Mercurial > libpst
annotate src/readpst.c @ 75:987aa872294e stable-0-6-11
Use ftello/fseeko to properly handle large files.
Document and properly use datasize field in b5 blocks.
Fix some MSVC compile issues and collect MSVC dependencies into one place.
author | Carl Byington <carl@five-ten-sg.com> |
---|---|
date | Tue, 03 Jun 2008 12:00:58 -0700 |
parents | 3cb02cb1e6cd |
children | 87216aefc6df |
rev | line source |
---|---|
16 | 1 /*** |
2 * readpst.c | |
3 * Part of the LibPST project | |
4 * Written by David Smith | |
43 | 5 * dave.s@earthcorp.com |
16 | 6 */ |
48 | 7 #include "define.h" |
59
7d5c637aaafb
General cleanup and code fixes.
Carl Byington <carl@five-ten-sg.com>
parents:
52
diff
changeset
|
8 #include "libstrfunc.h" |
16 | 9 #include "libpst.h" |
10 #include "common.h" | |
11 #include "timeconv.h" | |
12 #include "lzfu.h" | |
59
7d5c637aaafb
General cleanup and code fixes.
Carl Byington <carl@five-ten-sg.com>
parents:
52
diff
changeset
|
13 |
16 | 14 #define OUTPUT_TEMPLATE "%s" |
15 #define OUTPUT_KMAIL_DIR_TEMPLATE ".%s.directory" | |
16 #define KMAIL_INDEX ".%s.index" | |
25 | 17 #define SEP_MAIL_FILE_TEMPLATE "%i" /* "%09i" */ |
16 | 18 |
19 // max size of the c_time char*. It will store the date of the email | |
20 #define C_TIME_SIZE 500 | |
59
7d5c637aaafb
General cleanup and code fixes.
Carl Byington <carl@five-ten-sg.com>
parents:
52
diff
changeset
|
21 |
16 | 22 struct file_ll { |
43 | 23 char *name; |
24 char *dname; | |
25 FILE * output; | |
26 int32_t stored_count; | |
27 int32_t email_count; | |
28 int32_t skip_count; | |
29 int32_t type; | |
16 | 30 }; |
31 | 31 |
43 | 32 void process(pst_item *outeritem, pst_desc_ll *d_ptr); |
33 void write_email_body(FILE *f, char *body); | |
34 char* removeCR (char *c); | |
35 int usage(); | |
36 int version(); | |
37 char* mk_kmail_dir(char*); | |
38 int close_kmail_dir(); | |
39 char* mk_recurse_dir(char*); | |
40 int close_recurse_dir(); | |
41 char* mk_seperate_dir(char *dir); | |
42 int close_seperate_dir(); | |
43 int mk_seperate_file(struct file_ll *f); | |
44 char* my_stristr(char *haystack, char *needle); | |
45 void check_filename(char *fname); | |
46 char* skip_header_prologue(char *headers); | |
47 void write_separate_attachment(char f_name[], pst_item_attach* current_attach, int attach_num, pst_file* pst); | |
48 void write_inline_attachment(FILE* f_output, pst_item_attach* current_attach, char boundary[], pst_file* pst); | |
49 void write_normal_email(FILE* f_output, char f_name[], pst_item* item, int mode, int mode_MH, pst_file* pst, int save_rtf); | |
50 void write_vcard(FILE* f_output, pst_item_contact* contact, char comment[]); | |
51 void write_appointment(FILE* f_output, pst_item_appointment* appointment, | |
52 pst_item_email* email, FILETIME* create_date, FILETIME* modify_date); | |
53 void create_enter_dir(struct file_ll* f, pst_item *item); | |
54 void close_enter_dir(struct file_ll *f); | |
34
07177825c91b
fix signed/unsigned to allow very small pst files with only leaf nodes
carl
parents:
33
diff
changeset
|
55 |
07177825c91b
fix signed/unsigned to allow very small pst files with only leaf nodes
carl
parents:
33
diff
changeset
|
56 char* prog_name; |
07177825c91b
fix signed/unsigned to allow very small pst files with only leaf nodes
carl
parents:
33
diff
changeset
|
57 char* output_dir = "."; |
07177825c91b
fix signed/unsigned to allow very small pst files with only leaf nodes
carl
parents:
33
diff
changeset
|
58 char* kmail_chdir = NULL; |
16 | 59 // Normal mode just creates mbox format files in the current directory. Each file is named |
60 // the same as the folder's name that it represents | |
61 #define MODE_NORMAL 0 | |
62 // KMail mode creates a directory structure suitable for being used directly | |
63 // by the KMail application | |
64 #define MODE_KMAIL 1 | |
65 // recurse mode creates a directory structure like the PST file. Each directory | |
66 // contains only one file which stores the emails in mbox format. | |
67 #define MODE_RECURSE 2 | |
68 // seperate mode is similar directory structure to RECURSE. The emails are stored in | |
69 // seperate files, numbering from 1 upward. Attachments belonging to the emails are | |
70 // saved as email_no-filename (e.g. 1-samplefile.doc or 000001-Attachment2.zip) | |
71 #define MODE_SEPERATE 3 | |
43 | 72 // Decrypt the whole file (even the parts that aren't encrypted) and ralph it to stdout |
73 #define MODE_DECSPEW 4 | |
16 | 74 |
75 | |
76 // Output Normal just prints the standard information about what is going on | |
77 #define OUTPUT_NORMAL 0 | |
78 // Output Quiet is provided so that only errors are printed | |
79 #define OUTPUT_QUIET 1 | |
80 | |
81 // default mime-type for attachments that have a null mime-type | |
82 #define MIME_TYPE_DEFAULT "application/octet-stream" | |
83 | |
84 // output mode for contacts | |
85 #define CMODE_VCARD 0 | |
43 | 86 #define CMODE_LIST 1 |
16 | 87 |
88 // output settings for RTF bodies | |
89 // filename for the attachment | |
90 #define RTF_ATTACH_NAME "rtf-body.rtf" | |
91 // mime type for the attachment | |
92 #define RTF_ATTACH_TYPE "application/rtf" | |
34
07177825c91b
fix signed/unsigned to allow very small pst files with only leaf nodes
carl
parents:
33
diff
changeset
|
93 |
39 | 94 // global settings |
95 int mode = MODE_NORMAL; | |
96 int mode_MH = 0; | |
97 int output_mode = OUTPUT_NORMAL; | |
98 int contact_mode = CMODE_VCARD; | |
99 int overwrite = 0; | |
100 int save_rtf_body = 1; | |
101 pst_file pstfile; | |
102 | |
103 | |
104 | |
105 void process(pst_item *outeritem, pst_desc_ll *d_ptr) | |
106 { | |
43 | 107 struct file_ll ff; |
108 pst_item *item = NULL; | |
39 | 109 |
43 | 110 DEBUG_ENT("process"); |
111 memset(&ff, 0, sizeof(ff)); | |
112 create_enter_dir(&ff, outeritem); | |
39 | 113 |
43 | 114 while (d_ptr) { |
115 DEBUG_MAIN(("main: New item record\n")); | |
116 if (!d_ptr->desc) { | |
117 DEBUG_WARN(("main: ERROR ?? item's desc record is NULL\n")); | |
118 ff.skip_count++; | |
119 } | |
120 else { | |
121 DEBUG_MAIN(("main: Desc Email ID %#x [d_ptr->id = %#x]\n", d_ptr->desc->id, d_ptr->id)); | |
39 | 122 |
46 | 123 item = pst_parse_item(&pstfile, d_ptr); |
43 | 124 DEBUG_MAIN(("main: About to process item\n")); |
125 if (item && item->email && item->email->subject && item->email->subject->subj) { | |
126 DEBUG_EMAIL(("item->email->subject = %p\n", item->email->subject)); | |
127 DEBUG_EMAIL(("item->email->subject->subj = %p\n", item->email->subject->subj)); | |
128 } | |
129 if (item) { | |
130 if (item->message_store) { | |
131 // there should only be one message_store, and we have already done it | |
132 DIE(("main: A second message_store has been found. Sorry, this must be an error.\n")); | |
133 } | |
39 | 134 |
43 | 135 if (item->folder && d_ptr->child && strcasecmp(item->file_as, "Deleted Items")) { |
136 //if this is a non-empty folder other than deleted items, we want to recurse into it | |
137 if (output_mode != OUTPUT_QUIET) printf("Processing Folder \"%s\"\n", item->file_as); | |
138 process(item, d_ptr->child); | |
139 | |
140 } else if (item->contact && (item->type == PST_TYPE_CONTACT)) { | |
141 // deal with a contact | |
142 // write them to the file, one per line in this format | |
143 // Desc Name <email@address>\n | |
144 if (mode == MODE_SEPERATE) mk_seperate_file(&ff); | |
145 ff.email_count++; | |
146 DEBUG_MAIN(("main: Processing Contact\n")); | |
147 if (ff.type != PST_TYPE_CONTACT) { | |
148 DEBUG_MAIN(("main: I have a contact, but the folder isn't a contacts folder. Processing anyway\n")); | |
149 } | |
150 if (contact_mode == CMODE_VCARD) | |
151 write_vcard(ff.output, item->contact, item->comment); | |
152 else | |
153 fprintf(ff.output, "%s <%s>\n", item->contact->fullname, item->contact->address1); | |
39 | 154 |
43 | 155 } else if (item->email && (item->type == PST_TYPE_NOTE || item->type == PST_TYPE_REPORT)) { |
156 if (mode == MODE_SEPERATE) mk_seperate_file(&ff); | |
157 ff.email_count++; | |
158 DEBUG_MAIN(("main: Processing Email\n")); | |
159 if ((ff.type != PST_TYPE_NOTE) && (ff.type != PST_TYPE_REPORT)) { | |
160 DEBUG_MAIN(("main: I have an email, but the folder isn't an email folder. Processing anyway\n")); | |
161 } | |
162 write_normal_email(ff.output, ff.name, item, mode, mode_MH, &pstfile, save_rtf_body); | |
39 | 163 |
43 | 164 } else if (item->journal && (item->type == PST_TYPE_JOURNAL)) { |
165 // deal with journal items | |
166 if (mode == MODE_SEPERATE) mk_seperate_file(&ff); | |
167 ff.email_count++; | |
168 DEBUG_MAIN(("main: Processing Journal Entry\n")); | |
169 if (ff.type != PST_TYPE_JOURNAL) { | |
170 DEBUG_MAIN(("main: I have a journal entry, but the folder isn't a journal folder. Processing anyway\n")); | |
171 } | |
172 fprintf(ff.output, "BEGIN:VJOURNAL\n"); | |
50 | 173 if (item->email && item->email->subject && item->email->subject->subj) |
43 | 174 fprintf(ff.output, "SUMMARY:%s\n", pst_rfc2426_escape(item->email->subject->subj)); |
50 | 175 if (item->email && item->email->body) |
43 | 176 fprintf(ff.output, "DESCRIPTION:%s\n", pst_rfc2426_escape(item->email->body)); |
177 if (item->journal->start) | |
178 fprintf(ff.output, "DTSTART;VALUE=DATE-TIME:%s\n", pst_rfc2445_datetime_format(item->journal->start)); | |
179 fprintf(ff.output, "END:VJOURNAL\n\n"); | |
39 | 180 |
43 | 181 } else if (item->appointment && (item->type == PST_TYPE_APPOINTMENT)) { |
182 // deal with Calendar appointments | |
183 if (mode == MODE_SEPERATE) mk_seperate_file(&ff); | |
184 ff.email_count++; | |
185 DEBUG_MAIN(("main: Processing Appointment Entry\n")); | |
186 if (ff.type != PST_TYPE_APPOINTMENT) { | |
187 DEBUG_MAIN(("main: I have an appointment, but folder isn't specified as an appointment type. Processing...\n")); | |
188 } | |
189 write_appointment(ff.output, item->appointment, item->email, item->create_date, item->modify_date); | |
39 | 190 |
43 | 191 } else { |
63
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
192 // these all seem to be things that MS agrees are not included in the item count |
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
193 //ff.skip_count++; |
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
194 DEBUG_MAIN(("main: Unknown item type %i (%s) name (%s)\n", |
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
195 item->type, item->ascii_type, item->file_as)); |
43 | 196 } |
46 | 197 pst_freeItem(item); |
43 | 198 } else { |
199 ff.skip_count++; | |
200 DEBUG_MAIN(("main: A NULL item was seen\n")); | |
201 } | |
202 d_ptr = d_ptr->next; | |
203 } | |
204 } | |
205 close_enter_dir(&ff); | |
206 DEBUG_RET(); | |
39 | 207 } |
208 | |
209 | |
34
07177825c91b
fix signed/unsigned to allow very small pst files with only leaf nodes
carl
parents:
33
diff
changeset
|
210 |
16 | 211 int main(int argc, char** argv) { |
43 | 212 pst_item *item = NULL; |
213 pst_desc_ll *d_ptr; | |
214 char * fname = NULL; | |
48 | 215 char *d_log = NULL; |
43 | 216 int c,x; |
217 char *temp = NULL; //temporary char pointer | |
218 prog_name = argv[0]; | |
16 | 219 |
43 | 220 // command-line option handling |
221 while ((c = getopt(argc, argv, "bCc:d:hko:qrSMVw"))!= -1) { | |
222 switch (c) { | |
223 case 'b': | |
224 save_rtf_body = 0; | |
225 break; | |
226 case 'C': | |
227 mode = MODE_DECSPEW; | |
228 break; | |
229 case 'c': | |
230 if (optarg && optarg[0]=='v') | |
231 contact_mode=CMODE_VCARD; | |
232 else if (optarg && optarg[0]=='l') | |
233 contact_mode=CMODE_LIST; | |
234 else { | |
235 usage(); | |
236 exit(0); | |
237 } | |
238 break; | |
239 case 'd': | |
240 d_log = optarg; | |
241 break; | |
242 case 'h': | |
243 usage(); | |
244 exit(0); | |
245 break; | |
246 case 'V': | |
247 version(); | |
248 exit(0); | |
249 break; | |
250 case 'k': | |
251 mode = MODE_KMAIL; | |
252 break; | |
253 case 'M': | |
254 mode = MODE_SEPERATE; | |
255 mode_MH = 1; | |
256 break; | |
257 case 'o': | |
258 output_dir = optarg; | |
259 break; | |
260 case 'q': | |
261 output_mode = OUTPUT_QUIET; | |
262 break; | |
263 case 'r': | |
264 mode = MODE_RECURSE; | |
265 break; | |
266 case 'S': | |
267 mode = MODE_SEPERATE; | |
268 break; | |
269 case 'w': | |
270 overwrite = 1; | |
271 break; | |
272 default: | |
273 usage(); | |
274 exit(1); | |
275 break; | |
276 } | |
277 } | |
278 | |
279 if (argc > optind) { | |
280 fname = argv[optind]; | |
281 } else { | |
282 usage(); | |
283 exit(2); | |
284 } | |
285 | |
286 #ifdef DEBUG_ALL | |
287 // force a log file | |
288 if (!d_log) d_log = "readpst.log"; | |
289 #endif // defined DEBUG_ALL | |
290 DEBUG_INIT(d_log); | |
291 DEBUG_REGISTER_CLOSE(); | |
292 DEBUG_ENT("main"); | |
16 | 293 |
43 | 294 if (mode == MODE_DECSPEW) { |
70
b12f4e50e2e8
Patch from Joachim Metz <joachim.metz@gmail.com> for 64 bit compile.
Carl Byington <carl@five-ten-sg.com>
parents:
63
diff
changeset
|
295 FILE *fp; |
b12f4e50e2e8
Patch from Joachim Metz <joachim.metz@gmail.com> for 64 bit compile.
Carl Byington <carl@five-ten-sg.com>
parents:
63
diff
changeset
|
296 char buf[1024]; |
b12f4e50e2e8
Patch from Joachim Metz <joachim.metz@gmail.com> for 64 bit compile.
Carl Byington <carl@five-ten-sg.com>
parents:
63
diff
changeset
|
297 size_t l = 0; |
43 | 298 if (NULL == (fp = fopen(fname, "rb"))) { |
299 fprintf(stderr, "Couldn't open file %s\n", fname ); | |
52 | 300 DEBUG_RET(); |
43 | 301 return 1; |
302 } | |
39 | 303 |
73
3cb02cb1e6cd
Patch from Robert Simpson to fix doubly-linked list in the cache_ptr code, and allow arrays of unicode strings (without converting them).
Carl Byington <carl@five-ten-sg.com>
parents:
70
diff
changeset
|
304 while (0 != (l = fread(buf, 1, 1024, fp))) { |
3cb02cb1e6cd
Patch from Robert Simpson to fix doubly-linked list in the cache_ptr code, and allow arrays of unicode strings (without converting them).
Carl Byington <carl@five-ten-sg.com>
parents:
70
diff
changeset
|
305 if (0 != pst_decrypt(buf, l, PST_COMP_ENCRYPT)) |
46 | 306 fprintf(stderr, "pst_decrypt() failed (I'll try to continue)\n"); |
16 | 307 |
73
3cb02cb1e6cd
Patch from Robert Simpson to fix doubly-linked list in the cache_ptr code, and allow arrays of unicode strings (without converting them).
Carl Byington <carl@five-ten-sg.com>
parents:
70
diff
changeset
|
308 if (l != pst_fwrite(buf, 1, l, stdout)) { |
43 | 309 fprintf(stderr, "Couldn't output to stdout?\n"); |
52 | 310 DEBUG_RET(); |
43 | 311 return 1; |
312 } | |
313 } | |
52 | 314 DEBUG_RET(); |
43 | 315 return 0; |
316 } | |
16 | 317 |
43 | 318 if (output_mode != OUTPUT_QUIET) printf("Opening PST file and indexes...\n"); |
319 | |
59
7d5c637aaafb
General cleanup and code fixes.
Carl Byington <carl@five-ten-sg.com>
parents:
52
diff
changeset
|
320 RET_DERROR(pst_open(&pstfile, fname), 1, ("Error opening File\n")); |
43 | 321 RET_DERROR(pst_load_index(&pstfile), 2, ("Index Error\n")); |
322 | |
323 pst_load_extended_attributes(&pstfile); | |
16 | 324 |
43 | 325 if (chdir(output_dir)) { |
326 x = errno; | |
327 pst_close(&pstfile); | |
328 DEBUG_RET(); | |
329 DIE(("main: Cannot change to output dir %s: %s\n", output_dir, strerror(x))); | |
330 } | |
331 | |
332 if (output_mode != OUTPUT_QUIET) printf("About to start processing first record...\n"); | |
333 | |
334 d_ptr = pstfile.d_head; // first record is main record | |
46 | 335 item = pst_parse_item(&pstfile, d_ptr); |
43 | 336 if (!item || !item->message_store) { |
337 DEBUG_RET(); | |
338 DIE(("main: Could not get root record\n")); | |
339 } | |
16 | 340 |
43 | 341 // default the file_as to the same as the main filename if it doesn't exist |
342 if (!item->file_as) { | |
343 if (!(temp = strrchr(fname, '/'))) | |
344 if (!(temp = strrchr(fname, '\\'))) | |
345 temp = fname; | |
346 else | |
347 temp++; // get past the "\\" | |
348 else | |
349 temp++; // get past the "/" | |
350 item->file_as = (char*)xmalloc(strlen(temp)+1); | |
351 strcpy(item->file_as, temp); | |
352 DEBUG_MAIN(("file_as was blank, so am using %s\n", item->file_as)); | |
353 } | |
354 DEBUG_MAIN(("main: Root Folder Name: %s\n", item->file_as)); | |
16 | 355 |
43 | 356 d_ptr = pst_getTopOfFolders(&pstfile, item); |
357 if (!d_ptr) { | |
358 DEBUG_RET(); | |
359 DIE(("Top of folders record not found. Cannot continue\n")); | |
360 } | |
16 | 361 |
63
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
362 process(item, d_ptr->child); // do the children of TOPF |
46 | 363 pst_freeItem(item); |
43 | 364 pst_close(&pstfile); |
365 DEBUG_RET(); | |
366 return 0; | |
16 | 367 } |
31 | 368 |
369 | |
16 | 370 void write_email_body(FILE *f, char *body) { |
43 | 371 char *n = body; |
372 // DEBUG_MAIN(("write_email_body(): \"%s\"\n", body)); | |
373 DEBUG_ENT("write_email_body"); | |
374 while (n) { | |
375 if (strncmp(body, "From ", 5) == 0) | |
376 fprintf(f, ">"); | |
377 if ((n = strchr(body, '\n'))) { | |
378 n++; | |
73
3cb02cb1e6cd
Patch from Robert Simpson to fix doubly-linked list in the cache_ptr code, and allow arrays of unicode strings (without converting them).
Carl Byington <carl@five-ten-sg.com>
parents:
70
diff
changeset
|
379 pst_fwrite(body, n-body, 1, f); //write just a line |
43 | 380 body = n; |
381 } | |
382 } | |
73
3cb02cb1e6cd
Patch from Robert Simpson to fix doubly-linked list in the cache_ptr code, and allow arrays of unicode strings (without converting them).
Carl Byington <carl@five-ten-sg.com>
parents:
70
diff
changeset
|
383 pst_fwrite(body, strlen(body), 1, f); |
43 | 384 DEBUG_RET(); |
16 | 385 } |
31 | 386 |
387 | |
16 | 388 char *removeCR (char *c) { |
43 | 389 // converts /r/n to /n |
390 char *a, *b; | |
391 DEBUG_ENT("removeCR"); | |
392 a = b = c; | |
393 while (*a != '\0') { | |
394 *b = *a; | |
395 if (*a != '\r') | |
396 b++; | |
397 a++; | |
398 } | |
399 *b = '\0'; | |
400 DEBUG_RET(); | |
401 return c; | |
16 | 402 } |
31 | 403 |
404 | |
16 | 405 int usage() { |
43 | 406 DEBUG_ENT("usage"); |
407 version(); | |
408 printf("Usage: %s [OPTIONS] {PST FILENAME}\n", prog_name); | |
409 printf("OPTIONS:\n"); | |
63
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
410 printf("\t-C\t- Decrypt the entire file and output on stdout (not typically useful)\n"); |
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
411 printf("\t-M\t- MH. Write emails in the MH format\n"); |
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
412 printf("\t-S\t- Seperate. Write emails in the seperate format\n"); |
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
413 printf("\t-V\t- Version. Display program version\n"); |
43 | 414 printf("\t-b\t- Don't save RTF-Body attachments\n"); |
415 printf("\t-c[v|l]\t- Set the Contact output mode. -cv = VCard, -cl = EMail list\n"); | |
63
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
416 printf("\t-d <filename> \t- Debug to file. This is a binary log. Use readpstlog to print it\n"); |
43 | 417 printf("\t-h\t- Help. This screen\n"); |
418 printf("\t-k\t- KMail. Output in kmail format\n"); | |
63
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
419 printf("\t-o <dirname>\t- Output directory to write files to. CWD is changed *after* opening pst file\n"); |
43 | 420 printf("\t-q\t- Quiet. Only print error messages\n"); |
421 printf("\t-r\t- Recursive. Output in a recursive format\n"); | |
422 printf("\t-w\t- Overwrite any output mbox files\n"); | |
423 DEBUG_RET(); | |
424 return 0; | |
16 | 425 } |
31 | 426 |
427 | |
16 | 428 int version() { |
43 | 429 DEBUG_ENT("version"); |
50 | 430 printf("ReadPST / LibPST v%s\n", VERSION); |
16 | 431 #if BYTE_ORDER == BIG_ENDIAN |
43 | 432 printf("Big Endian implementation being used.\n"); |
16 | 433 #elif BYTE_ORDER == LITTLE_ENDIAN |
43 | 434 printf("Little Endian implementation being used.\n"); |
16 | 435 #else |
436 # error "Byte order not supported by this library" | |
437 #endif | |
438 #ifdef __GNUC__ | |
43 | 439 printf("GCC %d.%d : %s %s\n", __GNUC__, __GNUC_MINOR__, __DATE__, __TIME__); |
16 | 440 #endif |
43 | 441 DEBUG_RET(); |
442 return 0; | |
16 | 443 } |
31 | 444 |
445 | |
16 | 446 char *mk_kmail_dir(char *fname) { |
43 | 447 //change to that directory |
448 //make a directory based on OUTPUT_KMAIL_DIR_TEMPLATE | |
449 //allocate space for OUTPUT_TEMPLATE and form a char* with fname | |
450 //return that value | |
451 char *dir, *out_name, *index; | |
452 int x; | |
453 DEBUG_ENT("mk_kmail_dir"); | |
454 if (kmail_chdir && chdir(kmail_chdir)) { | |
455 x = errno; | |
456 DIE(("mk_kmail_dir: Cannot change to directory %s: %s\n", kmail_chdir, strerror(x))); | |
457 } | |
458 dir = malloc(strlen(fname)+strlen(OUTPUT_KMAIL_DIR_TEMPLATE)+1); | |
459 sprintf(dir, OUTPUT_KMAIL_DIR_TEMPLATE, fname); | |
460 check_filename(dir); | |
461 if (D_MKDIR(dir)) { | |
462 //error occured | |
463 if (errno != EEXIST) { | |
464 x = errno; | |
465 DIE(("mk_kmail_dir: Cannot create directory %s: %s\n", dir, strerror(x))); | |
466 } | |
467 } | |
468 kmail_chdir = realloc(kmail_chdir, strlen(dir)+1); | |
469 strcpy(kmail_chdir, dir); | |
470 free (dir); | |
16 | 471 |
43 | 472 //we should remove any existing indexes created by KMail, cause they might be different now |
473 index = malloc(strlen(fname)+strlen(KMAIL_INDEX)+1); | |
474 sprintf(index, KMAIL_INDEX, fname); | |
475 unlink(index); | |
476 free(index); | |
16 | 477 |
43 | 478 out_name = malloc(strlen(fname)+strlen(OUTPUT_TEMPLATE)+1); |
479 sprintf(out_name, OUTPUT_TEMPLATE, fname); | |
480 DEBUG_RET(); | |
481 return out_name; | |
16 | 482 } |
31 | 483 |
484 | |
16 | 485 int close_kmail_dir() { |
43 | 486 // change .. |
487 int x; | |
488 DEBUG_ENT("close_kmail_dir"); | |
489 if (kmail_chdir) { //only free kmail_chdir if not NULL. do not change directory | |
490 free(kmail_chdir); | |
491 kmail_chdir = NULL; | |
492 } else { | |
493 if (chdir("..")) { | |
494 x = errno; | |
495 DIE(("close_kmail_dir: Cannot move up dir (..): %s\n", strerror(x))); | |
496 } | |
497 } | |
498 DEBUG_RET(); | |
499 return 0; | |
16 | 500 } |
31 | 501 |
502 | |
16 | 503 // this will create a directory by that name, then make an mbox file inside |
504 // that dir. any subsequent dirs will be created by name, and they will | |
505 // contain mbox files | |
506 char *mk_recurse_dir(char *dir) { | |
43 | 507 int x; |
508 char *out_name; | |
509 DEBUG_ENT("mk_recurse_dir"); | |
510 check_filename(dir); | |
511 if (D_MKDIR (dir)) { | |
512 if (errno != EEXIST) { // not an error because it exists | |
513 x = errno; | |
514 DIE(("mk_recurse_dir: Cannot create directory %s: %s\n", dir, strerror(x))); | |
515 } | |
516 } | |
517 if (chdir (dir)) { | |
518 x = errno; | |
519 DIE(("mk_recurse_dir: Cannot change to directory %s: %s\n", dir, strerror(x))); | |
520 } | |
521 out_name = malloc(strlen("mbox")+1); | |
522 strcpy(out_name, "mbox"); | |
523 DEBUG_RET(); | |
524 return out_name; | |
16 | 525 } |
31 | 526 |
527 | |
16 | 528 int close_recurse_dir() { |
43 | 529 int x; |
530 DEBUG_ENT("close_recurse_dir"); | |
531 if (chdir("..")) { | |
532 x = errno; | |
533 DIE(("close_recurse_dir: Cannot go up dir (..): %s\n", strerror(x))); | |
534 } | |
535 DEBUG_RET(); | |
536 return 0; | |
16 | 537 } |
31 | 538 |
539 | |
39 | 540 char *mk_seperate_dir(char *dir) { |
43 | 541 size_t dirsize = strlen(dir) + 10; |
542 char dir_name[dirsize]; | |
543 int x = 0, y = 0; | |
16 | 544 |
46 | 545 DEBUG_ENT("mk_seperate_dir"); |
43 | 546 do { |
547 if (y == 0) | |
548 snprintf(dir_name, dirsize, "%s", dir); | |
549 else | |
550 snprintf(dir_name, dirsize, "%s" SEP_MAIL_FILE_TEMPLATE, dir, y); // enough for 9 digits allocated above | |
16 | 551 |
43 | 552 check_filename(dir_name); |
553 DEBUG_MAIN(("about to try creating %s\n", dir_name)); | |
554 if (D_MKDIR(dir_name)) { | |
555 if (errno != EEXIST) { // if there is an error, and it doesn't already exist | |
556 x = errno; | |
557 DIE(("mk_seperate_dir: Cannot create directory %s: %s\n", dir, strerror(x))); | |
558 } | |
559 } else { | |
560 break; | |
561 } | |
562 y++; | |
563 } while (overwrite == 0); | |
16 | 564 |
43 | 565 if (chdir(dir_name)) { |
566 x = errno; | |
75
987aa872294e
Use ftello/fseeko to properly handle large files.
Carl Byington <carl@five-ten-sg.com>
parents:
73
diff
changeset
|
567 DIE(("mk_seperate_dir: Cannot change to directory %s: %s\n", dir, strerror(x))); |
43 | 568 } |
16 | 569 |
43 | 570 if (overwrite) { |
571 // we should probably delete all files from this directory | |
16 | 572 #if !defined(WIN32) && !defined(__CYGWIN__) |
43 | 573 DIR * sdir = NULL; |
574 struct dirent *dirent = NULL; | |
575 struct stat filestat; | |
576 if (!(sdir = opendir("./"))) { | |
577 WARN(("mk_seperate_dir: Cannot open dir \"%s\" for deletion of old contents\n", "./")); | |
578 } else { | |
579 while ((dirent = readdir(sdir))) { | |
580 if (lstat(dirent->d_name, &filestat) != -1) | |
581 if (S_ISREG(filestat.st_mode)) { | |
582 if (unlink(dirent->d_name)) { | |
583 y = errno; | |
584 DIE(("mk_seperate_dir: unlink returned error on file %s: %s\n", dirent->d_name, strerror(y))); | |
585 } | |
586 } | |
587 } | |
588 } | |
26 | 589 #endif |
43 | 590 } |
16 | 591 |
43 | 592 // we don't return a filename here cause it isn't necessary. |
593 DEBUG_RET(); | |
594 return NULL; | |
16 | 595 } |
31 | 596 |
597 | |
16 | 598 int close_seperate_dir() { |
43 | 599 int x; |
600 DEBUG_ENT("close_seperate_dir"); | |
601 if (chdir("..")) { | |
602 x = errno; | |
603 DIE(("close_seperate_dir: Cannot go up dir (..): %s\n", strerror(x))); | |
604 } | |
605 DEBUG_RET(); | |
606 return 0; | |
16 | 607 } |
31 | 608 |
609 | |
16 | 610 int mk_seperate_file(struct file_ll *f) { |
43 | 611 const int name_offset = 1; |
612 DEBUG_ENT("mk_seperate_file"); | |
613 DEBUG_MAIN(("opening next file to save email\n")); | |
614 if (f->email_count > 999999999) { // bigger than nine 9's | |
615 DIE(("mk_seperate_file: The number of emails in this folder has become too high to handle")); | |
616 } | |
617 sprintf(f->name, SEP_MAIL_FILE_TEMPLATE, f->email_count + name_offset); | |
618 if (f->output) fclose(f->output); | |
619 f->output = NULL; | |
620 check_filename(f->name); | |
621 if (!(f->output = fopen(f->name, "w"))) { | |
622 DIE(("mk_seperate_file: Cannot open file to save email \"%s\"\n", f->name)); | |
623 } | |
624 DEBUG_RET(); | |
625 return 0; | |
16 | 626 } |
31 | 627 |
628 | |
16 | 629 char *my_stristr(char *haystack, char *needle) { |
43 | 630 // my_stristr varies from strstr in that its searches are case-insensitive |
631 char *x=haystack, *y=needle, *z = NULL; | |
632 DEBUG_ENT("my_stristr"); | |
52 | 633 if (!haystack || !needle) { |
634 DEBUG_RET(); | |
43 | 635 return NULL; |
52 | 636 } |
43 | 637 while (*y != '\0' && *x != '\0') { |
638 if (tolower(*y) == tolower(*x)) { | |
639 // move y on one | |
640 y++; | |
641 if (!z) { | |
642 z = x; // store first position in haystack where a match is made | |
643 } | |
644 } else { | |
645 y = needle; // reset y to the beginning of the needle | |
646 z = NULL; // reset the haystack storage point | |
647 } | |
648 x++; // advance the search in the haystack | |
649 } | |
650 DEBUG_RET(); | |
651 return z; | |
16 | 652 } |
31 | 653 |
654 | |
41
183ae993b9ad
security fix for potential buffer overrun in lz decompress
carl
parents:
39
diff
changeset
|
655 void check_filename(char *fname) { |
43 | 656 char *t = fname; |
657 DEBUG_ENT("check_filename"); | |
658 if (!t) { | |
659 DEBUG_RET(); | |
52 | 660 return; |
43 | 661 } |
662 while ((t = strpbrk(t, "/\\:"))) { | |
663 // while there are characters in the second string that we don't want | |
664 *t = '_'; //replace them with an underscore | |
665 } | |
666 DEBUG_RET(); | |
16 | 667 } |
31 | 668 |
669 | |
16 | 670 // The sole purpose of this function is to bypass the pseudo-header prologue |
671 // that Microsoft Outlook inserts at the beginning of the internet email | |
672 // headers for emails stored in their "Personal Folders" files. | |
673 char *skip_header_prologue(char *headers) { | |
43 | 674 const char *bad = "Microsoft Mail Internet Headers"; |
73
3cb02cb1e6cd
Patch from Robert Simpson to fix doubly-linked list in the cache_ptr code, and allow arrays of unicode strings (without converting them).
Carl Byington <carl@five-ten-sg.com>
parents:
70
diff
changeset
|
675 if (strncmp(headers, bad, strlen(bad)) == 0) { |
43 | 676 // Found the offensive header prologue |
677 char *pc = strchr(headers, '\n'); | |
678 return pc + 1; | |
679 } | |
680 return headers; | |
16 | 681 } |
682 | |
31 | 683 |
25 | 684 void write_separate_attachment(char f_name[], pst_item_attach* current_attach, int attach_num, pst_file* pst) |
685 { | |
43 | 686 FILE *fp = NULL; |
687 int x = 0; | |
688 char *temp = NULL; | |
31 | 689 |
43 | 690 // If there is a long filename (filename2) use that, otherwise |
691 // use the 8.3 filename (filename1) | |
692 char *attach_filename = (current_attach->filename2) ? current_attach->filename2 | |
693 : current_attach->filename1; | |
46 | 694 DEBUG_ENT("write_separate_attachment"); |
25 | 695 |
43 | 696 check_filename(f_name); |
697 if (!attach_filename) { | |
698 // generate our own (dummy) filename for the attachement | |
699 temp = xmalloc(strlen(f_name)+15); | |
700 sprintf(temp, "%s-attach%i", f_name, attach_num); | |
701 } else { | |
702 // have an attachment name, make sure it's unique | |
703 temp = xmalloc(strlen(f_name)+strlen(attach_filename)+15); | |
704 do { | |
705 if (fp) fclose(fp); | |
706 if (x == 0) | |
707 sprintf(temp, "%s-%s", f_name, attach_filename); | |
708 else | |
709 sprintf(temp, "%s-%s-%i", f_name, attach_filename, x); | |
710 } while ((fp = fopen(temp, "r")) && ++x < 99999999); | |
711 if (x > 99999999) { | |
712 DIE(("error finding attachment name. exhausted possibilities to %s\n", temp)); | |
713 } | |
714 } | |
715 DEBUG_EMAIL(("Saving attachment to %s\n", temp)); | |
716 if (!(fp = fopen(temp, "w"))) { | |
717 WARN(("write_separate_attachment: Cannot open attachment save file \"%s\"\n", temp)); | |
718 } else { | |
719 if (current_attach->data) | |
73
3cb02cb1e6cd
Patch from Robert Simpson to fix doubly-linked list in the cache_ptr code, and allow arrays of unicode strings (without converting them).
Carl Byington <carl@five-ten-sg.com>
parents:
70
diff
changeset
|
720 pst_fwrite(current_attach->data, 1, current_attach->size, fp); |
43 | 721 else { |
46 | 722 (void)pst_attach_to_file(pst, current_attach, fp); |
43 | 723 } |
724 fclose(fp); | |
725 } | |
726 if (temp) free(temp); | |
727 DEBUG_RET(); | |
25 | 728 } |
729 | |
31 | 730 |
25 | 731 void write_inline_attachment(FILE* f_output, pst_item_attach* current_attach, char boundary[], pst_file* pst) |
732 { | |
46 | 733 char *enc = NULL; // base64 encoded attachment |
43 | 734 DEBUG_ENT("write_inline_attachment"); |
735 DEBUG_EMAIL(("Attachment Size is %i\n", current_attach->size)); | |
736 DEBUG_EMAIL(("Attachment Pointer is %p\n", current_attach->data)); | |
737 if (current_attach->data) { | |
738 enc = base64_encode (current_attach->data, current_attach->size); | |
739 if (!enc) { | |
740 DEBUG_EMAIL(("ERROR base64_encode returned NULL. Must have failed\n")); | |
52 | 741 DEBUG_RET(); |
43 | 742 return; |
743 } | |
744 } | |
745 if (boundary) { | |
746 char *attach_filename; | |
747 fprintf(f_output, "\n--%s\n", boundary); | |
748 if (!current_attach->mimetype) { | |
749 fprintf(f_output, "Content-type: %s\n", MIME_TYPE_DEFAULT); | |
750 } else { | |
751 fprintf(f_output, "Content-type: %s\n", current_attach->mimetype); | |
752 } | |
753 fprintf(f_output, "Content-transfer-encoding: base64\n"); | |
754 // If there is a long filename (filename2) use that, otherwise | |
755 // use the 8.3 filename (filename1) | |
756 if (current_attach->filename2) { | |
48 | 757 attach_filename = current_attach->filename2; |
43 | 758 } else { |
48 | 759 attach_filename = current_attach->filename1; |
43 | 760 } |
761 if (!attach_filename) { | |
762 fprintf(f_output, "Content-Disposition: inline\n\n"); | |
763 } else { | |
764 fprintf(f_output, "Content-Disposition: attachment; filename=\"%s\"\n\n", attach_filename); | |
765 } | |
766 } | |
767 if (current_attach->data) { | |
73
3cb02cb1e6cd
Patch from Robert Simpson to fix doubly-linked list in the cache_ptr code, and allow arrays of unicode strings (without converting them).
Carl Byington <carl@five-ten-sg.com>
parents:
70
diff
changeset
|
768 pst_fwrite(enc, 1, strlen(enc), f_output); |
43 | 769 DEBUG_EMAIL(("Attachment Size after encoding is %i\n", strlen(enc))); |
770 free(enc); // caught by valgrind | |
771 } else { | |
46 | 772 (void)pst_attach_to_file_base64(pst, current_attach, f_output); |
43 | 773 } |
774 fprintf(f_output, "\n\n"); | |
775 DEBUG_RET(); | |
25 | 776 } |
777 | |
31 | 778 |
779 void write_normal_email(FILE* f_output, char f_name[], pst_item* item, int mode, int mode_MH, pst_file* pst, int save_rtf) | |
25 | 780 { |
43 | 781 char *boundary = NULL; // the boundary marker between multipart sections |
782 int boundary_created = 0; // we have not (yet) created a new boundary | |
783 char *temp = NULL; | |
784 int attach_num, base64_body = 0; | |
785 time_t em_time; | |
786 char *c_time; | |
787 pst_item_attach* current_attach; | |
46 | 788 DEBUG_ENT("write_normal_email"); |
25 | 789 |
43 | 790 // convert the sent date if it exists, or set it to a fixed date |
791 if (item->email->sent_date) { | |
792 em_time = fileTimeToUnixTime(item->email->sent_date, 0); | |
793 c_time = ctime(&em_time); | |
794 if (c_time) | |
795 c_time[strlen(c_time)-1] = '\0'; //remove end \n | |
796 else | |
797 c_time = "Fri Dec 28 12:06:21 2001"; | |
798 } else | |
799 c_time= "Fri Dec 28 12:06:21 2001"; | |
25 | 800 |
43 | 801 // we will always look at the header to discover some stuff |
802 if (item->email->header ) { | |
803 char *b1, *b2; | |
804 // see if there is a boundary variable there | |
805 // this search MUST be made case insensitive (DONE). | |
806 // Also, we should check to find out if we are looking | |
807 // at the boundary associated with content-type, and that | |
808 // the content type really is multipart | |
25 | 809 |
43 | 810 removeCR(item->email->header); |
25 | 811 |
43 | 812 if ((b2 = my_stristr(item->email->header, "boundary="))) { |
813 int len; | |
814 b2 += strlen("boundary="); // move boundary to first char of marker | |
31 | 815 |
43 | 816 if (*b2 == '"') { |
817 b2++; | |
818 b1 = strchr(b2, '"'); // find terminating quote | |
819 } else { | |
820 b1 = b2; | |
821 while (isgraph(*b1)) // find first char that isn't part of boundary | |
822 b1++; | |
823 } | |
824 len = b1 - b2; | |
825 boundary = malloc(len+1); //malloc that length | |
826 strncpy(boundary, b2, len); // copy boundary to another variable | |
827 boundary[len] = '\0'; | |
828 b1 = b2 = boundary; | |
829 while (*b2 != '\0') { // remove any CRs and Tabs | |
830 if (*b2 != '\n' && *b2 != '\r' && *b2 != '\t') { | |
831 *b1 = *b2; | |
832 b1++; | |
833 } | |
834 b2++; | |
835 } | |
836 *b1 = '\0'; | |
31 | 837 |
43 | 838 DEBUG_EMAIL(("Found boundary of - %s\n", boundary)); |
839 } else { | |
840 DEBUG_EMAIL(("boundary not found in header\n")); | |
841 } | |
25 | 842 |
43 | 843 // also possible to set 7bit encoding detection here. |
844 if ((b2 = my_stristr(item->email->header, "Content-Transfer-Encoding:"))) { | |
845 if ((b2 = strchr(b2, ':'))) { | |
846 b2++; // skip to the : at the end of the string | |
31 | 847 |
43 | 848 while (*b2 == ' ' || *b2 == '\t') |
849 b2++; | |
850 if (pst_strincmp(b2, "base64", 6)==0) { | |
851 DEBUG_EMAIL(("body is base64 encoded\n")); | |
852 base64_body = 1; | |
853 } | |
854 } else { | |
855 DEBUG_WARN(("found a ':' during the my_stristr, but not after that..\n")); | |
856 } | |
857 } | |
858 } | |
25 | 859 |
43 | 860 if (!boundary && (item->attach || (item->email->body && item->email->htmlbody) |
861 || item->email->rtf_compressed || item->email->encrypted_body | |
862 || item->email->encrypted_htmlbody)) { | |
863 // we need to create a boundary here. | |
864 DEBUG_EMAIL(("must create own boundary. oh dear.\n")); | |
865 boundary = malloc(50 * sizeof(char)); // allow 50 chars for boundary | |
866 boundary[0] = '\0'; | |
867 sprintf(boundary, "--boundary-LibPST-iamunique-%i_-_-", rand()); | |
868 DEBUG_EMAIL(("created boundary is %s\n", boundary)); | |
869 boundary_created = 1; | |
870 } | |
25 | 871 |
43 | 872 DEBUG_EMAIL(("About to print Header\n")); |
31 | 873 |
43 | 874 if (item && item->email && item->email->subject && item->email->subject->subj) { |
875 DEBUG_EMAIL(("item->email->subject->subj = %s\n", item->email->subject->subj)); | |
876 } | |
31 | 877 |
43 | 878 if (item->email->header) { |
879 int len; | |
880 char *soh = NULL; // real start of headers. | |
31 | 881 |
43 | 882 // some of the headers we get from the file are not properly defined. |
883 // they can contain some email stuff too. We will cut off the header | |
884 // when we see a \n\n or \r\n\r\n | |
885 removeCR(item->email->header); | |
886 temp = strstr(item->email->header, "\n\n"); | |
25 | 887 |
43 | 888 if (temp) { |
889 DEBUG_EMAIL(("Found body text in header\n")); | |
890 temp[1] = '\0'; // stop after first \n | |
891 } | |
31 | 892 |
43 | 893 // Now, write out the header... |
894 soh = skip_header_prologue(item->email->header); | |
895 if (mode != MODE_SEPERATE) { | |
896 // don't put rubbish in if we are doing seperate | |
897 if (strncmp(soh, "X-From_: ", 9) == 0 ) { | |
898 fputs("From ", f_output); | |
899 soh += 9; | |
900 } else | |
901 fprintf(f_output, "From \"%s\" %s\n", item->email->outlook_sender_name, c_time); | |
902 } | |
903 fprintf(f_output, "%s", soh); | |
904 len = strlen(soh); | |
905 if (!len || (soh[len-1] != '\n')) fprintf(f_output, "\n"); | |
31 | 906 |
43 | 907 } else { |
63
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
908 //make up our own headers |
43 | 909 if (mode != MODE_SEPERATE) { |
910 // don't want this first line for this mode | |
911 if (item->email->outlook_sender_name) { | |
912 temp = item->email->outlook_sender_name; | |
913 } else { | |
914 temp = "(readpst_null)"; | |
915 } | |
916 fprintf(f_output, "From \"%s\" %s\n", temp, c_time); | |
917 } | |
31 | 918 |
43 | 919 temp = item->email->outlook_sender; |
920 if (!temp) temp = ""; | |
921 fprintf(f_output, "From: \"%s\" <%s>\n", item->email->outlook_sender_name, temp); | |
31 | 922 |
43 | 923 if (item->email->subject) { |
924 fprintf(f_output, "Subject: %s\n", item->email->subject->subj); | |
925 } else { | |
926 fprintf(f_output, "Subject: \n"); | |
927 } | |
31 | 928 |
43 | 929 fprintf(f_output, "To: %s\n", item->email->sentto_address); |
930 if (item->email->cc_address) { | |
931 fprintf(f_output, "Cc: %s\n", item->email->cc_address); | |
932 } | |
31 | 933 |
43 | 934 if (item->email->sent_date) { |
63
cfd6175f9334
Start work on pst2dii to convert to Summation dii load file format.
Carl Byington <carl@five-ten-sg.com>
parents:
59
diff
changeset
|
935 char c_time[C_TIME_SIZE]; |
43 | 936 strftime(c_time, C_TIME_SIZE, "%a, %d %b %Y %H:%M:%S %z", gmtime(&em_time)); |
937 fprintf(f_output, "Date: %s\n", c_time); | |
938 } | |
939 } | |
25 | 940 |
43 | 941 fprintf(f_output, "MIME-Version: 1.0\n"); |
942 if (boundary && boundary_created) { | |
943 // if we created the boundary, then it has NOT already been printed | |
944 // in the headers above. | |
945 if (item->attach) { | |
946 // write the boundary stuff if we have attachments | |
947 fprintf(f_output, "Content-type: multipart/mixed;\n\tboundary=\"%s\"\n", boundary); | |
948 } else if (boundary) { | |
949 // else if we have multipart/alternative then tell it so | |
950 fprintf(f_output, "Content-type: multipart/alternative;\n\tboundary=\"%s\"\n", boundary); | |
951 } else if (item->email->htmlbody) { | |
952 fprintf(f_output, "Content-type: text/html\n"); | |
953 } | |
954 } | |
955 fprintf(f_output, "\n"); // start the body | |
956 DEBUG_EMAIL(("About to print Body\n")); | |
25 | 957 |
43 | 958 if (item->email->body) { |
959 if (boundary) { | |
960 fprintf(f_output, "\n--%s\n", boundary); | |
961 fprintf(f_output, "Content-type: text/plain\n"); | |
962 if (base64_body) | |
963 fprintf(f_output, "Content-Transfer-Encoding: base64\n"); | |
964 fprintf(f_output, "\n"); | |
965 } | |
966 removeCR(item->email->body); | |
967 if (base64_body) { | |
968 char *enc = base64_encode(item->email->body, strlen(item->email->body)); | |
969 if (enc) { | |
970 write_email_body(f_output, enc); | |
971 free(enc); | |
972 } | |
973 } | |
974 else { | |
975 write_email_body(f_output, item->email->body); | |
976 } | |
977 } | |
31 | 978 |
43 | 979 if (item->email->htmlbody) { |
980 if (boundary) { | |
981 fprintf(f_output, "\n--%s\n", boundary); | |
982 fprintf(f_output, "Content-type: text/html\n"); | |
983 if (base64_body) fprintf(f_output, "Content-Transfer-Encoding: base64\n"); | |
984 fprintf(f_output, "\n"); | |
985 } | |
986 removeCR(item->email->htmlbody); | |
987 if (base64_body) { | |
988 char *enc = base64_encode(item->email->htmlbody, strlen(item->email->htmlbody)); | |
989 if (enc) { | |
990 write_email_body(f_output, enc); | |
991 free(enc); | |
992 } | |
993 } | |
994 else { | |
995 write_email_body(f_output, item->email->htmlbody); | |
996 } | |
997 } | |
25 | 998 |
43 | 999 if (item->email->rtf_compressed && save_rtf) { |
1000 //int32_t tester; | |
1001 DEBUG_EMAIL(("Adding RTF body as attachment\n")); | |
1002 current_attach = (pst_item_attach*)xmalloc(sizeof(pst_item_attach)); | |
1003 memset(current_attach, 0, sizeof(pst_item_attach)); | |
1004 current_attach->next = item->attach; | |
1005 item->attach = current_attach; | |
1006 current_attach->data = lzfu_decompress(item->email->rtf_compressed, item->email->rtf_compressed_size, ¤t_attach->size); | |
1007 current_attach->filename2 = xmalloc(strlen(RTF_ATTACH_NAME)+2); | |
1008 strcpy(current_attach->filename2, RTF_ATTACH_NAME); | |
1009 current_attach->mimetype = xmalloc(strlen(RTF_ATTACH_TYPE)+2); | |
1010 strcpy(current_attach->mimetype, RTF_ATTACH_TYPE); | |
1011 //memcpy(&tester, item->email->rtf_compressed+sizeof(int32_t), sizeof(int32_t)); | |
1012 //LE32_CPU(tester); | |
1013 //printf("lz produced %d bytes, rtf claims %d bytes\n", current_attach->size, tester); | |
1014 } | |
31 | 1015 |
43 | 1016 if (item->email->encrypted_body || item->email->encrypted_htmlbody) { |
1017 // if either the body or htmlbody is encrypted, add them as attachments | |
1018 if (item->email->encrypted_body) { | |
1019 DEBUG_EMAIL(("Adding Encrypted Body as attachment\n")); | |
1020 current_attach = (pst_item_attach*) xmalloc(sizeof(pst_item_attach)); | |
1021 memset(current_attach, 0, sizeof(pst_item_attach)); | |
1022 current_attach->next = item->attach; | |
1023 item->attach = current_attach; | |
1024 current_attach->data = item->email->encrypted_body; | |
1025 current_attach->size = item->email->encrypted_body_size; | |
1026 item->email->encrypted_body = NULL; | |
1027 } | |
31 | 1028 |
43 | 1029 if (item->email->encrypted_htmlbody) { |
1030 DEBUG_EMAIL(("Adding encrypted HTML body as attachment\n")); | |
1031 current_attach = (pst_item_attach*) xmalloc(sizeof(pst_item_attach)); | |
1032 memset(current_attach, 0, sizeof(pst_item_attach)); | |
1033 current_attach->next = item->attach; | |
1034 item->attach = current_attach; | |
1035 current_attach->data = item->email->encrypted_htmlbody; | |
1036 current_attach->size = item->email->encrypted_htmlbody_size; | |
1037 item->email->encrypted_htmlbody = NULL; | |
1038 } | |
1039 write_email_body(f_output, "The body of this email is encrypted. This isn't supported yet, but the body is now an attachment\n"); | |
1040 } | |
31 | 1041 |
43 | 1042 // attachments |
1043 attach_num = 0; | |
1044 for (current_attach = item->attach; current_attach; current_attach = current_attach->next) { | |
1045 DEBUG_EMAIL(("Attempting Attachment encoding\n")); | |
1046 if (!current_attach->data) { | |
1047 DEBUG_EMAIL(("Data of attachment is NULL!. Size is supposed to be %i\n", current_attach->size)); | |
1048 } | |
1049 if (mode == MODE_SEPERATE && !mode_MH) | |
1050 write_separate_attachment(f_name, current_attach, ++attach_num, pst); | |
1051 else | |
1052 write_inline_attachment(f_output, current_attach, boundary, pst); | |
1053 } | |
1054 if (mode != MODE_SEPERATE) { /* do not add a boundary after the last attachment for mode_MH */ | |
1055 DEBUG_EMAIL(("Writing buffer between emails\n")); | |
1056 if (boundary) fprintf(f_output, "\n--%s--\n", boundary); | |
1057 fprintf(f_output, "\n\n"); | |
1058 } | |
1059 if (boundary) free (boundary); | |
1060 DEBUG_RET(); | |
25 | 1061 } |
1062 | |
31 | 1063 |
25 | 1064 void write_vcard(FILE* f_output, pst_item_contact* contact, char comment[]) |
1065 { | |
43 | 1066 // We can only call rfc escape once per printf, since the second call |
1067 // may free the buffer returned by the first call. | |
1068 // I had tried to place those into a single printf - Carl. | |
39 | 1069 |
43 | 1070 DEBUG_ENT("write_vcard"); |
50 | 1071 // the specification I am following is (hopefully) RFC2426 vCard Mime Directory Profile |
43 | 1072 fprintf(f_output, "BEGIN:VCARD\n"); |
1073 fprintf(f_output, "FN:%s\n", pst_rfc2426_escape(contact->fullname)); | |
39 | 1074 |
43 | 1075 //fprintf(f_output, "N:%s;%s;%s;%s;%s\n", |
1076 fprintf(f_output, "N:%s;", (!contact->surname) ? "" : pst_rfc2426_escape(contact->surname)); | |
1077 fprintf(f_output, "%s;", (!contact->first_name) ? "" : pst_rfc2426_escape(contact->first_name)); | |
1078 fprintf(f_output, "%s;", (!contact->middle_name) ? "" : pst_rfc2426_escape(contact->middle_name)); | |
1079 fprintf(f_output, "%s;", (!contact->display_name_prefix) ? "" : pst_rfc2426_escape(contact->display_name_prefix)); | |
1080 fprintf(f_output, "%s\n", (!contact->suffix) ? "" : pst_rfc2426_escape(contact->suffix)); | |
39 | 1081 |
43 | 1082 if (contact->nickname) |
1083 fprintf(f_output, "NICKNAME:%s\n", pst_rfc2426_escape(contact->nickname)); | |
1084 if (contact->address1) | |
1085 fprintf(f_output, "EMAIL:%s\n", pst_rfc2426_escape(contact->address1)); | |
1086 if (contact->address2) | |
1087 fprintf(f_output, "EMAIL:%s\n", pst_rfc2426_escape(contact->address2)); | |
1088 if (contact->address3) | |
1089 fprintf(f_output, "EMAIL:%s\n", pst_rfc2426_escape(contact->address3)); | |
1090 if (contact->birthday) | |
1091 fprintf(f_output, "BDAY:%s\n", pst_rfc2425_datetime_format(contact->birthday)); | |
39 | 1092 |
43 | 1093 if (contact->home_address) { |
1094 //fprintf(f_output, "ADR;TYPE=home:%s;%s;%s;%s;%s;%s;%s\n", | |
1095 fprintf(f_output, "ADR;TYPE=home:%s;", (!contact->home_po_box) ? "" : pst_rfc2426_escape(contact->home_po_box)); | |
1096 fprintf(f_output, "%s;", ""); // extended Address | |
1097 fprintf(f_output, "%s;", (!contact->home_street) ? "" : pst_rfc2426_escape(contact->home_street)); | |
1098 fprintf(f_output, "%s;", (!contact->home_city) ? "" : pst_rfc2426_escape(contact->home_city)); | |
1099 fprintf(f_output, "%s;", (!contact->home_state) ? "" : pst_rfc2426_escape(contact->home_state)); | |
1100 fprintf(f_output, "%s;", (!contact->home_postal_code) ? "" : pst_rfc2426_escape(contact->home_postal_code)); | |
1101 fprintf(f_output, "%s\n", (!contact->home_country) ? "" : pst_rfc2426_escape(contact->home_country)); | |
1102 fprintf(f_output, "LABEL;TYPE=home:%s\n", pst_rfc2426_escape(contact->home_address)); | |
1103 } | |
39 | 1104 |
43 | 1105 if (contact->business_address) { |
1106 //fprintf(f_output, "ADR;TYPE=work:%s;%s;%s;%s;%s;%s;%s\n", | |
1107 fprintf(f_output, "ADR;TYPE=work:%s;", (!contact->business_po_box) ? "" : pst_rfc2426_escape(contact->business_po_box)); | |
1108 fprintf(f_output, "%s;", ""); // extended Address | |
1109 fprintf(f_output, "%s;", (!contact->business_street) ? "" : pst_rfc2426_escape(contact->business_street)); | |
1110 fprintf(f_output, "%s;", (!contact->business_city) ? "" : pst_rfc2426_escape(contact->business_city)); | |
1111 fprintf(f_output, "%s;", (!contact->business_state) ? "" : pst_rfc2426_escape(contact->business_state)); | |
1112 fprintf(f_output, "%s;", (!contact->business_postal_code) ? "" : pst_rfc2426_escape(contact->business_postal_code)); | |
1113 fprintf(f_output, "%s\n", (!contact->business_country) ? "" : pst_rfc2426_escape(contact->business_country)); | |
1114 fprintf(f_output, "LABEL;TYPE=work:%s\n", pst_rfc2426_escape(contact->business_address)); | |
1115 } | |
39 | 1116 |
43 | 1117 if (contact->other_address) { |
1118 //fprintf(f_output, "ADR;TYPE=postal:%s;%s;%s;%s;%s;%s;%s\n", | |
1119 fprintf(f_output, "ADR;TYPE=postal:%s;",(!contact->other_po_box) ? "" : pst_rfc2426_escape(contact->other_po_box)); | |
1120 fprintf(f_output, "%s;", ""); // extended Address | |
1121 fprintf(f_output, "%s;", (!contact->other_street) ? "" : pst_rfc2426_escape(contact->other_street)); | |
1122 fprintf(f_output, "%s;", (!contact->other_city) ? "" : pst_rfc2426_escape(contact->other_city)); | |
1123 fprintf(f_output, "%s;", (!contact->other_state) ? "" : pst_rfc2426_escape(contact->other_state)); | |
1124 fprintf(f_output, "%s;", (!contact->other_postal_code) ? "" : pst_rfc2426_escape(contact->other_postal_code)); | |
1125 fprintf(f_output, "%s\n", (!contact->other_country) ? "" : pst_rfc2426_escape(contact->other_country)); | |
1126 fprintf(f_output, "LABEL;TYPE=postal:%s\n", pst_rfc2426_escape(contact->other_address)); | |
1127 } | |
39 | 1128 |
43 | 1129 if (contact->business_fax) fprintf(f_output, "TEL;TYPE=work,fax:%s\n", pst_rfc2426_escape(contact->business_fax)); |
1130 if (contact->business_phone) fprintf(f_output, "TEL;TYPE=work,voice:%s\n", pst_rfc2426_escape(contact->business_phone)); | |
1131 if (contact->business_phone2) fprintf(f_output, "TEL;TYPE=work,voice:%s\n", pst_rfc2426_escape(contact->business_phone2)); | |
1132 if (contact->car_phone) fprintf(f_output, "TEL;TYPE=car,voice:%s\n", pst_rfc2426_escape(contact->car_phone)); | |
1133 if (contact->home_fax) fprintf(f_output, "TEL;TYPE=home,fax:%s\n", pst_rfc2426_escape(contact->home_fax)); | |
1134 if (contact->home_phone) fprintf(f_output, "TEL;TYPE=home,voice:%s\n", pst_rfc2426_escape(contact->home_phone)); | |
1135 if (contact->home_phone2) fprintf(f_output, "TEL;TYPE=home,voice:%s\n", pst_rfc2426_escape(contact->home_phone2)); | |
1136 if (contact->isdn_phone) fprintf(f_output, "TEL;TYPE=isdn:%s\n", pst_rfc2426_escape(contact->isdn_phone)); | |
1137 if (contact->mobile_phone) fprintf(f_output, "TEL;TYPE=cell,voice:%s\n", pst_rfc2426_escape(contact->mobile_phone)); | |
1138 if (contact->other_phone) fprintf(f_output, "TEL;TYPE=msg:%s\n", pst_rfc2426_escape(contact->other_phone)); | |
1139 if (contact->pager_phone) fprintf(f_output, "TEL;TYPE=pager:%s\n", pst_rfc2426_escape(contact->pager_phone)); | |
1140 if (contact->primary_fax) fprintf(f_output, "TEL;TYPE=fax,pref:%s\n", pst_rfc2426_escape(contact->primary_fax)); | |
1141 if (contact->primary_phone) fprintf(f_output, "TEL;TYPE=phone,pref:%s\n", pst_rfc2426_escape(contact->primary_phone)); | |
1142 if (contact->radio_phone) fprintf(f_output, "TEL;TYPE=pcs:%s\n", pst_rfc2426_escape(contact->radio_phone)); | |
1143 if (contact->telex) fprintf(f_output, "TEL;TYPE=bbs:%s\n", pst_rfc2426_escape(contact->telex)); | |
1144 if (contact->job_title) fprintf(f_output, "TITLE:%s\n", pst_rfc2426_escape(contact->job_title)); | |
1145 if (contact->profession) fprintf(f_output, "ROLE:%s\n", pst_rfc2426_escape(contact->profession)); | |
1146 if (contact->assistant_name || contact->assistant_phone) { | |
1147 fprintf(f_output, "AGENT:BEGIN:VCARD\n"); | |
1148 if (contact->assistant_name) fprintf(f_output, "FN:%s\n", pst_rfc2426_escape(contact->assistant_name)); | |
1149 if (contact->assistant_phone) fprintf(f_output, "TEL:%s\n", pst_rfc2426_escape(contact->assistant_phone)); | |
1150 } | |
1151 if (contact->company_name) fprintf(f_output, "ORG:%s\n", pst_rfc2426_escape(contact->company_name)); | |
1152 if (comment) fprintf(f_output, "NOTE:%s\n", pst_rfc2426_escape(comment)); | |
25 | 1153 |
43 | 1154 fprintf(f_output, "VERSION: 3.0\n"); |
1155 fprintf(f_output, "END:VCARD\n\n"); | |
1156 DEBUG_RET(); | |
25 | 1157 } |
1158 | |
31 | 1159 |
25 | 1160 void write_appointment(FILE* f_output, pst_item_appointment* appointment, |
43 | 1161 pst_item_email* email, FILETIME* create_date, FILETIME* modify_date) |
25 | 1162 { |
43 | 1163 fprintf(f_output, "BEGIN:VEVENT\n"); |
1164 if (create_date) | |
1165 fprintf(f_output, "CREATED:%s\n", | |
1166 pst_rfc2445_datetime_format(create_date)); | |
1167 if (modify_date) | |
1168 fprintf(f_output, "LAST-MOD:%s\n", | |
1169 pst_rfc2445_datetime_format(modify_date)); | |
1170 if (email && email->subject) | |
1171 fprintf(f_output, "SUMMARY:%s\n", | |
1172 pst_rfc2426_escape(email->subject->subj)); | |
1173 if (email && email->body) | |
1174 fprintf(f_output, "DESCRIPTION:%s\n", | |
1175 pst_rfc2426_escape(email->body)); | |
1176 if (appointment && appointment->start) | |
1177 fprintf(f_output, "DTSTART;VALUE=DATE-TIME:%s\n", | |
1178 pst_rfc2445_datetime_format(appointment->start)); | |
1179 if (appointment && appointment->end) | |
1180 fprintf(f_output, "DTEND;VALUE=DATE-TIME:%s\n", | |
1181 pst_rfc2445_datetime_format(appointment->end)); | |
1182 if (appointment && appointment->location) | |
1183 fprintf(f_output, "LOCATION:%s\n", | |
1184 pst_rfc2426_escape(appointment->location)); | |
1185 if (appointment) { | |
1186 switch (appointment->showas) { | |
50 | 1187 case PST_FREEBUSY_TENTATIVE: |
1188 fprintf(f_output, "STATUS:TENTATIVE\n"); | |
1189 break; | |
1190 case PST_FREEBUSY_FREE: | |
1191 // mark as transparent and as confirmed | |
1192 fprintf(f_output, "TRANSP:TRANSPARENT\n"); | |
1193 case PST_FREEBUSY_BUSY: | |
1194 case PST_FREEBUSY_OUT_OF_OFFICE: | |
1195 fprintf(f_output, "STATUS:CONFIRMED\n"); | |
1196 break; | |
43 | 1197 } |
1198 switch (appointment->label) { | |
50 | 1199 case PST_APP_LABEL_NONE: |
1200 fprintf(f_output, "CATEGORIES:NONE\n"); | |
1201 break; | |
1202 case PST_APP_LABEL_IMPORTANT: | |
1203 fprintf(f_output, "CATEGORIES:IMPORTANT\n"); | |
1204 break; | |
1205 case PST_APP_LABEL_BUSINESS: | |
1206 fprintf(f_output, "CATEGORIES:BUSINESS\n"); | |
1207 break; | |
1208 case PST_APP_LABEL_PERSONAL: | |
1209 fprintf(f_output, "CATEGORIES:PERSONAL\n"); | |
1210 break; | |
1211 case PST_APP_LABEL_VACATION: | |
1212 fprintf(f_output, "CATEGORIES:VACATION\n"); | |
1213 break; | |
1214 case PST_APP_LABEL_MUST_ATTEND: | |
1215 fprintf(f_output, "CATEGORIES:MUST-ATTEND\n"); | |
1216 break; | |
1217 case PST_APP_LABEL_TRAVEL_REQ: | |
1218 fprintf(f_output, "CATEGORIES:TRAVEL-REQUIRED\n"); | |
1219 break; | |
1220 case PST_APP_LABEL_NEEDS_PREP: | |
1221 fprintf(f_output, "CATEGORIES:NEEDS-PREPARATION\n"); | |
1222 break; | |
1223 case PST_APP_LABEL_BIRTHDAY: | |
1224 fprintf(f_output, "CATEGORIES:BIRTHDAY\n"); | |
1225 break; | |
1226 case PST_APP_LABEL_ANNIVERSARY: | |
1227 fprintf(f_output, "CATEGORIES:ANNIVERSARY\n"); | |
1228 break; | |
1229 case PST_APP_LABEL_PHONE_CALL: | |
1230 fprintf(f_output, "CATEGORIES:PHONE-CALL\n"); | |
1231 break; | |
43 | 1232 } |
1233 } | |
1234 fprintf(f_output, "END:VEVENT\n\n"); | |
25 | 1235 } |
1236 | |
31 | 1237 |
39 | 1238 void create_enter_dir(struct file_ll* f, pst_item *item) |
25 | 1239 { |
43 | 1240 f->email_count = 0; |
1241 f->skip_count = 0; | |
1242 f->type = item->type; | |
1243 f->stored_count = (item->folder) ? item->folder->email_count : 0; | |
39 | 1244 |
43 | 1245 DEBUG_ENT("create_enter_dir"); |
1246 if (mode == MODE_KMAIL) | |
1247 f->name = mk_kmail_dir(item->file_as); //create directory and form filename | |
1248 else if (mode == MODE_RECURSE) | |
1249 f->name = mk_recurse_dir(item->file_as); | |
1250 else if (mode == MODE_SEPERATE) { | |
1251 // do similar stuff to recurse here. | |
1252 mk_seperate_dir(item->file_as); | |
1253 f->name = (char*) xmalloc(10); | |
1254 memset(f->name, 0, 10); | |
1255 // sprintf(f->name, SEP_MAIL_FILE_TEMPLATE, f->email_count); | |
1256 } else { | |
1257 f->name = (char*) xmalloc(strlen(item->file_as)+strlen(OUTPUT_TEMPLATE)+1); | |
1258 sprintf(f->name, OUTPUT_TEMPLATE, item->file_as); | |
1259 } | |
25 | 1260 |
43 | 1261 f->dname = (char*) xmalloc(strlen(item->file_as)+1); |
1262 strcpy(f->dname, item->file_as); | |
25 | 1263 |
43 | 1264 if (overwrite != 1) { |
1265 int x = 0; | |
1266 char *temp = (char*) xmalloc (strlen(f->name)+10); //enough room for 10 digits | |
25 | 1267 |
43 | 1268 sprintf(temp, "%s", f->name); |
1269 check_filename(temp); | |
1270 while ((f->output = fopen(temp, "r"))) { | |
1271 DEBUG_MAIN(("need to increase filename because one already exists with that name\n")); | |
1272 DEBUG_MAIN(("- increasing it to %s%d\n", f->name, x)); | |
1273 x++; | |
1274 sprintf(temp, "%s%08d", f->name, x); | |
1275 DEBUG_MAIN(("- trying \"%s\"\n", f->name)); | |
1276 if (x == 99999999) { | |
1277 DIE(("create_enter_dir: Why can I not create a folder %s? I have tried %i extensions...\n", f->name, x)); | |
1278 } | |
1279 fclose(f->output); | |
1280 } | |
1281 if (x > 0) { //then the f->name should change | |
1282 free (f->name); | |
1283 f->name = temp; | |
1284 } else { | |
1285 free(temp); | |
1286 } | |
1287 } | |
25 | 1288 |
43 | 1289 DEBUG_MAIN(("f->name = %s\nitem->folder_name = %s\n", f->name, item->file_as)); |
1290 if (mode != MODE_SEPERATE) { | |
1291 check_filename(f->name); | |
1292 if (!(f->output = fopen(f->name, "w"))) { | |
1293 DIE(("create_enter_dir: Could not open file \"%s\" for write\n", f->name)); | |
1294 } | |
1295 } | |
1296 DEBUG_RET(); | |
25 | 1297 } |
1298 | |
39 | 1299 |
1300 void close_enter_dir(struct file_ll *f) | |
1301 { | |
43 | 1302 DEBUG_MAIN(("main: Email Count for folder %s is %i\n", f->dname, f->email_count)); |
1303 if (output_mode != OUTPUT_QUIET) | |
1304 printf("\t\"%s\" - %i items done, skipped %i, should have been %i\n", | |
1305 f->dname, f->email_count, f->skip_count, f->stored_count); | |
1306 if (f->output) fclose(f->output); | |
1307 free(f->name); | |
1308 free(f->dname); | |
39 | 1309 |
43 | 1310 if (mode == MODE_KMAIL) |
1311 close_kmail_dir(); | |
1312 else if (mode == MODE_RECURSE) | |
1313 close_recurse_dir(); | |
1314 else if (mode == MODE_SEPERATE) | |
1315 close_seperate_dir(); | |
39 | 1316 } |
1317 |