Flatten the 'db' commands, e.g. 'db prefix' to 'db/prefix'

[joamaki]

The user experience with commands that have sub-commands is not great and it makes it harder to implement completion in 'cilium-dbg shell' if we so choose. There isn't really any good rational for nesting the commands like this, so let's go with Plan 9 style command names (https://9p.io/magic/man2html/8/auth), e.g. "db prefix" becomes "db/prefix".

[joamaki]

I went to fair bit of effort in #65 to document the "sub-commands" and I just realized today that it's kind of stupid to nest the commands like this. It makes them hard to find (help wont list them) and it makes it hard/impossible to implement command-line completion in cilium-dbg shell if we want to add one later. The script/shell thing should be as simple and dumb as possible, so let's simplify this.

[github-actions]

$ make test
go: downloading golang.org/x/sys v0.17.0
go: downloading golang.org/x/tools v0.17.0
go: downloading golang.org/x/exp v0.0.0-20240119083558-1b970713d09a
go: downloading github.com/fsnotify/fsnotify v1.7.0
go: downloading github.com/sagikazarmark/slog-shim v0.1.0
go: downloading github.com/spf13/afero v1.11.0
go: downloading github.com/spf13/cast v1.6.0
go: downloading github.com/subosito/gotenv v1.6.0
go: downloading github.com/hashicorp/hcl v1.0.0
go: downloading gopkg.in/ini.v1 v1.67.0
go: downloading github.com/magiconair/properties v1.8.7
go: downloading github.com/pelletier/go-toml/v2 v2.1.0
go: downloading golang.org/x/text v0.14.0
	github.com/cilium/statedb/reconciler/benchmark		coverage: 0.0% of statements
	github.com/cilium/statedb/reconciler/example		coverage: 0.0% of statements
ok  	github.com/cilium/statedb	29.988s	coverage: 82.7% of statements
ok  	github.com/cilium/statedb/index	0.006s	coverage: 28.7% of statements
ok  	github.com/cilium/statedb/internal	0.010s	coverage: 46.7% of statements
ok  	github.com/cilium/statedb/part	2.744s	coverage: 83.2% of statements
ok  	github.com/cilium/statedb/reconciler	0.279s	coverage: 89.7% of statements
-----
$ make bench
go test ./... -bench . -benchmem -test.run xxx
goos: linux
goarch: amd64
pkg: github.com/cilium/statedb
cpu: AMD EPYC 7763 64-Core Processor                
BenchmarkDB_WriteTxn_1-4                    	  444084	      2699 ns/op	    370544 objects/sec	    2816 B/op	      32 allocs/op
BenchmarkDB_WriteTxn_10-4                   	 1214216	       984.2 ns/op	   1016021 objects/sec	     743 B/op	      10 allocs/op
BenchmarkDB_WriteTxn_100-4                  	 1555790	       766.6 ns/op	   1304406 objects/sec	     597 B/op	       7 allocs/op
BenchmarkDB_WriteTxn_1000-4                 	 1460091	       830.2 ns/op	   1204556 objects/sec	     549 B/op	       7 allocs/op
BenchmarkDB_WriteTxn_100_SecondaryIndex-4   	  503707	      2323 ns/op	    430449 objects/sec	    1505 B/op	      37 allocs/op
BenchmarkDB_Modify-4                        	    1267	    937209 ns/op	   1066999 objects/sec	  774498 B/op	    8463 allocs/op
BenchmarkDB_GetInsert-4                     	    1194	   1010853 ns/op	    989265 objects/sec	  755673 B/op	    8461 allocs/op
BenchmarkDB_RandomInsert-4                  	    2355	    501711 ns/op	   1993180 objects/sec	  402233 B/op	    7097 allocs/op
BenchmarkDB_RandomReplace-4                 	     310	   3836403 ns/op	    260661 objects/sec	 2357567 B/op	   48575 allocs/op
BenchmarkDB_SequentialInsert-4              	    1494	    804130 ns/op	   1243580 objects/sec	  551363 B/op	    7285 allocs/op
BenchmarkDB_Changes_Baseline-4              	    1244	    957850 ns/op	   1044006 objects/sec	  553134 B/op	   10252 allocs/op
BenchmarkDB_Changes-4                       	     679	   1755950 ns/op	    569492 objects/sec	  993403 B/op	   14493 allocs/op
BenchmarkDB_RandomLookup-4                  	   22114	     53786 ns/op	  18592125 objects/sec	     160 B/op	       1 allocs/op
BenchmarkDB_SequentialLookup-4              	   27476	     44156 ns/op	  22647015 objects/sec	       0 B/op	       0 allocs/op
BenchmarkDB_Prefix_SecondaryIndex-4         	    7539	    149192 ns/op	   6702759 objects/sec	   93911 B/op	    1044 allocs/op
BenchmarkDB_FullIteration_All-4             	     672	   1727447 ns/op	  57889034 objects/sec	     480 B/op	      12 allocs/op
BenchmarkDB_FullIteration_Get-4             	     204	   5926269 ns/op	  16874053 objects/sec	       0 B/op	       0 allocs/op
BenchmarkDB_PropagationDelay-4              	  511288	      2395 ns/op	        20.00 50th_µs	        25.00 90th_µs	        90.00 99th_µs	    1590 B/op	      24 allocs/op
PASS
ok  	github.com/cilium/statedb	27.948s
PASS
ok  	github.com/cilium/statedb/index	0.004s
PASS
ok  	github.com/cilium/statedb/internal	0.003s
goos: linux
goarch: amd64
pkg: github.com/cilium/statedb/part
cpu: AMD EPYC 7763 64-Core Processor                
Benchmark_Insert_RootOnlyWatch-4    	    8776	    136047 ns/op	   7350388 objects/sec	  104164 B/op	    2041 allocs/op
Benchmark_Insert-4                  	    6258	    186430 ns/op	   5363932 objects/sec	  219055 B/op	    3064 allocs/op
Benchmark_Modify-4                  	    7784	    152581 ns/op	   6553923 objects/sec	  212697 B/op	    1205 allocs/op
Benchmark_GetInsert-4               	    6507	    187044 ns/op	   5346342 objects/sec	  212401 B/op	    1203 allocs/op
Benchmark_Replace-4                 	26830669	        44.17 ns/op	  22641985 objects/sec	       0 B/op	       0 allocs/op
Benchmark_Replace_RootOnlyWatch-4   	26879574	        44.71 ns/op	  22366624 objects/sec	       0 B/op	       0 allocs/op
Benchmark_txn_1-4                   	 2935118	       426.6 ns/op	   2344373 objects/sec	     448 B/op	       7 allocs/op
Benchmark_txn_10-4                  	 7056416	       168.2 ns/op	   5945356 objects/sec	     154 B/op	       2 allocs/op
Benchmark_txn_100-4                 	 7776567	       155.4 ns/op	   6434067 objects/sec	     224 B/op	       2 allocs/op
Benchmark_txn_1000-4                	 6955147	       168.6 ns/op	   5931286 objects/sec	     216 B/op	       2 allocs/op
Benchmark_txn_delete_1-4            	 3241438	       371.1 ns/op	   2694992 objects/sec	     856 B/op	       6 allocs/op
Benchmark_txn_delete_10-4           	 8821153	       138.9 ns/op	   7199447 objects/sec	     132 B/op	       1 allocs/op
Benchmark_txn_delete_100-4          	10738854	       108.8 ns/op	   9191937 objects/sec	      60 B/op	       1 allocs/op
Benchmark_txn_delete_1000-4         	11706182	       103.3 ns/op	   9684056 objects/sec	      26 B/op	       1 allocs/op
Benchmark_Get-4                     	   36836	     31712 ns/op	  31533869 objects/sec	       0 B/op	       0 allocs/op
Benchmark_Iterate-4                 	  167732	      7399 ns/op	 135144889 objects/sec	      80 B/op	       3 allocs/op
Benchmark_Hashmap_Insert-4          	   15656	     75771 ns/op	  13197731 objects/sec	   86549 B/op	      64 allocs/op
Benchmark_Hashmap_Get_Uint64-4      	  143612	      8419 ns/op	 118778410 objects/sec	       0 B/op	       0 allocs/op
Benchmark_Hashmap_Get_Bytes-4       	  152944	      7862 ns/op	 127190274 objects/sec	       0 B/op	       0 allocs/op
Benchmark_Uint64Map_Random-4        	    1327	    886615 ns/op	   1127886 items/sec	 2700474 B/op	    9024 allocs/op
Benchmark_Uint64Map_Sequential-4    	    1466	    830707 ns/op	   1203795 items/sec	 2492408 B/op	    9749 allocs/op
PASS
ok  	github.com/cilium/statedb/part	28.558s
PASS
ok  	github.com/cilium/statedb/reconciler	0.005s
?   	github.com/cilium/statedb/reconciler/benchmark	[no test files]
?   	github.com/cilium/statedb/reconciler/example	[no test files]
go run ./reconciler/benchmark -quiet
1000000 objects reconciled in 3.00 seconds (batch size 1000)
Throughput 332939.68 objects per second
Allocated 6011330 objects, 424778kB bytes, 525464kB bytes still in use

[ovidiutirla]

🚀

[pippolo84]

Nice to see that the (pretty complicated) logic to generate the usage details has gone away. I'm (mildly) concerned that the Plan 9 style might be kind of unexpected when using statedb in cilium where we rely on subcommands, but I'd say that it's worth to give this a try. 👍