initial commit

Makefile

@@ -0,0 +1,18 @@
+#
+# no need to modify anything below
+tool    = tablizer
+version = $(shell egrep "^var version = " cmd/root.go | cut -d'=' -f2 | cut -d'"' -f 2)
+archs   = android darwin freebsd linux netbsd openbsd windows
+
+all:
+	@echo "Type 'make install' to install $(tool)"
+
+install:
+	install -m 755 -d $(bindir)
+	install -m 755 -d $(linkdir)
+	install -m 755 $(tool) $(bindir)/$(tool)-$(version)
+	ln -sf $(bindir)/$(tool)-$(version) $(linkdir)/$(tool)
+
+release:
+	mkdir -p releases
+	$(foreach arch,$(archs), GOOS=$(arch) GOARCH=amd64 go build -x -o releases/$(tool)-$(arch)-amd64-$(version); sha256sum releases/$(tool)-$(arch)-amd64-$(version) | cut -d' ' -f1 > releases/$(tool)-$(arch)-amd64-$(version).sha256sum;)

README.md

@@ -1,2 +1,93 @@
-# tablizer
-Manipulate tabular output of other programs
+## tablizer - Manipulate tabular output of other programs
+
+Tablizer  can   be  used   to  re-format   tabular  output   of  other
+programs. While you  could do this using standard unix  tools, in some
+cases it's a hard job.
+
+Let's take this output:
+```
+% kubectl get pods -o wide
+NAME                        READY   STATUS    RESTARTS      AGE
+repldepl-7bcd8d5b64-7zq4l   1/1     Running   1 (69m ago)   5h26m
+repldepl-7bcd8d5b64-m48n8   1/1     Running   1 (69m ago)   5h26m
+repldepl-7bcd8d5b64-q2bf4   1/1     Running   1 (69m ago)   5h26m
+```
+
+But you're only interested in the  NAME and STATUS columns. Here's how
+to do this with tablizer:
+
+```
+% kubectl get pods | ./tablizer 
+NAME(1)                      READY(2) STATUS(3)  RESTARTS(4)    AGE(5)
+repldepl-7bcd8d5b64-7zq4l    1/1      Running    1 (69m ago)    5h26m
+repldepl-7bcd8d5b64-m48n8    1/1      Running    1 (69m ago)    5h26m
+repldepl-7bcd8d5b64-q2bf4    1/1      Running    1 (69m ago)    5h26m
+
+% kubectl get pods | ./tablizer -c 1,3
+NAME(1)                      STATUS(3)
+repldepl-7bcd8d5b64-7zq4l    Running
+repldepl-7bcd8d5b64-m48n8    Running
+repldepl-7bcd8d5b64-q2bf4    Running 
+```
+
+Another use case is when the tabular  output is so wide that lines are
+being broken and  the whole output is completely distorted.  In such a
+case you can use the `-x` flag to get an output similar to `\x` in `psql`:
+
+```
+% kubectl get pods | ./tablizer -x    
+    NAME: repldepl-7bcd8d5b64-7zq4l  
+   READY: 1/1    
+  STATUS: Running  
+RESTARTS: 1 (71m ago)  
+     AGE: 5h28m
+
+    NAME: repldepl-7bcd8d5b64-m48n8  
+   READY: 1/1    
+  STATUS: Running  
+RESTARTS: 1 (71m ago)  
+     AGE: 5h28m
+
+    NAME: repldepl-7bcd8d5b64-q2bf4  
+   READY: 1/1    
+  STATUS: Running  
+RESTARTS: 1 (71m ago)  
+     AGE: 5h28m
+```
+
+Tablize can read one or more files or - if none specified - from STDIN.
+
+You can also specify a regex pattern to reduce the output:
+
+```
+% kubectl get pods | ./tablizer q2bf4
+NAME(1)                      READY(2) STATUS(3)  RESTARTS(4)    AGE(5)
+repldepl-7bcd8d5b64-q2bf4    1/1      Running    1 (69m ago)    5h26m
+```
+
+
+## Installation
+
+Download the latest release file for your architecture and put it into
+a directory within your `$PATH`.
+
+## Getting help
+
+Although I'm happy to hear from udpxd users in private email,
+that's the best way for me to forget to do something.
+
+In order to report a bug, unexpected behavior, feature requests
+or to submit a patch, please open an issue on github:
+https://github.com/TLINDEN/tablizer/issues.
+
+## Copyright and license
+
+This software is licensed under the GNU GENERAL PUBLIC LICENSE version 3.
+
+## Authors
+
+T.v.Dein <tom AT vondein DOT org>
+
+## Project homepage
+
+https://github.com/TLINDEN/tablizer

TODO

@@ -0,0 +1,4 @@
+Add a mode like FreeBSD stat(1):
+
+stat -s dead.letter 
+st_dev=170671546954750497 st_ino=159667 st_mode=0100644 st_nlink=1 st_uid=1001 st_gid=1001 st_rdev=18446744073709551615 st_size=573 st_atime=1661994007 st_mtime=1661961878 st_ctime=1661961878 st_birthtime=1658394900 st_blksize=4096 st_blocks=3 st_flags=2048

cmd/parser.go

@@ -0,0 +1,152 @@
+package cmd
+
+import (
+	"bufio"
+	"fmt"
+	"io"
+	"os"
+	"regexp"
+	"strings"
+)
+
+// 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", ...]
+	entries        [][]string
+}
+
+func die(v ...interface{}) {
+	fmt.Fprintln(os.Stderr, v...)
+	os.Exit(1)
+}
+
+/*
+   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.
+*/
+func parseFile(input io.Reader, pattern string) Tabdata {
+	data := Tabdata{}
+
+	var scanner *bufio.Scanner
+	var spaces = `\s\s+|$`
+
+	if len(Separator) > 0 {
+		spaces = Separator
+	}
+
+	hadFirst := false
+	spacefinder := regexp.MustCompile(spaces)
+	beg := 0
+
+	scanner = bufio.NewScanner(input)
+
+	for scanner.Scan() {
+		line := scanner.Text()
+		values := []string{}
+
+		patternR, err := regexp.Compile(pattern)
+		if err != nil {
+			die(err)
+		}
+
+		if !hadFirst {
+			// header processing
+			parts := spacefinder.FindAllStringIndex(line, -1)
+			data.columns = len(parts)
+			// if Debug {
+			// 	fmt.Println(parts)
+			// }
+
+			// process all header fields
+			for _, part := range parts {
+				// if Debug {
+				// 	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)
+				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]
+
+				// done
+				hadFirst = true
+			}
+			// if Debug {
+			// 	fmt.Println(data.headerIndices)
+			// }
+		} else {
+			// data processing
+			if len(pattern) > 0 {
+				//fmt.Println(patternR.MatchString(line))
+				if !patternR.MatchString(line) {
+					continue
+				}
+			}
+
+			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))
+
+				if len(data.maxwidthPerCol)-1 < idx {
+					data.maxwidthPerCol = append(data.maxwidthPerCol, width)
+				} else {
+					if width > data.maxwidthPerCol[idx] {
+						data.maxwidthPerCol[idx] = width
+					}
+				}
+
+				// if Debug {
+				// 	fmt.Printf("<%s> ", value)
+				// }
+				values = append(values, value)
+
+				idx++
+			}
+			if Debug {
+				fmt.Println()
+			}
+			data.entries = append(data.entries, values)
+		}
+	}
+
+	if scanner.Err() != nil {
+		die(scanner.Err())
+	}
+
+	return data
+}

cmd/printer.go

@@ -0,0 +1,109 @@
+package cmd
+
+import (
+	"fmt"
+	"strings"
+)
+
+func printTable(data Tabdata) {
+	if XtendedOut {
+		printExtended(data)
+		return
+	}
+
+	// needed for data output
+	var formats []string
+
+	if len(data.entries) > 0 {
+		// headers
+		for i, head := range data.headers {
+			if len(Columns) > 0 {
+				if !contains(UseColumns, i+1) {
+					continue
+				}
+			}
+
+			// calculate column width
+			var width int
+			var iwidth int
+			var format string
+
+			// generate format string
+			if len(head) > data.maxwidthPerCol[i] {
+				width = len(head)
+			} else {
+				width = data.maxwidthPerCol[i]
+			}
+
+			if NoNumbering {
+				iwidth = 0
+			} else {
+				iwidth = len(fmt.Sprintf("%d", i)) // in case i > 9
+			}
+
+			format = fmt.Sprintf("%%-%ds", 3+iwidth+width)
+
+			if NoNumbering {
+				fmt.Printf(format, fmt.Sprintf("%s ", head))
+			} else {
+				fmt.Printf(format, fmt.Sprintf("%s(%d) ", head, i+1))
+			}
+
+			// register
+			formats = append(formats, format)
+		}
+		fmt.Println()
+
+		// entries
+		var idx int
+		for _, entry := range data.entries {
+			idx = 0
+			//fmt.Println(entry)
+			for i, value := range entry {
+				if len(Columns) > 0 {
+					if !contains(UseColumns, i+1) {
+						continue
+					}
+				}
+				fmt.Printf(formats[idx], strings.TrimSpace(value))
+				idx++
+			}
+			fmt.Println()
+		}
+	}
+}
+
+/*
+   We simulate the \x command of psql (the PostgreSQL client)
+*/
+func printExtended(data Tabdata) {
+	// needed for data output
+	format := fmt.Sprintf("%%%ds: %%s\n", data.maxwidthHeader) // FIXME: re-calculate if -c has been set
+
+	if len(data.entries) > 0 {
+		var idx int
+		for _, entry := range data.entries {
+			idx = 0
+			for i, value := range entry {
+				if len(Columns) > 0 {
+					if !contains(UseColumns, i+1) {
+						continue
+					}
+				}
+
+				fmt.Printf(format, data.headers[idx], value)
+				idx++
+			}
+			fmt.Println()
+		}
+	}
+}
+
+func contains(s []int, e int) bool {
+	for _, a := range s {
+		if a == e {
+			return true
+		}
+	}
+	return false
+}