My personal build of CMark ✏️
Replace invalid characters in XML output
Control characters, U+FFFE and U+FFFF aren't allowed in XML 1.0, so replace them with U+FFFD (replacement character). This doesn't solve the problem how to roundtrip these characters, but at least we don't produce invalid XML. See #365.
2 files changed, 68 insertions, 6 deletions
| Status | File Name | N° Changes | Insertions | Deletions |
| Modified | api_test/main.c | 13 | 11 | 2 |
| Modified | src/xml.c | 61 | 57 | 4 |
diff --git a/api_test/main.c b/api_test/main.c @@ -542,7 +542,10 @@ static void render_xml(test_batch_runner *runner) { static const char markdown[] = "foo *bar*\n" "\n" - "paragraph 2\n" + "control -\x0C-\n" + "fffe -\xEF\xBF\xBE-\n" + "ffff -\xEF\xBF\xBF-\n" + "escape <>&\"\n" "\n" "```\ncode\n```\n"; cmark_node *doc = @@ -559,7 +562,13 @@ static void render_xml(test_batch_runner *runner) { " </emph>\n" " </paragraph>\n" " <paragraph>\n" - " <text xml:space=\"preserve\">paragraph 2</text>\n" + " <text xml:space=\"preserve\">control -" UTF8_REPL "-</text>\n" + " <softbreak />\n" + " <text xml:space=\"preserve\">fffe -" UTF8_REPL "-</text>\n" + " <softbreak />\n" + " <text xml:space=\"preserve\">ffff -" UTF8_REPL "-</text>\n" + " <softbreak />\n" + " <text xml:space=\"preserve\">escape <>&"</text>\n" " </paragraph>\n" " <code_block xml:space=\"preserve\">code\n" "</code_block>\n"
diff --git a/src/xml.c b/src/xml.c @@ -7,16 +7,69 @@ #include "cmark.h" #include "node.h" #include "buffer.h" -#include "houdini.h" #define BUFFER_SIZE 100 #define MAX_INDENT 40 // Functions to convert cmark_nodes to XML strings. -static void escape_xml(cmark_strbuf *dest, const unsigned char *source, - bufsize_t length) { - houdini_escape_html0(dest, source, length, 0); +// C0 control characters, U+FFFE and U+FFF aren't allowed in XML. +static const char XML_ESCAPE_TABLE[256] = { + /* 0x00 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, + /* 0x10 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + /* 0x20 */ 0, 0, 2, 0, 0, 0, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x30 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 0, 5, 0, + /* 0x40 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x50 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x60 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x70 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x80 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x90 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xA0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xB0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 9, 9, + /* 0xC0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xD0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xE0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xF0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, +}; + +// U+FFFD Replacement Character encoded in UTF-8 +#define UTF8_REPL "\xEF\xBF\xBD" + +static const char *XML_ESCAPES[] = { + "", UTF8_REPL, """, "&", "<", ">" +}; + +static void escape_xml(cmark_strbuf *ob, const unsigned char *src, + bufsize_t size) { + bufsize_t i = 0, org, esc = 0; + + while (i < size) { + org = i; + while (i < size && (esc = XML_ESCAPE_TABLE[src[i]]) == 0) + i++; + + if (i > org) + cmark_strbuf_put(ob, src + org, i - org); + + if (i >= size) + break; + + if (esc == 9) { + // To replace U+FFFE and U+FFFF with U+FFFD, only the last byte has to + // be changed. + // We know that src[i] is 0xBE or 0xBF. + if (i >= 2 && src[i-2] == 0xEF && src[i-1] == 0xBF) { + cmark_strbuf_putc(ob, 0xBD); + } else { + cmark_strbuf_putc(ob, src[i]); + } + } else { + cmark_strbuf_puts(ob, XML_ESCAPES[esc]); + } + + i++; + } } static void escape_xml_str(cmark_strbuf *dest, const unsigned char *source) {