Timeline for All possible ways of merging two lists while keeping their order
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 20, 2016 at 2:59 | vote | accept | samlaf | ||
| Feb 20, 2016 at 2:41 | history | edited | WorBlux | CC BY-SA 3.0 |
Added Adendum
|
| Feb 19, 2016 at 22:47 | comment | added | WorBlux | Map, fold... and other higher-order function are good to use when they fit, the shape of the recursion and the data structures make it hard to do. And in scheme at least it's not built-in to map to data structures that aren't lists. | |
| Feb 19, 2016 at 22:37 | comment | added | WorBlux |
I actually tried append! in your version, and it didn't make much difference. I did you with-timing in fact. If you wrap the body of the thunk with a begin you can put in a second value like "done" to force it through the calculation but disregard the results.
|
|
| Feb 19, 2016 at 14:00 | comment | added | samlaf | Also, are you timing the procedures using 'with-timing? Is there a way to stop it from printing the value of the call? | |
| Feb 19, 2016 at 13:55 | comment | added | samlaf | Good point. I have been diving through SICP lately and got used to the idea that making code human readable by using "conventional interfaces" (map, filter, fold) should be preferred over performance gains. If performance is such an issue, perhaps using append! instead of append in ordered-merge2 would make things faster? | |
| Feb 18, 2016 at 20:34 | history | answered | WorBlux | CC BY-SA 3.0 |