tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

comments and indent

Browse Source
This commit is contained in:
Jean-Francois Dockes 2018-11-22 14:41:06 +01:00
parent 385a7b9547
commit 9a9ce69647
2 changed files with 271 additions and 271 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

520
src/internfile/mh_mail.cpp
View File

@ -70,16 +70,16 @@ void MimeHandlerMail::clear_impl()
{ {
delete m_bincdoc; m_bincdoc = 0; delete m_bincdoc; m_bincdoc = 0;
if (m_fd >= 0) { if (m_fd >= 0) {
close(m_fd); close(m_fd);
m_fd = -1; m_fd = -1;
} }
delete m_stream; m_stream = 0; delete m_stream; m_stream = 0;
m_idx = -1; m_idx = -1;
m_startoftext = 0; m_startoftext = 0;
m_subject.erase(); m_subject.erase();
for (vector<MHMailAttach*>::iterator it = m_attachments.begin(); for (vector<MHMailAttach*>::iterator it = m_attachments.begin();
it != m_attachments.end(); it++) { it != m_attachments.end(); it++) {
delete *it; delete *it;
} }
m_attachments.clear(); m_attachments.clear();
} }
@ -88,25 +88,25 @@ bool MimeHandlerMail::set_document_file_impl(const string& mt, const string &fn)
{ {
LOGDEB("MimeHandlerMail::set_document_file(" << fn << ")\n"); LOGDEB("MimeHandlerMail::set_document_file(" << fn << ")\n");
if (m_fd >= 0) { if (m_fd >= 0) {
close(m_fd); close(m_fd);
m_fd = -1; m_fd = -1;
} }
if (!m_forPreview) { if (!m_forPreview) {
// Yes, we read the file twice. It would be possible in theory // Yes, we read the file twice. It would be possible in theory
// to add the md5 computation to the mime analysis, but ... // to add the md5 computation to the mime analysis, but ...
string md5, xmd5, reason; string md5, xmd5, reason;
if (MD5File(fn, md5, &reason)) { if (MD5File(fn, md5, &reason)) {
m_metaData[cstr_dj_keymd5] = MD5HexPrint(md5, xmd5); m_metaData[cstr_dj_keymd5] = MD5HexPrint(md5, xmd5);
} else { } else {
LOGERR("MimeHandlerMail: md5 [" << fn << "]: " << reason << "\n"); LOGERR("MimeHandlerMail: md5 [" << fn << "]: " << reason << "\n");
} }
} }
m_fd = open(fn.c_str(), 0); m_fd = open(fn.c_str(), 0);
if (m_fd < 0) { if (m_fd < 0) {
LOGERR("MimeHandlerMail::set_document_file: open(" << fn << LOGERR("MimeHandlerMail::set_document_file: open(" << fn <<
") errno " << errno << "\n"); ") errno " << errno << "\n");
return false; return false;
} }
#if defined O_NOATIME && O_NOATIME != 0 #if defined O_NOATIME && O_NOATIME != 0
if (fcntl(m_fd, F_SETFL, O_NOATIME) < 0) { if (fcntl(m_fd, F_SETFL, O_NOATIME) < 0) {
@ -117,8 +117,8 @@ bool MimeHandlerMail::set_document_file_impl(const string& mt, const string &fn)
m_bincdoc = new Binc::MimeDocument; m_bincdoc = new Binc::MimeDocument;
m_bincdoc->parseFull(m_fd); m_bincdoc->parseFull(m_fd);
if (!m_bincdoc->isHeaderParsed() && !m_bincdoc->isAllParsed()) { if (!m_bincdoc->isHeaderParsed() && !m_bincdoc->isAllParsed()) {
LOGERR("MimeHandlerMail::mkDoc: mime parse error for " << fn << "\n"); LOGERR("MimeHandlerMail::mkDoc: mime parse error for " << fn << "\n");
return false; return false;
} }
m_havedoc = true; m_havedoc = true;
return true; return true;
@ -132,26 +132,26 @@ bool MimeHandlerMail::set_document_string_impl(const string& mt,
delete m_stream; delete m_stream;
if (!m_forPreview) { if (!m_forPreview) {
string md5, xmd5; string md5, xmd5;
MD5String(msgtxt, md5); MD5String(msgtxt, md5);
m_metaData[cstr_dj_keymd5] = MD5HexPrint(md5, xmd5); m_metaData[cstr_dj_keymd5] = MD5HexPrint(md5, xmd5);
} }
if ((m_stream = new stringstream(msgtxt)) == 0 || !m_stream->good()) { if ((m_stream = new stringstream(msgtxt)) == 0 || !m_stream->good()) {
LOGERR("MimeHandlerMail::set_document_string: stream create error." LOGERR("MimeHandlerMail::set_document_string: stream create error."
"msgtxt.size() " << msgtxt.size() << "\n"); "msgtxt.size() " << msgtxt.size() << "\n");
return false; return false;
} }
delete m_bincdoc; delete m_bincdoc;
if ((m_bincdoc = new Binc::MimeDocument) == 0) { if ((m_bincdoc = new Binc::MimeDocument) == 0) {
LOGERR("MimeHandlerMail::set_doc._string: new Binc:Document failed. " LOGERR("MimeHandlerMail::set_doc._string: new Binc:Document failed. "
"Out of memory?"); "Out of memory?");
return false; return false;
} }
m_bincdoc->parseFull(*m_stream); m_bincdoc->parseFull(*m_stream);
if (!m_bincdoc->isHeaderParsed() && !m_bincdoc->isAllParsed()) { if (!m_bincdoc->isHeaderParsed() && !m_bincdoc->isAllParsed()) {
LOGERR("MimeHandlerMail::set_document_string: mime parse error\n"); LOGERR("MimeHandlerMail::set_document_string: mime parse error\n");
return false; return false;
} }
m_havedoc = true; m_havedoc = true;
return true; return true;
@ -161,14 +161,14 @@ bool MimeHandlerMail::skip_to_document(const string& ipath)
{ {
LOGDEB("MimeHandlerMail::skip_to_document(" << ipath << ")\n"); LOGDEB("MimeHandlerMail::skip_to_document(" << ipath << ")\n");
if (m_idx == -1) { if (m_idx == -1) {
// No decoding done yet. If ipath is null need do nothing // No decoding done yet. If ipath is null need do nothing
if (ipath.empty() || ipath == "-1") if (ipath.empty() || ipath == "-1")
return true; return true;
// ipath points to attachment: need to decode message // ipath points to attachment: need to decode message
if (!next_document()) { if (!next_document()) {
LOGERR("MimeHandlerMail::skip_to_doc: next_document failed\n"); LOGERR("MimeHandlerMail::skip_to_doc: next_document failed\n");
return false; return false;
} }
} }
m_idx = atoi(ipath.c_str()); m_idx = atoi(ipath.c_str());
return true; return true;
@ -179,25 +179,25 @@ bool MimeHandlerMail::next_document()
LOGDEB("MimeHandlerMail::next_document m_idx " << m_idx << " m_havedoc " << LOGDEB("MimeHandlerMail::next_document m_idx " << m_idx << " m_havedoc " <<
m_havedoc << "\n"); m_havedoc << "\n");
if (!m_havedoc) if (!m_havedoc)
return false; return false;
bool res = false; bool res = false;
if (m_idx == -1) { if (m_idx == -1) {
m_metaData[cstr_dj_keymt] = cstr_textplain; m_metaData[cstr_dj_keymt] = cstr_textplain;
res = processMsg(m_bincdoc, 0); res = processMsg(m_bincdoc, 0);
LOGDEB1("MimeHandlerMail::next_document: mt " << LOGDEB1("MimeHandlerMail::next_document: mt " <<
m_metaData[cstr_dj_keymt] << ", att cnt " << m_metaData[cstr_dj_keymt] << ", att cnt " <<
m_attachments.size() << "\n"); m_attachments.size() << "\n");
const string& txt = m_metaData[cstr_dj_keycontent]; const string& txt = m_metaData[cstr_dj_keycontent];
if (m_startoftext < txt.size()) if (m_startoftext < txt.size())
m_metaData[cstr_dj_keyabstract] = m_metaData[cstr_dj_keyabstract] =
truncate_to_word(txt.substr(m_startoftext), 250); truncate_to_word(txt.substr(m_startoftext), 250);
if (m_attachments.size() > 0) { if (m_attachments.size() > 0) {
m_metaData[cstr_dj_keyanc] = "t"; m_metaData[cstr_dj_keyanc] = "t";
} }
} else { } else {
m_metaData[cstr_dj_keyabstract].clear(); m_metaData[cstr_dj_keyabstract].clear();
res = processAttach(); res = processAttach();
} }
m_idx++; m_idx++;
m_havedoc = m_idx < (int)m_attachments.size(); m_havedoc = m_idx < (int)m_attachments.size();
@ -211,30 +211,30 @@ bool MimeHandlerMail::next_document()
// which will be indicated by the *respp argument pointing to the original // which will be indicated by the *respp argument pointing to the original
// text on exit // text on exit
static bool decodeBody(const string& cte, // Content transfer encoding static bool decodeBody(const string& cte, // Content transfer encoding
const string& body, // Source text const string& body, // Source text
string& decoded, // Decoded text if actual decoding string& decoded, // Decoded text if actual decoding
const string** respp // Decoding Indicator const string** respp // Decoding Indicator
) )
{ {
// By default, there is no encoding (7bit,8bit,raw). Also in case of // By default, there is no encoding (7bit,8bit,raw). Also in case of
// decoding error // decoding error
*respp = &body; *respp = &body;
if (!stringlowercmp("quoted-printable", cte)) { if (!stringlowercmp("quoted-printable", cte)) {
if (!qp_decode(body, decoded)) { if (!qp_decode(body, decoded)) {
LOGERR("decodeBody: quoted-printable decoding failed !\n"); LOGERR("decodeBody: quoted-printable decoding failed !\n");
LOGDEB(" Body: \n" << body << "\n"); LOGDEB(" Body: \n" << body << "\n");
return false; return false;
} }
*respp = &decoded; *respp = &decoded;
} else if (!stringlowercmp("base64", cte)) { } else if (!stringlowercmp("base64", cte)) {
if (!base64_decode(body, decoded)) { if (!base64_decode(body, decoded)) {
// base64 encoding errors are actually relatively common // base64 encoding errors are actually relatively common
LOGERR("decodeBody: base64 decoding failed !\n"); LOGERR("decodeBody: base64 decoding failed !\n");
LOGDEB(" Body: \n" << body << "\n"); LOGDEB(" Body: \n" << body << "\n");
return false; return false;
} }
*respp = &decoded; *respp = &decoded;
} }
return true; return true;
} }
@ -243,10 +243,10 @@ bool MimeHandlerMail::processAttach()
{ {
LOGDEB("MimeHandlerMail::processAttach() m_idx " << m_idx << "\n"); LOGDEB("MimeHandlerMail::processAttach() m_idx " << m_idx << "\n");
if (!m_havedoc) if (!m_havedoc)
return false; return false;
if (m_idx >= (int)m_attachments.size()) { if (m_idx >= (int)m_attachments.size()) {
m_havedoc = false; m_havedoc = false;
return false; return false;
} }
MHMailAttach *att = m_attachments[m_idx]; MHMailAttach *att = m_attachments[m_idx];
@ -263,31 +263,31 @@ bool MimeHandlerMail::processAttach()
body.clear(); body.clear();
att->m_part->getBody(body, 0, att->m_part->bodylength); att->m_part->getBody(body, 0, att->m_part->bodylength);
{ {
string decoded; string decoded;
const string *bdp; const string *bdp;
if (!decodeBody(att->m_contentTransferEncoding, body, decoded, &bdp)) { if (!decodeBody(att->m_contentTransferEncoding, body, decoded, &bdp)) {
return false; return false;
} }
if (bdp != &body) if (bdp != &body)
body.swap(decoded); body.swap(decoded);
} }
// Special case for application/octet-stream: try to better // Special case for application/octet-stream: try to better
// identify content, using file name if set // identify content, using file name if set
if (m_metaData[cstr_dj_keymt] == "application/octet-stream" && if (m_metaData[cstr_dj_keymt] == "application/octet-stream" &&
!m_metaData[cstr_dj_keyfn].empty()) { !m_metaData[cstr_dj_keyfn].empty()) {
string mt = mimetype(m_metaData[cstr_dj_keyfn], 0, string mt = mimetype(m_metaData[cstr_dj_keyfn], 0,
m_config, false); m_config, false);
if (!mt.empty()) if (!mt.empty())
m_metaData[cstr_dj_keymt] = mt; m_metaData[cstr_dj_keymt] = mt;
} }
// Special case for text/plain content. Internfile should deal // Special case for text/plain content. Internfile should deal
// with this but it expects text/plain to be utf-8 already, so we // with this but it expects text/plain to be utf-8 already, so we
// handle the transcoding if needed. Same kind of issue for the MD5 // handle the transcoding if needed. Same kind of issue for the MD5
if (m_metaData[cstr_dj_keymt] == cstr_textplain) { if (m_metaData[cstr_dj_keymt] == cstr_textplain) {
if (!txtdcode("MimeHandlerMail::processAttach")) { if (!txtdcode("MimeHandlerMail::processAttach")) {
body.clear(); body.clear();
} else if (!m_forPreview) { } else if (!m_forPreview) {
string md5, xmd5; string md5, xmd5;
MD5String(body, md5); MD5String(body, md5);
@ -314,76 +314,76 @@ bool MimeHandlerMail::processMsg(Binc::MimePart *doc, int depth)
{ {
LOGDEB2("MimeHandlerMail::processMsg: depth " << depth << "\n"); LOGDEB2("MimeHandlerMail::processMsg: depth " << depth << "\n");
if (depth++ >= maxdepth) { if (depth++ >= maxdepth) {
// Have to stop somewhere // Have to stop somewhere
LOGINFO("MimeHandlerMail::processMsg: maxdepth " << maxdepth << LOGINFO("MimeHandlerMail::processMsg: maxdepth " << maxdepth <<
" exceeded\n"); " exceeded\n");
// Return true anyway, better to index partially than not at all // Return true anyway, better to index partially than not at all
return true; return true;
} }
// Handle some headers. // Handle some headers.
string& text = m_metaData[cstr_dj_keycontent]; string& text = m_metaData[cstr_dj_keycontent];
Binc::HeaderItem hi; Binc::HeaderItem hi;
string decoded; string decoded;
if (doc->h.getFirstHeader("From", hi)) { if (doc->h.getFirstHeader("From", hi)) {
rfc2047_decode(hi.getValue(), decoded); rfc2047_decode(hi.getValue(), decoded);
if (preview()) if (preview())
text += string("From: "); text += string("From: ");
text += decoded + cstr_newline; text += decoded + cstr_newline;
if (depth == 1) { if (depth == 1) {
m_metaData[cstr_dj_keyauthor] = decoded; m_metaData[cstr_dj_keyauthor] = decoded;
} }
} }
if (doc->h.getFirstHeader("To", hi)) { if (doc->h.getFirstHeader("To", hi)) {
rfc2047_decode(hi.getValue(), decoded); rfc2047_decode(hi.getValue(), decoded);
if (preview()) if (preview())
text += string("To: "); text += string("To: ");
text += decoded + cstr_newline; text += decoded + cstr_newline;
if (depth == 1) { if (depth == 1) {
m_metaData[cstr_dj_keyrecipient] = decoded; m_metaData[cstr_dj_keyrecipient] = decoded;
} }
} }
if (doc->h.getFirstHeader("Cc", hi)) { if (doc->h.getFirstHeader("Cc", hi)) {
rfc2047_decode(hi.getValue(), decoded); rfc2047_decode(hi.getValue(), decoded);
if (preview()) if (preview())
text += string("Cc: "); text += string("Cc: ");
text += decoded + cstr_newline; text += decoded + cstr_newline;
if (depth == 1) { if (depth == 1) {
m_metaData[cstr_dj_keyrecipient] += " " + decoded; m_metaData[cstr_dj_keyrecipient] += " " + decoded;
} }
} }
if (doc->h.getFirstHeader("Message-Id", hi)) { if (doc->h.getFirstHeader("Message-Id", hi)) {
if (depth == 1) { if (depth == 1) {
m_metaData[cstr_dj_keymsgid] = hi.getValue(); m_metaData[cstr_dj_keymsgid] = hi.getValue();
trimstring(m_metaData[cstr_dj_keymsgid], "<>"); trimstring(m_metaData[cstr_dj_keymsgid], "<>");
} }
} }
if (doc->h.getFirstHeader("Date", hi)) { if (doc->h.getFirstHeader("Date", hi)) {
rfc2047_decode(hi.getValue(), decoded); rfc2047_decode(hi.getValue(), decoded);
if (depth == 1) { if (depth == 1) {
time_t t = rfc2822DateToUxTime(decoded); time_t t = rfc2822DateToUxTime(decoded);
if (t != (time_t)-1) { if (t != (time_t)-1) {
char ascuxtime[100]; char ascuxtime[100];
sprintf(ascuxtime, "%ld", (long)t); sprintf(ascuxtime, "%ld", (long)t);
m_metaData[cstr_dj_keymd] = ascuxtime; m_metaData[cstr_dj_keymd] = ascuxtime;
} else { } else {
// Leave mtime field alone, ftime will be used instead. // Leave mtime field alone, ftime will be used instead.
LOGDEB("rfc2822Date...: failed: [" << decoded << "]\n"); LOGDEB("rfc2822Date...: failed: [" << decoded << "]\n");
} }
} }
if (preview()) if (preview())
text += string("Date: "); text += string("Date: ");
text += decoded + cstr_newline; text += decoded + cstr_newline;
} }
if (doc->h.getFirstHeader("Subject", hi)) { if (doc->h.getFirstHeader("Subject", hi)) {
rfc2047_decode(hi.getValue(), decoded); rfc2047_decode(hi.getValue(), decoded);
if (depth == 1) { if (depth == 1) {
m_metaData[cstr_dj_keytitle] = decoded; m_metaData[cstr_dj_keytitle] = decoded;
m_subject = decoded; m_subject = decoded;
} }
if (preview()) if (preview())
text += string("Subject: "); text += string("Subject: ");
text += decoded + cstr_newline; text += decoded + cstr_newline;
} }
// Check for the presence of configured additional headers and possibly // Check for the presence of configured additional headers and possibly
@ -424,58 +424,58 @@ void MimeHandlerMail::walkmime(Binc::MimePart* doc, int depth)
{ {
LOGDEB2("MimeHandlerMail::walkmime: depth " << depth << "\n"); LOGDEB2("MimeHandlerMail::walkmime: depth " << depth << "\n");
if (depth++ >= maxdepth) { if (depth++ >= maxdepth) {
LOGINFO("walkmime: max depth (" << maxdepth << ") exceeded\n"); LOGINFO("walkmime: max depth (" << maxdepth << ") exceeded\n");
return; return;
} }
string& out = m_metaData[cstr_dj_keycontent]; string& out = m_metaData[cstr_dj_keycontent];
if (doc->isMultipart()) { if (doc->isMultipart()) {
LOGDEB2("walkmime: ismultipart " << doc->isMultipart() << LOGDEB2("walkmime: ismultipart " << doc->isMultipart() <<
" subtype '" << doc->getSubType() << "'\n"); " subtype '" << doc->getSubType() << "'\n");
// We only handle alternative, related and mixed (no digests). // We only handle alternative, related and mixed (no digests).
std::vector<Binc::MimePart>::iterator it; std::vector<Binc::MimePart>::iterator it;
if (!stringicmp("mixed", doc->getSubType()) || if (!stringicmp("mixed", doc->getSubType()) ||
!stringicmp("signed", doc->getSubType()) || !stringicmp("signed", doc->getSubType()) ||
!stringicmp("related", doc->getSubType())) { !stringicmp("related", doc->getSubType())) {
// Multipart mixed and related: process each part. // Multipart mixed and related: process each part.
for (it = doc->members.begin(); it != doc->members.end();it++) { for (it = doc->members.begin(); it != doc->members.end();it++) {
walkmime(&(*it), depth); walkmime(&(*it), depth);
} }
} else if (!stringicmp("alternative", doc->getSubType())) { } else if (!stringicmp("alternative", doc->getSubType())) {
// Multipart/alternative: look for a text/plain part, then html. // Multipart/alternative: look for a text/plain part, then html.
// Process if found // Process if found
std::vector<Binc::MimePart>::iterator ittxt, ithtml; std::vector<Binc::MimePart>::iterator ittxt, ithtml;
ittxt = ithtml = doc->members.end(); ittxt = ithtml = doc->members.end();
int i = 1; int i = 1;
for (it = doc->members.begin(); for (it = doc->members.begin();
it != doc->members.end(); it++, i++) { it != doc->members.end(); it++, i++) {
// Get and parse content-type header // Get and parse content-type header
Binc::HeaderItem hi; Binc::HeaderItem hi;
if (!it->h.getFirstHeader("Content-Type", hi)) { if (!it->h.getFirstHeader("Content-Type", hi)) {
LOGDEB("walkmime:no ctent-type header for part "<<i<< "\n"); LOGDEB("walkmime:no ctent-type header for part "<<i<< "\n");
continue; continue;
} }
MimeHeaderValue content_type; MimeHeaderValue content_type;
parseMimeHeaderValue(hi.getValue(), content_type); parseMimeHeaderValue(hi.getValue(), content_type);
LOGDEB2("walkmime: C-type: " << content_type.value << "\n"); LOGDEB2("walkmime: C-type: " << content_type.value << "\n");
if (!stringlowercmp(cstr_textplain, content_type.value)) if (!stringlowercmp(cstr_textplain, content_type.value))
ittxt = it; ittxt = it;
else if (!stringlowercmp("text/html", content_type.value)) else if (!stringlowercmp("text/html", content_type.value))
ithtml = it; ithtml = it;
} }
if (ittxt != doc->members.end() && ittxt->bodylength) { if (ittxt != doc->members.end() && ittxt->bodylength) {
LOGDEB2("walkmime: alternative: choose text/plain. Size: " << LOGDEB2("walkmime: alternative: choose text/plain. Size: " <<
ittxt->bodylength << endl); ittxt->bodylength << endl);
walkmime(&(*ittxt), depth); walkmime(&(*ittxt), depth);
} else if (ithtml != doc->members.end()) { } else if (ithtml != doc->members.end()) {
LOGDEB2("walkmime: alternative: choose text/html.\n"); LOGDEB2("walkmime: alternative: choose text/html.\n");
walkmime(&(*ithtml), depth); walkmime(&(*ithtml), depth);
} }
} }
return; return;
} }
// Part is not multipart: it must be either simple or message. Take // Part is not multipart: it must be either simple or message. Take
@ -485,57 +485,60 @@ void MimeHandlerMail::walkmime(Binc::MimePart* doc, int depth)
Binc::HeaderItem hi; Binc::HeaderItem hi;
string ctt = cstr_textplain; string ctt = cstr_textplain;
if (doc->h.getFirstHeader("Content-Type", hi)) { if (doc->h.getFirstHeader("Content-Type", hi)) {
ctt = hi.getValue(); ctt = hi.getValue();
} }
LOGDEB2("walkmime:content-type: " << ctt << "\n"); LOGDEB2("walkmime:content-type: " << ctt << "\n");
MimeHeaderValue content_type; MimeHeaderValue content_type;
parseMimeHeaderValue(ctt, content_type); parseMimeHeaderValue(ctt, content_type);
// Get and parse Content-Disposition header // Get and parse Content-Disposition header
string ctd = "inline"; string ctd = "inline";
if (doc->h.getFirstHeader("Content-Disposition", hi)) { if (doc->h.getFirstHeader("Content-Disposition", hi)) {
ctd = hi.getValue(); ctd = hi.getValue();
} }
MimeHeaderValue content_disposition; MimeHeaderValue content_disposition;
parseMimeHeaderValue(ctd, content_disposition); parseMimeHeaderValue(ctd, content_disposition);
LOGDEB2("Content_disposition:[" << content_disposition.value << "]\n"); LOGDEB2("Content_disposition:[" << content_disposition.value << "]\n");
string dispindic; string dispindic;
if (stringlowercmp("inline", content_disposition.value)) if (stringlowercmp("inline", content_disposition.value))
dispindic = "Attachment"; dispindic = "Attachment";
else else
dispindic = "Inline"; dispindic = "Inline";
// See if we have a filename. // See if we have a filename.
string filename; string filename;
map<string,string>::const_iterator it; map<string,string>::const_iterator it;
it = content_disposition.params.find(string("filename")); it = content_disposition.params.find(string("filename"));
if (it != content_disposition.params.end()) if (it != content_disposition.params.end())
filename = it->second; filename = it->second;
if (filename.empty()) { if (filename.empty()) {
it = content_type.params.find(string("name")); it = content_type.params.find(string("name"));
if (it != content_type.params.end()) if (it != content_type.params.end())
filename = it->second; filename = it->second;
} }
// Note: I have never seen anything useful in the Content-Description field.
// So it's not processed. Use rfc2047 to decode if we ever do it.
if (doc->isMessageRFC822()) { if (doc->isMessageRFC822()) {
LOGDEB2("walkmime: message/RFC822 part\n"); LOGDEB2("walkmime: message/RFC822 part\n");
// The first part is the already parsed message. Call // The first part is the already parsed message. Call
// processMsg instead of walkmime so that mail headers get // processMsg instead of walkmime so that mail headers get
// printed. The depth will tell it what to do // printed. The depth will tell it what to do
if (doc->members.empty()) { if (doc->members.empty()) {
//?? //??
return; return;
} }
out += "\n"; out += "\n";
if (m_forPreview) if (m_forPreview)
out += "[" + dispindic + " " + content_type.value + ": "; out += "[" + dispindic + " " + content_type.value + ": ";
out += filename; out += filename;
if (m_forPreview) if (m_forPreview)
out += "]"; out += "]";
out += "\n\n"; out += "\n\n";
processMsg(&doc->members[0], depth); processMsg(&doc->members[0], depth);
return; return;
} }
// "Simple" part. // "Simple" part.
@ -550,13 +553,13 @@ void MimeHandlerMail::walkmime(Binc::MimePart* doc, int depth)
string charset; string charset;
it = content_type.params.find(cstr_mail_charset); it = content_type.params.find(cstr_mail_charset);
if (it != content_type.params.end()) if (it != content_type.params.end())
charset = it->second; charset = it->second;
if (charset.empty() || if (charset.empty() ||
!stringlowercmp("us-ascii", charset) || !stringlowercmp("us-ascii", charset) ||
!stringlowercmp("default", charset) || !stringlowercmp("default", charset) ||
!stringlowercmp("x-user-defined", charset) || !stringlowercmp("x-user-defined", charset) ||
!stringlowercmp("x-unknown", charset) || !stringlowercmp("x-unknown", charset) ||
!stringlowercmp("unknown", charset) ) { !stringlowercmp("unknown", charset) ) {
if (!m_config->getConfParam("maildefcharset", charset)) if (!m_config->getConfParam("maildefcharset", charset))
charset = "CP1252"; charset = "CP1252";
} }
@ -564,7 +567,7 @@ void MimeHandlerMail::walkmime(Binc::MimePart* doc, int depth)
// Content transfer encoding // Content transfer encoding
string cte = "7bit"; string cte = "7bit";
if (doc->h.getFirstHeader("Content-Transfer-Encoding", hi)) { if (doc->h.getFirstHeader("Content-Transfer-Encoding", hi)) {
cte = hi.getValue(); cte = hi.getValue();
} }
// If the Content-Disposition is not inline, we treat it as // If the Content-Disposition is not inline, we treat it as
@ -572,34 +575,34 @@ void MimeHandlerMail::walkmime(Binc::MimePart* doc, int depth)
// If it is inline but not text or html, same thing. // If it is inline but not text or html, same thing.
// Some early MIME msgs have "text" instead of "text/plain" as type // Some early MIME msgs have "text" instead of "text/plain" as type
if (stringlowercmp("inline", content_disposition.value) || if (stringlowercmp("inline", content_disposition.value) ||
(stringlowercmp(cstr_textplain, content_type.value) && (stringlowercmp(cstr_textplain, content_type.value) &&
stringlowercmp("text", content_type.value) && stringlowercmp("text", content_type.value) &&
stringlowercmp("text/html", content_type.value)) ) { stringlowercmp("text/html", content_type.value)) ) {
if (!filename.empty()) { if (!filename.empty()) {
out += "\n"; out += "\n";
if (m_forPreview) if (m_forPreview)
out += "[" + dispindic + " " + content_type.value + ": "; out += "[" + dispindic + " " + content_type.value + ": ";
out += filename; out += filename;
if (m_forPreview) if (m_forPreview)
out += "]"; out += "]";
out += "\n\n"; out += "\n\n";
} }
MHMailAttach *att = new MHMailAttach; MHMailAttach *att = new MHMailAttach;
if (att == 0) { if (att == 0) {
LOGERR("Out of memory\n"); LOGERR("Out of memory\n");
return; return;
} }
att->m_contentType = content_type.value; att->m_contentType = content_type.value;
stringtolower(att->m_contentType); stringtolower(att->m_contentType);
att->m_filename = filename; att->m_filename = filename;
att->m_charset = charset; att->m_charset = charset;
att->m_contentTransferEncoding = cte; att->m_contentTransferEncoding = cte;
att->m_part = doc; att->m_part = doc;
LOGDEB("walkmime: attachmnt: ct [" << att->m_contentType << LOGDEB("walkmime: attachmnt: ct [" << att->m_contentType <<
"] cte [" << att->m_contentTransferEncoding << "] cs [" << "] cte [" << att->m_contentTransferEncoding << "] cs [" <<
att->m_charset << "] fn [" << filename << "]\n"); att->m_charset << "] fn [" << filename << "]\n");
m_attachments.push_back(att); m_attachments.push_back(att);
return; return;
} }
// We are dealing with an inline part of text/plain or text/html // We are dealing with an inline part of text/plain or text/html
@ -612,43 +615,42 @@ void MimeHandlerMail::walkmime(Binc::MimePart* doc, int depth)
string body; string body;
doc->getBody(body, 0, doc->bodylength); doc->getBody(body, 0, doc->bodylength);
{ {
string decoded; string decoded;
const string *bdp; const string *bdp;
if (!decodeBody(cte, body, decoded, &bdp)) { if (!decodeBody(cte, body, decoded, &bdp)) {
LOGERR("MimeHandlerMail::walkmime: failed decoding body\n"); LOGERR("MimeHandlerMail::walkmime: failed decoding body\n");
} }
if (bdp != &body) if (bdp != &body)
body.swap(decoded); body.swap(decoded);
} }
// Handle html stripping and transcoding to utf8 // Handle html stripping and transcoding to utf8
if (!stringlowercmp("text/html", content_type.value)) { if (!stringlowercmp("text/html", content_type.value)) {
MimeHandlerHtml mh(m_config, "1234"); MimeHandlerHtml mh(m_config, "1234");
mh.set_property(Dijon::Filter::OPERATING_MODE, mh.set_property(Dijon::Filter::OPERATING_MODE,
m_forPreview ? "view" : "index"); m_forPreview ? "view" : "index");
mh.set_property(Dijon::Filter::DEFAULT_CHARSET, charset); mh.set_property(Dijon::Filter::DEFAULT_CHARSET, charset);
mh.set_document_string("text/html", body); mh.set_document_string("text/html", body);
mh.next_document(); mh.next_document();
map<string, string>::const_iterator it = map<string, string>::const_iterator it =
mh.get_meta_data().find(cstr_dj_keycontent); mh.get_meta_data().find(cstr_dj_keycontent);
if (it != mh.get_meta_data().end()) if (it != mh.get_meta_data().end())
out += it->second; out += it->second;
} else { } else {
string utf8; string utf8;
// Transcode to utf-8 // Transcode to utf-8
LOGDEB1("walkmime: transcoding from " << charset << " to UTF-8\n"); LOGDEB1("walkmime: transcoding from " << charset << " to UTF-8\n");
if (!transcode(body, utf8, charset, cstr_utf8)) { if (!transcode(body, utf8, charset, cstr_utf8)) {
LOGERR("walkmime: transcode failed from cs '" << charset << LOGERR("walkmime: transcode failed from cs '" << charset <<
"' to UTF-8\n"); "' to UTF-8\n");
out += body; out += body;
} else { } else {
out += utf8; out += utf8;
} }
} }
if (out.length() && out[out.length()-1] != '\n') if (out.length() && out[out.length()-1] != '\n')
out += '\n'; out += '\n';
LOGDEB2("walkmime: out now: [" << out << "]\n"); LOGDEB2("walkmime: out now: [" << out << "]\n");
} }

22
src/internfile/mh_mail.h
View File

@ -24,25 +24,23 @@
#include "mimehandler.h" #include "mimehandler.h"
namespace Binc { namespace Binc {
class MimeDocument; class MimeDocument;
class MimePart; class MimePart;
} }
class MHMailAttach; class MHMailAttach;
/** /**
* Translate a mail folder file into internal documents (also works * Process a mail message (rfc822) into internal documents.
* for maildir files). This has to keep state while parsing a mail folder
* file.
*/ */
class MimeHandlerMail : public RecollFilter { class MimeHandlerMail : public RecollFilter {
public: public:
MimeHandlerMail(RclConfig *cnf, const std::string &id); MimeHandlerMail(RclConfig *cnf, const std::string &id);
virtual ~MimeHandlerMail(); virtual ~MimeHandlerMail();
virtual bool is_data_input_ok(DataInput input) const { virtual bool is_data_input_ok(DataInput input) const {
if (input == DOCUMENT_FILE_NAME || input == DOCUMENT_STRING) if (input == DOCUMENT_FILE_NAME || input == DOCUMENT_STRING)
return true; return true;
return false; return false;
} }
virtual bool next_document(); virtual bool next_document();
virtual bool skip_to_document(const std::string& ipath); virtual bool skip_to_document(const std::string& ipath);
@ -76,10 +74,10 @@ private:
class MHMailAttach { class MHMailAttach {
public: public:
std::string m_contentType; std::string m_contentType;
std::string m_filename; std::string m_filename;
std::string m_charset; std::string m_charset;
std::string m_contentTransferEncoding; std::string m_contentTransferEncoding;
Binc::MimePart *m_part; Binc::MimePart *m_part;
}; };