turning smu into a library, thanks arg. - smu - smu - simple markup (Markdown) … | |
git clone git://git.codemadness.org/smu | |
Log | |
Files | |
Refs | |
README | |
LICENSE | |
--- | |
commit b2712c1480271f92a7c2667b6be398c33973edf7 | |
parent 66219165d078ff5c1f859548b5fa2b2cf0b49449 | |
Author: Enno Boland (Gottox) <[email protected]> | |
Date: Wed, 18 Jun 2008 08:00:18 +0200 | |
turning smu into a library, thanks arg. | |
Diffstat: | |
M Makefile | 20 +++++++++++++------- | |
M config.mk | 1 + | |
A libsmu.c | 557 +++++++++++++++++++++++++++++… | |
M smu.c | 566 +----------------------------… | |
A smu.h | 19 +++++++++++++++++++ | |
5 files changed, 602 insertions(+), 561 deletions(-) | |
--- | |
diff --git a/Makefile b/Makefile | |
@@ -1,12 +1,14 @@ | |
-# smu - simple markup | |
+# libsmu - simple markup | |
# (c) 2007, 2008 Enno Boland | |
include config.mk | |
-SRC = smu.c | |
-OBJ = ${SRC:.c=.o} | |
+SRC = smu.c | |
+LIBSRC = libsmu.c | |
+OBJ = ${SRC:.c=.o} | |
+LIBOBJ = ${LIBSRC:.c=.o} | |
-all: options smu | |
+all: options libsmu.a smu | |
options: | |
@echo smu build options: | |
@@ -20,19 +22,23 @@ options: | |
${OBJ}: config.mk | |
+libsmu.a: ${LIBOBJ} | |
+ @echo AR $@ | |
+ @${AR} $@ ${LIBOBJ} | |
+ | |
smu: ${OBJ} | |
@echo LD $@ | |
- @${CC} -o $@ ${OBJ} ${LDFLAGS} | |
+ @${CC} -o $@ ${OBJ} ${LDFLAGS} -L. -lsmu | |
clean: | |
@echo cleaning | |
- @rm -f smu ${OBJ} smu-${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} | |
+ smu.1 ${SRC} ${LIBSRC} smu.h smu-${VERSION} | |
@tar -cf smu-${VERSION}.tar smu-${VERSION} | |
@gzip smu-${VERSION}.tar | |
@rm -rf smu-${VERSION} | |
diff --git a/config.mk b/config.mk | |
@@ -18,3 +18,4 @@ LDFLAGS = ${LIBS} | |
# compiler | |
CC = cc | |
+AR = ar crsv | |
diff --git a/libsmu.c b/libsmu.c | |
@@ -0,0 +1,557 @@ | |
+/* libsmu - simple markup library | |
+ * Copyright (C) <2007, 2008> Enno Boland <g s01 de> | |
+ * | |
+ * See LICENSE for further informations | |
+ */ | |
+#include <stdarg.h> | |
+#include <stdlib.h> | |
+#include <string.h> | |
+ | |
+#include "smu.h" | |
+ | |
+#define BUFFERSIZE 512 | |
+#define LENGTH(x) sizeof(x)/sizeof(x[0]) | |
+#define ADDC(b,i) if(i % BUFFERSIZE == 0) \ | |
+ { b = realloc(b, (i + BUFFERSIZE) * sizeof(b)); if(!b) eprint("Malloc … | |
+ | |
+typedef int (*Parser)(const char *, const char *, int); | |
+typedef struct { | |
+ char *search; | |
+ int process; | |
+ char *before, *after; | |
+} Tag; | |
+ | |
+static int doamp(const char *begin, const char *end, int newblock); /* P… | |
+static int dogtlt(const char *begin, const char *end, int newblock); /* P… | |
+static int dohtml(const char *begin, const char *end, int newblock); /* P… | |
+static int dolineprefix(const char *begin, const char *end, int newblock);/* P… | |
+static int dolink(const char *begin, const char *end, int newblock); /* P… | |
+static int dolist(const char *begin, const char *end, int newblock); /* P… | |
+static int doparagraph(const char *begin, const char *end, int newblock); /* P… | |
+static int doreplace(const char *begin, const char *end, int newblock); /* P… | |
+static int doshortlink(const char *begin, const char *end, int newblock); /* P… | |
+static int dosurround(const char *begin, const char *end, int newblock); /* P… | |
+static int dounderline(const char *begin, const char *end, int newblock); /* P… | |
+static void hprint(const char *begin, const char *end); /* e… | |
+static void process(const char *begin, const char *end, int isblock); /* P… | |
+ | |
+/* list of parsers */ | |
+static Parser parsers[] = { dounderline, dohtml, dolineprefix, dolist, | |
+ doparagraph, dogtlt, dosurround, dolink, | |
+ doshortlink, doamp, doreplace }; | |
+static FILE *output; | |
+static int nohtml = 1; | |
+ | |
+static Tag lineprefix[] = { | |
+ { " ", 0, "<pre><code>", "</code></pre>" }, | |
+ { "\t", 0, "<pre><code>", "</code></pre>" }, | |
+ { "> ", 2, "<blockquote>", "</blockquote>… | |
+ { "###### ", 1, "<h6>", "</h6>" }, | |
+ { "##### ", 1, "<h5>", "</h5>" }, | |
+ { "#### ", 1, "<h4>", "</h4>" }, | |
+ { "### ", 1, "<h3>", "</h3>" }, | |
+ { "## ", 1, "<h2>", "</h2>" }, | |
+ { "# ", 1, "<h1>", "</h1>" }, | |
+ { "- - -\n", 1, "<hr />", ""}, | |
+}; | |
+ | |
+static Tag underline[] = { | |
+ { "=", 1, "<h1>", "</h1>\n" }, | |
+ { "-", 1, "<h2>", "</h2>\n" }, | |
+}; | |
+ | |
+static Tag surround[] = { | |
+ { "``", 0, "<code>", "</code>" }, | |
+ { "`", 0, "<code>", "</code>" }, | |
+ { "___", 1, "<strong><em>", "</em></strong>" }, | |
+ { "***", 1, "<strong><em>", "</em></strong>" }, | |
+ { "__", 1, "<strong>", "</strong>" }, | |
+ { "**", 1, "<strong>", "</strong>" }, | |
+ { "_", 1, "<em>", "</em>" }, | |
+ { "*", 1, "<em>", "</em>" }, | |
+}; | |
+ | |
+static const char *replace[][2] = { | |
+ { "\\\\", "\\" }, | |
+ { "\\`", "`" }, | |
+ { "\\*", "*" }, | |
+ { "\\_", "_" }, | |
+ { "\\{", "{" }, | |
+ { "\\}", "}" }, | |
+ { "\\[", "[" }, | |
+ { "\\]", "]" }, | |
+ { "\\(", "(" }, | |
+ { "\\)", ")" }, | |
+ { "\\#", "#" }, | |
+ { "\\+", "+" }, | |
+ { "\\-", "-" }, | |
+ { "\\.", "." }, | |
+ { "\\!", "!" }, | |
+}; | |
+ | |
+static const char *insert[][2] = { | |
+ { " \n", "<br />" }, | |
+}; | |
+ | |
+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) { | |
+ const char *p; | |
+ | |
+ if(*begin != '&') | |
+ return 0; | |
+ if(!nohtml) { | |
+ for(p = begin + 1; p != end && !strchr("; \\\n\t", *p); p++); | |
+ if(p == end || *p == ';') | |
+ return 0; | |
+ } | |
+ fputs("&", output); | |
+ return 1; | |
+} | |
+ | |
+int | |
+dogtlt(const char *begin, const char *end, int newblock) { | |
+ int brpos; | |
+ char c; | |
+ | |
+ if(nohtml || begin + 1 >= end) | |
+ return 0; | |
+ brpos = begin[1] == '>'; | |
+ if(!brpos && *begin != '<') | |
+ return 0; | |
+ c = begin[brpos ? 0 : 1]; | |
+ if(!brpos && (c < 'a' || c > 'z') && (c < 'A' || c > 'Z')) { | |
+ fputs("<", output); | |
+ return 1; | |
+ } | |
+ else if(brpos && (c < 'a' || c > 'z') && (c < 'A' || c > 'Z') && !strc… | |
+ fprintf(output, "%c>",c); | |
+ return 2; | |
+ } | |
+ return 0; | |
+} | |
+ | |
+int | |
+dohtml(const char *begin, const char *end, int newblock) { | |
+ const char *p, *tag, *tagend; | |
+ | |
+ if(nohtml || !newblock || *begin == '\n' || begin + 2 >= end) | |
+ return 0; | |
+ p = begin; | |
+ if(p[1] == '\n') | |
+ p++; | |
+ if(p[1] != '<' || strchr(" /\n\t\\", p[2])) | |
+ return 0; | |
+ tag = p + 2; | |
+ p += 2; | |
+ for(; !strchr(" >", *p); p++); | |
+ tagend = p; | |
+ while((p = strstr(p, "\n</")) && p < end) { | |
+ p += 3; | |
+ if(strncmp(p, tag, tagend - tag) == 0 && p[tagend - tag] == '>… | |
+ p++; | |
+ fwrite(begin, sizeof(char), p - begin + tagend - tag, … | |
+ puts("\n"); | |
+ return -(p - begin + tagend - tag); | |
+ } | |
+ } | |
+ return 0; | |
+} | |
+ | |
+int | |
+dolineprefix(const char *begin, const char *end, int newblock) { | |
+ unsigned int i, j, l; | |
+ char *buffer; | |
+ const char *p; | |
+ | |
+ if(newblock) | |
+ p = begin; | |
+ else if(*begin == '\n') | |
+ p = begin + 1; | |
+ else | |
+ return 0; | |
+ for(i = 0; i < LENGTH(lineprefix); i++) { | |
+ l = strlen(lineprefix[i].search); | |
+ if(end - p < l) | |
+ continue; | |
+ if(strncmp(lineprefix[i].search, p, l)) | |
+ continue; | |
+ if(*begin == '\n') | |
+ fputc('\n', output); | |
+ fputs(lineprefix[i].before, output); | |
+ if(lineprefix[i].search[l-1] == '\n') { | |
+ fputc('\n', output); | |
+ return l; | |
+ } | |
+ if(!(buffer = malloc(BUFFERSIZE))) | |
+ eprint("Malloc failed."); | |
+ buffer[0] = '\0'; | |
+ for(j = 0, p += l; p < end; p++, j++) { | |
+ ADDC(buffer, j) = *p; | |
+ if(*p == '\n' && p + l < end) { | |
+ if(strncmp(lineprefix[i].search, p + 1, l) != … | |
+ break; | |
+ p += l; | |
+ } | |
+ } | |
+ ADDC(buffer, j) = '\0'; | |
+ if(lineprefix[i].process) | |
+ process(buffer, buffer + strlen(buffer), lineprefix[i]… | |
+ else | |
+ hprint(buffer, buffer + strlen(buffer)); | |
+ puts(lineprefix[i].after); | |
+ free(buffer); | |
+ return -(p - begin); | |
+ } | |
+ return 0; | |
+} | |
+ | |
+int | |
+dolink(const char *begin, const char *end, int newblock) { | |
+ int img; | |
+ const char *desc, *link, *p, *q, *descend, *linkend; | |
+ | |
+ if(*begin == '[') | |
+ img = 0; | |
+ else if(strncmp(begin, ") || p > end) | |
+ return 0; | |
+ for(q = strstr(desc, ") || p > end) | |
+ return 0; | |
+ descend = p; | |
+ link = p + 2; | |
+ if(!(p = strstr(link, ")")) || p > end) | |
+ return 0; | |
+ linkend = p; | |
+ if(img) { | |
+ fputs("<img src=\"", output); | |
+ hprint(link, linkend); | |
+ fputs("\" alt=\"", output); | |
+ hprint(desc, descend); | |
+ fputs("\" />", output); | |
+ } | |
+ else { | |
+ fputs("<a href=\"", output); | |
+ hprint(link, linkend); | |
+ fputs("\">", output); | |
+ process(desc, descend, 0); | |
+ fputs("</a>", output); | |
+ } | |
+ return p + 1 - begin; | |
+} | |
+ | |
+int | |
+dolist(const char *begin, const char *end, int newblock) { | |
+ unsigned int i, j, indent, run, ul, isblock; | |
+ const char *p, *q; | |
+ char *buffer; | |
+ | |
+ isblock = 0; | |
+ if(newblock) | |
+ p = begin; | |
+ else if(*begin == '\n') | |
+ p = begin + 1; | |
+ else | |
+ return 0; | |
+ q = p; | |
+ if(*p == '-' || *p == '*' || *p == '+') | |
+ ul = 1; | |
+ else { | |
+ ul = 0; | |
+ for(; p < end && *p >= '0' && *p <= '9'; p++); | |
+ if(p >= end || *p != '.') | |
+ return 0; | |
+ } | |
+ p++; | |
+ if(p >= end || !(*p == ' ' || *p == '\t')) | |
+ return 0; | |
+ for(p++; p != end && (*p == ' ' || *p == '\t'); p++); | |
+ indent = p - q; | |
+ if(!(buffer = malloc(BUFFERSIZE))) | |
+ eprint("Malloc failed."); | |
+ if(!newblock) | |
+ fputc('\n', output); | |
+ fputs(ul ? "<ul>\n" : "<ol>\n", output); | |
+ run = 1; | |
+ for(; p < end && run; p++) { | |
+ for(i = 0; p < end && run; p++, i++) { | |
+ if(*p == '\n') { | |
+ if(p + 1 == end) | |
+ break; | |
+ else if(p[1] == '\n') { | |
+ p++; | |
+ ADDC(buffer, i) = '\n'; | |
+ i++; | |
+ run = 0; | |
+ isblock++; | |
+ } | |
+ q = p + 1; | |
+ j = 0; | |
+ if(ul && (*q == '-' || *q == '*' || *q == '+')) | |
+ j = 1; | |
+ else if(!ul) { | |
+ for(; q + j != end && q[j] >= '0' && q… | |
+ if(q + j == end) | |
+ break; | |
+ if(j > 0 && q[j] == '.') | |
+ j++; | |
+ else | |
+ j = 0; | |
+ } | |
+ if(q + indent < end) | |
+ for(; (q[j] == ' ' || q[j] == '\t') &&… | |
+ if(j == indent) { | |
+ ADDC(buffer, i) = '\n'; | |
+ i++; | |
+ p += indent; | |
+ run = 1; | |
+ if(*q == ' ' || *q == '\t') | |
+ p++; | |
+ else | |
+ break; | |
+ } | |
+ } | |
+ ADDC(buffer, i) = *p; | |
+ } | |
+ ADDC(buffer, i) = '\0'; | |
+ fputs("<li>", output); | |
+ process(buffer, buffer + i, isblock > 1 || (isblock == 1 && ru… | |
+ fputs("</li>\n", output); | |
+ } | |
+ fputs(ul ? "</ul>\n" : "</ol>\n", output); | |
+ free(buffer); | |
+ p--; | |
+ while(*(--p) == '\n'); | |
+ return -(p - begin + 1); | |
+} | |
+ | |
+int | |
+doparagraph(const char *begin, const char *end, int newblock) { | |
+ const char *p; | |
+ | |
+ if(!newblock) | |
+ return 0; | |
+ p = strstr(begin, "\n\n"); | |
+ if(!p || p > end) | |
+ p = end; | |
+ if(p - begin <= 1) | |
+ return 0; | |
+ fputs("<p>\n", output); | |
+ process(begin, p, 0); | |
+ fputs("</p>\n", output); | |
+ return -(p - begin); | |
+} | |
+ | |
+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) | |
+ fputs(insert[i][1], output); | |
+ for(i = 0; i < LENGTH(replace); i++) { | |
+ l = strlen(replace[i][0]); | |
+ if(end - begin < l) | |
+ continue; | |
+ if(strncmp(replace[i][0], begin, l) == 0) { | |
+ fputs(replace[i][1], output); | |
+ return l; | |
+ } | |
+ } | |
+ return 0; | |
+} | |
+ | |
+int | |
+doshortlink(const char *begin, const char *end, int newblock) { | |
+ const char *p, *c; | |
+ int ismail = 0; | |
+ | |
+ if(*begin != '<') | |
+ return 0; | |
+ for(p = begin + 1; p != end; p++) { | |
+ switch(*p) { | |
+ case ' ': | |
+ case '\t': | |
+ case '\n': | |
+ return 0; | |
+ case '#': | |
+ case ':': | |
+ ismail = -1; | |
+ break; | |
+ case '@': | |
+ if(ismail == 0) | |
+ ismail = 1; | |
+ break; | |
+ case '>': | |
+ if(ismail == 0) | |
+ return 0; | |
+ fputs("<a href=\"", output); | |
+ if(ismail == 1) { | |
+ /* mailto: */ | |
+ fputs("mailto:", outp… | |
+ for(c = begin + 1; *c != '>'; c++) | |
+ fprintf(output, "&#%u;", *c); | |
+ fputs("\">", output); | |
+ for(c = begin + 1; *c != '>'; c++) | |
+ fprintf(output, "&#%u;", *c); | |
+ } | |
+ else { | |
+ hprint(begin + 1, p); | |
+ fputs("\">", output); | |
+ hprint(begin + 1, p); | |
+ } | |
+ fputs("</a>", output); | |
+ return p - begin + 1; | |
+ } | |
+ } | |
+ return 0; | |
+} | |
+ | |
+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++) { | |
+ l = strlen(surround[i].search); | |
+ if(end - begin < 2*l || strncmp(begin, surround[i].search, l) … | |
+ continue; | |
+ start = begin + l; | |
+ p = start - 1; | |
+ do { | |
+ p = strstr(p + 1, surround[i].search); | |
+ } while(p && p[-1] == '\\'); | |
+ if(!p || p >= end || | |
+ !(stop = strstr(start, surround[i].search)) ||… | |
+ continue; | |
+ fputs(surround[i].before, output); | |
+ if(surround[i].process) | |
+ process(start, stop, 0); | |
+ else | |
+ hprint(start, stop); | |
+ fputs(surround[i].after, output); | |
+ return stop - begin + l; | |
+ } | |
+ return 0; | |
+} | |
+ | |
+int | |
+dounderline(const char *begin, const char *end, int newblock) { | |
+ unsigned int i, j, l; | |
+ const char *p; | |
+ | |
+ if(!newblock) | |
+ return 0; | |
+ p = begin; | |
+ for(l = 0; p + l != end && p[l] != '\n'; l++); | |
+ p += l + 1; | |
+ 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… | |
+ if(j >= l) { | |
+ fputs(underline[i].before, output); | |
+ if(underline[i].process) | |
+ process(begin, begin + l, 0); | |
+ else | |
+ hprint(begin, begin + l); | |
+ fputs(underline[i].after, output); | |
+ return -(j + p - begin); | |
+ } | |
+ } | |
+ return 0; | |
+} | |
+ | |
+void | |
+hprint(const char *begin, const char *end) { | |
+ const char *p; | |
+ | |
+ for(p = begin; p != end; p++) { | |
+ if(*p == '&') | |
+ fputs("&", output); | |
+ else if(*p == '"') | |
+ fputs(""", output); | |
+ else if(*p == '>') | |
+ fputs(">", output); | |
+ else if(*p == '<') | |
+ fputs("<", output); | |
+ else | |
+ fputc(*p, output); | |
+ } | |
+} | |
+ | |
+void | |
+process(const char *begin, const char *end, int newblock) { | |
+ const char *p, *q; | |
+ int affected; | |
+ unsigned int i; | |
+ | |
+ for(p = begin; p != end;) { | |
+ if(newblock) | |
+ while(*p == '\n') | |
+ if(++p == end) | |
+ return; | |
+ affected = 0; | |
+ for(i = 0; i < LENGTH(parsers) && affected == 0; i++) | |
+ affected = parsers[i](p, end, newblock); | |
+ p += abs(affected); | |
+ if(!affected) { | |
+ if(nohtml) | |
+ hprint(p, p + 1); | |
+ else | |
+ fputc(*p, output); | |
+ p++; | |
+ } | |
+ for(q = p; q != end && *q == '\n'; q++); | |
+ if(q == end) | |
+ return; | |
+ else if(p[0] == '\n' && p + 1 != end && p[1] == '\n') | |
+ newblock = 1; | |
+ else | |
+ newblock = affected < 0; | |
+ } | |
+} | |
+ | |
+/** library call **/ | |
+ | |
+int | |
+smu_convert(FILE *out, FILE *in, int suppresshtml) { | |
+ char *buffer; | |
+ int s; | |
+ unsigned long len, bsize; | |
+ | |
+ nohtml = suppresshtml; | |
+ output = out; | |
+ | |
+ bsize = 2 * BUFFERSIZE; | |
+ if(!(buffer = malloc(bsize))) | |
+ eprint("Malloc failed."); | |
+ len = 0; | |
+ while((s = fread(buffer + len, 1, BUFFERSIZE, in))) { | |
+ len += s; | |
+ if(BUFFERSIZE + len + 1 > bsize) { | |
+ bsize += BUFFERSIZE; | |
+ if(!(buffer = realloc(buffer, bsize))) | |
+ eprint("Malloc failed."); | |
+ } | |
+ } | |
+ buffer[len] = '\0'; | |
+ process(buffer, buffer + len, 1); | |
+ free(buffer); | |
+ | |
+ return EXIT_SUCCESS; | |
+} | |
diff --git a/smu.c b/smu.c | |
@@ -1,571 +1,29 @@ | |
-/* smu - simple markup | |
- * Copyright (C) <2007, 2008> Enno boland <g s01 de> | |
+/* libsmu - simple markup library | |
+ * Copyright (C) <2007, 2008> Enno Boland <g s01 de> | |
* | |
* See LICENSE for further informations | |
*/ | |
- | |
#include <stdlib.h> | |
-#include <stdio.h> | |
-#include <stdarg.h> | |
#include <string.h> | |
-#define BUFFERSIZE 512 | |
-#define LENGTH(x) sizeof(x)/sizeof(x[0]) | |
-#define ADDC(b,i) if(i % BUFFERSIZE == 0) \ | |
- { b = realloc(b, (i + BUFFERSIZE) * sizeof(b)); if(!b) eprint("Malloc … | |
- | |
- | |
-typedef int (*Parser)(const char *, const char *, int); | |
-struct Tag { | |
- char *search; | |
- int process; | |
- char *before, *after; | |
-}; | |
- | |
- | |
-void eprint(const char *format, ...); /* Prints error a… | |
-int doamp(const char *begin, const char *end, int newblock); | |
- /* Parser for & */ | |
-int dogtlt(const char *begin, const char *end, int newblock); | |
- /* Parser for < and > … | |
-int dohtml(const char *begin, const char *end, int newblock); | |
- /* Parser for html */ | |
-int dolineprefix(const char *begin, const char *end, int newblock); | |
- /* Parser for line pre… | |
-int dolink(const char *begin, const char *end, int newblock); | |
- /* Parser for links an… | |
-int dolist(const char *begin, const char *end, int newblock); | |
- /* Parser for lists */ | |
-int doparagraph(const char *begin, const char *end, int newblock); | |
- /* Parser for paragrap… | |
-int doreplace(const char *begin, const char *end, int newblock); | |
- /* Parser for simple r… | |
-int doshortlink(const char *begin, const char *end, int newblock); | |
- /* Parser for links an… | |
-int dosurround(const char *begin, const char *end, int newblock); | |
- /* Parser for surround… | |
-int dounderline(const char *begin, const char *end, int newblock); | |
- /* Parser for underlin… | |
-void hprint(const char *begin, const char *end); /* escapes HTML and pr… | |
-void process(const char *begin, const char *end, int isblock); | |
- /* Processes range bet… | |
- | |
-Parser parsers[] = { dounderline, dohtml, dolineprefix, dolist, doparagraph, | |
- dogtlt, dosurround, dolink, doshortlink, doamp, doreplace }; | |
- /* list of parsers */ | |
-FILE *source; | |
-unsigned int nohtml = 0; | |
-struct Tag lineprefix[] = { | |
- { " ", 0, "<pre><code>", "</code></pre>" }, | |
- { "\t", 0, "<pre><code>", "</code></pre>" }, | |
- { "> ", 2, "<blockquote>", "</blockquote>… | |
- { "###### ", 1, "<h6>", "</h6>" }, | |
- { "##### ", 1, "<h5>", "</h5>" }, | |
- { "#### ", 1, "<h4>", "</h4>" }, | |
- { "### ", 1, "<h3>", "</h3>" }, | |
- { "## ", 1, "<h2>", "</h2>" }, | |
- { "# ", 1, "<h1>", "</h1>" }, | |
- { "- - -\n", 1, "<hr />", ""}, | |
-}; | |
-struct Tag underline[] = { | |
- { "=", 1, "<h1>", "</h1>\n" }, | |
- { "-", 1, "<h2>", "</h2>\n" }, | |
-}; | |
-struct Tag surround[] = { | |
- { "``", 0, "<code>", "</code>" }, | |
- { "`", 0, "<code>", "</code>" }, | |
- { "___", 1, "<strong><em>", "</em></strong>" }, | |
- { "***", 1, "<strong><em>", "</em></strong>" }, | |
- { "__", 1, "<strong>", "</strong>" }, | |
- { "**", 1, "<strong>", "</strong>" }, | |
- { "_", 1, "<em>", "</em>" }, | |
- { "*", 1, "<em>", "</em>" }, | |
-}; | |
-char * replace[][2] = { | |
- { "\\\\", "\\" }, | |
- { "\\`", "`" }, | |
- { "\\*", "*" }, | |
- { "\\_", "_" }, | |
- { "\\{", "{" }, | |
- { "\\}", "}" }, | |
- { "\\[", "[" }, | |
- { "\\]", "]" }, | |
- { "\\(", "(" }, | |
- { "\\)", ")" }, | |
- { "\\#", "#" }, | |
- { "\\+", "+" }, | |
- { "\\-", "-" }, | |
- { "\\.", "." }, | |
- { "\\!", "!" }, | |
-}; | |
-char * insert[][2] = { | |
- { " \n", "<br />" }, | |
-}; | |
- | |
-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) { | |
- const char *p; | |
- | |
- if(*begin != '&') | |
- return 0; | |
- if(!nohtml) { | |
- for(p = begin + 1; p != end && !strchr("; \\\n\t", *p); p++); | |
- if(p == end || *p == ';') | |
- return 0; | |
- } | |
- fputs("&", stdout); | |
- return 1; | |
-} | |
- | |
-int | |
-dogtlt(const char *begin, const char *end, int newblock) { | |
- int brpos; | |
- char c; | |
- | |
- if(nohtml || begin + 1 >= end) | |
- return 0; | |
- brpos = begin[1] == '>'; | |
- if(!brpos && *begin != '<') | |
- return 0; | |
- c = begin[brpos ? 0 : 1]; | |
- if(!brpos && (c < 'a' || c > 'z') && (c < 'A' || c > 'Z')) { | |
- fputs("<",stdout); | |
- return 1; | |
- } | |
- else if(brpos && (c < 'a' || c > 'z') && (c < 'A' || c > 'Z') && !strc… | |
- printf("%c>",c); | |
- return 2; | |
- } | |
- return 0; | |
-} | |
- | |
-int | |
-dohtml(const char *begin, const char *end, int newblock) { | |
- const char *p, *tag, *tagend; | |
- | |
- if(nohtml || !newblock || *begin == '\n' || begin + 2 >= end) | |
- return 0; | |
- p = begin; | |
- if(p[1] == '\n') | |
- p++; | |
- if(p[1] != '<' || strchr(" /\n\t\\", p[2])) | |
- return 0; | |
- tag = p + 2; | |
- p += 2; | |
- for(; !strchr(" >", *p); p++); | |
- tagend = p; | |
- while((p = strstr(p, "\n</")) && p < end) { | |
- p += 3; | |
- if(strncmp(p, tag, tagend - tag) == 0 && p[tagend - tag] == '>… | |
- p++; | |
- fwrite(begin, sizeof(char), p - begin + tagend - tag, … | |
- puts("\n"); | |
- return -(p - begin + tagend - tag); | |
- } | |
- } | |
- return 0; | |
-} | |
- | |
-int | |
-dolineprefix(const char *begin, const char *end, int newblock) { | |
- unsigned int i, j, l; | |
- char *buffer; | |
- const char *p; | |
- | |
- if(newblock) | |
- p = begin; | |
- else if(*begin == '\n') | |
- p = begin + 1; | |
- else | |
- return 0; | |
- for(i = 0; i < LENGTH(lineprefix); i++) { | |
- l = strlen(lineprefix[i].search); | |
- if(end - p < l) | |
- continue; | |
- if(strncmp(lineprefix[i].search, p, l)) | |
- continue; | |
- if(*begin == '\n') | |
- fputc('\n', stdout); | |
- fputs(lineprefix[i].before, stdout); | |
- if(lineprefix[i].search[l-1] == '\n') { | |
- fputc('\n', stdout); | |
- return l; | |
- } | |
- if(!(buffer = malloc(BUFFERSIZE))) | |
- eprint("Malloc failed."); | |
- buffer[0] = '\0'; | |
- for(j = 0, p += l; p < end; p++, j++) { | |
- ADDC(buffer, j) = *p; | |
- if(*p == '\n' && p + l < end) { | |
- if(strncmp(lineprefix[i].search, p + 1, l) != … | |
- break; | |
- p += l; | |
- } | |
- } | |
- ADDC(buffer, j) = '\0'; | |
- if(lineprefix[i].process) | |
- process(buffer, buffer + strlen(buffer), lineprefix[i]… | |
- else | |
- hprint(buffer, buffer + strlen(buffer)); | |
- puts(lineprefix[i].after); | |
- free(buffer); | |
- return -(p - begin); | |
- } | |
- return 0; | |
-} | |
- | |
-int | |
-dolink(const char *begin, const char *end, int newblock) { | |
- int img; | |
- const char *desc, *link, *p, *q, *descend, *linkend; | |
- | |
- if(*begin == '[') | |
- img = 0; | |
- else if(strncmp(begin, ") || p > end) | |
- return 0; | |
- for(q = strstr(desc, ") || p > end) | |
- return 0; | |
- descend = p; | |
- link = p + 2; | |
- if(!(p = strstr(link, ")")) || p > end) | |
- return 0; | |
- linkend = p; | |
- if(img) { | |
- fputs("<img src=\"", stdout); | |
- hprint(link, linkend); | |
- fputs("\" alt=\"", stdout); | |
- hprint(desc, descend); | |
- fputs("\" />", stdout); | |
- } | |
- else { | |
- fputs("<a href=\"", stdout); | |
- hprint(link, linkend); | |
- fputs("\">", stdout); | |
- process(desc, descend, 0); | |
- fputs("</a>", stdout); | |
- } | |
- return p + 1 - begin; | |
-} | |
- | |
-int | |
-dolist(const char *begin, const char *end, int newblock) { | |
- unsigned int i, j, indent, run, ul, isblock; | |
- const char *p, *q; | |
- char *buffer; | |
- | |
- isblock = 0; | |
- if(newblock) | |
- p = begin; | |
- else if(*begin == '\n') | |
- p = begin + 1; | |
- else | |
- return 0; | |
- q = p; | |
- if(*p == '-' || *p == '*' || *p == '+') | |
- ul = 1; | |
- else { | |
- ul = 0; | |
- for(; p < end && *p >= '0' && *p <= '9'; p++); | |
- if(p >= end || *p != '.') | |
- return 0; | |
- } | |
- p++; | |
- if(p >= end || !(*p == ' ' || *p == '\t')) | |
- return 0; | |
- for(p++; p != end && (*p == ' ' || *p == '\t'); p++); | |
- indent = p - q; | |
- if(!(buffer = malloc(BUFFERSIZE))) | |
- eprint("Malloc failed."); | |
- if(!newblock) | |
- putchar('\n'); | |
- 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) | |
- break; | |
- else if(p[1] == '\n') { | |
- p++; | |
- ADDC(buffer, i) = '\n'; | |
- i++; | |
- run = 0; | |
- isblock++; | |
- } | |
- q = p + 1; | |
- j = 0; | |
- if(ul && (*q == '-' || *q == '*' || *q == '+')) | |
- j = 1; | |
- else if(!ul) { | |
- for(; q + j != end && q[j] >= '0' && q… | |
- if(q + j == end) | |
- break; | |
- if(j > 0 && q[j] == '.') | |
- j++; | |
- else | |
- j = 0; | |
- } | |
- if(q + indent < end) | |
- for(; (q[j] == ' ' || q[j] == '\t') &&… | |
- if(j == indent) { | |
- ADDC(buffer, i) = '\n'; | |
- i++; | |
- p += indent; | |
- run = 1; | |
- if(*q == ' ' || *q == '\t') | |
- p++; | |
- else | |
- break; | |
- } | |
- } | |
- ADDC(buffer, i) = *p; | |
- } | |
- ADDC(buffer, i) = '\0'; | |
- fputs("<li>", stdout); | |
- process(buffer, buffer + i, isblock > 1 || (isblock == 1 && ru… | |
- fputs("</li>\n", stdout); | |
- } | |
- fputs(ul ? "</ul>\n" : "</ol>\n", stdout); | |
- free(buffer); | |
- p--; | |
- while(*(--p) == '\n'); | |
- return -(p - begin + 1); | |
-} | |
- | |
-int | |
-doparagraph(const char *begin, const char *end, int newblock) { | |
- const char *p; | |
- | |
- if(!newblock) | |
- return 0; | |
- p = strstr(begin, "\n\n"); | |
- if(!p || p > end) | |
- p = end; | |
- if(p - begin <= 1) | |
- return 0; | |
- fputs("<p>\n", stdout); | |
- process(begin, p, 0); | |
- fputs("</p>\n", stdout); | |
- return -(p - begin); | |
-} | |
- | |
-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) | |
- fputs(insert[i][1], stdout); | |
- for(i = 0; i < LENGTH(replace); i++) { | |
- l = strlen(replace[i][0]); | |
- if(end - begin < l) | |
- continue; | |
- if(strncmp(replace[i][0], begin, l) == 0) { | |
- fputs(replace[i][1], stdout); | |
- return l; | |
- } | |
- } | |
- return 0; | |
-} | |
- | |
-int | |
-doshortlink(const char *begin, const char *end, int newblock) { | |
- const char *p, *c; | |
- int ismail = 0; | |
- | |
- if(*begin != '<') | |
- return 0; | |
- for(p = begin + 1; p != end; p++) { | |
- switch(*p) { | |
- case ' ': | |
- case '\t': | |
- case '\n': | |
- return 0; | |
- case '#': | |
- case ':': | |
- ismail = -1; | |
- break; | |
- case '@': | |
- if(ismail == 0) | |
- ismail = 1; | |
- break; | |
- case '>': | |
- if(ismail == 0) | |
- return 0; | |
- fputs("<a href=\"", stdout); | |
- if(ismail == 1) { | |
- /* mailto: */ | |
- fputs("mailto:", stdo… | |
- for(c = begin + 1; *c != '>'; c++) | |
- printf("&#%u;", *c); | |
- fputs("\">", stdout); | |
- for(c = begin + 1; *c != '>'; c++) | |
- printf("&#%u;", *c); | |
- } | |
- else { | |
- hprint(begin + 1, p); | |
- fputs("\">", stdout); | |
- hprint(begin + 1, p); | |
- } | |
- fputs("</a>", stdout); | |
- return p - begin + 1; | |
- } | |
- } | |
- return 0; | |
-} | |
- | |
-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++) { | |
- l = strlen(surround[i].search); | |
- if(end - begin < 2*l || strncmp(begin, surround[i].search, l) … | |
- continue; | |
- start = begin + l; | |
- p = start - 1; | |
- do { | |
- p = strstr(p + 1, surround[i].search); | |
- } while(p && p[-1] == '\\'); | |
- if(!p || p >= end || | |
- !(stop = strstr(start, surround[i].search)) ||… | |
- continue; | |
- fputs(surround[i].before, stdout); | |
- if(surround[i].process) | |
- process(start, stop, 0); | |
- else | |
- hprint(start, stop); | |
- fputs(surround[i].after, stdout); | |
- return stop - begin + l; | |
- } | |
- return 0; | |
-} | |
- | |
-int | |
-dounderline(const char *begin, const char *end, int newblock) { | |
- unsigned int i, j, l; | |
- const char *p; | |
- | |
- if(!newblock) | |
- return 0; | |
- p = begin; | |
- for(l = 0; p + l != end && p[l] != '\n'; l++); | |
- p += l + 1; | |
- 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… | |
- if(j >= l) { | |
- fputs(underline[i].before, stdout); | |
- if(underline[i].process) | |
- process(begin, begin + l, 0); | |
- else | |
- hprint(begin, begin + l); | |
- fputs(underline[i].after, stdout); | |
- return -(j + p - begin); | |
- } | |
- } | |
- return 0; | |
-} | |
- | |
-void | |
-hprint(const char *begin, const char *end) { | |
- const char *p; | |
- | |
- for(p = begin; p != end; p++) { | |
- if(*p == '&') | |
- fputs("&", stdout); | |
- else if(*p == '"') | |
- fputs(""", stdout); | |
- else if(*p == '>') | |
- fputs(">", stdout); | |
- else if(*p == '<') | |
- fputs("<", stdout); | |
- else | |
- putchar(*p); | |
- } | |
-} | |
- | |
-void | |
-process(const char *begin, const char *end, int newblock) { | |
- const char *p, *q; | |
- int affected; | |
- unsigned int i; | |
- | |
- for(p = begin; p != end;) { | |
- if(newblock) | |
- while(*p == '\n') | |
- if (++p == end) | |
- return; | |
- affected = 0; | |
- for(i = 0; i < LENGTH(parsers) && affected == 0; i++) | |
- affected = parsers[i](p, end, newblock); | |
- p += abs(affected); | |
- if(!affected) { | |
- if(nohtml) | |
- hprint(p, p + 1); | |
- else | |
- putchar(*p); | |
- p++; | |
- } | |
- for(q = p; q != end && *q == '\n'; q++); | |
- if(q == end) | |
- return; | |
- else if(p[0] == '\n' && p + 1 != end && p[1] == '\n') | |
- newblock = 1; | |
- else | |
- newblock = affected < 0; | |
- } | |
-} | |
+#include "smu.h" | |
int | |
main(int argc, char *argv[]) { | |
- char *buffer; | |
- int s; | |
- unsigned long len, bsize; | |
+ int no = 0; | |
+ FILE *in = stdin; | |
- source = stdin; | |
if(argc > 1 && strcmp("-v", argv[1]) == 0) | |
eprint("simple markup %s (C) Enno Boland\n",VERSION); | |
else if(argc > 1 && strcmp("-h", argv[1]) == 0) | |
eprint("Usage %s [-n] [file]\n -n escape html strictly\n",argv… | |
if(argc > 1 && strcmp("-n", argv[1]) == 0) | |
- nohtml = 1; | |
- if(argc > 1 + nohtml && strcmp("-", argv[1 + nohtml]) != 0 | |
- && !(source = fopen(argv[1 + nohtml],"r"))) | |
- eprint("Cannot open file `%s`\n",argv[1 + nohtml]); | |
- bsize = 2 * BUFFERSIZE; | |
- if(!(buffer = malloc(bsize))) | |
- eprint("Malloc failed."); | |
- len = 0; | |
- while((s = fread(buffer + len, 1, BUFFERSIZE, source))) { | |
- len += s; | |
- if(BUFFERSIZE + len + 1 > bsize) { | |
- bsize += BUFFERSIZE; | |
- if(!(buffer = realloc(buffer, bsize))) | |
- eprint("Malloc failed."); | |
- } | |
- } | |
- buffer[len] = '\0'; | |
- process(buffer, buffer + len, 1); | |
- fclose(source); | |
- free(buffer); | |
+ no = 1; | |
+ if(argc > 1 + no | |
+ && strcmp("-", argv[1 + no]) != 0 | |
+ && !(in = fopen(argv[1 + no],"r"))) | |
+ eprint("Cannot open file `%s`\n",argv[1 + no]); | |
+ smu_convert(stdout, in, no); | |
+ fclose(in); | |
return EXIT_SUCCESS; | |
} | |
diff --git a/smu.h b/smu.h | |
@@ -0,0 +1,19 @@ | |
+/* libsmu - simple markup library | |
+ * Copyright (C) <2007, 2008> Enno Boland <g s01 de> | |
+ * | |
+ * See LICENSE for further informations | |
+ */ | |
+#include <stdio.h> | |
+ | |
+/** | |
+ * Converts contents of a simple markup stream (in) and prints them to out. | |
+ * If suppresshtml == 1, it will create plain text of the simple markup instead | |
+ * of HTML. | |
+ * | |
+ * Returns 0 on success. | |
+ */ | |
+int smu_convert(FILE *out, FILE *in, int suppresshtml); | |
+ | |
+/** utility */ | |
+void eprint(const char *format, ...); | |
+ |