In this edit: https://en.wikipedia.org/w/index.php?diff=prev&oldid=773768262 the bot corrupted a reference by incorrectly assuming that the link in the {{webarchive}} template was the same source as the previous URL. The URL's weren't even identical. Even if they were, it's conceivable that the {{webarchive}} template could serve the purpose of going back in time. I think you should be more careful when making your bot "smart".
Description
Description
Event Timeline
Comment Actions
I do not appreciate your tone in this bug report. Please stick to the facts and do not assume I just carelessly work on my bot. If you have any idea how much time I have invested in this project you would not be saying something so rude and presumptuous.
Comment Actions
I did not mean to be rude, I was just honest. If the bot makes errors when trying to be smart, that's a problem. If it sticks to what bots are good at, then you won't have that problem. I would be doing you a well-intentioned disservice if I didn't bring attention to (what I conceive to be) the elephant in the room. My judgment may be wrong.
Comment Actions
IABot's intelligent parser is capable of correctly recognizing, analyzing, and modifying 99% of all currently formatted references and external links outside of references.
Comment Actions
To add on, the parser used to be a basic parser to only handle obvious cases, but that was only 60% of all formatting cases. The aim of the No 404 Project, is to be able to save all URLs that are dying. That means slowly expanding the bot to interpret more and more formatting.
I have been developing IABot for almost 2 years.
Comment Actions
Ah I see it now. The archive URL and the original URLs, don't match. But they're seen together in the same reference. This is something that will confuse the bot, and nothing can be done about this. Best to just tag the source with {{cbignore}}.