Your crawling experience and on about mirage's name
Yesterday was a very productive day for mirage project. In brief I touched the main.c source file giving a rekon of a real security tool. I added some time handling routines and I bumped the version number to 0.10.
I’m very happy because with this fix, mirage is ready for a real source code crawling capability, that it’s the easiest thing to add to a static analyzer.
I’m not sure I’ll add some context aware capability since this change will add a lot of difficulties in writing libcrawel lex grammar. I don’t want to move there the intelligence to detect comments or programmin language specific stuff that it will change crawling results.
So, blame at me but source crawling will have a lot of false positives.
I know and I really don’t care about this since I use crawling just to address manual review. So please use mirage crawling results just to spotlight the source code areas where you might find something interesting. But please, check this out manually.
Don’t trust crawling.
Always double check.
Owasp has it’s own page about source code crawling
Of course, mirage will implement the recommendiations you can find on the Owasp site.
When I first designed Owasp Orizon I introduced mirage as just a application modeling engine choosing that name with this meaning: building an error prone source code modeling engine is a mirage.
I still think this one even today that mirage is going to be Owasp Orizon successor.
I don’t think mirage will be the final name and I want to use this period of time that the project is mostly unknown to find a good name with a good logo to address the Orizon succession.
The only decided issue is that the Owasp Orizon project is frozen and you have to consider no longer mantained.