testdoc: Add ```

Makefile

@@ -1,12 +1,17 @@
 # libsmu - simple markup
 # (c) 2007, 2008 Enno Boland
 
-include config.mk
-
+VERSION = 1.5
+PREFIX ?= /usr/local
+MANPREFIX ?= ${PREFIX}/share/man
+CFLAGS += -DVERSION=\"${VERSION}\"
+PGO_GEN = ${CFLAGS} -fprofile-generate
+PGO_USE = ${CFLAGS} -fprofile-use -fprofile-correction
+LDFLAGS ?=
+CC ?= cc
 SRC = smu.c
-OBJ    = ${SRC:.c=.o}
 
-all: options smu
+all: smu
 
 options:
 	@echo smu build options:
@@ -14,42 +19,51 @@ options:
 	@echo "LDFLAGS  = ${LDFLAGS}"
 	@echo "CC       = ${CC}"
 
-.c.o:
-	@echo CC $<
-	@${CC} -c ${CFLAGS} $<
+smu: ${SRC}
+	@echo CC $@
+	@${CC} -o $@ ${SRC} ${CFLAGS} ${LDFLAGS}
 
-${OBJ}: config.mk
+pgo-build: ${SRC}
+	@echo CC smu
+	@${CC} -o smu ${SRC} ${PGO_GEN} ${LDFLAGS}
+	@echo Generating pgo data
+	@printf .
+	@./smu testdoc > /dev/null
+	@printf .
+	@cat testdoc | ./smu > /dev/null
+	@printf .
+	@./smu -n testdoc > /dev/null
+	@printf .
+	@cat testdoc | ./smu -n > /dev/null
+	@printf .
+	@rm smu
+	@printf '\n'
+	@echo Recompiling with pgo data
+	@echo CC smu
+	@${CC} -o smu ${SRC} ${PGO_USE} ${LDFLAGS}
+	@rm smu.gcda
 
-smu: ${OBJ}
-	@echo LD $@
-	@${CC} -o $@ ${OBJ} ${LDFLAGS}
 
 clean:
-	@echo cleaning
-	@rm -f smu ${OBJ} ${LIBOBJ} smu-${VERSION}.tar.gz
+	rm -f smu ${OBJ} ${LIBOBJ} smu-${VERSION}.tar.gz
 
 dist: clean
-	@echo creating dist tarball
-	@mkdir -p smu-${VERSION}
-	@cp -R LICENSE Makefile config.mk smu.1 ${SRC} smu-${VERSION}
-	@tar -cf smu-${VERSION}.tar smu-${VERSION}
-	@gzip smu-${VERSION}.tar
-	@rm -rf smu-${VERSION}
+	mkdir -p smu-${VERSION}
+	cp -R LICENSE Makefile config.mk smu.1 ${SRC} smu-${VERSION}
+	tar -cf smu-${VERSION}.tar smu-${VERSION}
+	gzip smu-${VERSION}.tar
+	rm -rf smu-${VERSION}
 
 install: all
-	@echo installing executable file to ${DESTDIR}${PREFIX}/bin
-	@mkdir -p ${DESTDIR}${PREFIX}/bin
-	@cp -f smu ${DESTDIR}${PREFIX}/bin
-	@chmod 755 ${DESTDIR}${PREFIX}/bin/smu
-	@echo installing manual page to ${DESTDIR}${MANPREFIX}/man1
-	@mkdir -p ${DESTDIR}${MANPREFIX}/man1
-	@sed "s/VERSION/${VERSION}/g" < smu.1 > ${DESTDIR}${MANPREFIX}/man1/smu.1
-	@chmod 644 ${DESTDIR}${MANPREFIX}/man1/smu.1
+	mkdir -p ${DESTDIR}${PREFIX}/bin
+	cp -f smu ${DESTDIR}${PREFIX}/bin
+	chmod 755 ${DESTDIR}${PREFIX}/bin/smu
+	mkdir -p ${DESTDIR}${MANPREFIX}/man1
+	sed "s/VERSION/${VERSION}/g" < smu.1 > ${DESTDIR}${MANPREFIX}/man1/smu.1
+	chmod 644 ${DESTDIR}${MANPREFIX}/man1/smu.1
 
 uninstall:
-	@echo removing executable file from ${DESTDIR}${PREFIX}/bin
-	@rm -f ${DESTDIR}${PREFIX}/bin/smu
-	@echo removing manual page from ${DESTDIR}${MANPREFIX}/man1
-	@rm -f ${DESTDIR}${MANPREFIX}/man1/smu.1
+	rm -f ${DESTDIR}${PREFIX}/bin/smu
+	rm -f ${DESTDIR}${MANPREFIX}/man1/smu.1
 
-.PHONY: all options clean dist install uninstall
+.PHONY: all options clean dist install uninstall pgo-build

config.mk

@@ -1,20 +0,0 @@
-# smu version
-VERSION = 1.5
-
-# paths
-PREFIX = /usr/local
-MANPREFIX = ${PREFIX}/share/man
-
-# includes and libs
-INCS = -I. -I/usr/include
-LIBS = -L/usr/lib
-
-# flags
-CFLAGS = -g -O0 -Wall -Werror -ansi ${INCS} -DVERSION=\"${VERSION}\"
-#CFLAGS = -fprofile-arcs -ftest-coverage -pg -g -O0 -Wall -Werror -ansi ${INCS} -DVERSION=\"${VERSION}\"
-#CFLAGS = -Os -Wall -Werror -ansi ${INCS} -DVERSION=\"${VERSION}\"
-#LDFLAGS = -fprofile-arcs -ftest-coverage -pg ${LIBS}
-LDFLAGS = ${LIBS}
-
-# compiler
-CC = cc

smu.1

@@ -20,4 +20,4 @@ prints usage information to standard error, then exits.
 .B \-n
 escapes all HTML Tags.
 .SH BUGS
-Please report any Bugs to https://github.com/Gottox/smu/issues or via mail.
+Report any bugs to tenno+smu@suij.in

smu.c

@@ -1,48 +1,78 @@
 /* smu - simple markup
  * Copyright (C) <2007, 2008> Enno Boland <g s01 de>
+ * Copyright (C) 2025 Enno Tensing <tenno+smu@suij.in>
  *
  * See LICENSE for further informations
  */
+#define _LARGEFILE64_SOURCE
 #include <stdarg.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
 #include <ctype.h>
+#include <unistd.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <fcntl.h>
+#include <errno.h>
 
-#define LENGTH(x)  sizeof(x)/sizeof(x[0])
-#define ADDC(b,i)  if(i % BUFSIZ == 0) { b = realloc(b, (i + BUFSIZ) * sizeof(char)); if(!b) eprint("Malloc failed."); } b[i]
+#ifndef PACKAGE
+#define PACKAGE "smu"
+#endif
+
+#ifndef VERSION
+#define VERSION "0.0"
+#endif
+
+#define CHARWIDTH 4
+#define LENGTH(x) sizeof(x) / sizeof(x[0])
+#define ADDC(b, i, a)                                                \
+	do {                                                         \
+		if (i % BUFSIZ == 0) {                               \
+			b = realloc(b, (i + BUFSIZ) * sizeof(char)); \
+			if (!b) {                                    \
+				eprint("Malloc failed.");            \
+				return -1;                           \
+			}                                            \
+		}                                                    \
+		b[i] = a;                                            \
+	} while (0)
 
 typedef int (*Parser)(const char *, const char *, int);
-typedef struct {
+struct tag {
 	char *search;
 	int process;
-	char *before, *after;
-} Tag;
+	char *before;
+	char *after;
+};
 
-static int doamp(const char *begin, const char *end, int newblock);       /* Parser for & */
-static int docomment(const char *begin, const char *end, int newblock);   /* Parser for html-comments */
-static int dogtlt(const char *begin, const char *end, int newblock);      /* Parser for < and > */
-static int dohtml(const char *begin, const char *end, int newblock);      /* Parser for html */
-static int dolineprefix(const char *begin, const char *end, int newblock);/* Parser for line prefix tags */
-static int dolink(const char *begin, const char *end, int newblock);      /* Parser for links and images */
-static int dolist(const char *begin, const char *end, int newblock);      /* Parser for lists */
-static int doparagraph(const char *begin, const char *end, int newblock); /* Parser for paragraphs */
-static int doreplace(const char *begin, const char *end, int newblock);   /* Parser for simple replaces */
-static int doshortlink(const char *begin, const char *end, int newblock); /* Parser for links and images */
-static int dosurround(const char *begin, const char *end, int newblock);  /* Parser for surrounding tags */
-static int dounderline(const char *begin, const char *end, int newblock); /* Parser for underline tags */
+off64_t get_file_size(const char *);
+char *read_file(const char *, off64_t);
+
+static int doamp(const char *begin, const char *end, int newblock);
+static int docomment(const char *begin, const char *end, int newblock);
+static int dogtlt(const char *begin, const char *end, int newblock);
+static int dohtml(const char *begin, const char *end, int newblock);
+static int dolineprefix(const char *begin, const char *end, int newblock);
+static int dolink(const char *begin, const char *end, int newblock);
+static int dolist(const char *begin, const char *end, int newblock);
+static int doparagraph(const char *begin, const char *end, int newblock);
+static int doreplace(const char *begin, const char *end, int newblock);
+static int doshortlink(const char *begin, const char *end, int newblock);
+static int dosurround(const char *begin, const char *end, int newblock);
+static int dounderline(const char *begin, const char *end, int newblock);
 static void *ereallocz(void *p, size_t size);
 static void eprint(const char *format, ...);
-static void hprint(const char *begin, const char *end);                   /* escapes HTML and prints it to output */
-static void process(const char *begin, const char *end, int isblock);     /* Processes range between begin and end. */
+static void hprint(const char *begin, const char *end);
+static int process(const char *begin, const char *end, int isblock);
 
 /* list of parsers */
-static Parser parsers[] = { dounderline, docomment, dolineprefix,
-                            dolist, doparagraph, dogtlt, dosurround, dolink,
+static Parser parsers[] = { dounderline, docomment, dolineprefix, dolist,
+			    doparagraph, dogtlt,    dosurround,	  dolink,
 			    doshortlink, dohtml,    doamp,	  doreplace };
 static int nohtml = 0;
 
-static Tag lineprefix[] = {
+static struct tag lineprefix[] = {
 	{ "    ", 0, "<pre><code>", "\n</code></pre>" },
 	{ "\t", 0, "<pre><code>", "\n</code></pre>" },
 	{ ">", 2, "<blockquote>", "</blockquote>" },
@@ -52,15 +82,16 @@ static Tag lineprefix[] = {
 	{ "### ", 1, "<h3>", "</h3>" },
 	{ "## ", 1, "<h2>", "</h2>" },
 	{ "# ", 1, "<h1>", "</h1>" },
-	{ "- - -\n",	1,	"<hr />",	""},
+	{ "- - -\n", 1, "<hr />", "" },
 };
 
-static Tag underline[] = {
+static struct tag underline[] = {
 	{ "=", 1, "<h1>", "</h1>\n" },
 	{ "-", 1, "<h2>", "</h2>\n" },
 };
 
-static Tag surround[] = {
+static struct tag surround[] = {
+	{ "```", 0, "<code>", "</code>" },
 	{ "``", 0, "<code>", "</code>" },
 	{ "`", 0, "<code>", "</code>" },
 	{ "___", 1, "<strong><em>", "</em></strong>" },
@@ -72,174 +103,203 @@ static Tag surround[] = {
 };
 
 static const char *replace[][2] = {
-	{ "\\\\",	"\\" },
-	{ "\\`",	"`" },
-	{ "\\*",	"*" },
-	{ "\\_",	"_" },
-	{ "\\{",	"{" },
-	{ "\\}",	"}" },
-	{ "\\[",	"[" },
-	{ "\\]",	"]" },
-	{ "\\(",	"(" },
-	{ "\\)",	")" },
-	{ "\\#",	"#" },
-	{ "\\+",	"+" },
-	{ "\\-",	"-" },
-	{ "\\.",	"." },
-	{ "\\!",	"!" },
+	{ "\\\\", "\\" }, { "\\`", "`" }, { "\\*", "*" }, { "\\_", "_" },
+	{ "\\{", "{" },	  { "\\}", "}" }, { "\\[", "[" }, { "\\]", "]" },
+	{ "\\(", "(" },	  { "\\)", ")" }, { "\\#", "#" }, { "\\+", "+" },
+	{ "\\-", "-" },	  { "\\.", "." }, { "\\!", "!" },
 };
 
 static const char *insert[][2] = {
 	{ "  \n", "<br />" },
 };
 
-void
-eprint(const char *format, ...) {
+off64_t get_file_size(const char *path)
+{
+	struct stat st;
+
+	if (stat(path, &st) == 0)
+		return st.st_size;
+
+	return -1;
+}
+
+char *read_file(const char *path, off64_t file_size)
+{
+	int fd = open(path, O_LARGEFILE | O_NONBLOCK);
+
+	ssize_t bytes;
+	char *buf = calloc(file_size + CHARWIDTH, sizeof(char));
+
+	if (!buf) {
+		perror(PACKAGE);
+		close(fd);
+		return NULL;
+	}
+
+	bytes = read(fd, buf, file_size);
+	if (bytes != file_size) {
+		perror(PACKAGE);
+		close(fd);
+		free(buf);
+		return NULL;
+	}
+
+	close(fd);
+
+	return buf;
+}
+
+void eprint(const char *format, ...)
+{
 	va_list ap;
 
 	va_start(ap, format);
 	vfprintf(stderr, format, ap);
 	va_end(ap);
-	exit(EXIT_FAILURE);
 }
 
-int
-doamp(const char *begin, const char *end, int newblock) {
+int doamp(const char *begin, const char *end, int newblock)
+{
 	const char *p;
 
-	if(*begin != '&')
+	if (*begin != '&')
 		return 0;
-	if(!nohtml) {
-		for(p = begin + 1; p != end && !strchr("; \\\n\t", *p); p++);
-		if(p == end || *p == ';')
+	if (!nohtml) {
+		for (p = begin + 1; p != end && !strchr("; \\\n\t", *p); p++)
+			;
+		if (p == end || *p == ';')
 			return 0;
 	}
 	fputs("&amp;", stdout);
 	return 1;
 }
 
-int
-dogtlt(const char *begin, const char *end, int newblock) {
+int dogtlt(const char *begin, const char *end, int newblock)
+{
 	int brpos;
 	char c;
 
-	if(nohtml || begin + 1 >= end)
+	if (nohtml || begin + 1 >= end)
 		return 0;
 	brpos = begin[1] == '>';
-	if(!brpos && *begin != '<')
+	if (!brpos && *begin != '<')
 		return 0;
 	c = begin[brpos ? 0 : 1];
-	if(!brpos && (c < 'a' || c > 'z') && (c < 'A' || c > 'Z')) {
+	if (!brpos && (c < 'a' || c > 'z') && (c < 'A' || c > 'Z')) {
 		fputs("&lt;", stdout);
 		return 1;
-	}
-	else if(brpos && (c < 'a' || c > 'z') && (c < 'A' || c > 'Z') && !strchr("/\"'",c)) {
-		fprintf(stdout, "%c&gt;",c);
+	} else if (brpos && (c < 'a' || c > 'z') && (c < 'A' || c > 'Z') &&
+		   !strchr("/\"'", c)) {
+		fprintf(stdout, "%c&gt;", c);
 		return 2;
 	}
 	return 0;
 }
 
-int
-docomment(const char *begin, const char *end, int newblock) {
+int docomment(const char *begin, const char *end, int newblock)
+{
 	char *p;
 
-	if(nohtml || strncmp("<!--", begin, 4))
+	if (nohtml || strncmp("<!--", begin, 4))
 		return 0;
 	p = strstr(begin, "-->");
-	if(!p || p + 3 >= end)
+	if (!p || p + 3 >= end)
 		return 0;
 	fprintf(stdout, "%.*s\n", (int)(p + 3 - begin), begin);
 	return (p + 3 - begin) * (newblock ? -1 : 1);
 }
 
-int
-dohtml(const char *begin, const char *end, int newblock) {
+int dohtml(const char *begin, const char *end, int newblock)
+{
 	const char *p, *tag, *tagend;
 
-	if(nohtml || begin + 2 >= end)
+	if (nohtml || begin + 2 >= end)
 		return 0;
 	p = begin;
-	if(p[0] != '<' || !isalpha(p[1]))
+	if (p[0] != '<' || !isalpha(p[1]))
 		return 0;
 	p++;
 	tag = p;
-	for(; isalnum(*p) && p < end; p++);
+	for (; isalnum(*p) && p < end; p++)
+		;
 	tagend = p;
-	if(p > end || tag == tagend)
+	if (p > end || tag == tagend)
 		return 0;
-	while((p = strstr(p, "</")) && p < end) {
+	while ((p = strstr(p, "</")) && p < end) {
 		p += 2;
-		if(strncmp(p, tag, tagend - tag) == 0 && p[tagend - tag] == '>') {
+		if (strncmp(p, tag, tagend - tag) == 0 &&
+		    p[tagend - tag] == '>') {
 			p++;
-			fwrite(begin, sizeof(char), p - begin + tagend - tag - 1, stdout);
+			fwrite(begin, sizeof(char),
+			       p - begin + tagend - tag - 1, stdout);
 			return p - begin + tagend - tag - 1;
 		}
 	}
 	p = strchr(tagend, '>');
-	if(p) {
+	if (p) {
 		fwrite(begin, sizeof(char), p - begin + 1, stdout);
 		return p - begin + 1;
-	}
-	else
+	} else
 		return 0;
 }
 
-int
-dolineprefix(const char *begin, const char *end, int newblock) {
+int dolineprefix(const char *begin, const char *end, int newblock)
+{
 	unsigned int i, j, l;
 	char *buffer;
 	const char *p;
 
-	if(newblock)
+	if (newblock)
 		p = begin;
-	else if(*begin == '\n')
+	else if (*begin == '\n')
 		p = begin + 1;
 	else
 		return 0;
-	for(i = 0; i < LENGTH(lineprefix); i++) {
+	for (i = 0; i < LENGTH(lineprefix); i++) {
 		l = strlen(lineprefix[i].search);
-		if(end - p < l)
+		if (end - p < l)
 			continue;
-		if(strncmp(lineprefix[i].search, p, l))
+		if (strncmp(lineprefix[i].search, p, l))
 			continue;
-		if(*begin == '\n')
+		if (*begin == '\n')
 			fputc('\n', stdout);
 		fputs(lineprefix[i].before, stdout);
-		if(lineprefix[i].search[l-1] == '\n') {
+		if (lineprefix[i].search[l - 1] == '\n') {
 			fputc('\n', stdout);
 			return l - 1;
 		}
-		if(!(buffer = malloc(BUFSIZ)))
+		if (!(buffer = malloc(BUFSIZ)))
 			eprint("Malloc failed.");
 		buffer[0] = '\0';
 
 		/* Collect lines into buffer while they start with the prefix */
 		j = 0;
-		while((strncmp(lineprefix[i].search, p, l) == 0) && p + l < end) {
+		while ((strncmp(lineprefix[i].search, p, l) == 0) &&
+		       p + l < end) {
 			p += l;
 
 			/* Special case for blockquotes: optional space after > */
-			if(lineprefix[i].search[0] == '>' && *p == ' ') {
+			if (lineprefix[i].search[0] == '>' && *p == ' ') {
 				p++;
 			}
 
-			while(p < end) {
-				ADDC(buffer, j) = *p;
+			while (p < end) {
+				ADDC(buffer, j, *p);
 				j++;
-				if(*(p++) == '\n')
+				if (*(p++) == '\n')
 					break;
 			}
 		}
 
 		/* Skip empty lines in block */
-		while(buffer + j - 1 >= buffer && *(buffer + j - 1) == '\n') {
+		while (buffer + j - 1 >= buffer && *(buffer + j - 1) == '\n') {
 			j--;
 		}
 
-		ADDC(buffer, j) = '\0';
-		if(lineprefix[i].process)
-			process(buffer, buffer + strlen(buffer), lineprefix[i].process >= 2);
+		ADDC(buffer, j, '\0');
+		if (lineprefix[i].process)
+			process(buffer, buffer + strlen(buffer),
+				lineprefix[i].process >= 2);
 		else
 			hprint(buffer, buffer + strlen(buffer));
 		puts(lineprefix[i].after);
@@ -249,79 +309,82 @@ dolineprefix(const char *begin, const char *end, int newblock) {
 	return 0;
 }
 
-int
-dolink(const char *begin, const char *end, int newblock) {
+int dolink(const char *begin, const char *end, int newblock)
+{
 	int img, len, sep, parens_depth = 1;
 	const char *desc, *link, *p, *q, *descend, *linkend;
 	const char *title = NULL, *titleend = NULL;
 
-	if(*begin == '[')
+	if (*begin == '[')
 		img = 0;
-	else if(strncmp(begin, "![", 2) == 0)
+	else if (strncmp(begin, "![", 2) == 0)
 		img = 1;
 	else
 		return 0;
 	p = desc = begin + 1 + img;
-	if(!(p = strstr(desc, "](")) || p > end)
+	if (!(p = strstr(desc, "](")) || p > end)
 		return 0;
-	for(q = strstr(desc, "!["); q && q < end && q < p; q = strstr(q + 1, "!["))
-		if(!(p = strstr(p + 1, "](")) || p > end)
+	for (q = strstr(desc, "!["); q && q < end && q < p;
+	     q = strstr(q + 1, "!["))
+		if (!(p = strstr(p + 1, "](")) || p > end)
 			return 0;
 	descend = p;
 	link = p + 2;
 
 	/* find end of link while handling nested parens */
 	q = link;
-	while(parens_depth) {
-		if(!(q = strpbrk(q, "()")) || q > end)
+	while (parens_depth) {
+		if (!(q = strpbrk(q, "()")) || q > end)
 			return 0;
-		if(*q == '(')
+		if (*q == '(')
 			parens_depth++;
 		else
 			parens_depth--;
-		if(parens_depth && q < end)
+		if (parens_depth && q < end)
 			q++;
 	}
 
-	if((p = strpbrk(link, "\"'")) && p < end && q - 1 > p + 1) {
+	if ((p = strpbrk(link, "\"'")) && p < end && q - 1 > p + 1) {
 		sep = p[0]; /* separator: can be " or ' */
 		title = p + 1;
 		/* strip trailing whitespace */
-		for(linkend = p; linkend > link && isspace(*(linkend - 1)); linkend--);
-		for(titleend = q - 1; titleend > title && isspace(*(titleend)); titleend--);
-		if(*titleend != sep) {
+		for (linkend = p; linkend > link && isspace(*(linkend - 1));
+		     linkend--)
+			;
+		for (titleend = q - 1; titleend > title && isspace(*(titleend));
+		     titleend--)
+			;
+		if (*titleend != sep) {
 			return 0;
 		}
-	}
-	else {
+	} else {
 		linkend = q;
 	}
 
 	/* Links can be given in angular brackets */
-	if(*link == '<' && *(linkend - 1) == '>') {
+	if (*link == '<' && *(linkend - 1) == '>') {
 		link++;
 		linkend--;
 	}
 
 	len = q + 1 - begin;
-	if(img) {
+	if (img) {
 		fputs("<img src=\"", stdout);
 		hprint(link, linkend);
 		fputs("\" alt=\"", stdout);
 		hprint(desc, descend);
 		fputs("\" ", stdout);
-		if(title && titleend) {
+		if (title && titleend) {
 			fputs("title=\"", stdout);
 			hprint(title, titleend);
 			fputs("\" ", stdout);
 		}
 		fputs("/>", stdout);
-	}
-	else {
+	} else {
 		fputs("<a href=\"", stdout);
 		hprint(link, linkend);
 		fputs("\"", stdout);
-		if(title && titleend) {
+		if (title && titleend) {
 			fputs(" title=\"", stdout);
 			hprint(title, titleend);
 			fputs("\"", stdout);
@@ -333,50 +396,56 @@ dolink(const char *begin, const char *end, int newblock) {
 	return len;
 }
 
-int
-dolist(const char *begin, const char *end, int newblock) {
+int dolist(const char *begin, const char *end, int newblock)
+{
 	unsigned int i, j, indent, run, ul, isblock;
 	const char *p, *q;
 	char *buffer = NULL;
 	char marker = 0;
 
 	isblock = 0;
-	if(newblock)
+	if (newblock)
 		p = begin;
-	else if(*begin == '\n')
+	else if (*begin == '\n')
 		p = begin + 1;
 	else
 		return 0;
 	q = p;
-	if(*p == '-' || *p == '*' || *p == '+') {
+	if (*p == '-' || *p == '*' || *p == '+') {
 		ul = 1;
 		marker = *p;
 	} else {
 		ul = 0;
-		for(; p < end && *p >= '0' && *p <= '9'; p++);
-		if(p >= end || *p != '.')
+		for (; p < end && *p >= '0' && *p <= '9'; p++)
+			;
+		if (p >= end || *p != '.')
 			return 0;
 	}
 	p++;
-	if(p >= end || !(*p == ' ' || *p == '\t'))
+	if (p >= end || !(*p == ' ' || *p == '\t'))
 		return 0;
-	for(p++; p != end && (*p == ' ' || *p == '\t'); p++);
+	for (p++; p != end && (*p == ' ' || *p == '\t'); p++)
+		;
 	indent = p - q;
 	buffer = ereallocz(buffer, BUFSIZ);
-	if(!newblock)
+	if (!newblock)
 		fputc('\n', stdout);
 	fputs(ul ? "<ul>\n" : "<ol>\n", stdout);
 	run = 1;
-	for(; p < end && run; p++) {
-		for(i = 0; p < end && run; p++, i++) {
-			if(*p == '\n') {
-				if(p + 1 == end)
+	for (; p < end && run; p++) {
+		for (i = 0; p < end && run; p++, i++) {
+			if (*p == '\n') {
+				if (p + 1 == end)
 					break;
 				else {
 					/* Handle empty lines */
-					for(q = p + 1; (*q == ' ' || *q == '\t') && q < end; q++);
-					if(*q == '\n') {
-						ADDC(buffer, i) = '\n';
+					for (q = p + 1;
+					     (*q == ' ' || *q == '\t') &&
+					     q < end;
+					     q++)
+						;
+					if (*q == '\n') {
+						ADDC(buffer, i, '\0');
 						i++;
 						run = 0;
 						isblock++;
@@ -385,56 +454,63 @@ dolist(const char *begin, const char *end, int newblock) {
 				}
 				q = p + 1;
 				j = 0;
-				if(ul && *q == marker)
+				if (ul && *q == marker)
 					j = 1;
-				else if(!ul) {
-					for(; q + j != end && q[j] >= '0' && q[j] <= '9' && j < indent; j++);
-					if(q + j == end)
+				else if (!ul) {
+					for (; q + j != end && q[j] >= '0' &&
+					       q[j] <= '9' && j < indent;
+					     j++)
+						;
+					if (q + j == end)
 						break;
-					if(j > 0 && q[j] == '.')
+					if (j > 0 && q[j] == '.')
 						j++;
 					else
 						j = 0;
 				}
-				if(q + indent < end)
-					for(; (q[j] == ' ' || q[j] == '\t') && j < indent; j++);
-				if(j == indent) {
-					ADDC(buffer, i) = '\n';
+				if (q + indent < end)
+					for (; (q[j] == ' ' || q[j] == '\t') &&
+					       j < indent;
+					     j++)
+						;
+				if (j == indent) {
+					ADDC(buffer, i, '\n');
 					i++;
 					p += indent;
 					run = 1;
-					if(*q == ' ' || *q == '\t')
+					if (*q == ' ' || *q == '\t')
 						p++;
 					else
 						break;
-				}
-				else if (j < indent)
+				} else if (j < indent)
 					run = 0;
 			}
-			ADDC(buffer, i) = *p;
+			ADDC(buffer, i, *p);
 		}
-		ADDC(buffer, i) = '\0';
+		ADDC(buffer, i, '\0');
 		fputs("<li>", stdout);
-		process(buffer, buffer + i, isblock > 1 || (isblock == 1 && run));
+		process(buffer, buffer + i,
+			isblock > 1 || (isblock == 1 && run));
 		fputs("</li>\n", stdout);
 	}
 	fputs(ul ? "</ul>\n" : "</ol>\n", stdout);
 	free(buffer);
 	p--;
-	while(*(--p) == '\n');
+	while (*(--p) == '\n')
+		;
 	return -(p - begin + 1);
 }
 
-int
-doparagraph(const char *begin, const char *end, int newblock) {
+int doparagraph(const char *begin, const char *end, int newblock)
+{
 	const char *p;
 
-	if(!newblock)
+	if (!newblock)
 		return 0;
 	p = strstr(begin, "\n\n");
-	if(!p || p > end)
+	if (!p || p > end)
 		p = end;
-	if(p - begin <= 1)
+	if (p - begin <= 1)
 		return 0;
 	fputs("<p>", stdout);
 	process(begin, p, 0);
@@ -442,18 +518,18 @@ doparagraph(const char *begin, const char *end, int newblock) {
 	return -(p - begin);
 }
 
-int
-doreplace(const char *begin, const char *end, int newblock) {
+int doreplace(const char *begin, const char *end, int newblock)
+{
 	unsigned int i, l;
 
-	for(i = 0; i < LENGTH(insert); i++)
-		if(strncmp(insert[i][0], begin, strlen(insert[i][0])) == 0)
+	for (i = 0; i < LENGTH(insert); i++)
+		if (strncmp(insert[i][0], begin, strlen(insert[i][0])) == 0)
 			fputs(insert[i][1], stdout);
-	for(i = 0; i < LENGTH(replace); i++) {
+	for (i = 0; i < LENGTH(replace); i++) {
 		l = strlen(replace[i][0]);
-		if(end - begin < l)
+		if (end - begin < l)
 			continue;
-		if(strncmp(replace[i][0], begin, l) == 0) {
+		if (strncmp(replace[i][0], begin, l) == 0) {
 			fputs(replace[i][1], stdout);
 			return l;
 		}
@@ -461,15 +537,15 @@ doreplace(const char *begin, const char *end, int newblock) {
 	return 0;
 }
 
-int
-doshortlink(const char *begin, const char *end, int newblock) {
+int doshortlink(const char *begin, const char *end, int newblock)
+{
 	const char *p, *c;
 	int ismail = 0;
 
-	if(*begin != '<')
+	if (*begin != '<')
 		return 0;
-	for(p = begin + 1; p != end; p++) {
-		switch(*p) {
+	for (p = begin + 1; p != end; p++) {
+		switch (*p) {
 		case ' ':
 		case '\t':
 		case '\n':
@@ -479,23 +555,23 @@ doshortlink(const char *begin, const char *end, int newblock) {
 			ismail = -1;
 			break;
 		case '@':
-			if(ismail == 0)
+			if (ismail == 0)
 				ismail = 1;
 			break;
 		case '>':
-			if(ismail == 0)
+			if (ismail == 0)
 				return 0;
 			fputs("<a href=\"", stdout);
-			if(ismail == 1) {
+			if (ismail == 1) {
 				/* mailto: */
-				fputs("&#x6D;&#x61;i&#x6C;&#x74;&#x6F;:", stdout);
-				for(c = begin + 1; *c != '>'; c++)
+				fputs("&#x6D;&#x61;i&#x6C;&#x74;&#x6F;:",
+				      stdout);
+				for (c = begin + 1; *c != '>'; c++)
 					fprintf(stdout, "&#%u;", *c);
 				fputs("\">", stdout);
-				for(c = begin + 1; *c != '>'; c++)
+				for (c = begin + 1; *c != '>'; c++)
 					fprintf(stdout, "&#%u;", *c);
-			}
-			else {
+			} else {
 				hprint(begin + 1, p);
 				fputs("\">", stdout);
 				hprint(begin + 1, p);
@@ -507,24 +583,25 @@ doshortlink(const char *begin, const char *end, int newblock) {
 	return 0;
 }
 
-int
-dosurround(const char *begin, const char *end, int newblock) {
+int dosurround(const char *begin, const char *end, int newblock)
+{
 	unsigned int i, l;
 	const char *p, *start, *stop;
 
-	for(i = 0; i < LENGTH(surround); i++) {
+	for (i = 0; i < LENGTH(surround); i++) {
 		l = strlen(surround[i].search);
-		if(end - begin < 2*l || strncmp(begin, surround[i].search, l) != 0)
+		if (end - begin < 2 * l ||
+		    strncmp(begin, surround[i].search, l) != 0)
 			continue;
 		start = begin + l;
 		p = start - 1;
 		do {
 			stop = p;
 			p = strstr(p + 1, surround[i].search);
-		} while(p && p[-1] == '\\');
+		} while (p && p[-1] == '\\');
 		if (p && p[-1] != '\\')
 			stop = p;
-		if(!stop || stop < start || stop >= end)
+		if (!stop || stop < start || stop >= end)
 			continue;
 		fputs(surround[i].before, stdout);
 
@@ -535,7 +612,7 @@ dosurround(const char *begin, const char *end, int newblock) {
 			l++;
 		}
 
-		if(surround[i].process)
+		if (surround[i].process)
 			process(start, stop, 0);
 		else
 			hprint(start, stop);
@@ -545,23 +622,27 @@ dosurround(const char *begin, const char *end, int newblock) {
 	return 0;
 }
 
-int
-dounderline(const char *begin, const char *end, int newblock) {
+int dounderline(const char *begin, const char *end, int newblock)
+{
 	unsigned int i, j, l;
 	const char *p;
 
-	if(!newblock)
+	if (!newblock)
 		return 0;
 	p = begin;
-	for(l = 0; p + l + 1 != end && p[l] != '\n'; l++);
+	for (l = 0; p + l + 1 != end && p[l] != '\n'; l++)
+		;
 	p += l + 1;
-	if(l == 0)
+	if (l == 0)
 		return 0;
-	for(i = 0; i < LENGTH(underline); i++) {
-		for(j = 0; p + j != end && p[j] != '\n' && p[j] == underline[i].search[0]; j++);
-		if(j >= l) {
+	for (i = 0; i < LENGTH(underline); i++) {
+		for (j = 0; p + j != end && p[j] != '\n' &&
+			    p[j] == underline[i].search[0];
+		     j++)
+			;
+		if (j >= l) {
 			fputs(underline[i].before, stdout);
-			if(underline[i].process)
+			if (underline[i].process)
 				process(begin, begin + l, 0);
 			else
 				hprint(begin, begin + l);
@@ -572,106 +653,143 @@ dounderline(const char *begin, const char *end, int newblock) {
 	return 0;
 }
 
-void *
-ereallocz(void *p, size_t size) {
+void *ereallocz(void *p, size_t size)
+{
 	void *res;
-	if(p)
-		res = realloc(p , size);
+	if (p)
+		res = realloc(p, size);
 	else
 		res = calloc(1, size);
 
-	if(!res)
+	if (!res)
 		eprint("fatal: could not malloc() %u bytes\n", size);
 	return res;
 }
 
-void
-hprint(const char *begin, const char *end) {
+void hprint(const char *begin, const char *end)
+{
 	const char *p;
 
-	for(p = begin; p != end; p++) {
-		if(*p == '&')
+	for (p = begin; p != end; p++) {
+		if (*p == '&')
 			fputs("&amp;", stdout);
-		else if(*p == '"')
+		else if (*p == '"')
 			fputs("&quot;", stdout);
-		else if(*p == '>')
+		else if (*p == '>')
 			fputs("&gt;", stdout);
-		else if(*p == '<')
+		else if (*p == '<')
 			fputs("&lt;", stdout);
 		else
 			fputc(*p, stdout);
 	}
 }
 
-void
-process(const char *begin, const char *end, int newblock) {
-	const char *p, *q;
+int process(const char *begin, const char *end, int newblock)
+{
+	const char *q;
+	const char *p;
 	int affected;
 	unsigned int i;
 
-	for(p = begin; p < end;) {
-		if(newblock)
-			while(*p == '\n')
-				if(++p == end)
-					return;
+	for (p = begin; p < end;) {
+		if (newblock)
+			while (*p == '\n')
+				if (++p == end)
+					return 1;
 		affected = 0;
-		for(i = 0; i < LENGTH(parsers) && !affected; i++)
+		for (i = 0; i < LENGTH(parsers) && !affected; i++)
 			affected = parsers[i](p, end, newblock);
 		p += abs(affected);
-		if(!affected) {
-			if(nohtml)
+		if (affected == -1)
+			return 0;
+		if (!affected) {
+			if (nohtml)
 				hprint(p, p + 1);
 			else
 				fputc(*p, stdout);
 			p++;
 		}
-		for(q = p; q != end && *q == '\n'; q++);
-		if(q == end)
-			return;
-		else if(p[0] == '\n' && p + 1 != end && p[1] == '\n')
+		for (q = p; q != end && *q == '\n'; q++)
+			;
+		if (q == end)
+			return 1;
+		else if (p[0] == '\n' && p + 1 != end && p[1] == '\n')
 			newblock = 1;
 		else
 			newblock = affected < 0;
 	}
+	return 1;
 }
 
-int
-main(int argc, char *argv[]) {
+int main(int argc, char *argv[])
+{
 	char *buffer = NULL;
-	int s, i;
-	unsigned long len, bsize;
-	FILE *source = stdin;
+	const char *path = "STDIN";
+	int i;
+	int ret = EXIT_SUCCESS;
 
-	for(i = 1; i < argc; i++) {
-		if(!strcmp("-v", argv[i]))
-			eprint("simple markup %s (C) Enno Boland\n",VERSION);
-		else if(!strcmp("-n", argv[i]))
+	for (i = 1; i < argc; i++) {
+		if (!strcmp("-v", argv[i]))
+			eprint("simple markup %s (C) Enno Boland\n", VERSION);
+		else if (!strcmp("-n", argv[i]))
 			nohtml = 1;
-		else if(argv[i][0] != '-')
+		else if (argv[i][0] != '-')
 			break;
-		else if(!strcmp("--", argv[i])) {
+		else if (!strcmp("--", argv[i])) {
 			i++;
 			break;
+		} else
+			eprint("Usage %s [-n] [file]\n -n escape html strictly\n",
+			       argv[0]);
 	}
-		else
-			eprint("Usage %s [-n] [file]\n -n escape html strictly\n", argv[0]);
-	}
-	if(i < argc && !(source = fopen(argv[i], "r")))
-		eprint("Cannot open file `%s`\n",argv[i]);
-	bsize = 2 * BUFSIZ;
-	buffer = ereallocz(buffer, bsize);
-	len = 0;
-	while((s = fread(buffer + len, 1, BUFSIZ, source))) {
-		len += s;
-		if(BUFSIZ + len + 1 > bsize) {
-			bsize += BUFSIZ;
-			if(!(buffer = realloc(buffer, bsize)))
-				eprint("realloc failed.");
+
+	if (i < argc) {
+		path = argv[i];
+
+		off64_t len = get_file_size(path);
+		if (len == -1) {
+			eprint("%s: %s: %s\n", argv[0], path, strerror(errno));
+			ret = EXIT_FAILURE;
+			goto exit;
 		}
+		buffer = read_file(path, len);
+		if (!buffer) {
+			perror(PACKAGE);
+			ret = EXIT_FAILURE;
+			goto exit;
 		}
 		buffer[len] = '\0';
 		process(buffer, buffer + len, 1);
-	fclose(source);
 		free(buffer);
-	return EXIT_SUCCESS;
+	} else {
+		size_t buffer_size = 1024 * CHARWIDTH;
+		buffer = calloc(buffer_size + CHARWIDTH, sizeof(char));
+		if (!buffer) {
+			perror(PACKAGE);
+			ret = EXIT_FAILURE;
+			goto exit;
+		}
+
+		size_t read_bytes;
+		while (1) {
+			read_bytes = read(STDIN_FILENO, buffer, buffer_size);
+			if (read_bytes <= 0) {
+				if (errno) {
+					perror(PACKAGE);
+					ret = EXIT_FAILURE;
+				}
+				free(buffer);
+				break;
+			}
+
+			buffer[read_bytes] = '\0';
+			if (!process(buffer, buffer + read_bytes, 1)) {
+				ret = EXIT_FAILURE;
+				free(buffer);
+				break;
+			}
+		}
+	}
+exit:
+	return ret;
 }

smu.h

@@ -16,4 +16,3 @@ int smu_convert(FILE *out, FILE *in, int suppresshtml);
 
 /** utility */
 void eprint(const char *format, ...);
-

testdoc

@@ -47,9 +47,11 @@ list in list:
 entity: &, <, >
 
 code:
+```
 	int powerof2(unsigned int n) {
 		return !((n - 1) & n) && n > 0;
 	}
+```
 
 links
 -----