const SkString escape_xml(const SkString& input,
const char* before = nullptr,
const char* after = nullptr) {
if (input.size() == 0) {
return input;
}
// "&" --> "&" and "<" --> "<"
// text is assumed to be in UTF-8
// all strings are xml content, not attribute values.
size_t beforeLen = before ? strlen(before) : 0;
size_t afterLen = after ? strlen(after) : 0;
int extra = count_xml_escape_size(input);
SkString output(input.size() + extra + beforeLen + afterLen);
char* out = output.writable_str();
if (before) {
strncpy(out, before, beforeLen);
out += beforeLen;
}
static const char kAmp[] = "&";
static const char kLt[] = "<";
for (size_t i = 0; i < input.size(); ++i) {
if (input[i] == '&') {
strncpy(out, kAmp, strlen(kAmp));
out += strlen(kAmp);
} else if (input[i] == '<') {
strncpy(out, kLt, strlen(kLt));
out += strlen(kLt);
} else {
*out++ = input[i];
}
}
if (after) {
strncpy(out, after, afterLen);
out += afterLen;
}
// Validate that we haven't written outside of our string.
SkASSERT(out == &output.writable_str()[output.size()]);
*out = '\0';
return output;
}
sk_sp<SkPDFObject> SkPDFMetadata::MakeXMPObject(
const SkDocument::PDFMetadata& metadata,
const UUID& doc,
const UUID& instance) {
static const char templateString[] =
"<?xpacket begin=\"\" id=\"W5M0MpCehiHzreSzNTczkc9d\"?>\n"
"<x:xmpmeta xmlns:x=\"adobe:ns:meta/\"\n"
" x:xmptk=\"Adobe XMP Core 5.4-c005 78.147326, "
"2012/08/23-13:03:03\">\n"
"<rdf:RDF "
"xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n"
"<rdf:Description rdf:about=\"\"\n"
" xmlns:xmp=\"http://ns.adobe.com/xap/1.0/\"\n"
" xmlns:dc=\"http://purl.org/dc/elements/1.1/\"\n"
" xmlns:xmpMM=\"http://ns.adobe.com/xap/1.0/mm/\"\n"
" xmlns:pdf=\"http://ns.adobe.com/pdf/1.3/\"\n"
" xmlns:pdfaid=\"http://www.aiim.org/pdfa/ns/id/\">\n"
"<pdfaid:part>2</pdfaid:part>\n"
"<pdfaid:conformance>B</pdfaid:conformance>\n"
"%s" // ModifyDate
"%s" // CreateDate
"%s" // xmp:CreatorTool
"<dc:format>application/pdf</dc:format>\n"
"%s" // dc:title
"%s" // dc:description
"%s" // author
"%s" // keywords
"<xmpMM:DocumentID>uuid:%s</xmpMM:DocumentID>\n"
"<xmpMM:InstanceID>uuid:%s</xmpMM:InstanceID>\n"
"%s" // pdf:Producer
"%s" // pdf:Keywords
"</rdf:Description>\n"
"</rdf:RDF>\n"
"</x:xmpmeta>\n" // Note: the standard suggests 4k of padding.
"<?xpacket end=\"w\"?>\n";
SkString creationDate;
SkString modificationDate;
if (metadata.fCreation.fEnabled) {
SkString tmp;
metadata.fCreation.fDateTime.toISO8601(&tmp);
SkASSERT(0 == count_xml_escape_size(tmp));
// YYYY-mm-ddTHH:MM:SS[+|-]ZZ:ZZ; no need to escape
creationDate = SkStringPrintf("<xmp:CreateDate>%s</xmp:CreateDate>\n",
tmp.c_str());
}
if (metadata.fModified.fEnabled) {
SkString tmp;
metadata.fModified.fDateTime.toISO8601(&tmp);
SkASSERT(0 == count_xml_escape_size(tmp));
modificationDate = SkStringPrintf(
"<xmp:ModifyDate>%s</xmp:ModifyDate>\n", tmp.c_str());
}
SkString title =
escape_xml(metadata.fTitle,
"<dc:title><rdf:Alt><rdf:li xml:lang=\"x-default\">",
"</rdf:li></rdf:Alt></dc:title>\n");
SkString author =
escape_xml(metadata.fAuthor, "<dc:creator><rdf:Bag><rdf:li>",
"</rdf:li></rdf:Bag></dc:creator>\n");
// TODO: in theory, XMP can support multiple authors. Split on a delimiter?
SkString subject = escape_xml(
metadata.fSubject,
"<dc:description><rdf:Alt><rdf:li xml:lang=\"x-default\">",
"</rdf:li></rdf:Alt></dc:description>\n");
SkString keywords1 =
escape_xml(metadata.fKeywords, "<dc:subject><rdf:Bag><rdf:li>",
"</rdf:li></rdf:Bag></dc:subject>\n");
SkString keywords2 = escape_xml(metadata.fKeywords, "<pdf:Keywords>",
"</pdf:Keywords>\n");
// TODO: in theory, keywords can be a list too.
SkString producer("<pdf:Producer>" SKPDF_PRODUCER "</pdf:Producer>\n");
if (!metadata.fProducer.isEmpty()) {
// TODO: register a developer prefix to make
// <skia:SKPDF_CUSTOM_PRODUCER_KEY> a real XML tag.
producer = escape_xml(
metadata.fProducer, "<pdf:Producer>",
"</pdf:Producer>\n<!-- <skia:" SKPDF_CUSTOM_PRODUCER_KEY ">"
SKPDF_PRODUCER "</skia:" SKPDF_CUSTOM_PRODUCER_KEY "> -->\n");
}
SkString creator = escape_xml(metadata.fCreator, "<xmp:CreatorTool>",
"</xmp:CreatorTool>\n");
SkString documentID = uuid_to_string(doc); // no need to escape
SkASSERT(0 == count_xml_escape_size(documentID));
SkString instanceID = uuid_to_string(instance);
SkASSERT(0 == count_xml_escape_size(instanceID));
return sk_make_sp<PDFXMLObject>(SkStringPrintf(
templateString, modificationDate.c_str(), creationDate.c_str(),
creator.c_str(), title.c_str(), subject.c_str(), author.c_str(),
keywords1.c_str(), documentID.c_str(), instanceID.c_str(),
producer.c_str(), keywords2.c_str()));
}
SkPDFObject* SkPDFMetadata::createXMPObject(const UUID& doc,
const UUID& instance) const {
static const char templateString[] =
"<?xpacket begin=\"\" id=\"W5M0MpCehiHzreSzNTczkc9d\"?>\n"
"<x:xmpmeta xmlns:x=\"adobe:ns:meta/\"\n"
" x:xmptk=\"Adobe XMP Core 5.4-c005 78.147326, "
"2012/08/23-13:03:03\">\n"
"<rdf:RDF "
"xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n"
"<rdf:Description rdf:about=\"\"\n"
" xmlns:xmp=\"http://ns.adobe.com/xap/1.0/\"\n"
" xmlns:dc=\"http://purl.org/dc/elements/1.1/\"\n"
" xmlns:xmpMM=\"http://ns.adobe.com/xap/1.0/mm/\"\n"
" xmlns:pdf=\"http://ns.adobe.com/pdf/1.3/\"\n"
" xmlns:pdfaid=\"http://www.aiim.org/pdfa/ns/id/\">\n"
"<pdfaid:part>2</pdfaid:part>\n"
"<pdfaid:conformance>B</pdfaid:conformance>\n"
"%s" // ModifyDate
"%s" // CreateDate
"%s" // MetadataDate
"%s" // xmp:CreatorTool
"<dc:format>application/pdf</dc:format>\n"
"%s" // dc:title
"%s" // dc:description
"%s" // author
"%s" // keywords
"<xmpMM:DocumentID>uuid:%s</xmpMM:DocumentID>\n"
"<xmpMM:InstanceID>uuid:%s</xmpMM:InstanceID>\n"
"<pdf:Producer>Skia/PDF</pdf:Producer>\n"
"%s" // pdf:Keywords
"</rdf:Description>\n"
"</rdf:RDF>\n"
"</x:xmpmeta>\n" // Note: the standard suggests 4k of padding.
"<?xpacket end=\"w\"?>\n";
SkString creationDate;
SkString modificationDate;
SkString metadataDate;
if (fCreation) {
SkString tmp;
fCreation->toISO8601(&tmp);
SkASSERT(0 == count_xml_escape_size(tmp));
// YYYY-mm-ddTHH:MM:SS[+|-]ZZ:ZZ; no need to escape
creationDate = sk_string_printf("<xmp:CreateDate>%s</xmp:CreateDate>\n",
tmp.c_str());
}
if (fModified) {
SkString tmp;
fModified->toISO8601(&tmp);
SkASSERT(0 == count_xml_escape_size(tmp));
modificationDate = sk_string_printf(
"<xmp:ModifyDate>%s</xmp:ModifyDate>\n", tmp.c_str());
metadataDate = sk_string_printf(
"<xmp:MetadataDate>%s</xmp:MetadataDate>\n", tmp.c_str());
}
SkString title =
escape_xml(get(fInfo, "Title"), "<dc:title><rdf:Alt><rdf:li>",
"</rdf:li></rdf:Alt></dc:title>\n");
SkString author =
escape_xml(get(fInfo, "Author"), "<dc:creator><rdf:Bag><rdf:li>",
"</rdf:li></rdf:Bag></dc:creator>\n");
// TODO: in theory, XMP can support multiple authors. Split on a delimiter?
SkString subject = escape_xml(get(fInfo, "Subject"),
"<dc:description><rdf:Alt><rdf:li>",
"</rdf:li></rdf:Alt></dc:description>\n");
SkString keywords1 =
escape_xml(get(fInfo, "Keywords"), "<dc:subject><rdf:Bag><rdf:li>",
"</rdf:li></rdf:Bag></dc:subject>\n");
SkString keywords2 = escape_xml(get(fInfo, "Keywords"), "<pdf:Keywords>",
"</pdf:Keywords>\n");
// TODO: in theory, keywords can be a list too.
SkString creator = escape_xml(get(fInfo, "Creator"), "<xmp:CreatorTool>",
"</xmp:CreatorTool>\n");
SkString documentID = uuid_to_string(doc); // no need to escape
SkASSERT(0 == count_xml_escape_size(documentID));
SkString instanceID = uuid_to_string(instance);
SkASSERT(0 == count_xml_escape_size(instanceID));
return new PDFXMLObject(sk_string_printf(
templateString, modificationDate.c_str(), creationDate.c_str(),
metadataDate.c_str(), creator.c_str(), title.c_str(),
subject.c_str(), author.c_str(), keywords1.c_str(),
documentID.c_str(), instanceID.c_str(), keywords2.c_str()));
}
