Merge branch 'development'

cmd/root.go

@@ -87,7 +87,7 @@ func init() {
 	rootCmd.PersistentFlags().BoolVarP(&lib.ShowVersion, "version", "V", false, "Print program version")
 	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", "", "Custom field separator")
+	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

lib/common.go

@@ -19,25 +19,26 @@ package lib
 
 var (
 	// command line flags
-	Debug           bool
-	XtendedOut      bool
-	NoNumbering     bool
-	ShowVersion     bool
-	Columns         string
-	UseColumns      []int
-	Separator       string
-	OutflagExtended bool
-	OutflagMarkdown bool
-	OutflagOrgtable bool
-	OutflagShell    bool
-	OutputMode      string
-	InvertMatch     bool
+	Debug            bool
+	XtendedOut       bool
+	NoNumbering      bool
+	ShowVersion      bool
+	Columns          string
+	UseColumns       []int
+	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 = "v1.0.5"
+	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

@@ -29,49 +29,36 @@ import (
 
 // contains a whole parsed table
 type Tabdata struct {
-	maxwidthHeader int   // longest header
-	maxwidthPerCol []int // max width per column
-	columns        int
-	headerIndices  []map[string]int // [ {beg=>0, end=>17}, ... ]
-	headers        []string         // [ "ID", "NAME", ...]
+	maxwidthHeader int      // longest header
+	maxwidthPerCol []int    // max width per column
+	columns        int      // count
+	headers        []string // [ "ID", "NAME", ...]
 	entries        [][]string
 }
 
 /*
-   Parse tabular input. We split the  header (first line) by 2 or more
-   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.
+   Parse tabular input.
 */
 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)
-	beg := 0
+	separate := regexp.MustCompile(Separator)
+	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())
-		values := []string{}
-
-		patternR, err := regexp.Compile(pattern)
-		if err != nil {
-			return data, errors.Unwrap(fmt.Errorf("Regexp pattern %s is invalid: %w", pattern, err))
-		}
+		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, head)
-
-				// end of current field == begin of next one
-				beg = part[1]
+				data.headers = append(data.headers, strings.TrimSpace(part))
 
 				// done
 				hadFirst = true
@@ -124,16 +95,9 @@ func parseFile(input io.Reader, pattern string) (Tabdata, error) {
 			}
 
 			idx := 0 // we cannot use the header index, because we could exclude columns
-			for _, index := range data.headerIndices {
-				value := ""
-
-				if index["end"] == 0 {
-					value = string(line[index["beg"]:])
-				} else {
-					value = string(line[index["beg"]:index["end"]])
-				}
-
-				width := len(strings.TrimSpace(value))
+			values := []string{}
+			for _, part := range parts {
+				width := len(strings.TrimSpace(part))
 
 				if len(data.maxwidthPerCol)-1 < idx {
 					data.maxwidthPerCol = append(data.maxwidthPerCol, width)
@@ -146,7 +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++
 			}

lib/parser_test.go

@@ -28,40 +28,18 @@ func TestParser(t *testing.T) {
 	data := Tabdata{
 		maxwidthHeader: 5,
 		maxwidthPerCol: []int{
-			5,
-			5,
-			8,
+			5, 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",
-			"TWO",
-			"THREE",
+			"ONE", "TWO", "THREE",
 		},
 		entries: [][]string{
 			[]string{
-				"asd",
-				"igig",
-				"cxxxncnc",
+				"asd", "igig", "cxxxncnc",
 			},
 			[]string{
-				"19191",
-				"EDD 1",
-				"X",
+				"19191", "EDD 1", "X",
 			},
 		},
 	}
@@ -72,13 +50,14 @@ 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)
 	}
 }
 
@@ -91,9 +70,7 @@ func TestParserPatternmatching(t *testing.T) {
 		{
 			entries: [][]string{
 				[]string{
-					"asd",
-					"igig",
-					"cxxxncnc",
+					"asd", "igig", "cxxxncnc",
 				},
 			},
 			pattern: "ig",
@@ -102,9 +79,7 @@ func TestParserPatternmatching(t *testing.T) {
 		{
 			entries: [][]string{
 				[]string{
-					"19191",
-					"EDD 1",
-					"X",
+					"19191", "EDD 1", "X",
 				},
 			},
 			pattern: "ig",