<?xml version="1.0" encoding="utf-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: src on web</title>
	<atom:link href="http://www.e-pig.org/epilogue/?feed=rss2&#038;p=17" rel="self" type="application/rss+xml" />
	<link>http://www.e-pig.org/epilogue/?p=17</link>
	<description>for Epigram</description>
	<lastBuildDate>Sun,  6 Nov 2011 09:25:45 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.4</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: pcc</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-68</link>
		<dc:creator>pcc</dc:creator>
		<pubDate>Tue, 25 Jan 2005 13:14:52 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-68</guid>
		<description>oh cr*p - the regexp has been severely edited 
mail me if you want the uncensored version</description>
		<content:encoded><![CDATA[<p>oh cr*p &#8211; the regexp has been severely edited<br />
mail me if you want the uncensored version</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: pcc</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-67</link>
		<dc:creator>pcc</dc:creator>
		<pubDate>Tue, 25 Jan 2005 13:12:27 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-67</guid>
		<description>This does the job for http urls (only partially tested)
Put this all on one line, or change it into a shell script

perl -pe &#039;BEGIN {$l=shift;$/=&quot;nn&quot;;} s/&lt;a (s+)href=&quot;(?!http://)([^&gt;]+)&gt;/&lt;a 1href=$l2&gt;/gis&#039; &quot;http://sneezy.cs.nott.ac.uk/&quot; file1 file2 ...&lt;/a&gt;&lt;/a&gt;
It does update in place, putting sneezy etc before links which don&#039;t have a http:// prefix. 


A quick hack for the tools is just to extend the syntax so it recognises infix tycons, but stores them as conventional tycons.</description>
		<content:encoded><![CDATA[<p>This does the job for http urls (only partially tested)<br />
Put this all on one line, or change it into a shell script</p>
<p>perl -pe &#8216;BEGIN {$l=shift;$/=&#8221;nn&#8221;;} s/<a (s+)href="(?!http://)([^>]+)>/</a><a 1href=$l2>/gis&#8217; &#8220;http://sneezy.cs.nott.ac.uk/&#8221; file1 file2 &#8230;</a><br />
It does update in place, putting sneezy etc before links which don&#8217;t have a http:// prefix. </p>
<p>A quick hack for the tools is just to extend the syntax so it recognises infix tycons, but stores them as conventional tycons.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Conor</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-66</link>
		<dc:creator>Conor</dc:creator>
		<pubDate>Tue, 25 Jan 2005 11:40:30 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-66</guid>
		<description>Frankly, for pretty much ephemeral things like infix type constructors, I don&#039;t think its worth any of our efforts fiddling with the tools. By the way, does anyone have a script which prepends a given path to every relative link in an html file? I&#039;m just thinking about how we might process chunks of programatica output to make something bloggable. Also search and replace, so &#92; becomes &#955;, -&gt; becomes &#8594; and idiom brackets don&#039;t look totally stupid: I guess that&#039;s a no- brainer.</description>
		<content:encoded><![CDATA[<p>Frankly, for pretty much ephemeral things like infix type constructors, I don&#8217;t think its worth any of our efforts fiddling with the tools. By the way, does anyone have a script which prepends a given path to every relative link in an html file? I&#8217;m just thinking about how we might process chunks of programatica output to make something bloggable. Also search and replace, so &#92; becomes &lambda;, -&gt; becomes &rarr; and idiom brackets don&#8217;t look totally stupid: I guess that&#8217;s a no- brainer.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: pcc</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-63</link>
		<dc:creator>pcc</dc:creator>
		<pubDate>Tue, 25 Jan 2005 09:42:27 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-63</guid>
		<description>Using &quot;ghc -E&quot;, the simplest way to embed haddock info is to write comments with -- and prefix them with &gt;. Such comments survive de-litting.

Rather than remove infix type constrs etc, it isn&#039;t impossible to get a Ph^H^Hvolunteer to extend the relevant tools (or moan to the authors) about the missing syntax. Haddock objects to (n+1) patterns! Failing that, a perl script might be able to rewrite the 
offending symbols.</description>
		<content:encoded><![CDATA[<p>Using &#8220;ghc -E&#8221;, the simplest way to embed haddock info is to write comments with &#8212; and prefix them with >. Such comments survive de-litting.</p>
<p>Rather than remove infix type constrs etc, it isn&#8217;t impossible to get a Ph^H^Hvolunteer to extend the relevant tools (or moan to the authors) about the missing syntax. Haddock objects to (n+1) patterns! Failing that, a perl script might be able to rewrite the<br />
offending symbols.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Conor</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-62</link>
		<dc:creator>Conor</dc:creator>
		<pubDate>Tue, 25 Jan 2005 08:37:45 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-62</guid>
		<description>So ghc spits out the illiterate files for haddock? That&#039;s handy. What I still don&#039;t get is how haddock&#039;s special comment formats go in a .lhs file. I suppose I&#039;m not so very attached to .lhs that I couldn&#039;t just switch. Maybe we could start a top-level src directory with .hs files, no infix type constructors, no pattern guards. How terribly responsible.</description>
		<content:encoded><![CDATA[<p>So ghc spits out the illiterate files for haddock? That&#8217;s handy. What I still don&#8217;t get is how haddock&#8217;s special comment formats go in a .lhs file. I suppose I&#8217;m not so very attached to .lhs that I couldn&#8217;t just switch. Maybe we could start a top-level src directory with .hs files, no infix type constructors, no pattern guards. How terribly responsible.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: pcc</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-61</link>
		<dc:creator>pcc</dc:creator>
		<pubDate>Mon, 24 Jan 2005 19:23:09 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-61</guid>
		<description>Take a look at the old Makefile, under the &quot;html&quot; target. That uses ghc to de-lit the .lhs and pass it safely through haddock.

So, no change needed.</description>
		<content:encoded><![CDATA[<p>Take a look at the old Makefile, under the &#8220;html&#8221; target. That uses ghc to de-lit the .lhs and pass it safely through haddock.</p>
<p>So, no change needed.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Conor</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-60</link>
		<dc:creator>Conor</dc:creator>
		<pubDate>Mon, 24 Jan 2005 19:06:28 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-60</guid>
		<description>Not that I&#039;m aware of. Haddock provides a useful complement. Does this mean I need to switch to using .hs files instead of .lhs? I&#039;ve never got Haddock to work with the latter.</description>
		<content:encoded><![CDATA[<p>Not that I&#8217;m aware of. Haddock provides a useful complement. Does this mean I need to switch to using .hs files instead of .lhs? I&#8217;ve never got Haddock to work with the latter.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: pcc</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-59</link>
		<dc:creator>pcc</dc:creator>
		<pubDate>Mon, 24 Jan 2005 18:49:14 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-59</guid>
		<description>Can this produce interface-only output? (I&#039;d find it useful)

Else, shall I add a makefile which generates Haddock output to some convenient place?</description>
		<content:encoded><![CDATA[<p>Can this produce interface-only output? (I&#8217;d find it useful)</p>
<p>Else, shall I add a makefile which generates Haddock output to some convenient place?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Conor</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-57</link>
		<dc:creator>Conor</dc:creator>
		<pubDate>Mon, 24 Jan 2005 16:53:37 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-57</guid>
		<description>I&#039;ve just had a go at making the Programatica front end eat the story so far. You can see how I got on &lt;a href=&quot;http://www.cs.nott.ac.uk/~ctm/pfetest/TermParser.html&quot;&gt;here&lt;/a&gt;. I had to drop some infix type constructors (no great loss, I suppose) and some pattern guards (more of an issue with me), but it&#039;s better than a poke in the eye. Thanks &lt;a href=&quot;http://www.cse.ogi.edu/~hallgren/&quot;&gt;Thomas&lt;/a&gt;!</description>
		<content:encoded><![CDATA[<p>I&#8217;ve just had a go at making the Programatica front end eat the story so far. You can see how I got on <a href="http://www.cs.nott.ac.uk/~ctm/pfetest/TermParser.html">here</a>. I had to drop some infix type constructors (no great loss, I suppose) and some pattern guards (more of an issue with me), but it&#8217;s better than a poke in the eye. Thanks <a href="http://www.cse.ogi.edu/~hallgren/">Thomas</a>!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: pwm</title>
		<link>http://www.e-pig.org/epilogue/?p=17&#038;cpage=1#comment-55</link>
		<dc:creator>pwm</dc:creator>
		<pubDate>Mon, 24 Jan 2005 15:53:38 +0000</pubDate>
		<guid isPermaLink="false">http://www.e-pig.org/epilogue/?p=17#comment-55</guid>
		<description>Done</description>
		<content:encoded><![CDATA[<p>Done</p>
]]></content:encoded>
	</item>
</channel>
</rss>
