<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Hi Chen,</p>
<p>thank you for your ideas!</p>
<p>Actually I cannot see what is "safer" in your proposal, but maybe
I am missing to see a hidden risk in instanceof. Can you please
outline the potential risk you actually see in "if (appendable
implements Flushable f) f.flush();"?</p>
<p>I mean, Flushable and Closable are simply *mix-ins* existing for
exactly the purpose of "flushing-if-flusing-is-supported" and
"closing-if-closing-is-supported", which is what we do need right
here. Nobody wants to pass in a standalone "flusher" or standalone
"closer" in addition to the actual object to flush and close, i.
e., the Appendable. In particular, nobody actually reported the
need to build a Writer from three distrinct implementation objects
(or I missed this need). Explicitly passing "null" feels rather
unintuitive and IMHO is doubtful. Why should someone want to do
that? Again, apparently you see that use case, so if you really
have strong feelings, then please make me understand who needs
that and for what actual purpose. :-)<br>
</p>
<p>To be all on the same side, again, please always share the core
idea that this API more or less solely is the combination of
"Writer.of(StringBuilder)" with "Writer.of(StringBuffer)" and
"Writer.of(CharSet)".<br>
</p>
<p></p>
<p>Note that the sole target still is to pass in a StringBuilder,
StringWriter, or CharBuffer, as wrapping *them* is *the driver*
for the new API. While someone *can* do that, it is *not the
target* of this API to pass in any Writer or any arbitrary
Appendable. Therefore, we just need to be able *to deal with that
case* once it happens -- which is why it is IMHO absolutely fine
to directly return Writers *non-wrapped*. The API so far just
says, "passing a Writer in turn returns a Writer", but it does
*not* propose to enhance or limit that Writer in any way, and that
is why it is (IMHO absolutely) safe to check all other Appendables
for *their* actual ability to get flushed or to get closed.
Remember, the target of *this* API proposal is *not* to be able to
write into any Flushable-and-Closable-Appendable *without*
flushing or closing it. Having that said, *I do not veto* adding
an *additional* method like Writer.of(Appendable, boolean
preventFlush, preventClose) *later* **if needed**, but IMHO that
should rather be *separate* wrappers like
Writer.withoutFlushing(Writer) and Writer.withoutClosing(Writer)
(either you have the need to not-flush/not-close, or you don't
have it, so it is not a special case of *this* API), or something
like that, which both are, again, *non-targets* of my current
proposal. In fact I still do not see *any* benefit of passing in a
Writer into Writer.of(), neither as a single reference, nor split
up into three interfaces (and BTW, I did *not* say a Writer is a
combination of Appendable, Flushable and Closable). Neither do I
see *any benefit* of being able to pass in in three different
implementation objects. But what I do see in your proposal
actually is:</p>
<p>* It would make up a can of worms due to the possibility of
providing three different implementation objects for that three
parameters. Someone could do Writer.of(new StringBuilder(),
Files.newBufferedWriter(), new CharBuffer()) and the outcome would
be rather dubious (and mostly useless but confusing).<br>
</p>
<p>* As the sole target is to allow wrapping StringBuilder,
StringWriter, and CharBuffer, and as we solely came to Flushable
and Closable due to the question about "How to call flush and
close ON THE PASSED REFERENCE, IFF the Appendable implements
them?" it would be a real pain for alle users to be FORCED to
repeat the same object three times.</p>
<p>Having said that, my proposal is (as this is it what is IMHO
mostly intuitive and most wanted):</p>
<p>* Let's have solely Writer.of(Appendable) without any other
parameters *in the first PR*; discuss the use case of more
parameters *in subsequent PRs* IFF NEEDED as these should be
*additional* method signatures to not torture the 90% standard
case users with parameters they never need.<br>
</p>
<p>* Let's return Writer non-wrapped, and clearly document that in
the JavaDocs. Have separate discussions about
Writer.withoutFlushing(Writer) and Writer.withourClosing(Writer)
*in subsequent threads* IFF NEEDED.<br>
</p>
<p>* Let's use "if (appendable instanceof Flushable f) f.flush()"
and "if (appendable instanceof Closebale c) c.close()", and
clearly document that in the JavaDocs. In case users do really
want non-flushed, non-closed appendables wrapped as Writer, they
do not lose something, but have to wait for the outcome of
*subsequent* discussions about *additional* wrappers.</p>
<p>I think that could be a clean, safe and straightforward way
towards the replacement of StringWriter.<br>
</p>
<p>Regards and a happy new year! :-)</p>
<p>-Markus</p>
<p><br>
</p>
<div class="moz-cite-prefix">Am 31.12.2024 um 06:42 schrieb Chen
Liang:<br>
</div>
<blockquote type="cite"
cite="mid:CABe8uE0Err3uUtkDqMxeVArp0LHKzeSr5k03zFG7VwswwdWZ3A@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">Hi Markus,
<div>Thanks for your analysis that a Writer can be seen as a
composition as an Appendable, a Flushable, and a Closeable.</div>
<div>Given this view, I think we should add a
Writer.of(Appenable, Flushable, Closeable) to specify the 3
component behaviors of the returned writer.</div>
<div>Each of the 3 arguments can be null, so that component will
be no-op (Writer's Appendable methods only need to trivially
return the Writer itself; all other methods return void).</div>
<div>We will always require all 3 arguments to be passed; a null
component means the caller knowingly demands no-op behavior
for that component.</div>
<div>I believe this approach would be safer, and avoids the
accidental delegation of unwanted features from a given input
Appendable when it happens to duck type.</div>
<div><br>
</div>
<div>Regards,</div>
<div>Chen Liang</div>
</div>
<br>
<div class="gmail_quote gmail_quote_container">
<div dir="ltr" class="gmail_attr">On Sat, Dec 28, 2024 at
10:41 PM Markus KARG <<a
href="mailto:markus@headcrashing.eu" moz-do-not-send="true"
class="moz-txt-link-freetext">markus@headcrashing.eu</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div>
<p>Chen,</p>
<p>thank you for your comments! My ideas to address them
are:</p>
<p>* flush(): If the Appendable implements Flushable, then
perform Flushable.flush() on it. Otherwise, Writer.flush()
will be a no-op (besides checking if Writer is open).</p>
<p>* close(): If the Appendable implements Closeable, then
perform Closeable.close() on it. Otherwise, Writer.close()
will be a no-op (besides calling this.flush() if open, and
internally marking itself as closed).</p>
<p>* Writer.of(Writer): The original sense of the new API is
to create a Writer wrapping non-Writers like
StringBuilder, CharBuffer etc., but not to reduce a Writer
to an Appendable (that would rather be
Appendable.narrow(Writer) or so). IMHO there is neither
any need nor benefit to return a limited Writer instead of
the actual writer. So actually I would plea for directly
returning the given writer itself, so Writer.of(Writer) is
a no-op. I do not see why someone would intentionally pass
in a Writer in the hope to get back a more limited,
non-flushing / non-closing variant of it, and I have a bad
feeling about returning a Writer which is deliberately
cutting away the ability to flush and close without any
technical need. Maybe you could elaborate on your idea if
you have strong feelings about that use case?</p>
<p>* StringWriter: Writer.of() is -by intention- not a "fire
and forget" drop-in replacement, but a "real" Writer. It
comes with a price, but in do not see a big problem here.
If one is such happy with StringWriter that dealing with
IOException would be a no-go, then simply keep the app
as-is. But if one really wants the benefits provided by
Writer.of(), then dealing with IOExcpetion should be worth
it. This is a (IMHO very) low price the programmer has to
pay for the benefit of gaining non-sync, non-copy
behavior. In most code using StringWriter I have seen so
far, IOException was dealt with anyways, as the code was
mostly IO-bound already (it expects "some" Writer, not a
StringWriter, as it wants to perform I/O, but the target
is "by incident" a String).</p>
<p>To sum up: IMHO still it sounds feasible and the benefits
outweigh the costs. :-)<br>
</p>
<p>-Markus</p>
<p><br>
</p>
<div>Am 28.12.2024 um 01:51 schrieb Chen Liang:<br>
</div>
<blockquote type="cite">
<div dir="ltr">
<p dir="ltr">Hi Markus,<br>
I think the idea makes sense, but it comes with more
difficulties than in the case of Reader.of. An
Appendable is a higher abstraction modeling only the
character writing aspects, without concerns with
resource control (such as flush or close).</p>
<p dir="ltr">One detail of note is that Writer itself
implements Appendable, but I don't think the new
method should return a Writer as-is; I think it should
return another writer whose close will not close the
underlying writer as we are only modelling the
appendable behavior without exporting the resource
control methods. Not sure about flush.</p>
<p>One use case you have mentioned is StringWriter.
StringWriter is distinct from StringReader: its write
and append methods do not throw IOE while the base
Writer does. So Writer.of cannot adequately replace
StringWriter without use-site ugliness, until we have
generic types that represent the bottom type.</p>
<p>Regards,</p>
<p>Chen Liang</p>
</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Fri, Dec 20, 2024,
11:12 PM Markus KARG <<a
href="mailto:markus@headcrashing.eu" target="_blank"
moz-do-not-send="true" class="moz-txt-link-freetext">markus@headcrashing.eu</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear
Sirs,<br>
<br>
JDK 24 comes with Reader.of(CharSequence), now let's
provide the <br>
symmetrical counterpart Writer.of(Appendable) in JDK
25! :-)<br>
<br>
For performance reasons, hereby I like to propose the
new public factory <br>
method Writer.of(Appendable). This will provide the
same benefits for <br>
writing, that Reader.of(CharSequence) provides for
reading since JDK 24 <br>
(see JDK-8341566). Before sharing a pull request, I'd
kindly like to <br>
request for comments.<br>
<br>
Since Java 1.1 we have the StringWriter class. Since
Java 1.5 we have <br>
the Appendable interface. StringBuilder, StringBuffer
and CharBuffer are <br>
first-class implementations of it in the JDK, and
there might exist <br>
third-party implementations of non-String text sinks.
Until today, <br>
however, we do not have a Writer for Appendables, but
need to go costly <br>
detours.<br>
<br>
Text sinks in Java are expected to implement the
Writer interface. <br>
Libraries and frameworks expect application code to
provide Writers to <br>
consume text produced by the library or framework, for
example. <br>
Application code often wants to modify the received
text, e. g. embed <br>
received SVG text into in a larger HTML text document,
or simply forward <br>
the text as-is to I/O, so StringBuilder or CharBuffer
is what the <br>
application code actually uses, but not Strings! In
such cases, taking <br>
the StringWriter.toString() detour is common but
inefficient: It implies <br>
duplicating the COMPLETE text for the sole sake of
creating a temporary <br>
String, while the subsequent processing will copy the
data anyways or <br>
just uses a small piece of it. This eats up time and
memory uselessly, <br>
and increases GC pressure. Also, StringWriter is
synchronized (not <br>
explicitly, but de-facto, as it uses StringBuffer),
which implies <br>
another needless slowdown. In many cases, the
synchronization has no use <br>
at all, as in real-world applications least Writers
are actually <br>
accessed concurrently. As a result, today the major
benefit of <br>
StringBuilder over StringBuffer (being
non-synchronized) vanishes as <br>
soon as a StringWriter is used to provide its content.
This means, <br>
"stringBuilder.append(stringWriter.toString())"
imposes slower <br>
performance than essentially needed, in two ways:
toString(), synchronized.<br>
<br>
In an attempt to improve performance of this rather
typical use case, I <br>
like to contribute a pull request providing the new
public factory <br>
method java.io.Writer.of(Appendable). This is
symmetrical to the <br>
solution we implemented in JDK-8341566 for the
reversed case: <br>
java.io.Reader.of(CharSequence).<br>
<br>
My idea is to mostly copy the existing code of
StringWriter, but wrap a <br>
caller-provided Appendable instead of an internally
created <br>
StringBuilder; this strips synchronization; then add
optimized use for <br>
the StringBuffer, StringBuilder and CharBuffer
implementations (in the <br>
sense of write(char[],start,end) to prevent a
char-by-char loop in these <br>
cases).<br>
<br>
Alternatives:<br>
<br>
- Applications could use Apache Commons IO's
StringBuilderWriter, which <br>
is limited to StringBuilder, so is not usable for the
CharBuffer or <br>
custom Appendable case. As it is an open-source
third-party dependency, <br>
some authors might not be allowed to use it, or may
not want to carry <br>
this additional burden just for the sake of this
single performance <br>
improvement. In addition, this library is not actively
modernized; its <br>
Java baseline still is Java 8. There is no commercial
support.<br>
<br>
- Applications could write their own Writer
implementation. Given the <br>
assumption that this is a rather common use case, this
imposes <br>
unjustified additional work for the authors of
thousands of <br>
applications. It is hard to justify why there is a
StringWriter but not <br>
a Writer for other Appendables.<br>
<br>
- Instead of writing a new Writer factory method, we
could slightly <br>
modify StringWriter, so it uses StringBuilder (instead
of StringBuffer). <br>
This (still) results in unnecessary duplication of the
full text at <br>
toString() and (now also) at getBuffer(), and it will
break existing <br>
applications due the missing synchronization.<br>
<br>
- Instead of writing a new Writer factory method, we
could write a new <br>
AppendableWriter class. This piles up the amount of
public classes, <br>
which was the main reason in JDK-8341566 to go with
the <br>
"Reader.of(CharSequence)" factory method instead of
the <br>
"CharSequenceReader" class. Also it would be confusing
to have <br>
Reader.of(...) but not Writer.of(...) in the API.<br>
<br>
- We could go with a specific Appendable class (like
StringBuilder) <br>
instead of supporting all Appendable implementations.
This would reduce <br>
the number of applicable use cases daramatically (in
particular as <br>
CharBuffer is not supported any more) without
providing any considerable <br>
benefit (other than making the OpenJDK-internal source
code a bit <br>
shorter). In particular it makes it impossible to
opt-in for the below <br>
option:<br>
<br>
Option:<br>
<br>
- Once we have Writer.of(Appendable), we could replace
the full <br>
implementation of StringWriter by synchronized calls
to the new Writer. <br>
This would reduce duplicate code.<br>
<br>
Kindly requesting comments.<br>
<br>
-Markus Karg<br>
<br>
</blockquote>
</div>
</blockquote>
</div>
</blockquote>
</div>
</blockquote>
</body>
</html>