This is just to document it somewhere... After every new release submission here, the archive is downloaded by BOTs -- I imagine it is not just this one but all Freecode hosted archives (or perhaps those in certain categories). It is easy to distinguish between human and bot, these bots just get the archive and make no other requests. The IP address is different each time. There is one repeated user agent, and one slightly varying user agent. It sure is weird.
The next release will have (the start of) a really cool thing: The comment HTML form will be "designable" by data, not by the code, which means that if someone wanted to add a FORM input, only some data would be changed, not the code.
This is the start of a significant advance, which means that, ultimately, upgrading the code will not undo any modifications a user may have implemented.
This is not plugin type stuff, but simply data -- INI/JSON/XML or whatever we decide to support -- will be all that is needed to customize the code.
Well, that was embarrassing. (As my daughter reminds from time to time, "Your PTSD is showing." sigh)
But the debug code does have flaws in that applying it as an API - which is hosted here at Freecode as 'PHP Debug' - I forgot several important settings so the code does not work as documented. Mistake number one.
In fixing all that, needing to call the function config() within the function debug() I accidentally typed it in as debug(). Mistake number two.
Mistake number three was a missed dose of Lorazepam...
(Why am I going into such detail? Here? I have no idea.)
Version 1.3.3 made a core data change that we no longer like! Therefore, there is going to be some code "un-winding". However, there are some improvements in 1.3.3, especially in Admin, that are really nice. So, we do not really want to recommend going back to 1.3.2.
We will try to get this resolved soon.
"Um... There was a problem with the last release."
"What? And why are you whispering?"
"I hope nobody will hear us. There was a function missing in the testing module."
"What! Who... Never mind. What did..."
"I fixed it. And I just re-uploaded the archive."
"Maybe no one will notice."
"Yeah, right. Probably just got several more people thinking we are idiots!"
Wicked cool update!
Oops, forgot to add tags to the latest release notes! (Sorry Freecode!)
tags: Update, Bug Fix
P.S. Anyone from here going to THIS' website getting 500 Server Errors? The logs show people are but not the 500.shtml that would normally be associated with it, but the full bytes count of the site (as well as 200s for the related files like .css).
Something had been bugging me about THIS code for a few months. A few days ago I decided the time was right to fix it. So, I gathered the developers together in one room, locked the door, and said...
"And nows you can't leave."
The result was all that inefficient code has been replaced.
All "web templates" have been placed into one PHP file and the configuration code is now one-quarter the size.
What a relief!
Sometimes you feel like a nut, sometimes... you feel like a nut!
A security firm contacted us the other day informing us that our code is a piece of s....
We disagree! There was that one SQL Injection bug, yeah! Bug it was just one! Sheesh!
BUG NOTICE! I just found a terrible BUG!
I develop this in a directory named "this", and I hardcoded that directory in at least two places! In the .htaccess file and for the cookie setting code!
Sorry about that!
We* are going to wait a few days before next release... We have much more documentation (like explaining why we are keeping an inefficiency in our database design), a bug fix, several tweaks that reduced code size, and more... But we found a way to reduce a particularly vexing web template problem that we, until now, could not see a "neat way" of doing it. So we are going to do that and several other things over the next week. The result should be a few more features and a much less complex implementation in several areas!
Also, we have the post/markup code spun-off into a separate API here here: https://freecode.com/projects/gmlp and is called the GMLP Markup Language Processor.
*I say "we" as I have a team! There's one full time coder, one assistant who does the stuff like archiving, and me! And I spend most of my time with lawyers and bankers as I look for investors!
An open, cross-platform journaling program.
A scientific plotting package.