Parent: [a5af2b] (diff)

Child: [78bd8d] (diff)

Download this file

mh_execm.cpp    309 lines (281 with data), 10.4 kB

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
/* Copyright (C) 2005 J.F.Dockes
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the
* Free Software Foundation, Inc.,
* 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*/
#include <stdio.h>
#include <iostream>
#include <sstream>
#include "cstr.h"
#include "mh_execm.h"
#include "mh_html.h"
#include "debuglog.h"
#include "cancelcheck.h"
#include "smallut.h"
#include "md5.h"
#include "rclconfig.h"
#include "mimetype.h"
#include "idfile.h"
#include <sys/types.h>
#include <sys/wait.h>
#ifndef NO_NAMESPACES
using namespace std;
#endif /* NO_NAMESPACES */
bool MimeHandlerExecMultiple::startCmd()
{
LOGDEB(("MimeHandlerExecMultiple::startCmd\n"));
if (params.empty()) {
// Hu ho
LOGERR(("MHExecMultiple::startCmd: empty params\n"));
m_reason = "RECFILTERROR BADCONFIG";
return false;
}
// Command name
string cmd = params.front();
// Build parameter list: delete cmd name
list<string>::iterator it = params.begin();
list<string>myparams(++it, params.end());
// Start filter
m_cmd.putenv(m_forPreview ? "RECOLL_FILTER_FORPREVIEW=yes" :
"RECOLL_FILTER_FORPREVIEW=no");
if (m_cmd.startExec(cmd, myparams, 1, 1) < 0) {
m_reason = string("RECFILTERROR HELPERNOTFOUND ") + cmd;
missingHelper = true;
return false;
}
return true;
}
// Note: data is not used if this is the "document:" field: it goes
// directly to m_metaData[cstr_dj_keycontent] to avoid an extra copy
//
// Messages are made of data elements. Each element is like:
// name: len\ndata
// An empty line signals the end of the message, so the whole thing
// would look like:
// Name1: Len1\nData1Name2: Len2\nData2\n
bool MimeHandlerExecMultiple::readDataElement(string& name, string &data)
{
string ibuf;
// Read name and length
if (m_cmd.getline(ibuf) <= 0) {
LOGERR(("MHExecMultiple: getline error\n"));
return false;
}
// Empty line (end of message) ?
if (!ibuf.compare("\n")) {
LOGDEB(("MHExecMultiple: Got empty line\n"));
name.clear();
return true;
}
// Filters will sometimes abort before entering the real protocol, ie if
// a module can't be loaded. Check the special filter error first word:
if (ibuf.find("RECFILTERROR ") == 0) {
m_reason = ibuf;
if (ibuf.find("HELPERNOTFOUND") != string::npos)
missingHelper = true;
return false;
}
// We're expecting something like Name: len\n
vector<string> tokens;
stringToTokens(ibuf, tokens);
if (tokens.size() != 2) {
LOGERR(("MHExecMultiple: bad line in filter output: [%s]\n",
ibuf.c_str()));
return false;
}
vector<string>::iterator it = tokens.begin();
name = *it++;
string& slen = *it;
int len;
if (sscanf(slen.c_str(), "%d", &len) != 1) {
LOGERR(("MHExecMultiple: bad line in filter output: [%s]\n",
ibuf.c_str()));
return false;
}
LOGDEB1(("MHExecMultiple: got name [%s] len: %d\n", name.c_str(), len));
// Hack: check for 'Document:' and read directly the document data
// to m_metaData[cstr_dj_keycontent] to avoid an extra copy of the bulky
// piece
string *datap = &data;
if (!stringlowercmp("document:", name)) {
datap = &m_metaData[cstr_dj_keycontent];
} else {
datap = &data;
}
// Read element data
datap->erase();
if (len > 0 && m_cmd.receive(*datap, len) != len) {
LOGERR(("MHExecMultiple: expected %d bytes of data, got %d\n",
len, datap->length()));
return false;
}
return true;
}
bool MimeHandlerExecMultiple::next_document()
{
LOGDEB(("MimeHandlerExecMultiple::next_document(): [%s]\n", m_fn.c_str()));
if (m_havedoc == false)
return false;
if (missingHelper) {
LOGDEB(("MHExecMultiple::next_document(): helper known missing\n"));
return false;
}
if (m_cmd.getChildPid() < 0 && !startCmd()) {
return false;
}
// Send request to child process. This maybe the first/only
// request for a given file, or a continuation request. We send an
// empty file name in the latter case.
ostringstream obuf;
if (m_filefirst) {
obuf << "FileName: " << m_fn.length() << "\n" << m_fn;
// m_filefirst is set to true by set_document_file()
m_filefirst = false;
} else {
obuf << "Filename: " << 0 << "\n";
}
if (!m_ipath.empty()) {
LOGDEB(("next_doc: sending len %d val [%s]\n", m_ipath.length(),
m_ipath.c_str()));
obuf << "Ipath: " << m_ipath.length() << "\n" << m_ipath;
}
if (!m_dfltInputCharset.empty()) {
obuf << "DflInCS: " << m_dfltInputCharset.length() << "\n"
<< m_dfltInputCharset;
}
obuf << "Mimetype: " << m_mimeType.length() << "\n" << m_mimeType;
obuf << "\n";
if (m_cmd.send(obuf.str()) < 0) {
m_cmd.zapChild();
LOGERR(("MHExecMultiple: send error\n"));
return false;
}
// Read answer (multiple elements)
LOGDEB1(("MHExecMultiple: reading answer\n"));
bool eofnext_received = false;
bool eofnow_received = false;
bool fileerror_received = false;
bool subdocerror_received = false;
string ipath;
string mtype;
string charset;
for (int loop=0;;loop++) {
string name, data;
if (!readDataElement(name, data)) {
m_cmd.zapChild();
return false;
}
if (name.empty())
break;
if (!stringlowercmp("eofnext:", name)) {
LOGDEB(("MHExecMultiple: got EOFNEXT\n"));
eofnext_received = true;
} else if (!stringlowercmp("eofnow:", name)) {
LOGDEB(("MHExecMultiple: got EOFNOW\n"));
eofnow_received = true;
} else if (!stringlowercmp("fileerror:", name)) {
LOGDEB(("MHExecMultiple: got FILEERROR\n"));
fileerror_received = true;
} else if (!stringlowercmp("subdocerror:", name)) {
LOGDEB(("MHExecMultiple: got SUBDOCERROR\n"));
subdocerror_received = true;
} else if (!stringlowercmp("ipath:", name)) {
ipath = data;
LOGDEB(("MHExecMultiple: got ipath [%s]\n", data.c_str()));
} else if (!stringlowercmp("charset:", name)) {
charset = data;
LOGDEB(("MHExecMultiple: got charset [%s]\n", data.c_str()));
} else if (!stringlowercmp("mimetype:", name)) {
mtype = data;
LOGDEB(("MHExecMultiple: got mimetype [%s]\n", data.c_str()));
}
if (loop == 10) {
// ??
LOGERR(("MHExecMultiple: filter sent too many parameters\n"));
return false;
}
}
if (eofnow_received || fileerror_received) {
// No more docs
m_havedoc = false;
return false;
}
if (subdocerror_received) {
return false;
}
// It used to be that eof could be signalled just by an empty document, but
// this was wrong. Empty documents can be found ie in zip files and should
// not be interpreted as eof.
if (m_metaData[cstr_dj_keycontent].empty()) {
LOGDEB0(("MHExecMultiple: got empty document inside [%s]: [%s]\n",
m_fn.c_str(), ipath.c_str()));
}
// If this has an ipath, it is an internal doc from a
// multi-document file. In this case, either the filter supplies the
// mimetype, or the ipath MUST be a filename-like string which we can use
// to compute a mime type
if (!ipath.empty()) {
m_metaData[cstr_dj_keyipath] = ipath;
if (mtype.empty()) {
LOGDEB0(("MHExecMultiple: no mime type from filter, "
"using ipath for a guess\n"));
mtype = mimetype(ipath, 0, m_config, false);
if (mtype.empty()) {
// mimetype() won't call idFile when there is no file. Do it
mtype = idFileMem(m_metaData[cstr_dj_keycontent]);
if (mtype.empty()) {
// Note this happens for example for directory zip members
// We could recognize them by the end /, but wouldn't know
// what to do with them anyway.
LOGINFO(("MHExecMultiple: cant guess mime type\n"));
mtype = "application/octet-stream";
}
}
}
m_metaData[cstr_dj_keymt] = mtype;
string md5, xmd5;
MD5String(m_metaData[cstr_dj_keycontent], md5);
m_metaData[cstr_dj_keymd5] = MD5HexPrint(md5, xmd5);
} else {
m_metaData[cstr_dj_keymt] = mtype.empty() ? "text/html" : mtype;
m_metaData.erase(cstr_dj_keyipath);
string md5, xmd5, reason;
if (MD5File(m_fn, md5, &reason)) {
m_metaData[cstr_dj_keymd5] = MD5HexPrint(md5, xmd5);
} else {
LOGERR(("MimeHandlerExecM: cant compute md5 for [%s]: %s\n",
m_fn.c_str(), reason.c_str()));
}
}
// Charset. For many document types it doesn't matter. For text
// and html it does. We supply a default from the configuration.
if (charset.empty()) {
charset = cfgFilterOutputCharset.empty() ? "utf-8" :
cfgFilterOutputCharset;
if (!stringlowercmp("default", charset)) {
charset = m_dfltInputCharset;
}
}
m_metaData[cstr_dj_keyorigcharset] = charset;
m_metaData[cstr_dj_keycharset] = charset;
if (!m_metaData[cstr_dj_keymt].compare(cstr_textplain)) {
(void)txtdcode("mh_execm");
}
if (eofnext_received)
m_havedoc = false;
LOGDEB0(("MHExecMultiple: returning %d bytes of content,"
" mtype [%s] charset [%s]\n", m_metaData[cstr_dj_keycontent].size(),
m_metaData[cstr_dj_keymt].c_str(), m_metaData[cstr_dj_keycharset].c_str()));
return true;
}