diff --git a/docs/src/reference-verbs.md b/docs/src/reference-verbs.md
index cfa66dd826..142b90ce04 100644
--- a/docs/src/reference-verbs.md
+++ b/docs/src/reference-verbs.md
@@ -3126,6 +3126,25 @@ a b c
9 8 7
+## sparsify
+
+
+mlr sparsify --help
+
+
+Usage: mlr sparsify [options]
+Unsets fields for which the key is the empty string (or, optionally, another
+specified value). Only makes sense with output format not being CSV or TSV.
+Options:
+-s {filler string} What values to remove. Defaults to the empty string.
+-f {a,b,c} Specify field names to be operated on; any other fields won't be
+ modified. The default is to modify all fields.
+-h|--help Show this message.
+Example: if the input is two records, one being 'a=1,b=2' and the other
+being 'b=3,c=4', then the output is the two records 'a=1,b=2,c=' and
+'a=,b=3,c=4'.
+
+
## split
@@ -3409,14 +3428,14 @@ fields, optionally categorized by one or more fields.
data/medium
-x_y_cov 0.000042574820827444476
-x_y_corr 0.0005042001844467462
-y_y_cov 0.08461122467974003
+x_y_cov 0.00004257482082749404
+x_y_corr 0.0005042001844473328
+y_y_cov 0.08461122467974005
y_y_corr 1
-x2_xy_cov 0.04188382281779374
-x2_xy_corr 0.630174342037994
-x2_y2_cov -0.00030953725962542085
-x2_y2_corr -0.0034249088761121966
+x2_xy_cov 0.041883822817793716
+x2_xy_corr 0.6301743420379936
+x2_y2_cov -0.0003095372596253918
+x2_y2_corr -0.003424908876111875
@@ -3425,12 +3444,12 @@ x2_y2_corr -0.0034249088761121966
data/medium
-a x_y_ols_m x_y_ols_b x_y_ols_n x_y_r2 y_y_ols_m y_y_ols_b y_y_ols_n y_y_r2 xy_y2_ols_m xy_y2_ols_b xy_y2_ols_n xy_y2_r2
-pan 0.01702551273681908 0.5004028922897639 2081 0.00028691820445814767 1 0 2081 1 0.8781320866715662 0.11908230147563566 2081 0.41749827377311266
-eks 0.0407804923685586 0.48140207967651016 1965 0.0016461239223448587 1 0 1965 1 0.8978728611690183 0.10734054433612333 1965 0.45563223864254526
-wye -0.03915349075204814 0.5255096523974456 1966 0.0015051268704373607 1 0 1966 1 0.8538317334220835 0.1267454301662969 1966 0.38991721818599295
-zee 0.0027812364960399147 0.5043070448033061 2047 0.000007751652858786137 1 0 2047 1 0.8524439912011013 0.12401684308018937 2047 0.39356598090006495
-hat -0.018620577041095078 0.5179005397264935 1941 0.0003520036646055585 1 0 1941 1 0.8412305086345014 0.13557328318623216 1941 0.3687944261732265
+a x_y_ols_m x_y_ols_b x_y_ols_n x_y_r2 y_y_ols_m y_y_ols_b y_y_ols_n y_y_r2 xy_y2_ols_m xy_y2_ols_b xy_y2_ols_n xy_y2_r2
+pan 0.017025512736819345 0.500402892289764 2081 0.00028691820445815624 1 -0.00000000000000002890430283104539 2081 1 0.8781320866715664 0.11908230147563569 2081 0.4174982737731127
+eks 0.04078049236855813 0.4814020796765104 1965 0.0016461239223448218 1 0.00000000000000017862676354313703 1965 1 0.897872861169018 0.1073405443361234 1965 0.4556322386425451
+wye -0.03915349075204785 0.5255096523974457 1966 0.0015051268704373377 1 0.00000000000000004464425401127647 1966 1 0.8538317334220837 0.1267454301662969 1966 0.3899172181859931
+zee 0.0027812364960401333 0.5043070448033061 2047 0.000007751652858787357 1 0.00000000000000004819404567023685 2047 1 0.8524439912011011 0.12401684308018947 2047 0.39356598090006495
+hat -0.018620577041095272 0.5179005397264937 1941 0.00035200366460556604 1 -0.00000000000000003400445761787692 1941 1 0.8412305086345017 0.13557328318623207 1941 0.3687944261732266
Here's an example simple line-fit. The `x` and `y`
@@ -3516,11 +3535,11 @@ upsec_count_pca_quality 0.9999590846136102
donesec 92.33051350964094
color purple
-upsec_count_pca_m -39.03009744795354
-upsec_count_pca_b 979.9883413064914
+upsec_count_pca_m -39.030097447953594
+upsec_count_pca_b 979.9883413064917
upsec_count_pca_n 21
upsec_count_pca_quality 0.9999908956206317
-donesec 25.10852919630297
+donesec 25.108529196302943
## step
@@ -3797,9 +3816,9 @@ distinct_count 5 5 10000 10000 10000
mode pan wye 1 0.3467901443380824 0.7268028627434533
sum 0 0 50005000 4986.019681679581 5062.057444929905
mean - - 5000.5 0.49860196816795804 0.5062057444929905
-stddev - - 2886.8956799071675 0.2902925151144007 0.290880086426933
-var - - 8334166.666666667 0.08426974433144456 0.08461122467974003
-skewness - - 0 -0.0006899591185521965 -0.017849760120133784
+stddev - - 2886.8956799071675 0.29029251511440074 0.2908800864269331
+var - - 8334166.666666667 0.08426974433144457 0.08461122467974005
+skewness - - 0 -0.0006899591185517494 -0.01784976012013298
minlen 3 3 1 15 13
maxlen 3 3 5 22 22
min eks eks 1 0.00004509679127584487 0.00008818962627266114
diff --git a/docs/src/reference-verbs.md.in b/docs/src/reference-verbs.md.in
index 44feda3deb..8959ebf6bb 100644
--- a/docs/src/reference-verbs.md.in
+++ b/docs/src/reference-verbs.md.in
@@ -995,6 +995,12 @@ GENMD-RUN-COMMAND
mlr --ijson --opprint sort-within-records data/sort-within-records.json
GENMD-EOF
+## sparsify
+
+GENMD-RUN-COMMAND
+mlr sparsify --help
+GENMD-EOF
+
## split
GENMD-RUN-COMMAND
diff --git a/pkg/transformers/aaa_transformer_table.go b/pkg/transformers/aaa_transformer_table.go
index ed98af07f2..34a5b6ea85 100644
--- a/pkg/transformers/aaa_transformer_table.go
+++ b/pkg/transformers/aaa_transformer_table.go
@@ -62,6 +62,7 @@ var TRANSFORMER_LOOKUP_TABLE = []TransformerSetup{
SkipTrivialRecordsSetup,
SortSetup,
SortWithinRecordsSetup,
+ SparsifySetup,
SplitSetup,
SsubSetup,
Stats1Setup,
diff --git a/pkg/transformers/sparsify.go b/pkg/transformers/sparsify.go
new file mode 100644
index 0000000000..be1a6de73a
--- /dev/null
+++ b/pkg/transformers/sparsify.go
@@ -0,0 +1,194 @@
+package transformers
+
+import (
+ "container/list"
+ "fmt"
+ "os"
+ "strings"
+
+ "github.com/johnkerl/miller/pkg/cli"
+ "github.com/johnkerl/miller/pkg/lib"
+ "github.com/johnkerl/miller/pkg/mlrval"
+ "github.com/johnkerl/miller/pkg/types"
+)
+
+// ----------------------------------------------------------------
+const verbNameSparsify = "sparsify"
+
+var SparsifySetup = TransformerSetup{
+ Verb: verbNameSparsify,
+ UsageFunc: transformerSparsifyUsage,
+ ParseCLIFunc: transformerSparsifyParseCLI,
+ IgnoresInput: false,
+}
+
+func transformerSparsifyUsage(
+ o *os.File,
+) {
+ fmt.Fprintf(o, "Usage: %s %s [options]\n", "mlr", verbNameSparsify)
+ fmt.Fprint(o,
+ `Unsets fields for which the key is the empty string (or, optionally, another
+specified value). Only makes sense with output format not being CSV or TSV.
+`)
+
+ fmt.Fprintf(o, "Options:\n")
+ fmt.Fprintf(o, "-s {filler string} What values to remove. Defaults to the empty string.\n")
+ fmt.Fprintf(o, "-f {a,b,c} Specify field names to be operated on; any other fields won't be\n")
+ fmt.Fprintf(o, " modified. The default is to modify all fields.\n")
+ fmt.Fprintf(o, "-h|--help Show this message.\n")
+
+ fmt.Fprint(o,
+ `Example: if the input is two records, one being 'a=1,b=2' and the other
+being 'b=3,c=4', then the output is the two records 'a=1,b=2,c=' and
+'a=,b=3,c=4'.
+`)
+}
+
+func transformerSparsifyParseCLI(
+ pargi *int,
+ argc int,
+ args []string,
+ _ *cli.TOptions,
+ doConstruct bool, // false for first pass of CLI-parse, true for second pass
+) IRecordTransformer {
+
+ // Skip the verb name from the current spot in the mlr command line
+ argi := *pargi
+ verb := args[argi]
+ argi++
+
+ fillerString := ""
+ var specifiedFieldNames []string = nil
+
+ for argi < argc /* variable increment: 1 or 2 depending on flag */ {
+ opt := args[argi]
+ if !strings.HasPrefix(opt, "-") {
+ break // No more flag options to process
+ }
+ if args[argi] == "--" {
+ break // All transformers must do this so main-flags can follow verb-flags
+ }
+ argi++
+
+ if opt == "-h" || opt == "--help" {
+ transformerSparsifyUsage(os.Stdout)
+ os.Exit(0)
+
+ } else if opt == "-s" {
+ fillerString = cli.VerbGetStringArgOrDie(verb, opt, args, &argi, argc)
+
+ } else if opt == "-f" {
+ specifiedFieldNames = cli.VerbGetStringArrayArgOrDie(verb, opt, args, &argi, argc)
+
+ } else {
+ transformerSparsifyUsage(os.Stderr)
+ os.Exit(1)
+ }
+ }
+
+ *pargi = argi
+ if !doConstruct { // All transformers must do this for main command-line parsing
+ return nil
+ }
+
+ transformer, err := NewTransformerSparsify(
+ fillerString,
+ specifiedFieldNames,
+ )
+ if err != nil {
+ fmt.Fprintln(os.Stderr, err)
+ os.Exit(1)
+ }
+
+ return transformer
+}
+
+// ----------------------------------------------------------------
+type TransformerSparsify struct {
+ fillerString string
+ fieldNamesSet map[string]bool
+ recordTransformerFunc RecordTransformerFunc
+}
+
+func NewTransformerSparsify(
+ fillerString string,
+ specifiedFieldNames []string,
+) (*TransformerSparsify, error) {
+
+ tr := &TransformerSparsify{
+ fillerString: fillerString,
+ fieldNamesSet: lib.StringListToSet(specifiedFieldNames),
+ }
+ if specifiedFieldNames == nil {
+ tr.recordTransformerFunc = tr.transformAll
+ } else {
+ tr.recordTransformerFunc = tr.transformSome
+ }
+
+ return tr, nil
+}
+
+func (tr *TransformerSparsify) Transform(
+ inrecAndContext *types.RecordAndContext,
+ outputRecordsAndContexts *list.List, // list of *types.RecordAndContext
+ inputDownstreamDoneChannel <-chan bool,
+ outputDownstreamDoneChannel chan<- bool,
+) {
+ HandleDefaultDownstreamDone(inputDownstreamDoneChannel, outputDownstreamDoneChannel)
+
+ if !inrecAndContext.EndOfStream {
+ tr.recordTransformerFunc(
+ inrecAndContext,
+ outputRecordsAndContexts,
+ inputDownstreamDoneChannel,
+ outputDownstreamDoneChannel,
+ )
+ } else {
+ outputRecordsAndContexts.PushBack(inrecAndContext) // end-of-stream marker
+ }
+}
+
+func (tr *TransformerSparsify) transformAll(
+ inrecAndContext *types.RecordAndContext,
+ outputRecordsAndContexts *list.List, // list of *types.RecordAndContext
+ inputDownstreamDoneChannel <-chan bool,
+ outputDownstreamDoneChannel chan<- bool,
+) {
+ inrec := inrecAndContext.Record
+ outrec := mlrval.NewMlrmapAsRecord()
+
+ for pe := inrec.Head; pe != nil; pe = pe.Next {
+ if pe.Value.String() != tr.fillerString {
+ // Reference OK because ownership transfer
+ outrec.PutReference(pe.Key, pe.Value)
+ }
+ }
+
+ outrecAndContext := types.NewRecordAndContext(outrec, &inrecAndContext.Context)
+ outputRecordsAndContexts.PushBack(outrecAndContext)
+}
+
+// ----------------------------------------------------------------
+func (tr *TransformerSparsify) transformSome(
+ inrecAndContext *types.RecordAndContext,
+ outputRecordsAndContexts *list.List, // list of *types.RecordAndContext
+ inputDownstreamDoneChannel <-chan bool,
+ outputDownstreamDoneChannel chan<- bool,
+) {
+ inrec := inrecAndContext.Record
+ outrec := mlrval.NewMlrmapAsRecord()
+
+ for pe := inrec.Head; pe != nil; pe = pe.Next {
+ if tr.fieldNamesSet[pe.Key] {
+ if pe.Value.String() != tr.fillerString {
+ // Reference OK because ownership transfer
+ outrec.PutReference(pe.Key, pe.Value)
+ }
+ } else {
+ outrec.PutReference(pe.Key, pe.Value)
+ }
+ }
+
+ outrecAndContext := types.NewRecordAndContext(outrec, &inrecAndContext.Context)
+ outputRecordsAndContexts.PushBack(outrecAndContext)
+}
diff --git a/test/cases/cli-help/0001/expout b/test/cases/cli-help/0001/expout
index b25e4a56d1..07ca9d0961 100644
--- a/test/cases/cli-help/0001/expout
+++ b/test/cases/cli-help/0001/expout
@@ -988,6 +988,20 @@ Options:
-r Recursively sort subobjects/submaps, e.g. for JSON input.
-h|--help Show this message.
+================================================================
+sparsify
+Usage: mlr sparsify [options]
+Unsets fields for which the key is the empty string (or, optionally, another
+specified value). Only makes sense with output format not being CSV or TSV.
+Options:
+-s {filler string} What values to remove. Defaults to the empty string.
+-f {a,b,c} Specify field names to be operated on; any other fields won't be
+ modified. The default is to modify all fields.
+-h|--help Show this message.
+Example: if the input is two records, one being 'a=1,b=2' and the other
+being 'b=3,c=4', then the output is the two records 'a=1,b=2,c=' and
+'a=,b=3,c=4'.
+
================================================================
split
Usage: mlr split [options] {filename}
diff --git a/test/input/sparsify-input.csv b/test/input/sparsify-input.csv
new file mode 100644
index 0000000000..16916596e0
--- /dev/null
+++ b/test/input/sparsify-input.csv
@@ -0,0 +1,5 @@
+a,b,c
+1,2,3
+4,5,
+,,
+7,8,9