan9/p9p version, but I’m working with him on
that).

https://git.sr.ht/~eli_oat/rawk/

A blog post by the author introducing it:

https://eli.li/make-awk-rawk

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.awk

On 09.08.2025 04:38, Kaz Kylheku wrote:

On 2025-08-08, Janis Papanagnou <janis_papanagnou+ng@hotmail.com> wrote:

Your post appears to me to be completely misleading! According to the
examples it's NOT an Awk library at all. It's a new language with new
syntax (running with Awk just as its vehicle).

To be fair, there are languages in which such a thing is considered
a library, like Lisps!

Well, people use all sorts of [(IMO) inappropriate] terminology and
flimsy comparisons to advertise their babies.

Rawk has a translator which is written in Awk (directly, so if you know nothing but Awk, you can understand the translator and work on it),

(...and "work on" the translator??? - To create yet another dialect
of a new language or dialect?)

and
the target language is Awk. You need an Awk program to run both the translator and the resulting code.

But so what? - Cim or Genie are also written in "C" and also produce
"C" [intermediate] code, yet clearly creating own languages (Simula
and Algol 68, respectively). - It's not exactly the same here since
they are not extending the "C" language, but the underlying language
platform is hardly an argument against this being a new language.

The only thing that's missing is that you can't launch an interactive
Awk session where you load Rawk, and then write Rawk programs; it is
batch translated. But otherwise it is locked squarely into your Awk ecosystem.

Moreover, the goal of this language is to implement Awk, but with
extensions, and not an entirely different language. Awk programs are supposedly valid Rawk programs. It's kind of like Objective C to C,
... or something.

The point is not that you can still use Awk and ignore the extensions.

The point is that, as opposed to an Awk library with Awk functions, a
new syntax (for this new language) has to be used. That got not clear
from the post. But from the examples and statements on the referenced
pages it's quite obvious, e.g.

$greet = (name) -> { return ...; };

There they also write
"... provides a quick snapshot of rawk's syntax and language features."
here clearly exposing its nature!

I was not objecting to yet another new language/dialect introduced[*],
but just that I considered the post completely misleading.

Janis

[*] Although I personally dislike attitudes of "write a new X language
to make the Y language appear more like the Z language". (But mileages
likely vary.)

It has fewer non-Awk dependencies than that cppawk think I wrote
a few years ago, which has a shell script driver, and requires the GNU preprocessor cpp.

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.awk

In article <1077fht$1a1tc$1@dont-email.me>,
Janis Papanagnou <janis_papanagnou+ng@hotmail.com> wrote:
...

The point is that, as opposed to an Awk library with Awk functions, a
new syntax (for this new language) has to be used. That got not clear
from the post. But from the examples and statements on the referenced
pages it's quite obvious, e.g.

$greet = (name) -> { return ...; };

Hmmm. Now, what language does this look like???

There they also write
"... provides a quick snapshot of rawk's syntax and language features."
here clearly exposing its nature!

I was not objecting to yet another new language/dialect introduced[*],
but just that I considered the post completely misleading.

Welcome to the 21st century - the age of bullshit.
Everywhere you look, there is bullshit. The existence of this thread is
just one more example.

[*] Although I personally dislike attitudes of "write a new X language
to make the Y language appear more like the Z language". (But mileages
likely vary.)

I assume X = "rawk", Y = AWK, and Z = P***.

And yes, I get your point and agree 100%.
--
The difference between communism and capitalism?
In capitalism, man exploits man. In communism, it's the other way around.

- Daniel Bell, The End of Ideology (1960) -
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.awk

On 09.08.2025 05:06, Anthony wrote:

Janis Papanagnou <janis_papanagnou+ng@hotmail.com> wrote:

Your post appears to me to be completely misleading! According to the
examples it's NOT an Awk library at all. It's a new language with new
syntax (running with Awk just as its vehicle).

I don’t think that’s a fair description. Rawk is entirely additive; all existing awk is valid rawk.

No doubt.

The question is: How do we obtain access to the (as you called it)
"standard library" [functions]?

And any new bits you’d write using the new
syntax are nicely contained in the RAWK {} block.

Sure. If you don't want to use the new language - and thus have no
gain of the new language's "standard library" functions! - you can
simply omit this block (and continue using plain *standard* Awk).

You wrote
"adds some niceties—and a reasonably large “standard library”—to awk"

But (as I understood) you'd have to use the _new language_ (with its
_new syntax_) to use the library.

Don't get me wrong; I'd appreciate _standard library functions_ for
Awk - there's indeed a gap to be filled! - but not if I'm forced to
use a non-Awk language syntax.[*]

Janis

[*] That's one reason, BTW, why I'm only scarcely using some of GNU
Awk's features that require using a new syntax, although these have
been incorporated quite sensibly without a rather disruptive syntax
as seen in the language advertised here.

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.awk

On 09.08.2025 14:51, Kenny McCormack wrote:

In article <1077fht$1a1tc$1@dont-email.me>,
Janis Papanagnou <janis_papanagnou+ng@hotmail.com> wrote:
...

The point is that, as opposed to an Awk library with Awk functions, a
new syntax (for this new language) has to be used. That got not clear
from the post. But from the examples and statements on the referenced
pages it's quite obvious, e.g.

$greet = (name) -> { return ...; };

Hmmm. Now, what language does this look like???

My first association ('$greet' or '->') was Perl, but then I read on
the posted links something about Rust (that I don't know, so I can't
tell).

In another group someone recently said: >> EVERYBODY had their "better
idea" << (concerning ad hoc inventions of programming languages). This
is probably true, and obviously not an unlikely explanation also here.

[...]

[...]

[*] Although I personally dislike attitudes of "write a new X language
to make the Y language appear more like the Z language". (But mileages
likely vary.)

I assume X = "rawk", Y = AWK, and Z = P***.

I had meant it abstractly as a general observation on some language
inventions (or evolution). Here Z might be "P***", but I don't know.

Generally I think it's not a bad idea to focus on good concepts and
(clear) syntax also from other language paragons. (My severe doubts
are more concerning the mass of new unnecessary languages. And then
the often poor choice of language concepts, including syntax, etc.)

Janis

And yes, I get your point and agree 100%.

--- Synchronet 3.21a-Linux NewsLink 1.2