KAZ/depollueur
7
1
Fork 0
Code Issues Pull Requests Projects Releases Activity

fix rfc2047 / filter log / filterTest options

Browse Source
This commit is contained in:
François
2022-12-25 06:57:44 +01:00
parent 596ae82fe4
commit 61bdc3a4ba
9 changed files with 193 additions and 101 deletions
Download Patch File Download Diff File Expand all files Collapse all files

53
src/cpp/Attachment.cpp
View File

@ -65,17 +65,9 @@ const string Attachment::ALTERNATIVE ("alternative");
const string Attachment::KAZ_ATTACH_NAME (".---KazAttachment---.html");
const string Attachment::MULTIPART ("multipart/");
const regex Attachment::nameCharsetRegEx (".*name\\*=[ \t]*(.*)");
const regex Attachment::nameRegEx (".*name=[ \t]*((\"(\\\\.|[^\\\\\r])*\")|[^\r; ]*);?.*");
// boundary="----=_Part_796779_1154936629.1668080348646"
// boundary="------------040709000505010508040808"
// boundary="----------=_1668606031-941125-91"
// boundary="_004_PAVPR10MB6792713B313048E3A259B215B2079PAVPR10MB6792EURP_";
// boundary="_000_PAVPR10MB6792713B313048E3A259B215B2079PAVPR10MB6792EURP_"
// boundary=--boundary_1351_64006126-2b0e-4a3b-98ac-4797d1634188
// boundary=--boundary_1352_7e294c9a-cfab-44a0-bfb3-7310380ac7cb;
const regex Attachment::boundaryRegEx (".*boundary=[ \t]*((\"(\\\\.|[^\\\\])*\")|[^; ]*);?.*");
const regex Attachment::nameCharsetRegEx (".*name\\*=\\s*([; \t]*)");
const regex Attachment::nameRegEx ( ".*name=\\s*((\"(\\\\.|[^\\\\])*\")|[^; \t]*).*");
const regex Attachment::boundaryRegEx (".*boundary=\\s*((\"(\\\\.|[^\\\\])*\")|[^; \t]*).*");
const regex Attachment::cidDefRegEx (".*<([^>]*)>.*");
const regex Attachment::textRegEx (".*text/("+PLAIN+"|"+HTML+").*");
const regex Attachment::multiRegEx ("\\s*"+MULTIPART+"(mixed|"+RELATED+"|"+ALTERNATIVE+"|"+SIGNED+").*");
@ -188,23 +180,49 @@ Attachment::getAttachName () const {
static string tokens [] = {contentTypeToken, contentDispositionToken};
DEF_LOG ("Attachment::getAttachName", "");
for (string token : tokens) {
// name=
string result = getProp (token, nameRegEx);
removeQuote (result);
if (result.length ()) {
LOG ("name=: " << result);
encodedWord (result);
encodedWordDecode (result);
return result;
}
// name*x=
for (int id = 0; ; ++id) {
string item = getProp (token, regex (".*name\\*"+to_string (id)+"=\\s*((\"(\\\\.|[^\\\\])*\")|[; \t]*).*"));
if (item.empty ())
break;
result += item;
}
removeQuote (result);
if (result.length ()) {
LOG ("name*x=: " << result);
encodedWordDecode (result);
return result;
}
// name*=
result = getProp (token, nameCharsetRegEx);
removeQuote (result);
if (result.length ()) {
LOG ("name*=: " << result);
charsetValue (result);
charsetValueDecode (result);
return result;
}
// name*x*=
for (int id = 0; ; ++id) {
string item = getProp (token, regex (".*name\\*"+to_string (id)+"\\*=\\s*([^; ]*)"));
if (item.empty ())
break;
result += item;
}
removeQuote (result);
if (result.length ()) {
LOG ("name*x*=: " << result);
encodedWordDecode (result);
return result;
}
}
// XXX il faut composer s'il y a plusieurs ligne filename*x=
// XXX il faut composer s'il y a plusieurs ligne filename*x*=
return getUnknown (getContentType ());
}
@ -257,7 +275,7 @@ Attachment::isDefProp (const string &token, const string &val) const {
if (it == env.end ())
return false;
// XXX case insensitive ??
return it->second.find (val) != string::npos;
return caseInsensitiveFind (it->second, val) != string::npos;
}
// ================================================================================
@ -312,7 +330,8 @@ Attachment::readMime (ifstream &mbox, streamoff &curPos) {
lastVar = line.substr (0, colonPos);
toLower (lastVar);
LOG ("find var: " << lastVar);
string val (cleanString (line.length () >= colonPos+2 ? line.substr (colonPos+2) : "")); // XXX check RFC " " after ": "
// XXX check in RFC if " " after ": " (=> +2 or +1)
string val (cleanString (line.length () >= colonPos+2 ? line.substr (colonPos+2) : ""));
LOG ("new var: <" << lastVar << " <=> " << val << ">");
env [lastVar] = val;
}

64
src/cpp/MainAttachment.cpp
View File

@ -61,7 +61,7 @@ static const string TMPL_FILENAME ("{{FILENAME}}");
static const string CID ("cid:");
// "l=/" => v1 compatibility
static const regex archiveURLSignature (".*(([&?]g=)|([&?]l=/)).*");
static const regex archiveURLRegex (".*(([&?]g=)|([&?]l=/)).*");
static const string KAZ_PLAIN_HR ("______________________________________________________________________________");
static const string KAZ_PLAIN_START ("~~ PJ-KAZ !"); // don't end whith space
@ -423,7 +423,7 @@ MainAttachment::removePreviousArchive () {
vector<string> toRemove;
for (map <string, string>::const_iterator it = previousLinks.begin (); it != previousLinks.end (); ++it) {
const string key (it->first);
if (regex_match (key, archiveURLSignature))
if (regex_match (key, archiveURLRegex))
toRemove.push_back (key);
}
for (string old : toRemove)
@ -567,7 +567,7 @@ MainAttachment::extract (ifstream &mbox, const SizeArg &minSize) const {
// ================================================================================
void
MainAttachment::substitute (ifstream &mbox, ofstream &outbox, const SizeArg &minSize, const AttachMode &attachMode) {
MainAttachment::substitute (ifstream &mbox, ofstream &outbox, const SizeArg &minSize, AttachMode attachMode) {
DEF_LOG ("MainAttachment::substitute", "minSize: " << minSize << " AttachMode: " << attachMode);
// preparation
@ -601,30 +601,34 @@ MainAttachment::substitute (ifstream &mbox, ofstream &outbox, const SizeArg &min
getDisclaim (plainDisclaim, htmlDisclaim);
// copy email
if (plainDisclaim.size () && emptyEMail && boundary.empty ()) {
// only one attachment must be replace
cerr << "eMailShrinker: force one attachment" << endl;
string mime (getMime (mbox));
string::size_type startPos = (0);
for (string token : {string ("Content-Transfer-Encoding"), Attachment::contentTypeToken}) {
startPos = caseInsensitiveFind (mime, "Content-Transfer-Encoding");
for (string::size_type stopPos (startPos);
(stopPos = mime.find ("\n", stopPos)) != string::npos;
) {
if (string (" \t").find (mime [stopPos+1]) == string::npos) {
mime.erase (startPos, stopPos-startPos);
break;
if (!boundary.size () && plainDisclaim.size ()) {
if (attachMode & ATTACHMENT)
attachMode = FOOTER;
if (emptyEMail) {
// only one attachment must be replace
cerr << "eMailShrinker: force one attachment" << endl;
string mime (getMime (mbox));
string::size_type startPos = (0);
for (string token : {string ("Content-Transfer-Encoding"), Attachment::contentTypeToken}) {
startPos = caseInsensitiveFind (mime, "Content-Transfer-Encoding");
for (string::size_type stopPos (startPos);
(stopPos = mime.find ("\n", stopPos)) != string::npos;
) {
if (string (" \t").find (mime [stopPos+1]) == string::npos) {
mime.erase (startPos, stopPos-startPos);
break;
}
}
}
mime.insert (startPos, KAZ_EMPTY_TEXT_PLAIN);
string content (plainDisclaim);
base64Encode (content);
outbox << mime
<< content << endl;
outbox.flush ();
outbox.close ();
return;
}
mime.insert (startPos, KAZ_EMPTY_TEXT_PLAIN);
string content (plainDisclaim);
base64Encode (content);
outbox << mime
<< content << endl;
outbox.flush ();
outbox.close ();
return;
}
streamoff curPos = 0;
copy (mbox, outbox, curPos, contentPos);
@ -694,8 +698,7 @@ MainAttachment::substitute (ifstream &mbox, ofstream &outbox, const SizeArg &min
}
removeSection (content, KAZ_HTML_START, KAZ_HTML_STOP);
removeSection (content, KAZ_PLAIN_START, KAZ_PLAIN_STOP);
// XXX case insensitive ??
if (content.find (CID) != string::npos)
if (caseInsensitiveFind (content, CID) != string::npos)
replaceAll (content, translateHtml);
attachP->replaceEmbedded (content);
}
@ -733,12 +736,9 @@ MainAttachment::substitute (ifstream &mbox, ofstream &outbox, const SizeArg &min
string content (KAZ_HTML_CONTENT+htmlDisclaim+BODY_END+HTML_END);
base64Encode (content);
if (boundary.size ())
outbox << boundary.substr (0, boundary.length () -2) << endl
<< KAZ_ATTACHMENT_TEXT_HTML << endl
<< content << endl;
else
outbox << "coucou No multipart" << endl;
outbox << boundary.substr (0, boundary.length () -2) << endl
<< KAZ_ATTACHMENT_TEXT_HTML << endl
<< content << endl;
outbox.flush ();
}
copy (mbox, outbox, curPos, endPos);

6
src/cpp/eMailShrinker.cpp
View File

@ -33,8 +33,8 @@
////////////////////////////////////////////////////////////////////////////
#include "version.hpp"
const std::string kaz::LAST_VERSION_NUM ("2.8");
const std::string kaz::LAST_VERSION_DATE ("2022-12-23");
const std::string kaz::LAST_VERSION_NUM ("2.9");
const std::string kaz::LAST_VERSION_DATE ("2022-12-24");
const std::string kaz::LAST_VERSION (LAST_VERSION_NUM+" "+LAST_VERSION_DATE+" eMailShrinker");
#include <iostream>
@ -106,7 +106,7 @@ static const char *const inputFileC = inputFile.c_str ();
int
main (int argc, char** argv) {
// XXX debug before parse options
// uncomment next line in case of debug parse options
// Log::debug = true;
DEF_LOG ("main:", "");
prog = argv [0];

2
src/cpp/jirafeauAPI.cpp
View File

@ -106,7 +106,7 @@ static const char *const inputFileC = inputFile.c_str ();
int
main (int argc, char** argv) {
// XXX debug before parse options
// uncomment next line in case of debug parse options
// Log::debug = true;
DEF_LOG ("main:", "");
prog = argv [0];

65
src/cpp/kazMisc.cpp
View File

@ -65,6 +65,14 @@ const string kaz::availableURLChars =
"abcdefghijklmnopqrstuvwxyz"
"~";
const regex kaz::encodedWordRegex ("\\s*=\\?" // flag begin
"([0-9A-Za-z!#$%&'+^_`{}~-]+)" // charset
"\\?" // flag sep
"([QqBb])" // quoted our base64
"\\?" // flag sep
"([^ ?]+)" // encoded string
"\\?=\\s*"); // flag end
// ================================================================================
uint16_t
@ -369,65 +377,66 @@ kaz::iso2utf (string &content) {
// ================================================================================
void
kaz::encodedWord (string &content) {
kaz::encodedWordDecode (string &content) {
// rfc2047
DEF_LOG ("kaz::extendedWord", "content: " << content);
DEF_LOG ("kaz::encodedWordDecode", "content: " << content);
string::size_type charsetPos = content.find ("=?");
if (charsetPos == string::npos)
return;
LOG ("charsetPos: " << charsetPos);
LOG_BUG (charsetPos != 0, return, "kazMisc::extendedWord bug: =? not at begin pos. (content: " << content << ")");
string result;
for ( ;
(charsetPos = content.find ("=?", charsetPos)) != string::npos;
) {
string::size_type modePos = content.find ("?", charsetPos+2);
auto pos (0);
sregex_iterator ewItEnd;
for (sregex_iterator ewIt (content.begin (), content.end (), encodedWordRegex);
ewIt != ewItEnd;
++ewIt) {
smatch m = *ewIt;
if (pos != m.position ()) {
result += content.substr (pos, m.position () - pos);
LOG ("stantad " << content.substr (pos, m.position () - pos));
}
string encoded (m[3]);
replace (encoded.begin (), encoded.end (), '_', ' ');
LOG_BUG (modePos == string::npos, return, "kazMisc::extendedWord bug: no end chartset. (content: " << content << ")");
string::size_type contentPos = content.find ("?", modePos+1);
LOG ("charset: " << m[1] << " mode: " << m[2] << " string: " << encoded);
LOG_BUG (contentPos != modePos+2, return, "kazMisc::extendedWord bug: no end chartset. (content: " << content << ")");
string::size_type endPos = content.find ("?=", contentPos+1);
LOG_BUG (endPos == string::npos, return, "kazMisc::extendedWord bug: no end chartset. (content: " << content << ")");
string tmp (content.substr (contentPos+1, endPos-contentPos-1));
switch (content [modePos+1]) {
switch (m[2].str ()[0]) {
case 'B':
case 'b':
base64Decode (tmp);
base64Decode (encoded);
break;
case 'Q':
case 'q':
quotedDecode (tmp);
quotedDecode (encoded);
break;
default:
LOG_BUG (true, return, "kazMisc::extendedWord bug: unknown mode. (mode: " << content [modePos+1] << ")");
LOG_BUG (true, return, "kazMisc::encodedWordDecode bug: unknown mode. (mode: " << m[2] << ")");
}
LOG ("tmp: " << tmp);
string charset (content.substr (charsetPos, modePos-charsetPos-2));
LOG ("decoded: " << encoded);
string charset (m[1]);
toLower (charset);
if (! caseInsensitiveFind (charset, "ISO"))
iso2utf (tmp);
result += tmp;
charsetPos = endPos+2;
iso2utf (encoded);
result += encoded;
pos = m.position () + m.str ().length ();
}
content = result;
content = result + content.substr (pos);
LOG ("content: " << content);
}
// ================================================================================
void
kaz::charsetValue (string &content) {
kaz::charsetValueDecode (string &content) {
// rfc2184
DEF_LOG ("kaz::charsetValue", "content: " << content);
DEF_LOG ("kaz::charsetValueDecode", "content: " << content);
string::size_type langPos = content.find ("'");
LOG_BUG (langPos == string::npos, return, "kazMisc::charsetValue bug: no '. (content: " << content << ")");
LOG_BUG (langPos == string::npos, return, "kazMisc::charsetValueDecode bug: no '. (content: " << content << ")");
string::size_type contentPos = content.find ("'", langPos+1);
LOG_BUG (contentPos == string::npos, return, "kazMisc::charsetValue bug: no double '. (content: " << content << ")");
LOG_BUG (contentPos == string::npos, return, "kazMisc::charsetValueDecode bug: no double '. (content: " << content << ")");
string tmp (content.substr (contentPos+1));
quotedDecode<'%'> (tmp);
LOG ("tmp: " << tmp);