commonmarker 0.23.6 → 1.0.0.pre

data/ext/commonmarker/ext_scanners.h

@@ -1,24 +0,0 @@
-#include "chunk.h"
-#include "cmark-gfm.h"
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-bufsize_t _ext_scan_at(bufsize_t (*scanner)(const unsigned char *),
-                       unsigned char *ptr, int len, bufsize_t offset);
-bufsize_t _scan_table_start(const unsigned char *p);
-bufsize_t _scan_table_cell(const unsigned char *p);
-bufsize_t _scan_table_cell_end(const unsigned char *p);
-bufsize_t _scan_table_row_end(const unsigned char *p);
-bufsize_t _scan_tasklist(const unsigned char *p);
-
-#define scan_table_start(c, l, n) _ext_scan_at(&_scan_table_start, c, l, n)
-#define scan_table_cell(c, l, n) _ext_scan_at(&_scan_table_cell, c, l, n)
-#define scan_table_cell_end(c, l, n) _ext_scan_at(&_scan_table_cell_end, c, l, n)
-#define scan_table_row_end(c, l, n) _ext_scan_at(&_scan_table_row_end, c, l, n)
-#define scan_tasklist(c, l, n) _ext_scan_at(&_scan_tasklist, c, l, n)
-
-#ifdef __cplusplus
-}
-#endif

data/ext/commonmarker/footnotes.c

@@ -1,63 +0,0 @@
-#include "cmark-gfm.h"
-#include "parser.h"
-#include "footnotes.h"
-#include "inlines.h"
-#include "chunk.h"
-
-static void footnote_free(cmark_map *map, cmark_map_entry *_ref) {
-  cmark_footnote *ref = (cmark_footnote *)_ref;
-  cmark_mem *mem = map->mem;
-  if (ref != NULL) {
-    mem->free(ref->entry.label);
-    if (ref->node)
-      cmark_node_free(ref->node);
-    mem->free(ref);
-  }
-}
-
-void cmark_footnote_create(cmark_map *map, cmark_node *node) {
-  cmark_footnote *ref;
-  unsigned char *reflabel = normalize_map_label(map->mem, &node->as.literal);
-
-  /* empty footnote name, or composed from only whitespace */
-  if (reflabel == NULL)
-    return;
-
-  assert(map->sorted == NULL);
-
-  ref = (cmark_footnote *)map->mem->calloc(1, sizeof(*ref));
-  ref->entry.label = reflabel;
-  ref->node = node;
-  ref->entry.age = map->size;
-  ref->entry.next = map->refs;
-
-  map->refs = (cmark_map_entry *)ref;
-  map->size++;
-}
-
-cmark_map *cmark_footnote_map_new(cmark_mem *mem) {
-  return cmark_map_new(mem, footnote_free);
-}
-
-// Before calling `cmark_map_free` on a map with `cmark_footnotes`, first
-// unlink all of the footnote nodes before freeing their memory.
-//
-// Sometimes, two (unused) footnote nodes can end up referencing each other,
-// which as they get freed up by calling `cmark_map_free` -> `footnote_free` ->
-// etc, can lead to a use-after-free error.
-//
-// Better to `unlink` every footnote node first, setting their next, prev, and
-// parent pointers to NULL, and only then walk thru & free them up.
-void cmark_unlink_footnotes_map(cmark_map *map) {
-  cmark_map_entry *ref;
-  cmark_map_entry *next;
-
-  ref = map->refs;
-  while(ref) {
-    next = ref->next;
-    if (((cmark_footnote *)ref)->node) {
-      cmark_node_unlink(((cmark_footnote *)ref)->node);
-    }
-    ref = next;
-  }
-}

data/ext/commonmarker/footnotes.h

@@ -1,27 +0,0 @@
-#ifndef CMARK_FOOTNOTES_H
-#define CMARK_FOOTNOTES_H
-
-#include "map.h"
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-struct cmark_footnote {
-  cmark_map_entry entry;
-  cmark_node *node;
-  unsigned int ix;
-};
-
-typedef struct cmark_footnote cmark_footnote;
-
-void cmark_footnote_create(cmark_map *map, cmark_node *node);
-cmark_map *cmark_footnote_map_new(cmark_mem *mem);
-
-void cmark_unlink_footnotes_map(cmark_map *map);
-
-#ifdef __cplusplus
-}
-#endif
-
-#endif

data/ext/commonmarker/houdini.h

@@ -1,57 +0,0 @@
-#ifndef CMARK_HOUDINI_H
-#define CMARK_HOUDINI_H
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-#include <stdint.h>
-#include "config.h"
-#include "buffer.h"
-
-#ifdef HAVE___BUILTIN_EXPECT
-#define likely(x) __builtin_expect((x), 1)
-#define unlikely(x) __builtin_expect((x), 0)
-#else
-#define likely(x) (x)
-#define unlikely(x) (x)
-#endif
-
-#ifdef HOUDINI_USE_LOCALE
-#define _isxdigit(c) isxdigit(c)
-#define _isdigit(c) isdigit(c)
-#else
-/*
- * Helper _isdigit methods -- do not trust the current locale
- * */
-#define _isxdigit(c) (strchr("0123456789ABCDEFabcdef", (c)) != NULL)
-#define _isdigit(c) ((c) >= '0' && (c) <= '9')
-#endif
-
-#define HOUDINI_ESCAPED_SIZE(x) (((x)*12) / 10)
-#define HOUDINI_UNESCAPED_SIZE(x) (x)
-
-CMARK_GFM_EXPORT
-bufsize_t houdini_unescape_ent(cmark_strbuf *ob, const uint8_t *src,
-                                      bufsize_t size);
-CMARK_GFM_EXPORT
-int houdini_escape_html(cmark_strbuf *ob, const uint8_t *src,
-                               bufsize_t size);
-CMARK_GFM_EXPORT
-int houdini_escape_html0(cmark_strbuf *ob, const uint8_t *src,
-                                bufsize_t size, int secure);
-CMARK_GFM_EXPORT
-int houdini_unescape_html(cmark_strbuf *ob, const uint8_t *src,
-                                 bufsize_t size);
-CMARK_GFM_EXPORT
-void houdini_unescape_html_f(cmark_strbuf *ob, const uint8_t *src,
-                                    bufsize_t size);
-CMARK_GFM_EXPORT
-int houdini_escape_href(cmark_strbuf *ob, const uint8_t *src,
-                               bufsize_t size);
-
-#ifdef __cplusplus
-}
-#endif
-
-#endif

data/ext/commonmarker/houdini_href_e.c

@@ -1,100 +0,0 @@
-#include <assert.h>
-#include <stdio.h>
-#include <string.h>
-
-#include "houdini.h"
-
-/*
- * The following characters will not be escaped:
- *
- *		-_.+!*'(),%#@?=;:/,+&$~ alphanum
- *
- * Note that this character set is the addition of:
- *
- *	- The characters which are safe to be in an URL
- *	- The characters which are *not* safe to be in
- *	an URL because they are RESERVED characters.
- *
- * We assume (lazily) that any RESERVED char that
- * appears inside an URL is actually meant to
- * have its native function (i.e. as an URL
- * component/separator) and hence needs no escaping.
- *
- * There are two exceptions: the chacters & (amp)
- * and ' (single quote) do not appear in the table.
- * They are meant to appear in the URL as components,
- * yet they require special HTML-entity escaping
- * to generate valid HTML markup.
- *
- * All other characters will be escaped to %XX.
- *
- */
-static const char HREF_SAFE[] = {
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1,
-    0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
-    1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-};
-
-int houdini_escape_href(cmark_strbuf *ob, const uint8_t *src, bufsize_t size) {
-  static const uint8_t hex_chars[] = "0123456789ABCDEF";
-  bufsize_t i = 0, org;
-  uint8_t hex_str[3];
-
-  hex_str[0] = '%';
-
-  while (i < size) {
-    org = i;
-    while (i < size && HREF_SAFE[src[i]] != 0)
-      i++;
-
-    if (likely(i > org))
-      cmark_strbuf_put(ob, src + org, i - org);
-
-    /* escaping */
-    if (i >= size)
-      break;
-
-    switch (src[i]) {
-    /* amp appears all the time in URLs, but needs
-     * HTML-entity escaping to be inside an href */
-    case '&':
-      cmark_strbuf_puts(ob, "&amp;");
-      break;
-
-    /* the single quote is a valid URL character
-     * according to the standard; it needs HTML
-     * entity escaping too */
-    case '\'':
-      cmark_strbuf_puts(ob, "&#x27;");
-      break;
-
-/* the space can be escaped to %20 or a plus
- * sign. we're going with the generic escape
- * for now. the plus thing is more commonly seen
- * when building GET strings */
-#if 0
-		case ' ':
-			cmark_strbuf_putc(ob, '+');
-			break;
-#endif
-
-    /* every other character goes with a %XX escaping */
-    default:
-      hex_str[1] = hex_chars[(src[i] >> 4) & 0xF];
-      hex_str[2] = hex_chars[src[i] & 0xF];
-      cmark_strbuf_put(ob, hex_str, 3);
-    }
-
-    i++;
-  }
-
-  return 1;
-}

data/ext/commonmarker/houdini_html_e.c

@@ -1,66 +0,0 @@
-#include <assert.h>
-#include <stdio.h>
-#include <string.h>
-
-#include "houdini.h"
-
-/**
- * According to the OWASP rules:
- *
- * & --> &amp;
- * < --> &lt;
- * > --> &gt;
- * " --> &quot;
- * ' --> &#x27;     &apos; is not recommended
- * / --> &#x2F;     forward slash is included as it helps end an HTML entity
- *
- */
-static const char HTML_ESCAPE_TABLE[] = {
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 2, 3, 0, 0, 0, 0, 0, 0, 0, 4,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-};
-
-static const char *HTML_ESCAPES[] = {"",      "&quot;", "&amp;", "&#39;",
-                                     "&#47;", "&lt;",   "&gt;"};
-
-int houdini_escape_html0(cmark_strbuf *ob, const uint8_t *src, bufsize_t size,
-                         int secure) {
-  bufsize_t i = 0, org, esc = 0;
-
-  while (i < size) {
-    org = i;
-    while (i < size && (esc = HTML_ESCAPE_TABLE[src[i]]) == 0)
-      i++;
-
-    if (i > org)
-      cmark_strbuf_put(ob, src + org, i - org);
-
-    /* escaping */
-    if (unlikely(i >= size))
-      break;
-
-    /* The forward slash and single quote are only escaped in secure mode */
-    if ((src[i] == '/' || src[i] == '\'') && !secure) {
-      cmark_strbuf_putc(ob, src[i]);
-    } else {
-      cmark_strbuf_puts(ob, HTML_ESCAPES[esc]);
-    }
-
-    i++;
-  }
-
-  return 1;
-}
-
-int houdini_escape_html(cmark_strbuf *ob, const uint8_t *src, bufsize_t size) {
-  return houdini_escape_html0(ob, src, size, 1);
-}

data/ext/commonmarker/houdini_html_u.c

@@ -1,149 +0,0 @@
-#include <assert.h>
-#include <stdio.h>
-#include <string.h>
-
-#include "buffer.h"
-#include "houdini.h"
-#include "utf8.h"
-#include "entities.inc"
-
-/* Binary tree lookup code for entities added by JGM */
-
-static const unsigned char *S_lookup(int i, int low, int hi,
-                                     const unsigned char *s, int len) {
-  int j;
-  int cmp =
-      strncmp((const char *)s, (const char *)cmark_entities[i].entity, len);
-  if (cmp == 0 && cmark_entities[i].entity[len] == 0) {
-    return (const unsigned char *)cmark_entities[i].bytes;
-  } else if (cmp <= 0 && i > low) {
-    j = i - ((i - low) / 2);
-    if (j == i)
-      j -= 1;
-    return S_lookup(j, low, i - 1, s, len);
-  } else if (cmp > 0 && i < hi) {
-    j = i + ((hi - i) / 2);
-    if (j == i)
-      j += 1;
-    return S_lookup(j, i + 1, hi, s, len);
-  } else {
-    return NULL;
-  }
-}
-
-static const unsigned char *S_lookup_entity(const unsigned char *s, int len) {
-  return S_lookup(CMARK_NUM_ENTITIES / 2, 0, CMARK_NUM_ENTITIES - 1, s, len);
-}
-
-bufsize_t houdini_unescape_ent(cmark_strbuf *ob, const uint8_t *src,
-                               bufsize_t size) {
-  bufsize_t i = 0;
-
-  if (size >= 3 && src[0] == '#') {
-    int codepoint = 0;
-    int num_digits = 0;
-
-    if (_isdigit(src[1])) {
-      for (i = 1; i < size && _isdigit(src[i]); ++i) {
-        codepoint = (codepoint * 10) + (src[i] - '0');
-
-        if (codepoint >= 0x110000) {
-          // Keep counting digits but
-          // avoid integer overflow.
-          codepoint = 0x110000;
-        }
-      }
-
-      num_digits = i - 1;
-    }
-
-    else if (src[1] == 'x' || src[1] == 'X') {
-      for (i = 2; i < size && _isxdigit(src[i]); ++i) {
-        codepoint = (codepoint * 16) + ((src[i] | 32) % 39 - 9);
-
-        if (codepoint >= 0x110000) {
-          // Keep counting digits but
-          // avoid integer overflow.
-          codepoint = 0x110000;
-        }
-      }
-
-      num_digits = i - 2;
-    }
-
-    if (num_digits >= 1 && num_digits <= 8 && i < size && src[i] == ';') {
-      if (codepoint == 0 || (codepoint >= 0xD800 && codepoint < 0xE000) ||
-          codepoint >= 0x110000) {
-        codepoint = 0xFFFD;
-      }
-      cmark_utf8proc_encode_char(codepoint, ob);
-      return i + 1;
-    }
-  }
-
-  else {
-    if (size > CMARK_ENTITY_MAX_LENGTH)
-      size = CMARK_ENTITY_MAX_LENGTH;
-
-    for (i = CMARK_ENTITY_MIN_LENGTH; i < size; ++i) {
-      if (src[i] == ' ')
-        break;
-
-      if (src[i] == ';') {
-        const unsigned char *entity = S_lookup_entity(src, i);
-
-        if (entity != NULL) {
-          cmark_strbuf_puts(ob, (const char *)entity);
-          return i + 1;
-        }
-
-        break;
-      }
-    }
-  }
-
-  return 0;
-}
-
-int houdini_unescape_html(cmark_strbuf *ob, const uint8_t *src,
-                          bufsize_t size) {
-  bufsize_t i = 0, org, ent;
-
-  while (i < size) {
-    org = i;
-    while (i < size && src[i] != '&')
-      i++;
-
-    if (likely(i > org)) {
-      if (unlikely(org == 0)) {
-        if (i >= size)
-          return 0;
-
-        cmark_strbuf_grow(ob, HOUDINI_UNESCAPED_SIZE(size));
-      }
-
-      cmark_strbuf_put(ob, src + org, i - org);
-    }
-
-    /* escaping */
-    if (i >= size)
-      break;
-
-    i++;
-
-    ent = houdini_unescape_ent(ob, src + i, size - i);
-    i += ent;
-
-    /* not really an entity */
-    if (ent == 0)
-      cmark_strbuf_putc(ob, '&');
-  }
-
-  return 1;
-}
-
-void houdini_unescape_html_f(cmark_strbuf *ob, const uint8_t *src,
-                             bufsize_t size) {
-  if (!houdini_unescape_html(ob, src, size))
-    cmark_strbuf_put(ob, src, size);
-}