Merge branch 'development'

Separator and fixed #3

cmd/root.go

@@ -18,18 +18,17 @@ package cmd
 
 import (
 	"bytes"
-	"github.com/tlinden/tablizer/lib"
 	"fmt"
 	"github.com/spf13/cobra"
+	"github.com/tlinden/tablizer/lib"
 	"log"
 	"os"
 	"os/exec"
 )
 
-var helpCmd = &cobra.Command{
+var ShowManual = false
-	Use:   "help",
+
-	Short: "Show documentation",
+func man() {
-	Run: func(cmd *cobra.Command, args []string) {
 	man := exec.Command("less", "-")
 
 	var b bytes.Buffer
@@ -44,7 +43,6 @@ var helpCmd = &cobra.Command{
 	if err != nil {
 		log.Fatal(err)
 	}
-	},
 }
 
 var rootCmd = &cobra.Command{
@@ -57,6 +55,11 @@ var rootCmd = &cobra.Command{
 			return nil
 		}
 
+		if ShowManual {
+			man()
+			return nil
+		}
+
 		err := lib.PrepareColumns()
 		if err != nil {
 			return err
@@ -81,8 +84,10 @@ func Execute() {
 func init() {
 	rootCmd.PersistentFlags().BoolVarP(&lib.Debug, "debug", "d", false, "Enable debugging")
 	rootCmd.PersistentFlags().BoolVarP(&lib.NoNumbering, "no-numbering", "n", false, "Disable header numbering")
-	rootCmd.PersistentFlags().BoolVarP(&lib.ShowVersion, "version", "v", false, "Print program version")
+	rootCmd.PersistentFlags().BoolVarP(&lib.ShowVersion, "version", "V", false, "Print program version")
-	rootCmd.PersistentFlags().StringVarP(&lib.Separator, "separator", "s", "", "Custom field separator")
+	rootCmd.PersistentFlags().BoolVarP(&lib.InvertMatch, "invert-match", "v", false, "select non-matching rows")
+	rootCmd.PersistentFlags().BoolVarP(&ShowManual, "man", "m", false, "Display manual page")
+	rootCmd.PersistentFlags().StringVarP(&lib.Separator, "separator", "s", lib.DefaultSeparator, "Custom field separator")
 	rootCmd.PersistentFlags().StringVarP(&lib.Columns, "columns", "c", "", "Only show the speficied columns (separated by ,)")
 
 	// output flags, only 1 allowed, hidden, since just short cuts
@@ -98,11 +103,4 @@ func init() {
 
 	// same thing but more common, takes precedence over above group
 	rootCmd.PersistentFlags().StringVarP(&lib.OutputMode, "output", "o", "", "Output mode - one of: orgtbl, markdown, extended, shell, ascii(default)")
-
-	rootCmd.AddCommand(helpCmd)
-
-	rootCmd.SetHelpCommand(&cobra.Command{
-		Use:    "no-help",
-		Hidden: true,
-	})
 }

cmd/tablizer.go

@@ -11,6 +11,8 @@ SYNOPSIS
           -c, --columns string     Only show the speficied columns (separated by ,)
           -d, --debug              Enable debugging
           -h, --help               help for tablizer
+          -v, --invert-match       select non-matching rows
+          -m, --man                Display manual page
           -n, --no-numbering       Disable header numbering
           -o, --output string      Output mode - one of: orgtbl, markdown, extended, ascii(default)
           -X, --extended           Enable extended output
@@ -38,7 +40,8 @@ DESCRIPTION
 
     Without any options it reads its input from "STDIN", but you can also
     specify a file as a parameter. If you want to reduce the output by some
-    regular expression, just specify it as its first parameters. Hence:
+    regular expression, just specify it as its first parameter. You may also
+    use the -v option to exclude all rows which match the pattern. Hence:
 
        # read from STDIN
        kubectl get pods | tablizer

lib/common.go

@@ -25,18 +25,20 @@ var (
 	ShowVersion      bool
 	Columns          string
 	UseColumns       []int
-	Separator       string
+	DefaultSeparator string = `(\s\s+|\t)`
+	Separator        string = `(\s\s+|\t)`
 	OutflagExtended  bool
 	OutflagMarkdown  bool
 	OutflagOrgtable  bool
 	OutflagShell     bool
 	OutputMode       string
+	InvertMatch      bool
 
 	// used for validation
 	validOutputmodes = "(orgtbl|markdown|extended|ascii)"
 
 	// main program version
-	Version = "v1.0.4"
+	Version = "v1.0.6"
 
 	// generated  version string, used  by -v contains  lib.Version on
 	//  main  branch,   and  lib.Version-$branch-$lastcommit-$date  on

lib/helpers.go

@@ -48,8 +48,49 @@ func PrepareColumns() error {
 	return nil
 }
 
+func numberizeHeaders(data *Tabdata) {
+	// prepare headers: add numbers to headers
+	numberedHeaders := []string{}
+	for i, head := range data.headers {
+		if len(Columns) > 0 {
+			// -c specified
+			if !contains(UseColumns, i+1) {
+				// ignore this one
+				continue
+			}
+		}
+		if NoNumbering {
+			numberedHeaders = append(numberedHeaders, head)
+		} else {
+			numberedHeaders = append(numberedHeaders, fmt.Sprintf("%s(%d)", head, i+1))
+		}
+	}
+	data.headers = numberedHeaders
+}
+
+func reduceColumns(data *Tabdata) {
+	// exclude columns, if any
+	if len(Columns) > 0 {
+		reducedEntries := [][]string{}
+		reducedEntry := []string{}
+		for _, entry := range data.entries {
+			reducedEntry = nil
+			for i, value := range entry {
+				if !contains(UseColumns, i+1) {
+					continue
+				}
+
+				reducedEntry = append(reducedEntry, value)
+			}
+			reducedEntries = append(reducedEntries, reducedEntry)
+		}
+		data.entries = reducedEntries
+	}
+}
+
 func PrepareModeFlags() error {
 	if len(OutputMode) == 0 {
+		// associate short flags like -X with mode selector
 		switch {
 		case OutflagExtended:
 			OutputMode = "extended"

lib/helpers_test.go

@@ -52,6 +52,7 @@ func TestPrepareColumns(t *testing.T) {
 	}{
 		{"1,2,3", []int{1, 2, 3}, false},
 		{"1,2,", []int{}, true},
+		{"a,b", []int{}, true},
 	}
 
 	for _, tt := range tests {
@@ -71,3 +72,46 @@ func TestPrepareColumns(t *testing.T) {
 		})
 	}
 }
+
+func TestReduceColumns(t *testing.T) {
+	var tests = []struct {
+		expect  [][]string
+		columns []int
+	}{
+		{
+			expect:  [][]string{[]string{"a", "b"}},
+			columns: []int{1, 2},
+		},
+		{
+			expect:  [][]string{[]string{"a", "c"}},
+			columns: []int{1, 3},
+		},
+		{
+			expect:  [][]string{[]string{"a"}},
+			columns: []int{1},
+		},
+		{
+			expect:  [][]string{nil},
+			columns: []int{4},
+		},
+	}
+
+	input := [][]string{[]string{"a", "b", "c"}}
+
+	Columns = "y" // used as a flag with len(Columns)...
+
+	for _, tt := range tests {
+		testname := fmt.Sprintf("reduce-columns-by-%+v", tt.columns)
+		t.Run(testname, func(t *testing.T) {
+			UseColumns = tt.columns
+			data := Tabdata{entries: input}
+			reduceColumns(&data)
+			if !reflect.DeepEqual(data.entries, tt.expect) {
+				t.Errorf("reduceColumns returned invalid data:\ngot: %+v\nexp: %+v", data.entries, tt.expect)
+			}
+		})
+	}
+
+	Columns = "" // reset for other tests
+	UseColumns = nil
+}

lib/parser.go

@@ -31,47 +31,34 @@ import (
 type Tabdata struct {
 	maxwidthHeader int      // longest header
 	maxwidthPerCol []int    // max width per column
-	columns        int
+	columns        int      // count
-	headerIndices  []map[string]int // [ {beg=>0, end=>17}, ... ]
 	headers        []string // [ "ID", "NAME", ...]
 	entries        [][]string
 }
 
 /*
-   Parse tabular input. We split the  header (first line) by 2 or more
+   Parse tabular input.
-   spaces, remember the positions of  the header fields. We then split
-   the data (everything after the first line) by those positions. That
-   way we can turn "tabular data" (with fields containing whitespaces)
-   into real tabular data. We re-tabulate our input if you will.
 */
 func parseFile(input io.Reader, pattern string) (Tabdata, error) {
 	data := Tabdata{}
 
 	var scanner *bufio.Scanner
-	var spaces = `\s\s+|$`
-
-	if len(Separator) > 0 {
-		spaces = Separator
-	}
 
 	hadFirst := false
-	spacefinder := regexp.MustCompile(spaces)
+	separate := regexp.MustCompile(Separator)
-	beg := 0
-
-	scanner = bufio.NewScanner(input)
-
-	for scanner.Scan() {
-		line := strings.TrimSpace(scanner.Text())
-		values := []string{}
-
 	patternR, err := regexp.Compile(pattern)
 	if err != nil {
 		return data, errors.Unwrap(fmt.Errorf("Regexp pattern %s is invalid: %w", pattern, err))
 	}
 
+	scanner = bufio.NewScanner(input)
+
+	for scanner.Scan() {
+		line := strings.TrimSpace(scanner.Text())
+		parts := separate.Split(line, -1)
+
 		if !hadFirst {
 			// header processing
-			parts := spacefinder.FindAllStringIndex(line, -1)
 			data.columns = len(parts)
 			// if Debug {
 			// 	fmt.Println(parts)
@@ -83,30 +70,14 @@ func parseFile(input io.Reader, pattern string) (Tabdata, error) {
 				// 	fmt.Printf("Part: <%s>\n", string(line[beg:part[0]]))
 				//}
 
-				// current field
-				head := string(line[beg:part[0]])
-
-				// register begin and end of field within line
-				indices := make(map[string]int)
-				indices["beg"] = beg
-				if part[0] == part[1] {
-					indices["end"] = 0
-				} else {
-					indices["end"] = part[1] - 1
-				}
-
 				// register widest header field
-				headerlen := len(head)
+				headerlen := len(part)
 				if headerlen > data.maxwidthHeader {
 					data.maxwidthHeader = headerlen
 				}
 
 				// register fields data
-				data.headerIndices = append(data.headerIndices, indices)
+				data.headers = append(data.headers, strings.TrimSpace(part))
-				data.headers = append(data.headers, head)
-
-				// end of current field == begin of next one
-				beg = part[1]
 
 				// done
 				hadFirst = true
@@ -114,22 +85,19 @@ func parseFile(input io.Reader, pattern string) (Tabdata, error) {
 		} else {
 			// data processing
 			if len(pattern) > 0 {
-				if !patternR.MatchString(line) {
+				if patternR.MatchString(line) == InvertMatch {
+					// by default  -v is false, so if a  line does NOT
+					// match the pattern, we will ignore it. However,
+					// if the user specified -v, the matching is inverted,
+					// so we ignore all lines, which DO match.
 					continue
 				}
 			}
 
 			idx := 0 // we cannot use the header index, because we could exclude columns
-			for _, index := range data.headerIndices {
+			values := []string{}
-				value := ""
+			for _, part := range parts {
-
+				width := len(strings.TrimSpace(part))
-				if index["end"] == 0 {
-					value = string(line[index["beg"]:])
-				} else {
-					value = string(line[index["beg"]:index["end"]])
-				}
-
-				width := len(strings.TrimSpace(value))
 
 				if len(data.maxwidthPerCol)-1 < idx {
 					data.maxwidthPerCol = append(data.maxwidthPerCol, width)
@@ -142,7 +110,7 @@ func parseFile(input io.Reader, pattern string) (Tabdata, error) {
 				// if Debug {
 				// 	fmt.Printf("<%s> ", value)
 				// }
-				values = append(values, strings.TrimSpace(value))
+				values = append(values, strings.TrimSpace(part))
 
 				idx++
 			}
@@ -151,7 +119,7 @@ func parseFile(input io.Reader, pattern string) (Tabdata, error) {
 	}
 
 	if scanner.Err() != nil {
-		return data, errors.Unwrap(fmt.Errorf("Regexp pattern %s is invalid: %w", pattern, scanner.Err()))
+		return data, errors.Unwrap(fmt.Errorf("Failed to read from io.Reader: %w", scanner.Err()))
 	}
 
 	if Debug {

lib/parser_test.go

@@ -18,6 +18,7 @@ along with this program. If not, see <http://www.gnu.org/licenses/>.
 package lib
 
 import (
+	"fmt"
 	"reflect"
 	"strings"
 	"testing"
@@ -27,40 +28,18 @@ func TestParser(t *testing.T) {
 	data := Tabdata{
 		maxwidthHeader: 5,
 		maxwidthPerCol: []int{
-			5,
+			5, 5, 8,
-			5,
-			8,
 		},
 		columns: 3,
-		headerIndices: []map[string]int{
-			map[string]int{
-				"beg": 0,
-				"end": 6,
-			},
-			map[string]int{
-				"end": 13,
-				"beg": 7,
-			},
-			map[string]int{
-				"beg": 14,
-				"end": 0,
-			},
-		},
 		headers: []string{
-			"ONE",
+			"ONE", "TWO", "THREE",
-			"TWO",
-			"THREE",
 		},
 		entries: [][]string{
 			[]string{
-				"asd",
+				"asd", "igig", "cxxxncnc",
-				"igig",
-				"cxxxncnc",
 			},
 			[]string{
-				"19191",
+				"19191", "EDD 1", "X",
-				"EDD 1",
-				"X",
 			},
 		},
 	}
@@ -71,12 +50,63 @@ asd    igig   cxxxncnc
 
 	readFd := strings.NewReader(table)
 	gotdata, err := parseFile(readFd, "")
+	Separator = DefaultSeparator
 
 	if err != nil {
 		t.Errorf("Parser returned error: %s\nData processed so far: %+v", err, gotdata)
 	}
 
 	if !reflect.DeepEqual(data, gotdata) {
-		t.Errorf("Parser returned invalid data\nExp: %+v\nGot: %+v\n", data, gotdata)
+		t.Errorf("Parser returned invalid data, Regex: %s\nExp: %+v\nGot: %+v\n", Separator, data, gotdata)
+	}
+}
+
+func TestParserPatternmatching(t *testing.T) {
+	var tests = []struct {
+		entries [][]string
+		pattern string
+		invert  bool
+	}{
+		{
+			entries: [][]string{
+				[]string{
+					"asd", "igig", "cxxxncnc",
+				},
+			},
+			pattern: "ig",
+			invert:  false,
+		},
+		{
+			entries: [][]string{
+				[]string{
+					"19191", "EDD 1", "X",
+				},
+			},
+			pattern: "ig",
+			invert:  true,
+		},
+	}
+
+	table := `ONE    TWO    THREE  
+asd    igig   cxxxncnc  
+19191  EDD 1  X`
+
+	for _, tt := range tests {
+		testname := fmt.Sprintf("parse-with-inverted-pattern-%t", tt.invert)
+		t.Run(testname, func(t *testing.T) {
+			InvertMatch = tt.invert
+
+			readFd := strings.NewReader(table)
+			gotdata, err := parseFile(readFd, tt.pattern)
+
+			if err != nil {
+				t.Errorf("Parser returned error: %s\nData processed so far: %+v", err, gotdata)
+			}
+
+			if !reflect.DeepEqual(tt.entries, gotdata.entries) {
+				t.Errorf("Parser returned invalid data (pattern: %s, invert: %t)\nExp: %+v\nGot: %+v\n",
+					tt.pattern, tt.invert, tt.entries, gotdata.entries)
+			}
+		})
 	}
 }

lib/printer.go

@@ -26,41 +26,10 @@ import (
 )
 
 func printData(data *Tabdata) {
-	// prepare headers: add numbers to headers
+	if OutputMode != "shell" {
-	numberedHeaders := []string{}
+		numberizeHeaders(data)
-	for i, head := range data.headers {
-		if len(Columns) > 0 {
-			// -c specified
-			if !contains(UseColumns, i+1) {
-				// ignore this one
-				continue
-			}
-		}
-		if NoNumbering {
-			numberedHeaders = append(numberedHeaders, head)
-		} else {
-			numberedHeaders = append(numberedHeaders, fmt.Sprintf("%s(%d)", head, i+1))
-		}
-	}
-	data.headers = numberedHeaders
-
-	// prepare data
-	if len(Columns) > 0 {
-		reducedEntries := [][]string{}
-		reducedEntry := []string{}
-		for _, entry := range data.entries {
-			reducedEntry = nil
-			for i, value := range entry {
-				if !contains(UseColumns, i+1) {
-					continue
-				}
-
-				reducedEntry = append(reducedEntry, value)
-			}
-			reducedEntries = append(reducedEntries, reducedEntry)
-		}
-		data.entries = reducedEntries
 	}
+	reduceColumns(data)
 
 	switch OutputMode {
 	case "extended":
@@ -190,6 +159,7 @@ func printShellData(data *Tabdata) {
 		var idx int
 		for _, entry := range data.entries {
 			idx = 0
+			shentries := []string{}
 			for i, value := range entry {
 				if len(Columns) > 0 {
 					if !contains(UseColumns, i+1) {
@@ -197,10 +167,10 @@ func printShellData(data *Tabdata) {
 					}
 				}
 
-				fmt.Printf("%s=\"%s\" ", data.headers[idx], value)
+				shentries = append(shentries, fmt.Sprintf("%s=\"%s\"", data.headers[idx], value))
 				idx++
 			}
-			fmt.Println()
+			fmt.Println(strings.Join(shentries, " "))
 		}
 	}
 }

lib/printer_test.go

@@ -18,15 +18,33 @@ along with this program. If not, see <http://www.gnu.org/licenses/>.
 package lib
 
 import (
+	"fmt"
 	"os"
 	"strings"
 	"testing"
 )
 
 func TestPrinter(t *testing.T) {
-	table := `ONE    TWO    THREE  
+	startdata := Tabdata{
-asd    igig   cxxxncnc  
+		maxwidthHeader: 5,
-19191  EDD 1  X`
+		maxwidthPerCol: []int{
+			5,
+			5,
+			8,
+		},
+		columns: 3,
+		headers: []string{
+			"ONE", "TWO", "THREE",
+		},
+		entries: [][]string{
+			[]string{
+				"asd", "igig", "cxxxncnc",
+			},
+			[]string{
+				"19191", "EDD 1", "X",
+			},
+		},
+	}
 
 	expects := map[string]string{
 		"ascii": `ONE(1)	TWO(2)	THREE(3) 
@@ -42,6 +60,8 @@ asd   	igig  	cxxxncnc
 |--------|--------|----------|
 | asd    | igig   | cxxxncnc |
 |  19191 | EDD 1  | X        |`,
+		"shell": `ONE="asd" TWO="igig" THREE="cxxxncnc"
+ONE="19191" TWO="EDD 1" THREE="X"`,
 	}
 
 	r, w, err := os.Pipe()
@@ -52,14 +72,11 @@ asd   	igig  	cxxxncnc
 	os.Stdout = w
 
 	for mode, expect := range expects {
+		testname := fmt.Sprintf("print-%s", mode)
+		t.Run(testname, func(t *testing.T) {
+
 			OutputMode = mode
-		fd := strings.NewReader(table)
+			data := startdata // we need to reset our mock data, since it's being modified in printData()
-		data, err := parseFile(fd, "")
-
-		if err != nil {
-			t.Errorf("Parser returned error: %s\nData processed so far: %+v", err, data)
-		}
-
 			printData(&data)
 
 			buf := make([]byte, 1024)
@@ -73,6 +90,7 @@ asd   	igig  	cxxxncnc
 			if output != expect {
 				t.Errorf("output mode: %s, got:\n%s\nwant:\n%s\n (%d <=> %d)", mode, output, expect, len(output), len(expect))
 			}
+		})
 	}
 
 	// Restore

tablizer.1

@@ -133,7 +133,7 @@
 .\" ========================================================================
 .\"
 .IX Title "TABLIZER 1"
-.TH TABLIZER 1 "2022-10-04" "1" "User Commands"
+.TH TABLIZER 1 "2022-10-05" "1" "User Commands"
 .\" For nroff, turn off justification.  Always turn off hyphenation; it makes
 .\" way too many mistakes in technical documents.
 .if n .ad l
@@ -150,6 +150,8 @@ tablizer \- Manipulate tabular output of other programs
 \&      \-c, \-\-columns string     Only show the speficied columns (separated by ,)
 \&      \-d, \-\-debug              Enable debugging
 \&      \-h, \-\-help               help for tablizer
+\&      \-v, \-\-invert\-match       select non\-matching rows
+\&      \-m, \-\-man                Display manual page
 \&      \-n, \-\-no\-numbering       Disable header numbering
 \&      \-o, \-\-output string      Output mode \- one of: orgtbl, markdown, extended, ascii(default)
 \&      \-X, \-\-extended           Enable extended output
@@ -179,8 +181,9 @@ positions.
 .PP
 Without any options it reads its input from \f(CW\*(C`STDIN\*(C'\fR, but you can also
 specify a  file as a  parameter. If you want  to reduce the  output by
-some   regular   expression,   just    specify   it   as   its   first
+some regular expression,  just specify it as its  first parameter. You
-parameters. Hence:
+may also  use the  \fB\-v\fR option  to exclude all  rows which  match the
+pattern. Hence:
 .PP
 .Vb 2
 \&   # read from STDIN

tablizer.pod

@@ -11,6 +11,8 @@ tablizer - Manipulate tabular output of other programs
       -c, --columns string     Only show the speficied columns (separated by ,)
       -d, --debug              Enable debugging
       -h, --help               help for tablizer
+      -v, --invert-match       select non-matching rows
+      -m, --man                Display manual page
       -n, --no-numbering       Disable header numbering
       -o, --output string      Output mode - one of: orgtbl, markdown, extended, ascii(default)
       -X, --extended           Enable extended output
@@ -24,10 +26,10 @@ tablizer - Manipulate tabular output of other programs
 
 Many  programs generate  tabular output.   But sometimes  you need  to
 post-process these tables, you may need  to remove one or more columns
-or you may want to filter for  some pattern or you may need the output
+or you  may want to filter  for some pattern (See  L<PATTERNS>) or you
-in another program and need to  parse it somehow.  Standard unix tools
+may need the  output in another program and need  to parse it somehow.
-such as  awk(1), grep(1) or column(1)  may help, but sometimes  it's a
+Standard unix tools such as awk(1), grep(1) or column(1) may help, but
-tedious business.
+sometimes it's a tedious business.
 
 Let's take  the output of  the tool  kubectl.  It contains  cells with
 withespace and they do not separate columns by TAB characters. This is
@@ -41,8 +43,9 @@ positions.
 
 Without any options it reads its input from C<STDIN>, but you can also
 specify a  file as a  parameter. If you want  to reduce the  output by
-some   regular   expression,   just    specify   it   as   its   first
+some regular expression,  just specify it as its  first parameter. You
-parameters. Hence:
+may also  use the  B<-v> option  to exclude all  rows which  match the
+pattern. Hence:
 
    # read from STDIN
    kubectl get pods | tablizer
@@ -74,6 +77,34 @@ The numbering can be suppressed by using the B<-n> option.
 Finally the  B<-d> option  enables debugging  output which  is mostly
 usefull for the developer.
 
+=head2 PATTERNS
+
+You can reduce the rows being  displayed by using a regular expression
+pattern.  The  regexp is  PCRE compatible, refer  to the  syntax cheat
+sheet here: L<https://github.com/google/re2/wiki/Syntax>.  If you want
+to read  a more comprehensive  documentation about the topic  and have
+perl installed you can read it with:
+
+    perldoc perlre
+
+Or read it online: L<https://perldoc.perl.org/perlre>.
+
+A note on  modifiers: the regexp engine used in  tablizer uses another
+modifier syntax:
+
+    (?MODIFIER)
+
+The most important modifiers are:
+
+C<i> ignore case
+C<m> multiline mode
+C<s> single line mode
+
+Example for a case insensitve search:
+
+    kubectl get pods -A | tablizer "(?i)account"
+
+
 =head2 OUTPUT MODES
 
 There might be cases  when the tabular output of a  program is way too