static PRBool
nsShouldIgnoreFile(nsString& name)
{
PRUnichar firstChar=name.CharAt(0);
if (firstChar == '.' || firstChar == '#' || name.CharAt(name.Length() - 1) == '~')
return PR_TRUE;
if (name.LowerCaseEqualsLiteral("rules.dat") || name.LowerCaseEqualsLiteral("rulesbackup.dat"))
return PR_TRUE;
// don't add summary files to the list of folders;
// don't add popstate files to the list either, or rules (sort.dat).
if (StringEndsWith(name, NS_LITERAL_STRING(".snm")) ||
name.LowerCaseEqualsLiteral("popstate.dat") ||
name.LowerCaseEqualsLiteral("sort.dat") ||
name.LowerCaseEqualsLiteral("mailfilt.log") ||
name.LowerCaseEqualsLiteral("filters.js") ||
StringEndsWith(name, NS_LITERAL_STRING(".toc")) ||
StringEndsWith(name, NS_LITERAL_STRING(".sbd")))
return PR_TRUE;
return PR_FALSE;
}
void nsEudoraCompose::ExtractCharset(nsString& str)
{
int32_t idx = MsgFind(str, "charset=", true, 0);
if (idx != -1) {
str.Cut(0, idx + 8);
idx = str.FindChar(';');
if (idx != -1)
str.SetLength(idx);
str.Trim(kWhitespace);
if ((str.CharAt(0) == '"') && (str.Length() > 2)) {
str.SetLength(str.Length() - 1);
str.Cut(0, 1);
str.Trim(kWhitespace);
}
}
else
str.Truncate();
}
void nsEudoraCompose::ExtractType(nsString& str)
{
nsString tStr;
int32_t idx = str.FindChar(';');
if (idx != -1)
str.SetLength(idx);
str.Trim(kWhitespace);
if ((str.CharAt(0) == '"') && (str.Length() > 2)) {
str.SetLength(str.Length() - 1);
str.Cut(0, 1);
str.Trim(kWhitespace);
}
// if multipart then ignore it since no outlook message body is ever
// valid multipart!
if (StringBeginsWith(str, NS_LITERAL_STRING("multipart/"), nsCaseInsensitiveStringComparator()))
str.Truncate();
}
void nsEudoraCompose::ExtractCharset( nsString& str)
{
nsString tStr;
PRInt32 idx = str.Find( "charset=", PR_TRUE);
if (idx != -1) {
idx += 8;
str.Right( tStr, str.Length() - idx);
idx = tStr.FindChar( ';');
if (idx != -1)
tStr.Left( str, idx);
else
str = tStr;
str.Trim( kWhitespace);
if ((str.CharAt( 0) == '"') && (str.Length() > 2)) {
str.Mid( tStr, 1, str.Length() - 2);
str = tStr;
str.Trim( kWhitespace);
}
}
else
str.Truncate();
}
nsresult
nsRDFParserUtils::GetQuotedAttributeValue(const nsString& aSource,
const nsString& aAttribute,
nsString& aValue)
{
static const char kQuote = '\"';
static const char kApostrophe = '\'';
PRInt32 offset;
PRInt32 endOffset = -1;
nsresult result = NS_OK;
offset = aSource.Find(aAttribute, 0);
if (-1 != offset) {
offset = aSource.FindChar('=', offset);
PRUnichar next = aSource.CharAt(++offset);
if (kQuote == next) {
endOffset = aSource.FindChar(kQuote, ++offset);
}
else if (kApostrophe == next) {
endOffset = aSource.FindChar(kApostrophe, ++offset);
}
if (-1 != endOffset) {
aSource.Mid(aValue, offset, endOffset-offset);
}
else {
// Mismatched quotes - return an error
result = NS_ERROR_FAILURE;
}
}
else {
aValue.Truncate();
}
return result;
}
void nsEudoraCompose::ExtractType( nsString& str)
{
nsString tStr;
PRInt32 idx = str.FindChar( ';');
if (idx != -1) {
str.Left( tStr, idx);
str = tStr;
}
str.Trim( kWhitespace);
if ((str.CharAt( 0) == '"') && (str.Length() > 2)) {
str.Mid( tStr, 1, str.Length() - 2);
str = tStr;
str.Trim( kWhitespace);
}
// if multipart then ignore it since no outlook message body is ever
// valid multipart!
if (str.Length() > 10) {
str.Left( tStr, 10);
if (tStr.LowerCaseEqualsLiteral("multipart/"))
str.Truncate();
}
}
void
mozTXTToHTMLConv::ScanHTML(nsString& aInString, uint32_t whattodo, nsString &aOutString)
{
// some common variables we were recalculating
// every time inside the for loop...
int32_t lengthOfInString = aInString.Length();
const char16_t * uniBuffer = aInString.get();
#ifdef DEBUG_BenB_Perf
PRTime parsing_start = PR_IntervalNow();
#endif
// Look for simple entities not included in a tags and scan them.
// Skip all tags ("<[...]>") and content in an a link tag ("<a [...]</a>"),
// comment tag ("<!--[...]-->"), style tag, script tag or head tag.
// Unescape the rest (text between tags) and pass it to ScanTXT.
for (int32_t i = 0; i < lengthOfInString;)
{
if (aInString[i] == '<') // html tag
{
int32_t start = i;
if (Substring(aInString, i + 1, 2).LowerCaseEqualsASCII("a "))
// if a tag, skip until </a>.
// Make sure there's a space after, not to match "abbr".
{
i = aInString.Find("</a>", true, i);
if (i == kNotFound)
i = lengthOfInString;
else
i += 4;
}
else if (Substring(aInString, i + 1, 3).LowerCaseEqualsASCII("!--"))
// if out-commended code, skip until -->
{
i = aInString.Find("-->", false, i);
if (i == kNotFound)
i = lengthOfInString;
else
i += 3;
}
else if (Substring(aInString, i + 1, 5).LowerCaseEqualsASCII("style") &&
(aInString.CharAt(i + 6) == ' ' || aInString.CharAt(i + 6) == '>'))
// if style tag, skip until </style>
{
i = aInString.Find("</style>", true, i);
if (i == kNotFound)
i = lengthOfInString;
else
i += 8;
}
else if (Substring(aInString, i + 1, 6).LowerCaseEqualsASCII("script") &&
(aInString.CharAt(i + 7) == ' ' || aInString.CharAt(i + 7) == '>'))
// if script tag, skip until </script>
{
i = aInString.Find("</script>", true, i);
if (i == kNotFound)
i = lengthOfInString;
else
i += 9;
}
else if (Substring(aInString, i + 1, 4).LowerCaseEqualsASCII("head") &&
(aInString.CharAt(i + 5) == ' ' || aInString.CharAt(i + 5) == '>'))
// if head tag, skip until </head>
// Make sure not to match <header>.
{
i = aInString.Find("</head>", true, i);
if (i == kNotFound)
i = lengthOfInString;
else
i += 7;
}
else // just skip tag (attributes etc.)
{
i = aInString.FindChar('>', i);
if (i == kNotFound)
i = lengthOfInString;
else
i++;
}
aOutString.Append(&uniBuffer[start], i - start);
}
else
{
uint32_t start = uint32_t(i);
i = aInString.FindChar('<', i);
if (i == kNotFound)
i = lengthOfInString;
nsString tempString;
tempString.SetCapacity(uint32_t((uint32_t(i) - start) * growthRate));
UnescapeStr(uniBuffer, start, uint32_t(i) - start, tempString);
ScanTXT(tempString.get(), tempString.Length(), whattodo, aOutString);
}
}
#ifdef DEBUG_BenB_Perf
printf("ScanHTML time: %d ms\n", PR_IntervalToMilliseconds(PR_IntervalNow() - parsing_start));
#endif
}
// XXX Code copied from nsHTMLContentSink. It should be shared.
void
nsRDFParserUtils::StripAndConvert(nsString& aResult)
{
if ( !aResult.IsEmpty() ) {
// Strip quotes if present
PRUnichar first = aResult.First();
if ((first == '"') || (first == '\'')) {
if (aResult.Last() == first) {
aResult.Cut(0, 1);
PRInt32 pos = aResult.Length() - 1;
if (pos >= 0) {
aResult.Cut(pos, 1);
}
} else {
// Mismatched quotes - leave them in
}
}
}
// Reduce any entities
// XXX Note: as coded today, this will only convert well formed
// entities. This may not be compatible enough.
// XXX there is a table in navigator that translates some numeric entities
// should we be doing that? If so then it needs to live in two places (bad)
// so we should add a translate numeric entity method from the parser...
char cbuf[100];
PRUint32 i = 0;
while (i < aResult.Length()) {
// If we have the start of an entity (and it's not at the end of
// our string) then translate the entity into it's unicode value.
if ((aResult.CharAt(i++) == '&') && (i < aResult.Length())) {
PRInt32 start = i - 1;
PRUnichar e = aResult.CharAt(i);
if (e == '#') {
// Convert a numeric character reference
i++;
char* cp = cbuf;
char* limit = cp + sizeof(cbuf) - 1;
PRBool ok = PR_FALSE;
PRUint32 slen = aResult.Length();
while ((i < slen) && (cp < limit)) {
PRUnichar f = aResult.CharAt(i);
if (f == ';') {
i++;
ok = PR_TRUE;
break;
}
if ((f >= '0') && (f <= '9')) {
*cp++ = char(f);
i++;
continue;
}
break;
}
if (!ok || (cp == cbuf)) {
continue;
}
*cp = '\0';
if (cp - cbuf > 5) {
continue;
}
PRInt32 ch = PRInt32( ::atoi(cbuf) );
if (ch > 65535) {
continue;
}
// Remove entity from string and replace it with the integer
// value.
aResult.Cut(start, i - start);
aResult.Insert(PRUnichar(ch), start);
i = start + 1;
}
else if (((e >= 'A') && (e <= 'Z')) ||
((e >= 'a') && (e <= 'z'))) {
// Convert a named entity
i++;
char* cp = cbuf;
char* limit = cp + sizeof(cbuf) - 1;
*cp++ = char(e);
PRBool ok = PR_FALSE;
PRUint32 slen = aResult.Length();
while ((i < slen) && (cp < limit)) {
PRUnichar f = aResult.CharAt(i);
if (f == ';') {
i++;
ok = PR_TRUE;
break;
}
if (((f >= '0') && (f <= '9')) ||
((f >= 'A') && (f <= 'Z')) ||
((f >= 'a') && (f <= 'z'))) {
*cp++ = char(f);
i++;
continue;
}
break;
}
if (!ok || (cp == cbuf)) {
continue;
}
*cp = '\0';
PRInt32 ch;
// XXX Um, here's where we should be converting a
// named entity. I removed this to avoid a link-time
// dependency on core raptor.
ch = EntityToUnicode(cbuf);
if (ch < 0) {
continue;
}
// Remove entity from string and replace it with the integer
// value.
aResult.Cut(start, i - start);
aResult.Insert(PRUnichar(ch), start);
i = start + 1;
}
else if (e == '{') {
// Convert a script entity
// XXX write me!
NS_NOTYETIMPLEMENTED("convert a script entity");
}
}
}
}
