BlitzBasic Code Archives coming to SyntaxBomb

[Qube]

btw: There are still are a lot of '"' in the code archives.

There are? God damn MySQL!

Could you post a couple of links here please so I can see what I missed last time and fixy fix fix

[GW]

I shouldn't say there's a lot, it looks like just a few posts have them..

Here's one
http://www.syntaxbomb.com/index.php/topic,2902.msg3891.html

I've also found several archive posts that have no content at all
http://www.syntaxbomb.com/index.php/topic,706.msg1695.html#msg1695

[Qube]

Oh I see, thanks. That should be easily sorted via the joys of SQL. I'll do that later today on a mass search and replace to correct other entries too.

I've also found several archive posts that have no content at all

Funnily enough I spotted a couple today. I need to find out why that happened and if it's only a couple I'll manually add them in but if there are loads then I'll see if my code got confused or if the forum just spat the SQL back out again.

[xlsior]

Just noticed another problem, looking at one of my old postings:

http://www.syntaxbomb.com/index.php/topic,2973.0.html

instead of & it shows & in the source.

[Qube]

Code Select Expand

" instead of "
fixed all 130+ instances

Code Select Expand

& instead of &
fixed all 500+ instances

[Krischan]

I noticed that the remote image links are messed and additional code tags in the comments section are not being parsed, see http://www.syntaxbomb.com/index.php/topic,3176.msg4165.html#msg4165 for example.

The URL is in plain text <img src="../../blitzbasic.christianhart.de/bb/elite/elitescanner.png"> but should be blitzbasic.christianhart.de/bb/elite/elitescanner.png only, or with a http:// prefix.

[Qube]

The URL is in plain text

Fixed for all <img src=""> tags

[Krischan]

Hmm ok, perhaps you have time to take a look at more nasty bugs, see attached images?

bug_empty: The post contains nothing, even if the post is older
bug_imagelinebreak: the image should start a new paragraph and shoudl be displayed instead of the link
bug_opentags: there is an open italics tag left
bug_url1: could be the same problem like bug_opentags, looks messed
bug_url2: the reference looks strange, and could the same bug like the two previous bugs

[Qube]

Hmm ok, perhaps you have time to take a look at more nasty bugs, see attached images?

Time?. This isn't the time to argue about time, we don't have the time...

There are a few more HTML elements in the archives than I thought. When I first looked at a handful of entries they were pretty much all the same pattern. I coded the importer based on those. Sods law there were more.

Luckily it's nothing drastic and for a majority you can do a blanket find / replace to sort the issue. For the <a href> posts I'll have to do a tiny bit of coding to strip the double URL's and make it one URL wrapped in BB code.

Thanks for creating more work for me

[RonTek]

The forum version looks nice and integrated, but I think it would be much easier and more convenient to just get the static copy of the archives and query it the same way as the original archives, then modify the theme a bit to match with the site. I have seen there are additional comments and info that provides fixes or updates on most resources.

http://blitzbasic.com/codearcs.php?cat=13 Categories
http://blitzbasic.com/codearcs.php?code=3307 Resource
http://blitzbasic.com/codearcs.php?cat=13&order=&asc=&lang_id=2 Categories with Language Type (BB/BMX/ALL)

Here's a screenshot of the quick styling that I did using the wasted.nz archive copy..




Full Page
http://i.imgur.com/745VGzz.jpg

[Qube]

The forum version looks nice and integrated, but I think it would be much easier and more convenient to just get the static copy of the archives and query it the same way as the original archives, then modify the theme a bit to match with the site. I have seen there are additional comments and info that provides fixes or updates on most resources.

I was not provided with the MySQL DB for the code archives or else it would of been much easier. All I had to work with was static HTML pages. So I can't just add a static copy and make it queryable via the forums, hence integrating them into the forums. Also I want the code archives as part of Syntaxbomb so they can be added to over time.

There's only a couple of niggles to solve

[RonTek]

Ah got it. Actually, though it may take time but it's not that hard to pull url queries or content particularly if you have a few fields to deal with. For the code archives I'm seeing only 3: the item id, category and language type. Even without an existing MySQL db, you can still rebuild the archives with MySQL as long as the reference links are consistent and intact.

Well, hope all works ok.

Btw, I was wondering why you still need to get permission if the code is licensed as Public Domain..

This code has been declared by its author to be Public Domain code.

I would assume the permission includes the whole page or user comments/replies and other content like images?

[Qube]

you can still rebuild the archives with MySQL as long as the reference links are consistent and intact.

All the HTML pages are converted into the DB here. There are just a few issues to solve with in post links. They'll be sorted

Btw, I was wondering why you still need to get permission if the code is licensed as Public Domain..

I wanted to include the comments with original post. While the code is public domain, the comments are not. As I've mirrored 1000's of posts it was far easier to ask permission. Simples

[xlsior]

It looks like it still shows &amp; instead of & in the post titles, e.g.:

https://www.syntaxbomb.com/index.php?topic=2845.0

[Qube]

It looks like it still shows &amp; instead of & in the post titles, e.g.:

https://www.syntaxbomb.com/index.php?topic=2845.0

I only applied the changes to the body of the post. I didn't think the issue would be in the titles too. Will sort out this evening, thanks.