aboutsummaryrefslogtreecommitdiffstats
path: root/src
diff options
context:
space:
mode:
Diffstat (limited to 'src')
-rw-r--r--src/content-parser.c271
-rw-r--r--src/directives.c62
-rw-r--r--src/directives.h36
-rw-r--r--src/error.c3
-rw-r--r--src/error.h1
-rw-r--r--src/utils/trie.c2
6 files changed, 356 insertions, 19 deletions
diff --git a/src/content-parser.c b/src/content-parser.c
index ccb96ef..8dceb8b 100644
--- a/src/content-parser.c
+++ b/src/content-parser.c
@@ -15,6 +15,7 @@
#include "utils/utils.h"
#include "content-parser.h"
+#include "directives.h"
// this is a half ass implementation of a markdown-like syntax. bugs are
// expected. feel free to improve the parser and add new features.
@@ -77,7 +78,7 @@ blogc_htmlentities(const char *str)
typedef enum {
CONTENT_START_LINE = 1,
- CONTENT_EXCERPT,
+ CONTENT_EXCERPT_OR_DIRECTIVE,
CONTENT_EXCERPT_END,
CONTENT_HEADER,
CONTENT_HEADER_TITLE_START,
@@ -98,6 +99,18 @@ typedef enum {
CONTENT_ORDERED_LIST_SPACE,
CONTENT_ORDERED_LIST_START,
CONTENT_ORDERED_LIST_END,
+ CONTENT_DIRECTIVE_NAME_START,
+ CONTENT_DIRECTIVE_NAME,
+ CONTENT_DIRECTIVE_COLON,
+ CONTENT_DIRECTIVE_ARGUMENT_START,
+ CONTENT_DIRECTIVE_ARGUMENT,
+ CONTENT_DIRECTIVE_PARAM_PREFIX_START,
+ CONTENT_DIRECTIVE_PARAM_PREFIX,
+ CONTENT_DIRECTIVE_PARAM_KEY_START,
+ CONTENT_DIRECTIVE_PARAM_KEY,
+ CONTENT_DIRECTIVE_PARAM_VALUE_START,
+ CONTENT_DIRECTIVE_PARAM_VALUE,
+ CONTENT_DIRECTIVE_PARAM_END,
CONTENT_PARAGRAPH,
CONTENT_PARAGRAPH_END,
} blogc_content_parser_state_t;
@@ -447,6 +460,11 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
char *parsed = NULL;
char *slug = NULL;
+ char *directive_name = NULL;
+ char *directive_argument = NULL;
+ char *directive_key = NULL;
+ b_trie_t *directive_params = NULL;
+
// this isn't empty because we need some reasonable default value in the
// unlikely case that we need to print some line ending before evaluating
// the "real" value.
@@ -497,11 +515,9 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
break;
start = current;
if (c == '.') {
- if (end_excerpt != NULL) {
- eend = rv->len; // fuck it
- state = CONTENT_EXCERPT;
- break;
- }
+ eend = rv->len; // fuck it
+ state = CONTENT_EXCERPT_OR_DIRECTIVE;
+ break;
}
if (c == '#') {
header_level = 1;
@@ -536,26 +552,31 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
state = CONTENT_PARAGRAPH;
break;
- case CONTENT_EXCERPT:
- if (end_excerpt != NULL) {
- if (c == '.')
- break;
- if (c == '\n' || c == '\r') {
- state = CONTENT_EXCERPT_END;
- break;
+ case CONTENT_EXCERPT_OR_DIRECTIVE:
+ if (c == '.')
+ break;
+ if ((c == ' ' || c == '\t') && current - start == 2) {
+ state = CONTENT_DIRECTIVE_NAME_START;
+ if (is_last) {
+ state = CONTENT_PARAGRAPH;
+ continue;
}
+ break;
+ }
+ if (c == '\n' || c == '\r') {
+ state = CONTENT_EXCERPT_END;
+ break;
}
eend = 0;
state = CONTENT_PARAGRAPH;
break;
case CONTENT_EXCERPT_END:
- if (end_excerpt != NULL) {
- if (c == '\n' || c == '\r') {
+ if (c == '\n' || c == '\r') {
+ if (end_excerpt != NULL)
*end_excerpt = eend;
- state = CONTENT_START_LINE;
- break;
- }
+ state = CONTENT_START_LINE;
+ break;
}
eend = 0;
state = CONTENT_PARAGRAPH_END;
@@ -981,6 +1002,214 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
}
break;
+ case CONTENT_DIRECTIVE_NAME_START:
+ if (is_last) {
+ state = CONTENT_PARAGRAPH;
+ continue;
+ }
+ if (c >= 'a' && c <= 'z') {
+ start2 = current;
+ state = CONTENT_DIRECTIVE_NAME;
+ break;
+ }
+ state = CONTENT_PARAGRAPH;
+ break;
+
+ case CONTENT_DIRECTIVE_NAME:
+ if (is_last) {
+ state = CONTENT_PARAGRAPH;
+ continue;
+ }
+ if ((c >= 'a' && c <= 'z') || (c >= '0' && c <= '9') || c == '_')
+ break;
+ if (c == ':') {
+ end = current;
+ state = CONTENT_DIRECTIVE_COLON;
+ break;
+ }
+ state = CONTENT_PARAGRAPH;
+ break;
+
+ case CONTENT_DIRECTIVE_COLON:
+ if (c == ':') {
+ free(directive_name);
+ directive_name = b_strndup(src + start2, end - start2);
+ state = CONTENT_DIRECTIVE_ARGUMENT_START;
+ if (is_last) {
+ state = CONTENT_DIRECTIVE_PARAM_END;
+ continue;
+ }
+ break;
+ }
+ if (is_last) {
+ state = CONTENT_PARAGRAPH;
+ continue;
+ }
+ state = CONTENT_PARAGRAPH;
+ break;
+
+ case CONTENT_DIRECTIVE_ARGUMENT_START:
+ if (c == ' ' || c == '\t') {
+ if (is_last) {
+ state = CONTENT_DIRECTIVE_PARAM_END;
+ continue;
+ }
+ break;
+ }
+ if (c == '\n' || c == '\r' || is_last) {
+ state = CONTENT_DIRECTIVE_PARAM_PREFIX_START;
+ directive_argument = NULL;
+ if (is_last) {
+ state = CONTENT_DIRECTIVE_PARAM_END;
+ continue;
+ }
+ else
+ start2 = current + 1;
+ break;
+ }
+ start2 = current;
+ state = CONTENT_DIRECTIVE_ARGUMENT;
+ break;
+
+ case CONTENT_DIRECTIVE_ARGUMENT:
+ if (c == '\n' || c == '\r' || is_last) {
+ state = CONTENT_DIRECTIVE_PARAM_PREFIX_START;
+ end = is_last && c != '\n' && c != '\r' ? src_len :
+ (real_end != 0 ? real_end : current);
+ free(directive_argument);
+ directive_argument = b_strndup(src + start2, end - start2);
+ if (is_last) {
+ state = CONTENT_DIRECTIVE_PARAM_END;
+ continue;
+ }
+ else
+ start2 = current + 1;
+ }
+ break;
+
+ case CONTENT_DIRECTIVE_PARAM_PREFIX_START:
+ if (is_last) {
+ state = CONTENT_PARAGRAPH;
+ continue;
+ }
+ if (c == ' ' || c == '\t')
+ break;
+ if (c == '\n' || c == '\r') {
+ state = CONTENT_DIRECTIVE_PARAM_END;
+ continue;
+ }
+ prefix = b_strndup(src + start2, current - start2);
+ state = CONTENT_DIRECTIVE_PARAM_PREFIX;
+ current--;
+ break;
+
+ case CONTENT_DIRECTIVE_PARAM_PREFIX:
+ if (c == ' ' || c == '\t')
+ break;
+ if (c == ':' && b_str_starts_with(src + start2, prefix)) {
+ state = CONTENT_DIRECTIVE_PARAM_KEY_START;
+ break;
+ }
+ state = CONTENT_PARAGRAPH;
+ if (is_last)
+ continue;
+ break;
+
+ case CONTENT_DIRECTIVE_PARAM_KEY_START:
+ if (is_last) {
+ state = CONTENT_PARAGRAPH;
+ continue;
+ }
+ if (c >= 'a' && c <= 'z') {
+ start2 = current;
+ state = CONTENT_DIRECTIVE_PARAM_KEY;
+ break;
+ }
+ state = CONTENT_PARAGRAPH;
+ break;
+
+ case CONTENT_DIRECTIVE_PARAM_KEY:
+ if (is_last) {
+ state = CONTENT_PARAGRAPH;
+ continue;
+ }
+ if ((c >= 'a' && c <= 'z') || (c >= '0' && c <= '9') || c == '_')
+ break;
+ if (c == ':') {
+ free(directive_key);
+ directive_key = b_strndup(src + start2, current - start2);
+ state = CONTENT_DIRECTIVE_PARAM_VALUE_START;
+ break;
+ }
+ state = CONTENT_PARAGRAPH;
+ break;
+
+ case CONTENT_DIRECTIVE_PARAM_VALUE_START:
+ if (is_last) {
+ state = CONTENT_PARAGRAPH;
+ continue;
+ }
+ if (c == ' ' || c == '\t')
+ break;
+ start2 = current;
+ state = CONTENT_DIRECTIVE_PARAM_VALUE;
+ break;
+
+ case CONTENT_DIRECTIVE_PARAM_VALUE:
+ if (c == '\n' || c == '\r' || is_last) {
+ state = CONTENT_DIRECTIVE_PARAM_END;
+ end = is_last && c != '\n' && c != '\r' ? src_len :
+ (real_end != 0 ? real_end : current);
+ if (directive_params == NULL)
+ directive_params = b_trie_new(free);
+ b_trie_insert(directive_params, directive_key,
+ b_strndup(src + start2, end - start2));
+ free(directive_key);
+ directive_key = NULL;
+ if (!is_last)
+ start2 = current + 1;
+ }
+ if (!is_last)
+ break;
+
+ case CONTENT_DIRECTIVE_PARAM_END:
+ if (c == '\n' || c == '\r' || is_last) {
+ // FIXME: handle errors in the rest of the parser.
+ blogc_error_t *err = NULL;
+ blogc_directive_ctx_t *ctx = b_malloc(
+ sizeof(blogc_directive_ctx_t));
+ ctx->name = directive_name;
+ ctx->argument = directive_argument;
+ ctx->params = directive_params;
+ ctx->eol = line_ending;
+ char *rv_d = blogc_directive_loader(ctx, &err);
+ free(ctx);
+ blogc_error_print(err);
+ if (rv_d != NULL)
+ b_string_append(rv, rv_d);
+ free(rv_d);
+ state = CONTENT_START_LINE;
+ start = current;
+ free(directive_name);
+ directive_name = NULL;
+ free(directive_argument);
+ directive_argument = NULL;
+ b_trie_free(directive_params);
+ directive_params = NULL;
+ free(prefix);
+ prefix = NULL;
+ break;
+ }
+ if (c == ' ' || c == '\t') {
+ start2 = current;
+ state = CONTENT_DIRECTIVE_PARAM_PREFIX;
+ break;
+ }
+ state = CONTENT_PARAGRAPH;
+ if (is_last)
+ continue;
+ break;
+
case CONTENT_PARAGRAPH:
if (c == '\n' || c == '\r' || is_last) {
state = CONTENT_PARAGRAPH_END;
@@ -1012,5 +1241,11 @@ blogc_content_parse(const char *src, size_t *end_excerpt)
current++;
}
+ free(directive_name);
+ free(directive_argument);
+ free(directive_key);
+ b_trie_free(directive_params);
+ free(prefix);
+
return b_string_free(rv, false);
}
diff --git a/src/directives.c b/src/directives.c
new file mode 100644
index 0000000..b538825
--- /dev/null
+++ b/src/directives.c
@@ -0,0 +1,62 @@
+/*
+ * blogc: A blog compiler.
+ * Copyright (C) 2015 Rafael G. Martins <rafael@rafaelmartins.eng.br>
+ *
+ * This program can be distributed under the terms of the BSD License.
+ * See the file LICENSE.
+ */
+
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif /* HAVE_CONFIG_H */
+
+#include <string.h>
+
+#include "utils/utils.h"
+#include "directives.h"
+#include "error.h"
+
+
+const static blogc_directive_t registry[] = {
+ {"youtube", blogc_directive_youtube},
+ {NULL, NULL},
+};
+
+
+char*
+blogc_directive_loader(blogc_directive_ctx_t *ctx, blogc_error_t **err)
+{
+ if (ctx == NULL)
+ return NULL;
+ if (err == NULL || *err != NULL)
+ return NULL;
+ for (unsigned int i = 0; registry[i].name != NULL; i++)
+ if (0 == strcmp(ctx->name, registry[i].name))
+ return registry[i].callback(ctx, err);
+ *err = blogc_error_new_printf(BLOGC_WARNING_CONTENT_PARSER,
+ "Directive not found: %s", ctx->name);
+ return NULL;
+}
+
+
+char*
+blogc_directive_youtube(blogc_directive_ctx_t *ctx, blogc_error_t **err)
+{
+ if (ctx->argument == NULL) {
+ *err = blogc_error_new_printf(BLOGC_WARNING_CONTENT_PARSER,
+ "youtube: video ID must be provided as argument");
+ return NULL;
+ }
+
+ char *width = b_trie_lookup(ctx->params, "width");
+ char *height = b_trie_lookup(ctx->params, "height");
+
+ // using default 16:9 sizes provided by youtube as of 2015-11-04
+ return b_strdup_printf(
+ "<iframe width=\"%s\" height=\"%s\" "
+ "src=\"https://www.youtube.com/embed/%s\" frameborder=\"0\" "
+ "allowfullscreen></iframe>%s",
+ width == NULL ? "560" : width,
+ height == NULL ? "315" : height,
+ ctx->argument, ctx->eol);
+}
diff --git a/src/directives.h b/src/directives.h
new file mode 100644
index 0000000..dd27f03
--- /dev/null
+++ b/src/directives.h
@@ -0,0 +1,36 @@
+/*
+ * blogc: A blog compiler.
+ * Copyright (C) 2015 Rafael G. Martins <rafael@rafaelmartins.eng.br>
+ *
+ * This program can be distributed under the terms of the BSD License.
+ * See the file LICENSE.
+ */
+
+#ifndef _DIRECTIVES_H
+#define _DIRECTIVES_H
+
+#include "utils/utils.h"
+#include "error.h"
+
+typedef struct {
+ const char *name;
+ const char *argument;
+ b_trie_t *params;
+ const char *eol;
+} blogc_directive_ctx_t;
+
+typedef char* (*blogc_directive_func_t)(blogc_directive_ctx_t *ctx,
+ blogc_error_t **err);
+
+typedef struct {
+ const char *name;
+ blogc_directive_func_t callback;
+} blogc_directive_t;
+
+char* blogc_directive_loader(blogc_directive_ctx_t *ctx, blogc_error_t **err);
+
+
+// built-in directives (that are everything we support right now
+char* blogc_directive_youtube(blogc_directive_ctx_t *ctx, blogc_error_t **err);
+
+#endif /* _DIRECTIVES_H */
diff --git a/src/error.c b/src/error.c
index 6256873..59a85a3 100644
--- a/src/error.c
+++ b/src/error.c
@@ -125,6 +125,9 @@ blogc_error_print(blogc_error_t *err)
case BLOGC_WARNING_DATETIME_PARSER:
fprintf(stderr, "blogc: warning: datetime: %s\n", err->msg);
break;
+ case BLOGC_WARNING_CONTENT_PARSER:
+ fprintf(stderr, "blogc: warning: content: %s\n", err->msg);
+ break;
default:
fprintf(stderr, "blogc: error: %s\n", err->msg);
}
diff --git a/src/error.h b/src/error.h
index caa86b0..886d7d3 100644
--- a/src/error.h
+++ b/src/error.h
@@ -17,6 +17,7 @@ typedef enum {
BLOGC_ERROR_TEMPLATE_PARSER,
BLOGC_ERROR_LOADER,
BLOGC_WARNING_DATETIME_PARSER,
+ BLOGC_WARNING_CONTENT_PARSER,
} blogc_error_type_t;
typedef struct {
diff --git a/src/utils/trie.c b/src/utils/trie.c
index b8c1e63..2e6a0d8 100644
--- a/src/utils/trie.c
+++ b/src/utils/trie.c
@@ -110,7 +110,7 @@ clean:
void*
b_trie_lookup(b_trie_t *trie, const char *key)
{
- if (trie->root == NULL || key == NULL)
+ if (trie == NULL || trie->root == NULL || key == NULL)
return NULL;
b_trie_node_t *parent = trie->root;
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2024-11-14 08:59:45 +0000