Updated README and TODO

The program doesn't seem to work on Windows, and I don't have the time
right now to debug it. So, at the moment, the program isn't supported on Windows.

.gitignore

@@ -1,2 +1,3 @@
 ccat
+*.zip
 

README.md

@@ -0,0 +1,73 @@
+## ccat
+
+ccat is a file printing tool (like 'cat') which uses Regular Expressions to enable syntax highlighting.
+
+---
+
+### Features
+- 11 colors are defined out-of-the-box: RED, BLUE, GREEN, MAGENTA, CYAN, BLACK, WHITE, YELLOW, GRAY, ORANGE and DARKBLUE.
+- Support for defining custom colors via the `ccat.colors` file.
+- Regex-color mappings are stored in configuration files.
+- Uses the file extension to determine which configuration file to use.
+- Highly extensible - to add a config file for an specific file type, name the file `<extension>.conf`.
+- Support for printing line numbers with the `-n` flag.
+- Statically linked Go binary - no runtime dependencies, config files are distributed along with the binary.
+- Linux and MacOS supported.
+
+---
+
+### Installing
+
+Download the appropriate zip-file from the 'Releases' section. Place the executable in your PATH.
+
+NOTE: The releases are not available on the GitHub repo (which is a mirror of https://gitea.twomorecents.org/Rockingcool/ccat). Obtain the [releases](https://gitea.twomorecents.org/Rockingcool/ccat/releases) from there instead.
+
+---
+
+### Building from source
+
+If you have the `go` command installed, run `make` after cloning the repository.
+
+---
+
+### Supported Languages
+
+The following languages have config files included by default:
+
+- C
+- Go
+
+---
+
+### Getting Started
+The config files are embedded within the binary. They will automatically be installed to the correct location (`~/.config/ccat` on UNIX) when the program is first run.
+
+As written above, if provided a file with extension `.example`, the program will look for the config file named `example.conf`. If such a file doesn't exist, the file is printed out without any highlighting.
+
+For example, if you want to create syntax highlighting for Java, create a file named `java.conf` in your config directory. In this file, include regular-expressions for each of the langauges's keywords, and provide a corresponding color. Use the provided `c.conf` and `go.conf` files as a starting point.
+
+---
+
+### Config Files
+
+The config files are written in YAML. Each line has the following syntax:
+
+`"<regex>": COLOR`
+
+Note that the regex must be enclosed in double quotes, and the color must be capitalized.
+
+---
+
+### Custom Colors
+
+To define a color of your own, create a file named `ccat.colors` in the config directory (mentioned above). The syntax of this file is the following:
+
+`COLOR: <red> <green> <blue>`
+
+Note that the color name must be capitalized (and shouldn't contain spaces). The RGB values must each be from 0 to 255.
+
+---
+
+### TODO:
+- Windows support.
+- Allow users to provide a config file in the command-line, overriding the extension-based config file.

color.go

@@ -2,8 +2,12 @@ package main
 
 import (
 	"fmt"
+	"os"
+	"strconv"
+	"strings"
 
 	colorData "github.com/fatih/color"
+	"gopkg.in/yaml.v2"
 )
 
 // A color represents a possible color, which text can be printed out in.
@@ -14,6 +18,14 @@ type color struct {
 	colorObj *colorData.Color
 }
 
+// A RGB represents a Red, Blue, Green trio of values. Each value is represented as
+// an int.
+type RGB struct {
+	red   int
+	blue  int
+	green int
+}
+
 // The following is a list of all possible colors, stored in a map.
 var possibleColors map[string]color = map[string]color{
 	"BLACK":   {"BLACK", colorData.New(colorData.FgBlack)},
@@ -27,6 +39,7 @@ var possibleColors map[string]color = map[string]color{
 	"GRAY":    {"GRAY", colorData.New(colorData.FgWhite, colorData.Faint)},
 	// Last three numbers are RGB. Reference https://en.wikipedia.org/wiki/ANSI_escape_code for what the first two numbers mean.
 	"ORANGE":   {"ORANGE", colorData.New(38, 2, 255, 153, 28)},
+	"DARKBLUE": {"DARKBLUE", colorData.New(38, 2, 0, 112, 255)},
 	"NONE":     {"NONE", colorData.New()},
 }
 
@@ -52,9 +65,95 @@ func newColor(colorString string) (color, error) {
 // newColorMust is similar to newColor, but prints an error and exits if the given color isn't valid.
 func newColorMust(colorString string) color {
 	if clr, err := newColor(colorString); err != nil {
-		printAndExit(err.Error())
+		printErrAndExit(err.Error())
 		panic(err) // NEVER REACHED
 	} else {
 		return clr
 	}
 }
+
+// isValidColorName returns true if the given string only contains uppercase alphabetic
+// characters.
+func isValidColorName(colorName string) bool {
+	for _, ch := range colorName {
+		if ch > 'Z' || ch < 'A' {
+			return false
+		}
+	}
+	return true
+}
+
+// stringToRGB takes a string representing an RGB trio. It constructs and RGB type and
+// returns it. Any errors encountered are returned. If an error is returned, it is safe to
+// assume that the string doesn't represent an RGB trio.
+func stringToRGB(rgbString string) (*RGB, error) {
+	values := strings.Split(rgbString, " ")
+	// There must be three space-separated strings.
+	if len(values) != 3 {
+		// TODO: Instead of ignoring these errors and returning a generic error (as I do in the
+		// callee), wrap the error returned from this function, inside the error returned by the callee.
+		return nil, fmt.Errorf("Error parsing RGB trio.")
+	}
+	// If any of the strings doesn't represent an integer (or is out of bounds), return an error.
+	// WARNING: LAZY CODE INCOMING
+	var toReturn RGB
+	var err error
+	toReturn.red, err = strconv.Atoi(values[0])
+	if err != nil {
+		return nil, fmt.Errorf("Error parsing RED integer: Invalid value.")
+	}
+	if toReturn.red < 0 || toReturn.red > 255 {
+		return nil, fmt.Errorf("Error parsing RED integer: Out-of-bounds.")
+	}
+	toReturn.blue, err = strconv.Atoi(values[1])
+	if err != nil {
+		return nil, fmt.Errorf("Error parsing BLUE integer: Invalid value.")
+	}
+	if toReturn.blue < 0 || toReturn.blue > 255 {
+		return nil, fmt.Errorf("Error parsing BLUE integer: Out-of-bounds.")
+	}
+	toReturn.green, err = strconv.Atoi(values[2])
+	if err != nil {
+		return nil, fmt.Errorf("Error parsing GREEN integer: Invalid value.")
+	}
+	if toReturn.green < 0 || toReturn.green > 255 {
+		return nil, fmt.Errorf("Error parsing GREEN integer: Out-of-bounds.")
+	}
+	return &toReturn, nil
+}
+
+// loadColorsFromFile loads the colors defined in the given config file, and adds them to
+// the possibleColors map. This allows the user to define custom colors at run-time.
+// The colors config file has the following syntax:
+// COLOR: <RED> <GREEN> <BLUE>
+//
+// Note that the color must be capitalized (and not contain spaces), and the R, G and B
+// values must be from 0 to 255.
+func loadColorsFromFile(filepath string) error {
+	data, err := os.ReadFile(filepath)
+	if err != nil {
+		panic(err)
+	}
+	// Read color config file into a MapSlice
+	tempMapSlice := yaml.MapSlice{}
+	if err := yaml.Unmarshal(data, &tempMapSlice); err != nil {
+		return fmt.Errorf("Unable to read color config file: %s", filepath)
+	}
+
+	for _, item := range tempMapSlice {
+		if !(isValidColorName(item.Key.(string))) {
+			return fmt.Errorf("Invalid color name: %s", item.Key.(string))
+		}
+		var rgb *RGB
+		if rgb, err = stringToRGB(item.Value.(string)); err != nil {
+			return fmt.Errorf("Invalid RGB trio: %s", item.Value.(string))
+		}
+		// If we haven't returned an error yet, the color must be valid.
+		// Add it to the map. colorData.New() expects values of type colorData.Attribute,
+		// so we must cast our RGB values accordingly.
+		possibleColors[item.Key.(string)] = color{item.Key.(string), colorData.New(38, 2, colorData.Attribute(rgb.red), colorData.Attribute(rgb.blue), colorData.Attribute(rgb.green))}
+
+	}
+
+	return nil
+}

config.go

@@ -1,14 +1,67 @@
 package main
 
 import (
-	"ccat/stack"
+	"embed"
+	"errors"
+	"gitea.twomorecents.org/Rockingcool/ccat/stack"
+	"io/fs"
 	"os"
+	"path/filepath"
 	"regexp"
+	"runtime"
 	"strings"
 
 	"gopkg.in/yaml.v2"
 )
 
+//go:embed config
+var storedConfigs embed.FS // Embed the folder containing config files
+
+// runningOnWindows: At the moment this function isn't used. When Window support is added,
+// it will be used to determine if the program is being run on Windows.
+func runningOnWindows() bool {
+	return runtime.GOOS == "windows"
+}
+
+// generateDefaultConfigs is used to generate a folder of default config files
+// for common languages. These default config files are embedded into the program, and will
+// be outputted into the given directory.
+//
+// If there is an error encountered, the error is returned.
+func generateDefaultConfigs(configOutputPath string) error {
+	err := os.MkdirAll(configOutputPath, 0755)
+	if err != nil {
+		if os.IsExist(err) {
+			return errors.New("Directory already exists.")
+		} else {
+			return errors.New("Unable to create directory.")
+		}
+	}
+
+	// Copy each folder from the embedded filesystem, into the destination path
+	err = fs.WalkDir(storedConfigs, "config", func(path string, d fs.DirEntry, err error) error {
+		if err != nil {
+			return err
+		}
+		if d.IsDir() { // Skip directories
+			return nil
+		}
+		relPath, _ := filepath.Rel("config", path)
+		dstPath := filepath.Join(configOutputPath, relPath) // Destination path
+
+		data, err := os.ReadFile(path)
+		if err != nil {
+			return err
+		}
+
+		if err := os.WriteFile(dstPath, data, 0644); err != nil {
+			return err
+		}
+		return nil
+	})
+	return nil
+}
+
 // loadConfig takes in the filename of a config file. It reads the file,
 // and returns a stack of RegColors, with the item at the bottom being the one that
 // was read first. This ensures that, _when accessing the RegColors in the stack, the last

config/go.conf

@@ -0,0 +1,30 @@
+# Priority decreases going downward ie. If two regexes match the same piece of
+# text, the one defined earlier will take precedence over the one defined later.
+# Comments
+'//.*': GRAY
+'/\*[^*]*\*+(?:[^/*][^*]*\*+)*/': GRAY
+# Numbers and special values
+'\b\-?[0-9]*\b': MAGENTA
+'\b(true|false)\b': MAGENTA
+'\b(nil)\b': MAGENTA
+# Strings in double quotes and backticks
+'"(.*?)"': BLUE
+'`(.*?)`': BLUE
+# Bytes / Runes
+"'(.)'": BLUE
+"'\\\\(.)'": BLUE # The escape backslash needs to be escaped as well
+# Assignments and comparisons
+'(?:\s|\b)(=|==|!=|<=|>=)(\s|\b)' : CYAN
+'(&&)|(\|\|)': CYAN
+# Keywords
+'\b(if|else|for|range|go|func|return|break|continue)\b': CYAN
+'\b(import|var|const|type|struct)\b': CYAN
+# Built-in Functions
+'\b(panic|len)\b': DARKBLUE
+# Functions from packages (package name and function name separated by dot)
+'\b(\w*\.\w*)\b': DARKBLUE
+# Data Types
+'\b(bool|byte|rune|string|interface|map|chan)\b': YELLOW
+'\b(u?int)(8|16|32|64)?\b': YELLOW
+'\b(float)(32|64)\b': YELLOW
+'\b(complex)(64|128)\b': YELLOW

create_release_builds.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+
+set -euo pipefail
+
+POSSIBLE_GOOS=( "linux" "darwin" )
+POSSIBLE_GOARCH=( "amd64" "arm64" )
+
+for OS in "${POSSIBLE_GOOS[@]}"; do
+	for ARCH in "${POSSIBLE_GOARCH[@]}"; do
+		FOLDER_NAME="ccat-$OS-$ARCH"
+		mkdir "${FOLDER_NAME}"
+		GOOS=$OS GOARCH=$ARCH go build -o "${FOLDER_NAME}/"
+		zip -r "${FOLDER_NAME}" "${FOLDER_NAME}"
+		rm -r "${FOLDER_NAME}"
+	done
+done
+

errors.go

@@ -5,7 +5,7 @@ import (
 	"os"
 )
 
-func printAndExit(errorStr string) {
+func printErrAndExit(errorStr string) {
 	fmt.Printf("ERROR: %s\n", errorStr)
 	os.Exit(1)
 }

go.mod

@@ -1,4 +1,4 @@
-module ccat
+module gitea.twomorecents.org/Rockingcool/ccat
 
 go 1.22.5
 

main.go

@@ -1,9 +1,13 @@
 package main
 
 import (
+	"bytes"
 	"errors"
+	"flag"
 	"fmt"
+	"math"
 	"os"
+	"os/user"
 	"path/filepath"
 )
 
@@ -15,7 +19,7 @@ func fileExists(filename string) bool {
 	} else if errors.Is(err, os.ErrNotExist) {
 		return false
 	} else {
-		printAndExit(err.Error())
+		printErrAndExit(err.Error())
 		return false // NEVER REACHED
 	}
 }
@@ -24,7 +28,7 @@ func fileExists(filename string) bool {
 // the file doesn't exist.
 func mustExist(filename string) {
 	if fileExists(filename) != true {
-		printAndExit(os.ErrNotExist.Error())
+		printErrAndExit(os.ErrNotExist.Error())
 	}
 }
 
@@ -32,12 +36,12 @@ func mustExist(filename string) {
 // It returns two values: the first is true if the config file exists.
 // If it does, the second value is the config filename.
 // If it doesn't, the second value is blank and can be ignored.
-func getConfig(extension string) (bool, string) {
+func getConfig(configPath, extension string) (bool, string) {
 	if extension == "" {
 		return false, ""
 	}
 	// Assuming the file has an extension
-	fileName := "config/" + extension[1:] + ".conf"
+	fileName := filepath.Join(configPath, extension[1:]+".conf")
 	if exists := fileExists(fileName); exists == false {
 		return false, ""
 	} else {
@@ -52,41 +56,89 @@ func printFile(fileName string) {
 	mustExist(fileName)
 	data, err := os.ReadFile(fileName)
 	if err != nil {
-		printAndExit(err.Error())
+		printErrAndExit(err.Error())
 	}
 	fmt.Print(string(data))
 	return
 }
 
+// computeLineNumDigits computes the number of digits in the number of lines
+// in the given byte array.
+func computeLineNumDigits(data []byte) int {
+	numLines := bytes.Count(data, []byte{'\n'}) + 1
+	return int(math.Round(math.Log10(float64(numLines))))
+}
+
 func main() {
+	disableColorFlag := flag.Bool("d", false, "Disable color")
+	lineNumberFlag := flag.Bool("n", false, "Print line numbers")
+	// Used only if lineNumberFlag is true
+	var lineNumDigits int
+	var lineNum int
+	flag.Parse()
+
+	// Check if config exists. If it doesn't, generate the config files.
+	currentUser, err := user.Current() // Get current user, to determine config path
+	if err != nil {
+		panic(err)
+	}
+	configPath := filepath.Join("/home/" + currentUser.Username + "/.config/ccat/")
+	if _, err := os.Stat(configPath); os.IsNotExist(err) {
+		generateDefaultConfigs(configPath)
+	}
 
 	// Check if user has provided a file name
-	if len(os.Args) != 2 {
-		printAndExit("Invalid number of arguments")
+	if len(flag.Args()) < 1 {
+		printErrAndExit("No File specified")
 	}
-	fileName := os.Args[1]
+	fileName := flag.Args()[0]
+
+	// Check if file exists.
 	mustExist(fileName)
 
 	extension := filepath.Ext(fileName)
-	configExists, configFilename := getConfig(extension)
-	// If the given file has no corresponding config, print it out
-	// and exit.
+	configExists, configFilename := getConfig(configPath, extension)
+	// If the given file has no corresponding config, print the file out and exit.
 	if configExists == false {
 		printFile(fileName)
 		return
 	}
+
+	// To save computing time, determine here if the file is empty. If it is, exit
+	// the program.
+	finfo, err := os.Stat(fileName)
+	if err != nil {
+		panic(err)
+	}
+	if finfo.Size() == 0 {
+		os.Exit(0)
+	}
+
+	// Assuming the file is not empty...
+	// If a ccat.colors file exists in the config directory, load all the colors in it
+	if fileExists(filepath.Join(configPath, "ccat.colors")) {
+		err := loadColorsFromFile(filepath.Join(configPath, "ccat.colors"))
+		if err != nil {
+			printErrAndExit(err.Error())
+		}
+	}
 	// If the given file has a config, load the config into a stack of regColors.
 	regColorStack, err := loadConfig(configFilename)
 	if err != nil {
-		printAndExit(err.Error())
+		printErrAndExit(err.Error())
 	}
 
 	// Load the input file into a colorunit slice (units) and a byte slice (data)
 	units, data, err := loadInputFile(fileName)
 	if err != nil {
-		printAndExit(err.Error())
+		printErrAndExit(err.Error())
 	}
 
+	// If the '-n' flag is set, compute the number of digits in the number of lines
+	// in the file, to determine the padding for the line numbers.
+	if *lineNumberFlag {
+		lineNumDigits = computeLineNumDigits(data)
+	}
 	// For each regular expression in the stack, apply it to the byte slice. Find
 	// the first and last index of all matches of the regex. Then apply the corresponding color
 	// to every character within these indices.
@@ -94,7 +146,9 @@ func main() {
 	// The infinite for loop exists, because I couldn't figure out a way to pop an element from
 	// the stack inside the 'for' statement. The loop exits when the 'pop' call returns 'false',
 	// indicating that the stack is empty.
-	for {
+	//
+	// The loop is also only run if the 'disable color' flag is not set.
+	for *disableColorFlag == false {
 		regclr, ok := regColorStack.Pop()
 		// regColorStack.Pop() returns false when there are no more elements to pop
 		if ok != true {
@@ -115,7 +169,21 @@ func main() {
 	}
 
 	// After all possible regexes have been matched, print out the contents of 'units'.
-	for _, unit := range units {
+
+	// If the line number flag is set, initialize the lineNum variable and print the first line number
+	// with the appropriate padding.
+	if *lineNumberFlag {
+		lineNum = 1
+		fmt.Printf("   %*d  ", lineNumDigits, lineNum)
+		lineNum++
+	}
+	for idx, unit := range units {
 		unit.print()
+		// If the flag is set and we encounter a newline (and the newline isn't a trailing newline),
+		// then print the next line number.
+		if *lineNumberFlag && unit.ch == '\n' && idx != len(units)-1 {
+			fmt.Printf("   %*d  ", lineNumDigits, lineNum)
+			lineNum++
+		}
 	}
 }

test.c

@@ -1,173 +0,0 @@
-//go:build exclude
-#include "easysock.h"
-
-
-int create_socket(int network, char transport) {
-	int domain;
-	int type;
-
-	if (network == 4) {
-		domain = AF_INET;
-	} else if (network == 6) {
-		domain = AF_INET6;
-	} else {
-		return -1;
-	}
-
-	if (transport == 'T') {
-		type = SOCK_STREAM;
-	} else if (transport == 'U') {
-		type = SOCK_DGRAM;
-	} else {
-		return -1;
-	}
-
-	int newSock = socket(domain,type,0);
-	return newSock;
-}
-
-
-int create_addr(int network, char* address, int port,struct sockaddr* dest) {
-	if (network == 4) {
-		struct sockaddr_in listen_address;
-
-		listen_address.sin_family = AF_INET;
-		listen_address.sin_port = htons(port);
-		inet_pton(AF_INET,address,&listen_address.sin_addr);
-		memcpy(dest,&listen_address,sizeof(listen_address));
-		return 0;
-
-	} else if (network == 6) {
-		struct sockaddr_in6 listen_ipv6;
-		listen_ipv6.sin6_family = AF_INET6;
-		listen_ipv6.sin6_port = htons(port);
-		inet_pton(AF_INET6,address,&listen_ipv6.sin6_addr);
-		memcpy(dest,&listen_ipv6,sizeof(listen_ipv6));
-		return 0;
-
-	} else {
-		return -202;
-	}
-
-
-
-}
-
-int create_local (int network, char transport, char* address, int port,struct sockaddr* addr_struct) {
-	int socket = create_socket(network,transport);
-	if (socket < 0) {
-		return (-1 * errno);
-	}
-	create_addr(network,address,port,addr_struct);
-	int addrlen;
-	if (network == 4) {
-		addrlen = sizeof(struct sockaddr_in);
-	} else if (network == 6) {
-		addrlen = sizeof(struct sockaddr_in6);
-	} else {
-		return -202;
-	}
-
-	/* The value of addrlen should be the size of the 'sockaddr'.
-	This should be set to the size of 'sockaddr_in' for IPv4, and 'sockaddr_in6' for IPv6.
-	See https://stackoverflow.com/questions/73707162/socket-bind-failed-with-invalid-argument-error-for-program-running-on-macos */
-
-	int i = bind (socket,addr_struct,(socklen_t)addrlen);
-	if (i < 0) {
-		return (-1 * errno);
-	}
-	return socket;
-}
-
-int create_remote (int network,char transport,char* address,int port,struct sockaddr* remote_addr_struct) {
-
-	struct addrinfo hints;  /* Used to tell getaddrinfo what kind of address we want */
-	struct addrinfo* results; /* Used by getaddrinfo to store the addresses */
-
-
-	if (check_ip_ver(address) < 0) { /* If the address is a domain name */
-		int err_code;
-		char* port_str = malloc(10 * sizeof(char));
-
-		sprintf(port_str,"%d",port); /* getaddrinfo expects a string for its port */
-
-
-		memset(&hints,'\0',sizeof(hints));
-		hints.ai_socktype = char_to_socktype(transport);
-
-		err_code = getaddrinfo(address,port_str,&hints,&results);
-		if (err_code != 0) {
-			return (-1 * err_code);
-		}
-		remote_addr_struct = results->ai_addr;
-		network = inet_to_int(results->ai_family);
-	} else {
-	        create_addr(network,address,port,remote_addr_struct);
-	}
-
-	int socket = create_socket(network,transport);
-	if (socket < 0) {
-                return (-1 * errno);
-        }
-
-       	int addrlen;
-	if (network == 4) {
-		addrlen = sizeof(struct sockaddr_in);
-	} else if (network == 6) {
-		addrlen = sizeof(struct sockaddr_in6);
-	} else {
-		return (-202);
-	}
-
-	/* The value of addrlen should be the size of the 'sockaddr'.
-	This should be set to the size of 'sockaddr_in' for IPv4, and 'sockaddr_in6' for IPv6.
-	See https://stackoverflow.com/questions/73707162/socket-bind-failed-with-invalid-argument-error-for-program-running-on-macos */
-
-        int i = connect(socket,remote_addr_struct,(socklen_t)addrlen);
-	if (i < 0) {
-		return (-1 * errno);
-	}
-        return socket;
-}
-
-
-int check_ip_ver(char* address) {
-	char buffer[16]; /* 16 chars - 128 bits - is enough to hold an ipv6 address */
-	if (inet_pton(AF_INET,address,buffer) == 1) {
-		return 4;
-	} else if (inet_pton(AF_INET6,address,buffer) == 1) {
-		return 6;
-	} else {
-		return -1;
-	}
-}
-
-int int_to_inet(int network) {
-	if (network == 4) {
-		return AF_INET;
-	} else if (network == 6) {
-		return AF_INET6;
-	} else {
-		return -202;
-	}
-}
-
-int inet_to_int(int af_type) {
-	if (af_type == AF_INET) {
-		return 4;
-	} else if (af_type == AF_INET6) {
-		return 6;
-	} else {
-		return -207;
-	}
-}
-
-int char_to_socktype(char transport) {
-	if (transport == 'T') {
-		return SOCK_STREAM;
-	} else if (transport == 'U') {
-		return SOCK_DGRAM;
-	} else {
-		return -250;
-	}
-}