Quan, One suggestion for finding dependencies is to turn on autoload messages. Starting from a clean image you can find the list of domains that need to be autoloaded. Choosing one of the listed domains you can apply the same technique to see what it requires.
I suppose you could capture the code used for autoloading and use it to develop a standalone tool. Or you could use the full image, capture the messages, and automate a depth-first domain walk mentioned above. You'll discover that trying to write code to parse the algebra source code to be "a challenge" to put it mildly. Only Bill Burge managed to do that. It is VERY hard to decide where that "+" sign comes from by inspecting source code. Be aware that it took a whole summer to find that minimal boostrap set and a long time to create the SVG dependency graph. There is no such thing as a simple job. (Daly's law of programming) On Saturday, September 21, 2024 at 6:58:54 AM UTC-4 Tim Daly wrote: > Ralf, > > I chose the SVG format because it is web-scalable. The total graph is too > complex to fit on a web page without scaling. Useful scales are on the > fractions at the top of the page. > > > On Saturday, September 21, 2024 at 6:50:15 AM UTC-4 [email protected] > wrote: > >> Hi Qian, >> >> long time ago I had such a program for Aldor and its library, >> but I have not yet found it. Still looking ... >> >> However, fricas.github.io/api probably comes close to what you want. >> However that would only lead to a dependency graph among the categories. >> >> It is much harder to dig out whether some domain/package *uses* another >> domain/package. And such dependencies are probably also interesting. >> >> So for the category part, I suggest to look into src/doc/api.spad and >> tweak it a bit so that it outputs in a .dot format instead of .rst. >> >> Maybe it is also possible to use the tikz latex package, but honestly, I >> would be more happy if we had a dependency graph as a visualization of >> the APL, i.e. in some web-presentable format. >> >> Ralf >> > -- You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/fricas-devel/c20a8b58-4648-40d1-9bf4-49a3b4ffd091n%40googlegroups.com.
