Comments for PatronusCharm Techblog http://blog.patronuscharm.net What goes on under the hood of PatronusCharm.Net and current development. Wed, 17 Mar 2010 03:43:38 +0000 http://wordpress.org/?v=2.6 Comment on New Feature: Document Multi-Uploader by PatronusCharm 1.1 Beta Release - Midnight CST | PatronusCharm Techblog http://blog.patronuscharm.net/2008/07/new-feature-document-multi-uploader/#comment-189 PatronusCharm 1.1 Beta Release - Midnight CST | PatronusCharm Techblog Thu, 31 Dec 2009 04:57:31 +0000 http://blog.patronuscharm.net/?p=8#comment-189 [...] Document Multi-Uploader [...] [...] Document Multi-Uploader [...]

]]> Comment on Sphinx Search: MySQL Fulltext Search Engine Replacement by PatronusCharm 1.1 Beta Release - 11:00 CST | PatronusCharm Techblog http://blog.patronuscharm.net/2008/07/sphinx-search-mysql-fulltext-search-engine-replacement/#comment-188 PatronusCharm 1.1 Beta Release - 11:00 CST | PatronusCharm Techblog Thu, 31 Dec 2009 04:53:33 +0000 http://blog.patronuscharm.net/?p=11#comment-188 [...] as you may have seen from my post Sphinx Search: MySQL Fulltext Search Engine Replacement is a replacement for the search engine from PatronusCharm 1.0.1. It’s faster, finds more, and [...] [...] as you may have seen from my post Sphinx Search: MySQL Fulltext Search Engine Replacement is a replacement for the search engine from PatronusCharm 1.0.1. It’s faster, finds more, and [...]

]]> Comment on XML Syndication by Lord Ravenclaw http://blog.patronuscharm.net/2008/07/xml-syndication/#comment-13 Lord Ravenclaw Sun, 27 Jul 2008 20:37:57 +0000 http://blog.patronuscharm.net/?p=6#comment-13 I love feedback like this :) I was slightly confused by this as I purposefully removed any cache control header from the feeds long ago. I couldn't figure out why until I did a little digging. Apparently, using sessions as I do throughout the site by default adds the aggressive cache headers. I've since changed my local copy and will be synchronized with the development copy later on tonight once all my changes are at least mostly working. I've been reworking feeds completely as I said above, and this works into it well. Thanks for bringing it to my attention, I'll email you for some testing assistance later. I love feedback like this :)

I was slightly confused by this as I purposefully removed any cache control header from the feeds long ago. I couldn’t figure out why until I did a little digging.

Apparently, using sessions as I do throughout the site by default adds the aggressive cache headers. I’ve since changed my local copy and will be synchronized with the development copy later on tonight once all my changes are at least mostly working.

I’ve been reworking feeds completely as I said above, and this works into it well. Thanks for bringing it to my attention, I’ll email you for some testing assistance later.

]]> Comment on XML Syndication by wolf550e http://blog.patronuscharm.net/2008/07/xml-syndication/#comment-12 wolf550e Sun, 27 Jul 2008 18:18:13 +0000 http://blog.patronuscharm.net/?p=6#comment-12 Your user feeds come with "cache-control: no-store, no-cache" which is very bad as it disallows well behaved proxies and other caching systems to not repeatedly re-download an unchanged feed. Even "must-revalidate" is a bit too much, in my opinion. Please change the cache control header to something reasonable like "max-age=86400" so aggregators/feed readers won't hammer the site. Also, make it possible to use conditional get to verify the version in cache is still the latest without re-downloading it. You can email me with technical question and I'm willing to help with whatever you need design/code/testing-wise. Your user feeds come with “cache-control: no-store, no-cache” which is very bad as it disallows well behaved proxies and other caching systems to not repeatedly re-download an unchanged feed. Even “must-revalidate” is a bit too much, in my opinion.
Please change the cache control header to something reasonable like “max-age=86400″ so aggregators/feed readers won’t hammer the site.
Also, make it possible to use conditional get to verify the version in cache is still the latest without re-downloading it.
You can email me with technical question and I’m willing to help with whatever you need design/code/testing-wise.

]]> Comment on Sphinx Search: MySQL Fulltext Search Engine Replacement by Lord Ravenclaw http://blog.patronuscharm.net/2008/07/sphinx-search-mysql-fulltext-search-engine-replacement/#comment-11 Lord Ravenclaw Tue, 22 Jul 2008 22:28:18 +0000 http://blog.patronuscharm.net/?p=11#comment-11 The highlighter as since undergone a revision, and now does tag highlighting server-side. It should be HTML-friendly now and not break HTML tags. So far I haven't been able to get it to break. The highlighter as since undergone a revision, and now does tag highlighting server-side. It should be HTML-friendly now and not break HTML tags. So far I haven’t been able to get it to break.

]]> Comment on Sphinx Search: MySQL Fulltext Search Engine Replacement by Lord Ravenclaw http://blog.patronuscharm.net/2008/07/sphinx-search-mysql-fulltext-search-engine-replacement/#comment-10 Lord Ravenclaw Mon, 21 Jul 2008 21:03:47 +0000 http://blog.patronuscharm.net/?p=11#comment-10 Good to have some feedback. Search highlighting is actually done with JS, as at the time I couldn't really use any other method. I will however look into making it more HTML friendly...I thought I had closed that issue, but apparently not. And no, not really. Even if you could, it's only JS and not run server-side. It does need fixing. The highlighting was intentional, to make it readily apparent what was being matched. I suppose highlighting could use an overhaul. I've put it on my to-do list. Sphinx handles mapping of ö -> o, so it should be just fine. I'll look into making sure that it properly handles these. Good to have some feedback.

Search highlighting is actually done with JS, as at the time I couldn’t really use any other method. I will however look into making it more HTML friendly…I thought I had closed that issue, but apparently not. And no, not really. Even if you could, it’s only JS and not run server-side. It does need fixing.

The highlighting was intentional, to make it readily apparent what was being matched. I suppose highlighting could use an overhaul. I’ve put it on my to-do list.

Sphinx handles mapping of ö -> o, so it should be just fine. I’ll look into making sure that it properly handles these.

]]> Comment on Sphinx Search: MySQL Fulltext Search Engine Replacement by Alssn http://blog.patronuscharm.net/2008/07/sphinx-search-mysql-fulltext-search-engine-replacement/#comment-9 Alssn Mon, 21 Jul 2008 20:36:25 +0000 http://blog.patronuscharm.net/?p=11#comment-9 Huh. It seems that I can't use certain symbols in those commentaries. Anyway, in fifth paragraph I've meant usual angle brackets used in the most perverted combinations possible. Huh. It seems that I can’t use certain symbols in those commentaries. Anyway, in fifth paragraph I’ve meant usual angle brackets used in the most perverted combinations possible.

]]> Comment on Sphinx Search: MySQL Fulltext Search Engine Replacement by Alssn http://blog.patronuscharm.net/2008/07/sphinx-search-mysql-fulltext-search-engine-replacement/#comment-8 Alssn Mon, 21 Jul 2008 20:22:13 +0000 http://blog.patronuscharm.net/?p=11#comment-8 Gave it a try. Works fast and precise even with one letter to search by. Still, I've spent couple of minutes doing unnatural things to it, and got some results. Naturally, first word I tried to search for was "ass". And what do you think happened? I lol'ed. It gives you something like this: "ass="nlink" href="/s/120/1/" name="storytitle">Harry Potter and the Acts of Betrayal by ass="nlink" I had already tried three other obscenities, when it occurred to me that "ass" was just a part of "class". Pity. Still, it only affects results. Now, searching for "" – that's where real fun begins. Oh, the things it was showing to me... Try "< as", nearly killed my computer with this amount of fucked-up java-script. Anyway, you really need to close this one. Not sure if it is possible to execute scripts or php there, but just to be safe. Next thing is about non-ASCII symbols. The likes of "ö" and such. Now, I know that probability is low, but still — if someone names their story using one of those, you will have some difficulties finding it. And finally, strange highlighting – with blank spaces around search string. Doesn't really make for a clean page. That's all I've got for now. Enjoy your debugging, I'll be in touch. Gave it a try. Works fast and precise even with one letter to search by. Still, I’ve spent couple of minutes doing unnatural things to it, and got some results.

Naturally, first word I tried to search for was “ass”. And what do you think happened? I lol’ed.

It gives you something like this:
“ass=”nlink” href=”/s/120/1/” name=”storytitle”>Harry Potter and the Acts of Betrayal by ass=”nlink”

I had already tried three other obscenities, when it occurred to me that “ass” was just a part of “class”. Pity.

Still, it only affects results. Now, searching for “” – that’s where real fun begins. Oh, the things it was showing to me… Try “< as”, nearly killed my computer with this amount of fucked-up java-script.

Anyway, you really need to close this one. Not sure if it is possible to execute scripts or php there, but just to be safe.

Next thing is about non-ASCII symbols. The likes of “ö” and such. Now, I know that probability is low, but still — if someone names their story using one of those, you will have some difficulties finding it.

And finally, strange highlighting – with blank spaces around search string. Doesn’t really make for a clean page.

That’s all I’ve got for now. Enjoy your debugging, I’ll be in touch.

]]> Comment on Sphinx Search: MySQL Fulltext Search Engine Replacement by Database Management » Blog Archive » Sphinx Search: MySQL Fulltext Search Engine Replacement http://blog.patronuscharm.net/2008/07/sphinx-search-mysql-fulltext-search-engine-replacement/#comment-7 Database Management » Blog Archive » Sphinx Search: MySQL Fulltext Search Engine Replacement Mon, 21 Jul 2008 04:22:29 +0000 http://blog.patronuscharm.net/?p=11#comment-7 [...] Read the rest of this great post here [...] [...] Read the rest of this great post here [...]

]]> Comment on SSL and DLP: Little Known Secret by Gizmore http://blog.patronuscharm.net/2008/07/ssl-and-dlp-little-known-secret/#comment-6 Gizmore Tue, 01 Jul 2008 11:04:29 +0000 http://blog.patronuscharm.net/?p=5#comment-6 Heya Raven, I love this new dev blog, keep it up! :) Heya Raven, I love this new dev blog, keep it up! :)

]]>