To avoid confusion between static and dynamic: I will clarify that both methods I have shown are dynamic (loaded on runtime and not included in the binary), the difference between them is when the library is loaded, the first method causes the operating system to load the library when the program starts and the second method enables a program to load a library whenever it wants (after being started)
great content, specially how you casually use ldd and ltrace and even though I have been coding in C for quite some time already you just taught me something new. Thank you! One small tip: the font size is too small, specially if watching it while commuting on a handheld device. Try watching it at 240p and that will tell you how somebody without the gift of perfect eyesight will see it in a moving vehicle. But again, the content is superb, thanks again!
you shouldn't watch/read any thing in a moving vehicle in the first place _"how somebody without the gift of perfect eyesight will see it in a moving vehicle"_
Can you do a video covering dynamic linking on Windows, and how it differs (and is also similar to) Linux? One thing I'm slightly perplexed by with Windows is .lib files - how do they differ in function and purpose to .h files?
Nice, the two methods look very similar, as far as i understand the first one calls the object during compile time, and the 2nd one during runtime ? Good to know !
@@nirlichtman i see, the 2nd method seems more appropriate to use because you can specify how to load the library (in case its very big) as you can call only specific functions, for example you have a program where its not always necessary to load a shared object in which you can also specify lazy loading, i assume you didn't bother check the return values in your code to make the video more simple, but if you're gonna load the entire library anyways, why not just use the first method. Are there other implications of this ?
The first method is indeed more popular and fits in the simple case demonstrated in the video but there are some cases where the second method could be used for example if you want to load the shared object only in a specific flow in the code. Another advantage of the first method is that the shared object dependency can clearly be seen when running ldd - making it more user friendly.
Thanks! Now I can finally execute my evil plans to make the user require a .so file or else they cannot open my app. Why not make it harder for everyone? I like suffering.
Probably CS Operating Systems course would cover some of these stuff, but the best way to truly learn this is hands on coding along with reading the docs/man pages.
