more work on converting to lua

api/api.go

@@ -65,7 +65,7 @@ func (s *Source) LuaOp(v interfaces.IVariant, code string, vals []interface{}) s
 	value, err := s.Vm.Run(code, map[string]interface{}{
 		"chrom": v.Chrom(),
 		"start": v.Start(),
-		"end":   v.End(),
+		"stop":  v.End(),
 		"vals":  vals})
 	if err != nil {
 		return fmt.Sprintf("lua-error: %s", err)

bench_test.go

@@ -23,9 +23,9 @@ func benchmarkAnno(b *testing.B) {
 	}
 
 	out := bufio.NewWriter(ioutil.Discard)
-	Js, _ := xopen.Ropen("example/custom.js")
-	jbytes, _ := ioutil.ReadAll(Js)
-	js_string := string(jbytes)
+	Lua, _ := xopen.Ropen("example/custom.lua")
+	lbytes, _ := ioutil.ReadAll(Lua)
+	l_string := string(lbytes)
 
 	srcs, err := configs.Sources()
 	if err != nil {
@@ -36,7 +36,7 @@ func benchmarkAnno(b *testing.B) {
 		if err != nil {
 			log.Fatal(err)
 		}
-		a := api.NewAnnotator(srcs, js_string, false, true, empty)
+		a := api.NewAnnotator(srcs, l_string, false, true, empty)
 		qrdr, err := xopen.Ropen("example/query.vcf.gz")
 		if err != nil {
 			log.Fatal(err)

docs/CHANGES.md

@@ -8,6 +8,7 @@ v0.0.9
 + new [documentation site](http://brentp.github.io/vcfanno/)
 + [[postannotation]] allows modifying stuff in the query VCF after annotation (or instead).
   See examples on the documentation site.
++ convert scripting engine to lua from javascript
 
 
 v0.0.8

docs/index.md

@@ -30,7 +30,7 @@ Usage
 After downloading the [binary for your system](https://github.com/brentp/vcfanno/releases/) (see section below) usage looks like:
 
 ```Shell
-  ./vcfanno -js example/custom.js example/conf.toml example/query.vcf.gz
+  ./vcfanno -lua example/custom.lua example/conf.toml example/query.vcf.gz
 ```
 
 Where conf.toml looks like:
@@ -44,9 +44,9 @@ ops=["first", "first", "min"]
 [[annotation]]
 file="fitcons.bed"
 columns = [4, 4]
-names=["fitcons_mean", "js_sum"]
-# note the 2nd op here is javascript that has access to `vals`
-ops=["mean", "js:sum=0;for(i=0;i<vals.length;i++){sum+=vals[i]}; vals"]
+names=["fitcons_mean", "lua_sum"]
+# note the 2nd op here is lua that has access to `vals`
+ops=["mean", "lua:function sum(t) local sum = 0; for i=1,#t do sum = sum + t[i] end return sum / #t end"]
 
 [[annotation]]
 file="example/ex.bam"
@@ -77,7 +77,7 @@ from this directory.
 Then, you can annotate with:
 
 ```Shell
-GOMAXPROCS=4 ./vcfanno -js example/custom.js example/conf.toml example/query.vcf.gz > annotated.vcf
+GOMAXPROCS=4 ./vcfanno -lua example/custom.lua example/conf.toml example/query.vcf.gz > annotated.vcf
 ```
 
 An example INFO field row before annotation (pos 98683):
@@ -87,7 +87,7 @@ AB=0.282443;ABP=56.8661;AC=11;AF=0.34375;AN=32;AO=45;CIGAR=1X;TYPE=snp
 
 and after:
 ```
-AB=0.2824;ABP=56.8661;AC=11;AF=0.3438;AN=32;AO=45;CIGAR=1X;TYPE=snp;AC_AFR=0;AC_AMR=0;AC_EAS=0;fitcons_mean=0.061;js_sum=0.061
+AB=0.2824;ABP=56.8661;AC=11;AF=0.3438;AN=32;AO=45;CIGAR=1X;TYPE=snp;AC_AFR=0;AC_AMR=0;AC_EAS=0;fitcons_mean=0.061;lua_sum=0.061
 ```
 
 Operations
@@ -98,7 +98,7 @@ in the query VCF. However, it is possible that there will be multiple annotation
 from a single annotation file--in this case, the op determines how the many values
 are `reduced`. Valid operations are:
 
- + js:$javascript // see section below for more details
+ + lua:$lua // see section below for more details
  + mean
  + max
  + min
@@ -129,17 +129,8 @@ Development
 ===========
 
 This, and the associated go libraries ([vcfgo](https://github.com/brentp/vcfgo),
-[irelate](https://github.com/brentp/irelate), [xopen](https://github.com/brentp/xopen)) are
-under active development. The following are on our radar (most have been completed):
-
-- [x] allow annotating with bam fields, e.g. QUAL and SEQ.
-- [ ] decompose, normalize, and get allelic primitives for variants on the fly
-      (we have code to do this, it just needs to be integrated)
-- [ ] allow custom golang ops when using api.
-- [x] improve test coverage for vcfanno (still need more tests for bam)
-- [x] embed otto js engine to allow custom ops.
-- [x] support for annotating BED files.
-
+[irelate](https://github.com/brentp/irelate), [xopen](https://github.com/brentp/xopen),
+[goluaez](https://github.com/brentp/goluaez) are under active development.
 
 Additional Usage
 ================
@@ -172,38 +163,38 @@ REF/ALT are not required.
 Set to the number of processes that `vcfanno` can use during annotation. `vcfanno` parallelizes well
 up to 15 or so cores.
 
--js
----
+-lua
+----
 
-custom in ops (javascript). For use when the built-in `ops` don't supply the needed reduction.
+custom in ops (lua). For use when the built-in `ops` don't supply the needed reduction.
 
-we embed the javascript engine [otto](https://github.com/robertkrimen/otto) so that it's 
+we embed the lua engine [go-lua](https://github.com/yuin/gopher-lua) so that it's 
 possible to create a custom op if it is not provided. For example if the users wants to
 
-    "js:sum=0;for(i=0;i<vals.length;i++){sum+=vals[i]};sum"
+    "lua:function sum(t) local sum = 0; for i=1,#t do sum = sum + t[i] end return sum end"
 
 where the last value (in this case sum) is returned as the annotation value. It is encouraged
-to instead define javascript functions in separate `.js` file and point to it when calling
-`vcfanno` using the `-js` flag. So, in an external file, "some.js", instead put:
-
-```javascript
-function sum(vals) {
- 	s = 0;
-	for(i=0; i<vals.length; i++){
-		s+=vals[i]
-	}
-	return s
-}
+to instead define lua functions in separate `.lua` file and point to it when calling
+`vcfanno` using the `-lua` flag. So, in an external file, "some.lua", instead put:
+
+```lua
+function sum(t)
+    local sum = 0
+    for i=1,#t do
+        sum = sum + t[i]
+    end
+    return sum
+end
 ```
 
-And then the above custom op would be: "js:sum(vals)". (note that there's a sum op provided
+And then the above custom op would be: "lua:sum(vals)". (note that there's a sum op provided
 by `vcfanno` which will be faster).
 
 The variables `vals`, `chrom`, `start`, `end` from the current variant will all be available
-in the javascript code.
+in the lua code.
 
 
 See [example/conf.toml](https://github.com/brentp/vcfanno/blob/master/example/conf.toml)
-and [example/custom.js](https://github.com/brentp/vcfanno/blob/master/example/custom.js)
+and [example/custom.lua](https://github.com/brentp/vcfanno/blob/master/example/custom.lua)
 for more examples.
 

example/conf.toml

@@ -14,23 +14,22 @@ ops=["first", "first", "first", "first"]
 
 [[annotation]]
 file="example/fitcons.bed.gz"
-names=["otto_start", "otto_end", "otto_len", "otto_sum", "otto_mean", "otto_loc"]
-columns=[4, 4, 4, 4, 4, 4]
-# when a custom value is needed, specify arbitrary javascript after "js:"
+names=["lua_start", "lua_end", "lua_len", "lua_mean", "lua_loc"]
+columns=[4, 4, 4, 4, 4]
+# when a custom value is needed, specify arbitrary javascript after "lua:"
 # the variable available will be:
 # + vals: the list of values from this annotation
 # + chrom: the chrom of the variant being annotated
 # + start: the (0-based) start of the variant being annotated
-# + end: the end of the variant being annotated
+# + stop: the end of the variant being annotated
 # the result of the expression can be anything. it will be converted to a string.
-# loc() and mean() are defined in examples/custom.js
-ops=["js:start", "js:end", "js:vals.length", "js:sum=0;for(i=0;i<vals.length;i++){sum+=vals[i]};vals", "js:mean(vals)", "js:loc(chrom, start, end)"]
-# note the last 2 use functions declared in js above.
+# loc() and mean() are defined in examples/custom.lua
+ops=["lua:start", "lua:stop", "lua:#vals", "lua:mean(vals)", "lua:loc(chrom, start, stop)"]
+# note the last 2 use functions declared in lua above.
 
 # it is also possible to define some javascript functions that will be
-# available. This is specified as a .js file to the vcfanno executable as
-# ./vcfanno -js some.js ... see example/custom.js for an example.
-
+# available. This is specified as a .lua file to the vcfanno executable as
+# ./vcfanno -lua some.lua ... see example/custom.lua for an example.
 
 [[annotation]]
 file="example/cadd.sub.txt.gz"

example/custom.lua

@@ -0,0 +1,67 @@
+function mean(vals)
+    local sum=0
+	for i=1,#vals do
+		sum = sum + vals[i]
+	end
+	return sum / #vals
+end
+
+function loc(chrom, start, stop)
+    return chrom .. ":" .. start .. "-" .. stop
+end
+
+CLINVAR_LOOKUP = {}
+CLINVAR_LOOKUP['0'] = 'unknown'
+CLINVAR_LOOKUP['1'] = 'germline'
+CLINVAR_LOOKUP['2'] = 'somatic'
+CLINVAR_LOOKUP['4'] = 'inherited'
+CLINVAR_LOOKUP['8'] = 'paternal'
+CLINVAR_LOOKUP['16'] = 'maternal'
+CLINVAR_LOOKUP['32'] = 'de-novo'
+CLINVAR_LOOKUP['64'] = 'biparental'
+CLINVAR_LOOKUP['128'] = 'uniparental'
+CLINVAR_LOOKUP['256'] = 'not-tested'
+CLINVAR_LOOKUP['512'] = 'tested-inconclusive'
+CLINVAR_LOOKUP['1073741824'] = 'other'
+
+CLINVAR_SIG = {}
+CLINVAR_SIG['0'] = 'uncertain'
+CLINVAR_SIG['1'] = 'not-provided'
+CLINVAR_SIG['2'] = 'benign'
+CLINVAR_SIG['3'] = 'likely-benign'
+CLINVAR_SIG['4'] = 'likely-pathogenic'
+CLINVAR_SIG['5'] = 'pathogenic'
+CLINVAR_SIG['6'] = 'drug-response'
+CLINVAR_SIG['7'] = 'histocompatibility'
+CLINVAR_SIG['255'] = 'other'
+CLINVAR_SIG['.'] = '.'
+
+function clinvar_sig(vals) 
+    local t = type(vals)
+	-- just a single-value
+    if(t == "string" or t == "number") then
+		return CLINVAR_SIG[vals]
+	else
+        vals = {vals}
+	end
+    local ret = {}
+    for i=1,#vals do
+        if(index(vals[i], "|") == -1) then
+            ret[#ret+1] = CLINVAR_SIG[vals[i]]
+        else
+            local invals = split(vals[i], "|")
+            local inret = {}
+            for j=1,#invals do
+                inret[#inret+1] = CLINVAR_SIG[invals[j]]
+			end
+			ret[#ret+1] = table.concat(inret, "|")
+		end
+    end
+    return table.concat(ret, ",")
+end
+
+
+function div(a, b)
+	if(a == 0) then return 0.0 end
+	return (a / b).toFixed(9)
+end