<?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/"
	xmlns:creativeCommons="http://backend.userland.com/creativeCommonsRssModule"
	>
<channel>
	<title>Comments on: The future of search and replace in Kate</title>
	<atom:link href="http://blog.hartwork.org/?feed=rss2&#038;p=63" rel="self" type="application/rss+xml" />
	<link>http://blog.hartwork.org/?p=63</link>
	<description>Free Software and Music</description>
	<lastBuildDate>Fri, 10 Sep 2010 22:08:05 +0200</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.6</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Hillary</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-7852</link>
		<dc:creator>Hillary</dc:creator>
		<pubDate>Wed, 02 Jan 2008 17:49:58 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-7852</guid>
		<description>Hi! Definitely nice and neat site you got there.o</description>
		<content:encoded><![CDATA[<p>Hi! Definitely nice and neat site you got there.o</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: /home/liquidat New search and search-and-replace bars for kate &#171;</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1462</link>
		<dc:creator>/home/liquidat New search and search-and-replace bars for kate &#171;</dc:creator>
		<pubDate>Fri, 10 Aug 2007 13:26:54 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1462</guid>
		<description>[...] sping thought about that problem and decided to work at. A couple of days later he showed mockups. The first mockup [...]</description>
		<content:encoded><![CDATA[<p>[...] sping thought about that problem and decided to work at. A couple of days later he showed mockups. The first mockup [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Marcos Dione</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1461</link>
		<dc:creator>Marcos Dione</dc:creator>
		<pubDate>Fri, 10 Aug 2007 12:37:42 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1461</guid>
		<description>kate already has incremental search implemented as a plugin. it shows as a toolbar. maybe you just need to add support of toolbars at the bottom, write another plugin or builtin function as you wish an voilÃ¡!</description>
		<content:encoded><![CDATA[<p>kate already has incremental search implemented as a plugin. it shows as a toolbar. maybe you just need to add support of toolbars at the bottom, write another plugin or builtin function as you wish an voilÃ¡!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sping</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1435</link>
		<dc:creator>sping</dc:creator>
		<pubDate>Thu, 09 Aug 2007 05:17:03 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1435</guid>
		<description>@shaforostoff
Update: Leaving KFind as is but offer the Firefoxy bar as a kdeui dialog for it might be possible. I plan to talk to Eike Hein of Konversion about this.</description>
		<content:encoded><![CDATA[<p>@shaforostoff<br />
Update: Leaving KFind as is but offer the Firefoxy bar as a kdeui dialog for it might be possible. I plan to talk to Eike Hein of Konversion about this.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: sping</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1433</link>
		<dc:creator>sping</dc:creator>
		<pubDate>Thu, 09 Aug 2007 04:43:50 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1433</guid>
		<description>@timmmm
I turned on forced registration some time before I installed Spam Karma 2. I thought about the whole thing again and turned off registration again. Let&#039;s see how that works.

@jankoh
I like Adode Reader&#039;s interface, too. Maybe we can add this feature later. Let&#039;s go for the two less-space-taking dialogs first.

@shaforostoff
I had a closer look at &lt;a href=&quot;http://websvn.kde.org/trunk/KDE/kdelibs/kdeui/findreplace/&quot; target=&quot;_blank&quot; rel=&quot;nofollow&quot;&gt;KFind and KReplace&lt;/a&gt; and it seems to me that it can only work with linear string input (Kate works with vectors of lines), and does not support references in the replacement text when replacing with regular expressions. (Did I overlook it?) Another problem we recently fixed for KatePart which KFind/KReplace still has is the fixed coupling to QRegExp in the API. KatePart only works with QStrings in the API so it would be possible to use another Regex engine for the backend if desired. To summarize I think bringing these two worlds together will be a major extra task. If you think we should go for that I will need your help with it.</description>
		<content:encoded><![CDATA[<p>@timmmm<br />
I turned on forced registration some time before I installed Spam Karma 2. I thought about the whole thing again and turned off registration again. Let&#8217;s see how that works.</p>
<p>@jankoh<br />
I like Adode Reader&#8217;s interface, too. Maybe we can add this feature later. Let&#8217;s go for the two less-space-taking dialogs first.</p>
<p>@shaforostoff<br />
I had a closer look at <a href="http://websvn.kde.org/trunk/KDE/kdelibs/kdeui/findreplace/" target="_blank" rel="nofollow">KFind and KReplace</a> and it seems to me that it can only work with linear string input (Kate works with vectors of lines), and does not support references in the replacement text when replacing with regular expressions. (Did I overlook it?) Another problem we recently fixed for KatePart which KFind/KReplace still has is the fixed coupling to QRegExp in the API. KatePart only works with QStrings in the API so it would be possible to use another Regex engine for the backend if desired. To summarize I think bringing these two worlds together will be a major extra task. If you think we should go for that I will need your help with it.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: shaforostoff</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1429</link>
		<dc:creator>shaforostoff</dc:creator>
		<pubDate>Tue, 07 Aug 2007 23:02:51 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1429</guid>
		<description>&quot;Also the KReplace dialog would again be a dialog in the way when searching.&quot;
i didnt say anything about KReplaceDialog. I just asked if you&#039;re planning to write alternative GUI for KReplace (maybe making some changes to KReplace itself).

Anyways, it would be cool if your class were generic enough to be used in other apps (e.g.: searching/replacing on multiple pages -- like in okular, kword, and KAider has similar concept of msgen/msgtarget pairs too...)</description>
		<content:encoded><![CDATA[<p>&#8220;Also the KReplace dialog would again be a dialog in the way when searching.&#8221;<br />
i didnt say anything about KReplaceDialog. I just asked if you&#8217;re planning to write alternative GUI for KReplace (maybe making some changes to KReplace itself).</p>
<p>Anyways, it would be cool if your class were generic enough to be used in other apps (e.g.: searching/replacing on multiple pages &#8212; like in okular, kword, and KAider has similar concept of msgen/msgtarget pairs too&#8230;)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jankoh</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1428</link>
		<dc:creator>jankoh</dc:creator>
		<pubDate>Tue, 07 Aug 2007 11:14:10 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1428</guid>
		<description>IMHO, one of the best search interfaces
is the one used by Adobe (Acrobat) Reader.

The nicest thing about it is, that you see (in the right panel) lots of other matches, not just the current occurence of the searched pattern.

You can thus quickly jump to the right place(with little scrlling) if needed, and you see also the context, which helps you to jump there.

Such an interface might be really good especially for replace - you would see, what are you about to replace - also the other occurances, which might help you to prevent a mistake.</description>
		<content:encoded><![CDATA[<p>IMHO, one of the best search interfaces<br />
is the one used by Adobe (Acrobat) Reader.</p>
<p>The nicest thing about it is, that you see (in the right panel) lots of other matches, not just the current occurence of the searched pattern.</p>
<p>You can thus quickly jump to the right place(with little scrlling) if needed, and you see also the context, which helps you to jump there.</p>
<p>Such an interface might be really good especially for replace &#8211; you would see, what are you about to replace &#8211; also the other occurances, which might help you to prevent a mistake.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: webmaster</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1427</link>
		<dc:creator>webmaster</dc:creator>
		<pubDate>Mon, 06 Aug 2007 23:13:57 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1427</guid>
		<description>&gt; Registering (easy as it was) for a site just
&gt; to post a comment is stupid - you should at
&gt; least allow moderated comments with a captcha.

I didn&#039;t write this blogging software and I
have to fight tons of Spam. Even if this
is not the most user-friendly solution I don&#039;t
think it&#039;s bad enough to call it stupid.



&gt; â€œ* Has next and previous buttons, with â€œnextâ€ on the leftâ€
&gt; How illogical. Most people read left to right, so things
&gt; that are â€˜nextâ€™ are on the right. Also I think the icons
&gt; should be up/down arrows rather than left right - 2 matches
&gt; on the same line are much less common than matches above/below
&gt; each other.

It&#039;s not necessarily illogical: It&#039;s exactly what Firefox
is doing and they do it because you need &quot;Next&quot; 90% of the
time (not &quot;Prev&quot;) so they put it as close as possible.
I agree about the button icons: Firefox does that as well.



&gt; Just a suggestion - for non regex modes, why not use the
&gt; return symbol (or something that is in common fonts) to
&gt; avoid confusing people with â€˜\nâ€™ (only geeks have even
&gt; heard of escape sequences).

I don&#039;t think using characters that do not appear on the
user&#039;s keyboard is a good idea. I guess non-geeks will
not even try to search for multi-line stuff from a single
pattern edit box.



&gt; â€œ* Wrap around (to not have to ask)â€
&gt; I canâ€™t see why youâ€™d need to ask. I would have thought
&gt; you always wrap around unless you have â€˜from cursorâ€™
&gt; and use replace all.

The question is whether you want (A) to find any match
whereever it is or (B) a match from here on or further
down. The Firefoxy bar will give (A) only while
the power search will do (A) or (B) for you.</description>
		<content:encoded><![CDATA[<p>> Registering (easy as it was) for a site just<br />
> to post a comment is stupid &#8211; you should at<br />
> least allow moderated comments with a captcha.</p>
<p>I didn&#8217;t write this blogging software and I<br />
have to fight tons of Spam. Even if this<br />
is not the most user-friendly solution I don&#8217;t<br />
think it&#8217;s bad enough to call it stupid.</p>
<p>> â€œ* Has next and previous buttons, with â€œnextâ€ on the leftâ€<br />
> How illogical. Most people read left to right, so things<br />
> that are â€˜nextâ€™ are on the right. Also I think the icons<br />
> should be up/down arrows rather than left right &#8211; 2 matches<br />
> on the same line are much less common than matches above/below<br />
> each other.</p>
<p>It&#8217;s not necessarily illogical: It&#8217;s exactly what Firefox<br />
is doing and they do it because you need &#8220;Next&#8221; 90% of the<br />
time (not &#8220;Prev&#8221;) so they put it as close as possible.<br />
I agree about the button icons: Firefox does that as well.</p>
<p>> Just a suggestion &#8211; for non regex modes, why not use the<br />
> return symbol (or something that is in common fonts) to<br />
> avoid confusing people with â€˜\nâ€™ (only geeks have even<br />
> heard of escape sequences).</p>
<p>I don&#8217;t think using characters that do not appear on the<br />
user&#8217;s keyboard is a good idea. I guess non-geeks will<br />
not even try to search for multi-line stuff from a single<br />
pattern edit box.</p>
<p>> â€œ* Wrap around (to not have to ask)â€<br />
> I canâ€™t see why youâ€™d need to ask. I would have thought<br />
> you always wrap around unless you have â€˜from cursorâ€™<br />
> and use replace all.</p>
<p>The question is whether you want (A) to find any match<br />
whereever it is or (B) a match from here on or further<br />
down. The Firefoxy bar will give (A) only while<br />
the power search will do (A) or (B) for you.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Timmmm</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1426</link>
		<dc:creator>Timmmm</dc:creator>
		<pubDate>Mon, 06 Aug 2007 22:47:55 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1426</guid>
		<description>My thoughts:

Registering (easy as it was) for a site just to post a comment is stupid - you should at least allow moderated comments with a captcha.

&quot;* Has next and previous buttons, with &quot;next&quot; on the left&quot;

How illogical. Most people read left to right, so things that are &#039;next&#039; are on the right. Also I think the icons should be up/down arrows rather than left right - 2 matches on the same line are much less common than matches above/below each other.

&quot;* Search always starts from cursor&quot;

Good. KDE 3 always annoyed me when searching in a selection, and it asked if you wanted to start from the cursor.

&quot;* &quot;&gt;&quot; button right to each edit control
     with stuff like &quot;\n&quot; and &quot;.&quot; or to launch
     KRegexpEditor if installed&quot;

Just a suggestion - for non regex modes, why not use the return symbol (or something that is in common fonts) to avoid confusing people with &#039;\n&#039; (only geeks have even heard of escape sequences).

&quot;* Wrap around (to not have to ask)&quot;

I can&#039;t see why you&#039;d need to ask. I would have thought you always wrap around unless you have &#039;from cursor&#039; and use replace all.</description>
		<content:encoded><![CDATA[<p>My thoughts:</p>
<p>Registering (easy as it was) for a site just to post a comment is stupid &#8211; you should at least allow moderated comments with a captcha.</p>
<p>&#8220;* Has next and previous buttons, with &#8220;next&#8221; on the left&#8221;</p>
<p>How illogical. Most people read left to right, so things that are &#8216;next&#8217; are on the right. Also I think the icons should be up/down arrows rather than left right &#8211; 2 matches on the same line are much less common than matches above/below each other.</p>
<p>&#8220;* Search always starts from cursor&#8221;</p>
<p>Good. KDE 3 always annoyed me when searching in a selection, and it asked if you wanted to start from the cursor.</p>
<p>&#8220;* &#8220;&gt;&#8221; button right to each edit control<br />
     with stuff like &#8220;\n&#8221; and &#8220;.&#8221; or to launch<br />
     KRegexpEditor if installed&#8221;</p>
<p>Just a suggestion &#8211; for non regex modes, why not use the return symbol (or something that is in common fonts) to avoid confusing people with &#8216;\n&#8217; (only geeks have even heard of escape sequences).</p>
<p>&#8220;* Wrap around (to not have to ask)&#8221;</p>
<p>I can&#8217;t see why you&#8217;d need to ask. I would have thought you always wrap around unless you have &#8216;from cursor&#8217; and use replace all.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: webmaster</title>
		<link>http://blog.hartwork.org/?p=63&#038;cpage=1#comment-1425</link>
		<dc:creator>webmaster</dc:creator>
		<pubDate>Mon, 06 Aug 2007 22:40:00 +0000</pubDate>
		<guid isPermaLink="false">http://blog.hartwork.org/?p=63#comment-1425</guid>
		<description>With &quot;The engine inside will also be more powerful&quot; I meant stuff like support for multi-line patterns and escape sequences in the replacement text.

I do not plan to duplicate KReplace but I currently do not plan using KReplace either. The idea is to have a &quot;Replace&quot; and &quot;Replace all&quot; button. That together with &quot;Find next&quot; should give the same power as KReplace. Also the KReplace dialog would again be a dialog in the way when searching.</description>
		<content:encoded><![CDATA[<p>With &#8220;The engine inside will also be more powerful&#8221; I meant stuff like support for multi-line patterns and escape sequences in the replacement text.</p>
<p>I do not plan to duplicate KReplace but I currently do not plan using KReplace either. The idea is to have a &#8220;Replace&#8221; and &#8220;Replace all&#8221; button. That together with &#8220;Find next&#8221; should give the same power as KReplace. Also the KReplace dialog would again be a dialog in the way when searching.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
