Skip to content

Fix: implement CFXML entity escaping and unescaping #153

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
DTW-Thalion wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Fix: implement CFXML entity escaping and unescaping #153

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
148 changes: 146 additions & 2 deletions Source/CFXMLParser.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -381,13 +381,157 @@ CFStringRef
CFXMLCreateStringByEscapingEntities(CFAllocatorRef allocator,
CFStringRef string, CFDictionaryRef entitiesDictionary)
{
return NULL;
CFMutableStringRef result;
CFIndex length;
CFIndex idx;

if (string == NULL)
return NULL;

length = CFStringGetLength (string);
result = CFStringCreateMutable (allocator, 0);
for (idx = 0 ; idx < length ; ++idx)
{
UniChar c = CFStringGetCharacterAtIndex (string, idx);
switch (c)
{
case '&':
CFStringAppendCString (result, "&amp;", kCFStringEncodingASCII);
break;
case '<':
CFStringAppendCString (result, "&lt;", kCFStringEncodingASCII);
break;
case '>':
CFStringAppendCString (result, "&gt;", kCFStringEncodingASCII);
break;
case '"':
CFStringAppendCString (result, "&quot;", kCFStringEncodingASCII);
break;
case '\'':
CFStringAppendCString (result, "&apos;", kCFStringEncodingASCII);
break;
default:
CFStringAppendCharacters (result, &c, 1);
break;
}
}

return result;
}

CFStringRef
CFXMLCreateStringByUnescapingEntities(CFAllocatorRef allocator,
CFStringRef string, CFDictionaryRef entitiesDictionary)
{
return NULL;
CFMutableStringRef result;
CFIndex length;
CFIndex idx;

if (string == NULL)
return NULL;

length = CFStringGetLength (string);
result = CFStringCreateMutable (allocator, 0);
idx = 0;
while (idx < length)
{
UniChar c = CFStringGetCharacterAtIndex (string, idx);
Boolean handled = false;

if (c == '&')
{
CFIndex semi = idx + 1;

while (semi < length
&& CFStringGetCharacterAtIndex (string, semi) != ';')
semi += 1;

if (semi < length && semi > idx + 1)
{
if (CFStringGetCharacterAtIndex (string, idx + 1) == '#')
{
/* Numeric character reference, decimal or hexadecimal. */
CFIndex pos = idx + 2;
long value = 0;
int base = 10;

if (pos < semi
&& (CFStringGetCharacterAtIndex (string, pos) == 'x'
|| CFStringGetCharacterAtIndex (string, pos) == 'X'))
{
base = 16;
pos += 1;
}
for ( ; pos < semi ; ++pos)
{
UniChar d = CFStringGetCharacterAtIndex (string, pos);
int v;

if (d >= '0' && d <= '9')
v = d - '0';
else if (base == 16 && d >= 'a' && d <= 'f')
v = d - 'a' + 10;
else if (base == 16 && d >= 'A' && d <= 'F')
v = d - 'A' + 10;
else
{
value = -1;
break;
}
value = value * base + v;
}
if (value >= 0 && value <= 0xFFFF)
{
UniChar u = (UniChar)value;
CFStringAppendCharacters (result, &u, 1);
handled = true;
}
}
else
{
CFStringRef name;
const void *repl = NULL;

name = CFStringCreateWithSubstring (NULL, string,
CFRangeMake (idx + 1, semi - idx - 1));
if (CFStringCompare (name, CFSTR ("amp"), 0)
== kCFCompareEqualTo)
repl = CFSTR ("&");
else if (CFStringCompare (name, CFSTR ("lt"), 0)
== kCFCompareEqualTo)
repl = CFSTR ("<");
else if (CFStringCompare (name, CFSTR ("gt"), 0)
== kCFCompareEqualTo)
repl = CFSTR (">");
else if (CFStringCompare (name, CFSTR ("quot"), 0)
== kCFCompareEqualTo)
repl = CFSTR ("\"");
else if (CFStringCompare (name, CFSTR ("apos"), 0)
== kCFCompareEqualTo)
repl = CFSTR ("'");
else if (entitiesDictionary != NULL)
repl = CFDictionaryGetValue (entitiesDictionary, name);

if (repl != NULL)
{
CFStringAppend (result, repl);
handled = true;
}
CFRelease (name);
}

if (handled)
idx = semi + 1;
}
}

if (!handled)
{
CFStringAppendCharacters (result, &c, 1);
idx += 1;
}
}

return result;
}

Empty file added Tests/CFXMLParser/TestInfo
View file
Open in desktop
Empty file.
54 changes: 54 additions & 0 deletions Tests/CFXMLParser/escaping.m
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
#include "CoreFoundation/CFXMLParser.h"
#include "CoreFoundation/CFString.h"

#include "../CFTesting.h"

int
main (void)
{
CFStringRef esc;
CFStringRef un;

/* Each of the five predefined entities is escaped, other text is left
* alone.
*/
esc = CFXMLCreateStringByEscapingEntities (NULL, CFSTR ("a<b>c&d\"e'f"),
NULL);
PASS_CFEQ(esc, CFSTR ("a&lt;b&gt;c&amp;d&quot;e&apos;f"),
"The five predefined entities are escaped.");

/* Unescaping is the inverse. */
un = CFXMLCreateStringByUnescapingEntities (NULL, esc, NULL);
PASS_CFEQ(un, CFSTR ("a<b>c&d\"e'f"),
"Unescaping the escaped string round-trips.");
CFRelease (un);
CFRelease (esc);

/* A string with no markup characters is unchanged. */
esc = CFXMLCreateStringByEscapingEntities (NULL, CFSTR ("plain text 123"),
NULL);
PASS_CFEQ(esc, CFSTR ("plain text 123"),
"Text without markup characters is copied verbatim.");
CFRelease (esc);

/* The empty string escapes to the empty string. */
esc = CFXMLCreateStringByEscapingEntities (NULL, CFSTR (""), NULL);
PASS_CFEQ(esc, CFSTR (""), "The empty string escapes to itself.");
CFRelease (esc);

/* Unescaping resolves named entities and numeric character references. */
un = CFXMLCreateStringByUnescapingEntities (NULL,
CFSTR ("x &amp; y &lt;z&gt; &#65; &quot;q&quot; &apos;"), NULL);
PASS_CFEQ(un, CFSTR ("x & y <z> A \"q\" '"),
"Named entities and a decimal character reference are resolved.");
CFRelease (un);

/* A hexadecimal character reference resolves too. */
un = CFXMLCreateStringByUnescapingEntities (NULL, CFSTR ("&#x41;&#x42;"),
NULL);
PASS_CFEQ(un, CFSTR ("AB"),
"A hexadecimal character reference is resolved.");
CFRelease (un);

return 0;
}
Loading