factor/CHANGES.html

<!-- :noWordSep=+-*\=><;.?/'()%,_|: -->

<html>
<head><title>Factor change log</title></head>
<body>

<h1>Factor 0.77:</h1>

<ul>
<li>Compiler:
<ul>
<li>Optimizing out conditionals where the test value is a constant.</li>
<li>Optimizing out type checks that are always/never satisfied.</li>
<li>Inlining method bodies when generic words are called on values with known compile-time types.</li>
<li>Side-effect-free words that output immutable values are evaluated at compile time if all their inputs are literal. You can declare a word as having this condition by suffixing the definition with <code>foldable</code>, eg:
<pre>: cube dup dup * * ; foldable</pre></li>
<li>Various arithmetic identities such as <code>1 *</code> are optimized out.
</ul>

</li>

<li>Collections:

<ul>
<li><code>2each ( seq seq quot -- quot: elt -- elt )</code> combinator</li>
<li><code>join ( seq glue -- seq )</code> word. Takes a sequence of sequences, and constructs a new sequence with the glue in between each sequence. For example:
<pre>  [ "usr" "bin" "grep" ] "/" join
<b>"usr/bin/grep"</b></pre></li>
<li>Integers now support the sequence protocol. An integer is an increasing sequence of its predecessors. This means the <code>count ( n -- [ 0 ... n-1 ] )</code> word is gone; just use <code>&gt;vector</code> instead. Also, <code>project</code> has been made redundant by <code>map</code>.</li>
<li>The <code>seq-transpose ( seq -- seq )</code> word is now named <code>flip</code>.
</li>
<li>The matrices library has been greatly simplified. Matrices are now represented as vectors of vectors, and matrix words have been moved to the <code>math</code> vocabulary.</li>
<li>More descriptive "out of bounds" errors.</li>
<li>New <code>make-hash ( quot -- namespace )</code> combinator executes quotation in a new namespace, which is then pushed on the stack.</li>
<li>The <code>&lt;namespace&gt;</code> word is gone. It would create a hashtable with a default capacity. Now, just write <code>{{ }} clone</code>.
</ul>

</li>

<li>Development tools:

<ul>
<li>In the UI, object slots are now clickable in the inspector.</li>
<li>Inspector now supports a history and an interactive loop; it prints a brief help message when it starts describing usage.</li>
<li>The prettyprinter has been merged with the unparser. The <code>unparse ( object -- string )</code> word has been moved to the <code>prettyprint</code> vocabulary, and can now produce a parsable string for any class supported by the prettyprinter.</li>
<li>New <code>unparse-short ( object -- string )</code> returns a string no longer than a single line.</li>
<li>The prettyprinter now supports many more configuration variables. See the handbook for details.</li>
<li>New <code>profile ( word -- )</code> word. Causes the word's accumulative runtime to be stored in a global variable named by the word. This is done with the annotation facility, the word's definition is modified; use <code>reload ( word -- )</code> to get the old definition back from the source file.</li>
</ul>

</li>


<li>Everything else:

<ul>
<li>New <code>sleep ( ms -- )</code> word pauses current thread for a number of milliseconds.</li>
<li>New <code>with-datastack ( stack word -- stack )</code> combinator.</li>
<li>New <code>cond ( conditions -- )</code> combinator. It behaves like a set of nested <code>ifte</code>s; see its documentation comment for details. Note that it does not compile.</li>
<li>Erlang/Termite-style concurrency library in <code>contrib/concurrency</code> (Chris Double).</li>
<li>Completely redid infix algebra in <code>conrib/algebra/</code>. Now, vector operations are possible
and the syntax doesn't use so many spaces. New way to write the quadratic formula:
<pre>MATH: quadratic[a;b;c] =
    plusmin[(-b)/2*a;(sqrt(b^2)-4*a*c)/2*a] ;</pre>
(Daniel Ehrenberg)</li>
</ul>

</li>

</ul>

<h1>Factor 0.76:</h1>

<ul>
<li>
UI framework:
<ul>
<li>Now uses 3-dimensional co-ordinates throughout</li>
<li>Gradient paint, bevel border paint</li>
<li>Split pane gadget</li>
<li>Horizontal scroll bars and wheel mouse scrolling in scroller gadgets</li>
<li>Incremental layout improves listener responsiveness</li>
<li>The listener supports styled text output and presentations</li>
<li>Slide-show tutorial with live code examples</li>
<li>Performance improvements, code cleanups, bug fixes</li>
</ul>
</li>
<li>
Sequences:
<ul>
<li>The following formely list-specific words are now generic:
    
    <pre>all? ( seq quot -- ? | quot: elt -- ? )
all-with? ( obj seq quot -- ? | quot: elt -- ? )
subset ( seq quot -- seq | quot: elt -- ? )
subset-with ( obj seq quot -- seq | quot: obj elt -- ? )
fiber? ( seq quot -- ? | quot: elt elt -- ? )
prune ( seq -- seq )</pre>

<li> The <code>contains?</code> word for testing membership in a sequence has been
  renamed to <code>member? ( elt seq -- ? )</code>.

<li> The list-specific <code>some?</code> and <code>some-with?</code> combinators are gone. Their replacements are generic:
  
  <pre>contains? ( seq quot -- ? | quot: elt -- ? )
contains-with? ( obj seq quot -- ? | quot: obj elt -- ? )
find ( seq quot -- i elt | quot: elt -- ? )
find* ( i seq quot -- i elt | quot: elt -- ? )
find-with ( obj seq quot -- i elt | quot: elt -- ? )
find-with* ( obj i seq quot -- i elt | quot: elt -- ? )</pre>

  See the developer's handbook for details.

<li> The <code>nreverse ( seq -- )</code> word has been removed.

<li> <code>reverse-slice ( seq -- seq )</code> outputs a new sequence that shares
  structure with the given sequence, but presents elements in reverse
  order.

<li> The <code>string-compare</code> primitive has been replaced with the lexi word
  which now operates on any pair of sequences of numbers. The
  string> word has been replaced with <code>lexi></code>.

<li> The <code>,</code> word no longer accepts a string as input inside a <code>make-string</code>. In 0.75, the following
  two lines were equivalent:
  
  <pre>[ "Hello" , " world" , ] make-string
[ "Hello" % " world" % ] make-string</pre>
  
<li> Now, the former raises a type error. Use <code>,</code> with characters, and <code>%</code> with
  strings inside make-string.
</ul>

<li>Streams:

<ul>
<li>The following words have been renamed:
  
    <pre>stream-auto-flush ==> stream-finish ( stream -- )
stream-write-attr ==> stream-format ( string style stream -- )
write-attr ==> format ( string style -- )</pre>

<li>The following words no longer accept character arguments:
  
    <pre>stream-format ( string style stream -- )
format ( string style -- )
stream-write ( string stream -- )
write ( string -- )
stream-print ( string -- )
print ( string -- )</pre>

<li>Use the new words to write characters:
  
    <pre>stream-write1 ( char stream -- )
write1 ( char -- )</pre>

  Note that <code>stream-write1</code> is generic and your stream must implement it.

<li><code>with-string</code> word renamed to <code>string-out ( quot -- string )</code>

<li>New <code>string-in ( string quot -- )</code> word, calls <code>quot</code> with <code>stdio</code> bound to
  a stream that reads from the given string.
</ul>

<li>Everything else:

<ul>

<li>Many improvements to the matrices library.

<li>Improved inspector. Call it with <code>inspect ( obj -- )</code>.

<li>The number of generations used for garbage collection can now be set
  with the +G command line switch. You must specify at least 2
  generations.

<li>Only 2 generations are used by default now, since there seems to be no
  performance benefit to having 3 after running some brief benchmarks.

<li>Fixed bug where images saved from the jEdit plugin would fail to
  start.

<li>md5 hashing algorithm in <code>contrib/crypto/</code> (Doug Coleman).


</ul>

</ul>

</body>
</html>
document 0.77 changes, print tuple addr in debugger 2005-07-24 23:12:34 -04:00			`<!-- :noWordSep=+-*\=><;.?/'()%,_\|: -->`

handbook updates, HTML-format changelog 2005-07-21 21:43:37 -04:00			`<html>`
			`<head><title>Factor change log</title></head>`
			`<body>`

document 0.77 changes, print tuple addr in debugger 2005-07-24 23:12:34 -04:00			`<h1>Factor 0.77:</h1>`

			`<ul>`
CHANGES updates 2005-08-03 19:09:44 -04:00			`<li>Compiler:`
			`<ul>`
			`<li>Optimizing out conditionals where the test value is a constant.</li>`
			`<li>Optimizing out type checks that are always/never satisfied.</li>`
			`<li>Inlining method bodies when generic words are called on values with known compile-time types.</li>`
collapse slice of a slice; %fast-set-slot linearizer was buggy 2005-08-15 23:09:44 -04:00			`<li>Side-effect-free words that output immutable values are evaluated at compile time if all their inputs are literal. You can declare a word as having this condition by suffixing the definition with <code>foldable</code>, eg:`
			`<pre>: cube dup dup * * ; foldable</pre></li>`
			`<li>Various arithmetic identities such as <code>1 *</code> are optimized out.`
CHANGES updates 2005-08-03 19:09:44 -04:00			`</ul>`

			`</li>`

major cleanup 2005-08-25 15:27:38 -04:00			`<li>Collections:`
document 0.77 changes, print tuple addr in debugger 2005-07-24 23:12:34 -04:00
			`<ul>`
			`<li><code>2each ( seq seq quot -- quot: elt -- elt )</code> combinator</li>`
updating CHANGES.html 2005-08-23 15:02:00 -04:00			`<li><code>join ( seq glue -- seq )</code> word. Takes a sequence of sequences, and constructs a new sequence with the glue in between each sequence. For example:`
document 0.77 changes, print tuple addr in debugger 2005-07-24 23:12:34 -04:00			`<pre> [ "usr" "bin" "grep" ] "/" join`
			`<b>"usr/bin/grep"</b></pre></li>`
dataflow optimizer work, phasing out 3list/3unlist 2005-07-27 20:13:11 -04:00			`<li>Integers now support the sequence protocol. An integer is an increasing sequence of its predecessors. This means the <code>count ( n -- [ 0 ... n-1 ] )</code> word is gone; just use <code>>vector</code> instead. Also, <code>project</code> has been made redundant by <code>map</code>.</li>`
CHANGES updates 2005-08-03 19:09:44 -04:00			`<li>The <code>seq-transpose ( seq -- seq )</code> word is now named <code>flip</code>.`
\ foo is now a first-class type; this removes the need for inference-level partial eval, and simplifies a lot of other code 2005-08-03 23:56:28 -04:00			`</li>`
major cleanup 2005-08-25 15:27:38 -04:00			`<li>The matrices library has been greatly simplified. Matrices are now represented as vectors of vectors, and matrix words have been moved to the <code>math</code> vocabulary.</li>`
			`<li>More descriptive "out of bounds" errors.</li>`
			`<li>New <code>make-hash ( quot -- namespace )</code> combinator executes quotation in a new namespace, which is then pushed on the stack.</li>`
			`<li>The <code><namespace></code> word is gone. It would create a hashtable with a default capacity. Now, just write <code>{{ }} clone</code>.`
document 0.77 changes, print tuple addr in debugger 2005-07-24 23:12:34 -04:00			`</ul>`

			`</li>`

major cleanup 2005-08-25 15:27:38 -04:00			`<li>Development tools:`
more prettyprinter updates, unparser is gone 2005-08-21 20:50:14 -04:00
			`<ul>`
major cleanup 2005-08-25 15:27:38 -04:00			`<li>In the UI, object slots are now clickable in the inspector.</li>`
			`<li>Inspector now supports a history and an interactive loop; it prints a brief help message when it starts describing usage.</li>`
more prettyprinter updates, unparser is gone 2005-08-21 20:50:14 -04:00			`<li>The prettyprinter has been merged with the unparser. The <code>unparse ( object -- string )</code> word has been moved to the <code>prettyprint</code> vocabulary, and can now produce a parsable string for any class supported by the prettyprinter.</li>`
			`<li>New <code>unparse-short ( object -- string )</code> returns a string no longer than a single line.</li>`
			`<li>The prettyprinter now supports many more configuration variables. See the handbook for details.</li>`
major cleanup 2005-08-25 15:27:38 -04:00			`<li>New <code>profile ( word -- )</code> word. Causes the word's accumulative runtime to be stored in a global variable named by the word. This is done with the annotation facility, the word's definition is modified; use <code>reload ( word -- )</code> to get the old definition back from the source file.</li>`
more prettyprinter updates, unparser is gone 2005-08-21 20:50:14 -04:00			`</ul>`

			`</li>`

major cleanup 2005-08-25 15:27:38 -04:00
\ foo is now a first-class type; this removes the need for inference-level partial eval, and simplifies a lot of other code 2005-08-03 23:56:28 -04:00			`<li>Everything else:`

			`<ul>`
refactoring timers 2005-08-23 20:27:42 -04:00			`<li>New <code>sleep ( ms -- )</code> word pauses current thread for a number of milliseconds.</li>`
faster hashtable = method, dataflow optimizer cleanups 2005-08-11 19:08:22 -04:00			`<li>New <code>with-datastack ( stack word -- stack )</code> combinator.</li>`
			`<li>New <code>cond ( conditions -- )</code> combinator. It behaves like a set of nested <code>ifte</code>s; see its documentation comment for details. Note that it does not compile.</li>`
major cleanup 2005-08-25 15:27:38 -04:00			`<li>Erlang/Termite-style concurrency library in <code>contrib/concurrency</code> (Chris Double).</li>`
updating CHANGES.html 2005-08-23 15:02:00 -04:00			`<li>Completely redid infix algebra in <code>conrib/algebra/</code>. Now, vector operations are possible`
			`and the syntax doesn't use so many spaces. New way to write the quadratic formula:`
			`<pre>MATH: quadratic[a;b;c] =`
			`plusmin[(-b)/2a;(sqrt(b^2)-4ac)/2a] ;</pre>`
			`(Daniel Ehrenberg)</li>`
document 0.77 changes, print tuple addr in debugger 2005-07-24 23:12:34 -04:00			`</ul>`

\ foo is now a first-class type; this removes the need for inference-level partial eval, and simplifies a lot of other code 2005-08-03 23:56:28 -04:00			`</li>`

			`</ul>`

handbook updates, HTML-format changelog 2005-07-21 21:43:37 -04:00			`<h1>Factor 0.76:</h1>`

			`<ul>`
			`<li>`
various code cleanups, removing some redundant calls to >list, fix i/o timeouts 2005-07-22 22:18:47 -04:00			`UI framework:`
handbook updates, HTML-format changelog 2005-07-21 21:43:37 -04:00			`<ul>`
			`<li>Now uses 3-dimensional co-ordinates throughout</li>`
			`<li>Gradient paint, bevel border paint</li>`
various code cleanups, removing some redundant calls to >list, fix i/o timeouts 2005-07-22 22:18:47 -04:00			`<li>Split pane gadget</li>`
			`<li>Horizontal scroll bars and wheel mouse scrolling in scroller gadgets</li>`
			`<li>Incremental layout improves listener responsiveness</li>`
			`<li>The listener supports styled text output and presentations</li>`
handbook updates, HTML-format changelog 2005-07-21 21:43:37 -04:00			`<li>Slide-show tutorial with live code examples</li>`
			`<li>Performance improvements, code cleanups, bug fixes</li>`
			`</ul>`
			`</li>`
			`<li>`
			`Sequences:`
			`<ul>`
			`<li>The following formely list-specific words are now generic:`

			`<pre>all? ( seq quot -- ? \| quot: elt -- ? )`
			`all-with? ( obj seq quot -- ? \| quot: elt -- ? )`
			`subset ( seq quot -- seq \| quot: elt -- ? )`
			`subset-with ( obj seq quot -- seq \| quot: obj elt -- ? )`
			`fiber? ( seq quot -- ? \| quot: elt elt -- ? )`
			`prune ( seq -- seq )</pre>`

			`<li> The <code>contains?</code> word for testing membership in a sequence has been`
			`renamed to <code>member? ( elt seq -- ? )</code>.`

			`<li> The list-specific <code>some?</code> and <code>some-with?</code> combinators are gone. Their replacements are generic:`

			`<pre>contains? ( seq quot -- ? \| quot: elt -- ? )`
			`contains-with? ( obj seq quot -- ? \| quot: obj elt -- ? )`
			`find ( seq quot -- i elt \| quot: elt -- ? )`
			`find* ( i seq quot -- i elt \| quot: elt -- ? )`
			`find-with ( obj seq quot -- i elt \| quot: elt -- ? )`
			`find-with* ( obj i seq quot -- i elt \| quot: elt -- ? )</pre>`

			`See the developer's handbook for details.`

			`<li> The <code>nreverse ( seq -- )</code> word has been removed.`

			`<li> <code>reverse-slice ( seq -- seq )</code> outputs a new sequence that shares`
			`structure with the given sequence, but presents elements in reverse`
			`order.`

			`<li> The <code>string-compare</code> primitive has been replaced with the lexi word`
			`which now operates on any pair of sequences of numbers. The`
various code cleanups, removing some redundant calls to >list, fix i/o timeouts 2005-07-22 22:18:47 -04:00			`string> word has been replaced with <code>lexi></code>.`
handbook updates, HTML-format changelog 2005-07-21 21:43:37 -04:00
			`<li> The <code>,</code> word no longer accepts a string as input inside a <code>make-string</code>. In 0.75, the following`
			`two lines were equivalent:`

			`<pre>[ "Hello" , " world" , ] make-string`
			`[ "Hello" % " world" % ] make-string</pre>`

			`<li> Now, the former raises a type error. Use <code>,</code> with characters, and <code>%</code> with`
			`strings inside make-string.`
			`</ul>`

			`<li>Streams:`

			`<ul>`
			`<li>The following words have been renamed:`

			`<pre>stream-auto-flush ==> stream-finish ( stream -- )`
			`stream-write-attr ==> stream-format ( string style stream -- )`
			`write-attr ==> format ( string style -- )</pre>`

			`<li>The following words no longer accept character arguments:`

			`<pre>stream-format ( string style stream -- )`
			`format ( string style -- )`
			`stream-write ( string stream -- )`
			`write ( string -- )`
			`stream-print ( string -- )`
			`print ( string -- )</pre>`

			`<li>Use the new words to write characters:`

			`<pre>stream-write1 ( char stream -- )`
			`write1 ( char -- )</pre>`

			`Note that <code>stream-write1</code> is generic and your stream must implement it.`

			`<li><code>with-string</code> word renamed to <code>string-out ( quot -- string )</code>`

			`<li>New <code>string-in ( string quot -- )</code> word, calls <code>quot</code> with <code>stdio</code> bound to`
			`a stream that reads from the given string.`
			`</ul>`

various code cleanups, removing some redundant calls to >list, fix i/o timeouts 2005-07-22 22:18:47 -04:00			`<li>Everything else:`

			`<ul>`

handbook updates, HTML-format changelog 2005-07-21 21:43:37 -04:00			`<li>Many improvements to the matrices library.`

			`<li>Improved inspector. Call it with <code>inspect ( obj -- )</code>.`

			`<li>The number of generations used for garbage collection can now be set`
			`with the +G command line switch. You must specify at least 2`
			`generations.`

			`<li>Only 2 generations are used by default now, since there seems to be no`
			`performance benefit to having 3 after running some brief benchmarks.`

			`<li>Fixed bug where images saved from the jEdit plugin would fail to`
			`start.`

			`<li>md5 hashing algorithm in <code>contrib/crypto/</code> (Doug Coleman).`

updating CHANGES.html 2005-08-23 15:02:00 -04:00
handbook updates, HTML-format changelog 2005-07-21 21:43:37 -04:00			`</ul>`
various code cleanups, removing some redundant calls to >list, fix i/o timeouts 2005-07-22 22:18:47 -04:00
			`</ul>`
factorbot fix, starting 0.77 2005-07-24 19:08:32 -04:00
			`</body>`
			`</html>`