feat: add "expr_noreflectmethod" build tag

[vincentbernat]

By enabling this tag, two features are removed:

• functions in environment
• Eval() method

Fix #863.
Related to #956.

I still need to work on the tests to make them work with or without the tag.

[vincentbernat]

I have added a commit to make tests work with the tag (and updated one workflow to demonstrate that).

goos: linux
goarch: amd64
pkg: github.com/expr-lang/expr
cpu: AMD Ryzen 5 5600X 6-Core Processor
                            │ before.txt  │             after.txt              │
                            │   sec/op    │   sec/op     vs base               │
_expr-12                      85.63n ± 1%   77.28n ± 2%   -9.75% (p=0.002 n=6)
_expr_eval-12                 5.781µ ± 1%   5.677µ ± 1%   -1.80% (p=0.004 n=6)
_expr_reuseVm-12              60.66n ± 3%   53.05n ± 4%  -12.55% (p=0.002 n=6)
_len-12                       48.46n ± 4%   47.78n ± 1%   -1.39% (p=0.009 n=6)
_filter-12                    42.43µ ± 2%   43.76µ ± 3%   +3.12% (p=0.002 n=6)
_filterLen-12                 37.82µ ± 2%   38.17µ ± 2%        ~ (p=0.699 n=6)
_filterFirst-12               380.0n ± 1%   386.0n ± 1%   +1.58% (p=0.009 n=6)
_filterLast-12                716.7n ± 2%   677.8n ± 1%   -5.44% (p=0.002 n=6)
_filterMap-12                 3.861µ ± 3%   3.899µ ± 2%        ~ (p=0.558 n=6)
_arrayIndex-12                84.54n ± 2%   84.92n ± 2%        ~ (p=0.699 n=6)
_envStruct-12                 53.07n ± 1%   53.17n ± 2%        ~ (p=0.485 n=6)
_envStruct_noEnv-12           140.5n ± 2%   145.0n ± 1%   +3.28% (p=0.002 n=6)
_envMap-12                    64.05n ± 3%   61.66n ± 2%   -3.72% (p=0.009 n=6)
_callFunc-12                  348.0n ± 6%   350.7n ± 1%        ~ (p=0.394 n=6)
_callMethod-12                365.4n ± 1%   374.2n ± 1%   +2.41% (p=0.002 n=6)
_callField-12                 75.81n ± 2%   75.46n ± 1%        ~ (p=0.394 n=6)
_callFast-12                  76.38n ± 3%   76.01n ± 1%        ~ (p=0.394 n=6)
_callConstExpr-12             35.02n ± 1%   34.74n ± 1%   -0.77% (p=0.045 n=6)
_largeStructAccess-12         138.4n ± 1%   148.0n ± 1%   +6.98% (p=0.002 n=6)
_largeNestedStructAccess-12   148.0n ± 1%   157.0n ± 1%   +6.11% (p=0.002 n=6)
_largeNestedArrayAccess-12    1.073m ± 8%   1.002m ± 6%   -6.61% (p=0.026 n=6)
_sort-12                      3.948µ ± 1%   3.702µ ± 1%   -6.24% (p=0.002 n=6)
_sortBy-12                    11.33µ ± 3%   11.06µ ± 2%   -2.32% (p=0.004 n=6)
_groupBy-12                   8.259µ ± 1%   8.589µ ± 1%   +4.00% (p=0.002 n=6)
_reduce-12                    34.64n ± 1%   34.50n ± 1%        ~ (p=0.197 n=6)
_min-12                       500.6n ± 1%   537.0n ± 1%   +7.28% (p=0.002 n=6)
_max-12                       506.7n ± 1%   535.9n ± 1%   +5.76% (p=0.002 n=6)
_mean-12                      218.1n ± 2%   229.5n ± 2%   +5.20% (p=0.002 n=6)
_median-12                    1.599µ ± 1%   1.593µ ± 2%        ~ (p=0.290 n=6)
geomean                       564.7n        563.4n        -0.22%

pkg: github.com/expr-lang/expr/checker
                                            │ before.txt  │             after.txt             │
                                            │   sec/op    │   sec/op     vs base              │
Checker/name=function_calls-12                6.248µ ± 5%   5.970µ ± 2%  -4.44% (p=0.004 n=6)
Checker/name=unary_and_binary_operations-12   6.167µ ± 3%   6.163µ ± 2%       ~ (p=0.937 n=6)
Checker/name=deep_struct_access-12            16.48µ ± 2%   16.74µ ± 2%  +1.61% (p=0.009 n=6)
geomean                                       8.594µ        8.509µ       -1.00%

pkg: github.com/expr-lang/expr/internal/difflib
                   │ before.txt  │             after.txt             │
                   │   sec/op    │   sec/op     vs base              │
SplitLines100-12     905.9n ± 1%   866.7n ± 1%  -4.32% (p=0.002 n=6)
SplitLines10000-12   81.46µ ± 1%   81.59µ ± 1%       ~ (p=0.485 n=6)
geomean              8.590µ        8.409µ       -2.11%

pkg: github.com/expr-lang/expr/internal/testify/assert
              │ before.txt  │             after.txt             │
              │   sec/op    │   sec/op     vs base              │
BytesEqual-12   144.9n ± 5%   147.8n ± 2%       ~ (p=0.485 n=6)
NotNil-12       4.416n ± 1%   4.399n ± 2%       ~ (p=1.000 n=6)
geomean         25.29n        25.49n       +0.80%

pkg: github.com/expr-lang/expr/optimizer
                                 │ before.txt  │             after.txt              │
                                 │   sec/op    │   sec/op     vs base               │
CountGtZero-12                     1.438µ ± 3%   1.416µ ± 4%        ~ (p=0.260 n=6)
CountGtZeroLargeEarlyMatch-12      9.107µ ± 7%   9.841µ ± 6%   +8.06% (p=0.004 n=6)
CountGtZeroNoMatch-12              29.32µ ± 4%   30.53µ ± 3%   +4.13% (p=0.026 n=6)
CountGteOneEarlyMatch-12           1.452µ ± 2%   1.541µ ± 4%   +6.13% (p=0.002 n=6)
CountGteOneNoMatch-12              29.22µ ± 5%   29.50µ ± 2%        ~ (p=0.132 n=6)
CountThresholdEarlyMatch-12        15.00µ ± 3%   13.64µ ± 3%   -9.09% (p=0.002 n=6)
CountThresholdGteEarlyMatch-12     12.20µ ± 2%   11.59µ ± 4%   -4.99% (p=0.002 n=6)
CountThresholdNoEarlyExit-12       342.4µ ± 2%   336.8µ ± 1%   -1.64% (p=0.041 n=6)
CountThresholdLargeEarlyMatch-12   89.66µ ± 2%   85.10µ ± 1%   -5.09% (p=0.002 n=6)
CountThresholdLtEarlyExit-12       15.56µ ± 9%   14.13µ ± 3%   -9.19% (p=0.002 n=6)
CountThresholdLteEarlyExit-12      12.45µ ± 3%   11.79µ ± 7%   -5.35% (p=0.002 n=6)
CountThresholdLtNoEarlyExit-12     340.8µ ± 3%   343.2µ ± 1%        ~ (p=0.699 n=6)
CountThresholdLteNoEarlyExit-12    353.2µ ± 3%   341.4µ ± 1%   -3.33% (p=0.004 n=6)
SumArray-12                        187.1n ± 4%   148.6n ± 2%  -20.58% (p=0.002 n=6)
SumRange_Optimized-12              35.64n ± 7%   34.83n ± 2%        ~ (p=0.058 n=6)
ReduceRangeSum_Optimized-12        35.45n ± 6%   34.85n ± 1%   -1.71% (p=0.035 n=6)
SumRange_Unoptimized-12            3.267µ ± 4%   3.336µ ± 1%        ~ (p=0.093 n=6)
geomean                            7.854µ        7.640µ        -2.73%

pkg: github.com/expr-lang/expr/parser
          │  before.txt  │          after.txt           │
          │    sec/op    │   sec/op     vs base         │
Parser-12   618.2n ± 13%   605.2n ± 3%  ~ (p=0.240 n=6)

          │ before.txt │           after.txt           │
          │    B/op    │    B/op     vs base           │
Parser-12   256.0 ± 0%   256.0 ± 0%  ~ (p=1.000 n=6) ¹
¹ all samples are equal

          │ before.txt │           after.txt           │
          │ allocs/op  │ allocs/op   vs base           │
Parser-12   8.000 ± 0%   8.000 ± 0%  ~ (p=1.000 n=6) ¹
¹ all samples are equal

pkg: github.com/expr-lang/expr/patcher/value
                    │ before.txt  │             after.txt             │
                    │   sec/op    │   sec/op     vs base              │
_valueAdd-12          86.08n ± 1%   83.43n ± 4%  -3.08% (p=0.026 n=6)
_valueUntypedAdd-12   102.7n ± 2%   102.1n ± 1%       ~ (p=0.223 n=6)
_valueTypedAdd-12     84.59n ± 1%   82.58n ± 1%  -2.37% (p=0.002 n=6)
geomean               90.75n        88.95n       -1.98%

pkg: github.com/expr-lang/expr/test/bench
                  │ before.txt  │             after.txt             │
                  │   sec/op    │   sec/op     vs base              │
Call_callTyped-12   8.686µ ± 1%   8.770µ ± 1%  +0.97% (p=0.041 n=6)
Call_eval-12        1.619µ ± 1%   1.603µ ± 1%       ~ (p=0.058 n=6)
geomean             3.750µ        3.749µ       -0.03%

pkg: github.com/expr-lang/expr/vm
                          │ before.txt  │             after.txt             │
                          │   sec/op    │   sec/op     vs base              │
VM/name=function_calls-12   1.670µ ± 1%   1.617µ ± 1%  -3.17% (p=0.002 n=6)

pkg: github.com/expr-lang/expr/vm/runtime
                                          │ before.txt  │             after.txt             │
                                          │   sec/op    │   sec/op     vs base              │
Equal/int_==_int-12                         2.163n ± 1%   2.235n ± 3%  +3.28% (p=0.004 n=6)
Equal/int_!=_int-12                         2.165n ± 0%   2.193n ± 2%  +1.27% (p=0.009 n=6)
Equal/int_==_int8-12                        2.379n ± 0%   2.383n ± 0%       ~ (p=0.831 n=6)
Equal/int_==_int16-12                       2.381n ± 0%   2.430n ± 2%       ~ (p=0.061 n=6)
Equal/int_==_int32-12                       2.188n ± 2%   2.188n ± 2%       ~ (p=0.792 n=6)
Equal/int_==_int64-12                       2.399n ± 1%   2.379n ± 1%       ~ (p=0.091 n=6)
Equal/float_==_float-12                     2.384n ± 1%   2.403n ± 1%  +0.78% (p=0.041 n=6)
Equal/float_!=_float-12                     2.384n ± 0%   2.408n ± 1%  +0.96% (p=0.026 n=6)
Equal/float_==_int-12                       2.606n ± 0%   2.600n ± 1%       ~ (p=0.329 n=6)
Equal/float_!=_int-12                       2.607n ± 2%   2.597n ± 0%       ~ (p=0.190 n=6)
Equal/string_==_string-12                   3.244n ± 2%   3.242n ± 1%       ~ (p=0.677 n=6)
Equal/string_!=_string-12                   4.547n ± 1%   4.544n ± 0%       ~ (p=0.608 n=6)
Equal/bool_==_bool-12                       2.165n ± 0%   2.163n ± 1%       ~ (p=0.790 n=6)
Equal/bool_!=_bool-12                       2.167n ± 0%   2.163n ± 0%       ~ (p=0.084 n=6)
Equal/[]any_==_[]int-12                     15.18n ± 6%   14.83n ± 7%       ~ (p=0.784 n=6)
Equal/[]any_!=_[]int-12                     15.50n ± 8%   15.30n ± 7%       ~ (p=0.413 n=6)
Equal/deep_[]any_==_[]any-12                84.08n ± 1%   85.07n ± 2%  +1.18% (p=0.015 n=6)
Equal/deep_[]any_!=_[]any-12                61.08n ± 2%   61.08n ± 1%       ~ (p=0.937 n=6)
Equal/map[string]any_==_map[string]any-12   253.5n ± 1%   254.0n ± 1%       ~ (p=0.615 n=6)
Equal/map[string]any_!=_map[string]any-12   59.46n ± 1%   59.36n ± 1%       ~ (p=0.589 n=6)
geomean                                     6.188n        6.201n       +0.20%

[antonmedv]

Nice. Let me review this pull request and I will merge it and release new version shortly after.

[ofek-frame]

Any news, perhaps?

[antonmedv]

Currently on my vacation, please @materight take a look

[vincentbernat]

I have updated this pull request to "fix" tests. In the future, this may be the main hindrance with such a patch: tests using Eval() will fail and need to be manually skipped. Maybe people will more likely use Compile()+Run() instead of Eval() when writing tests. I don't think in Go, we can automatically skip tests using Eval() (like an autofixture in pytest).

[vincentbernat]

I think this would be doable by modifying Error() and NoError() helpers. Instead of panicking, when running inside tests, Eval() could return a specific error that would make the test skip. It does not cover all use cases (notably the ones using env methods), but it should reduce the need to skip tests. Let me know if this is something worth to do.