1
|
1 /***
|
|
2 * lspst.c
|
|
3 * Part of the LibPST project
|
|
4 * Author: Joe Nahmias <joe@nahmias.net>
|
|
5 * Based on readpst.c by by David Smith <dave.s@earthcorp.com>
|
|
6 *
|
|
7 */
|
|
8
|
|
9 // header file includes {{{1
|
|
10 #include <stdio.h>
|
|
11 #include <stdlib.h>
|
|
12 #include <time.h>
|
|
13 #include <string.h>
|
|
14 #include <ctype.h>
|
|
15 #include <errno.h>
|
|
16
|
|
17 #include "libpst.h"
|
|
18 #include "define.h"
|
|
19 #include "timeconv.h"
|
|
20 // }}}1
|
|
21 // struct file_ll {{{1
|
|
22 struct file_ll {
|
|
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;
|
|
30 struct file_ll *next;
|
|
31 };
|
|
32 // }}}1
|
|
33 // Function Declarations {{{1
|
|
34 void canonicalize_filename(char *fname);
|
|
35 int chr_count(char *str, char x);
|
|
36 void debug_print(char *fmt, ...);
|
|
37 char *rfc2426_escape(char *str);
|
|
38 char *rfc2445_datetime_format(FILETIME *ft);
|
|
39 // }}}1
|
|
40 #define DEBUG_MAIN(x) debug_print x;
|
|
41 // int main(int argc, char** argv) {{{1
|
|
42 int main(int argc, char** argv) {
|
|
43
|
|
44 // declarations {{{2
|
|
45 pst_item *item = NULL;
|
|
46 pst_file pstfile;
|
|
47 pst_desc_ll *d_ptr;
|
|
48 char *temp = NULL; //temporary char pointer
|
|
49 int skip_child = 0;
|
|
50 struct file_ll *f, *head;
|
|
51 // }}}2
|
|
52
|
|
53 if (argc <= 1)
|
|
54 DIE(("Missing PST filename.\n"));
|
|
55
|
|
56 // Open PST file
|
|
57 if ( pst_open(&pstfile, argv[1], "r") )
|
|
58 DIE(("Error opening File\n"));
|
|
59 // Load PST index
|
|
60 if ( pst_load_index(&pstfile) )
|
|
61 DIE(("Index Error\n"));
|
|
62 pst_load_extended_attributes(&pstfile);
|
|
63
|
|
64 d_ptr = pstfile.d_head; // first record is main record
|
|
65 if ((item = _pst_parse_item(&pstfile, d_ptr)) == NULL || item->message_store == NULL) {
|
|
66 DIE(("main: Could not get root record\n"));
|
|
67 }
|
|
68
|
|
69 // default the file_as to the same as the main filename if it doesn't exist
|
|
70 if (item->file_as == NULL) {
|
|
71 if ((temp = strrchr(argv[1], '/')) == NULL)
|
|
72 if ((temp = strrchr(argv[1], '\\')) == NULL)
|
|
73 temp = argv[1];
|
|
74 else
|
|
75 temp++; // get past the "\\"
|
|
76 else
|
|
77 temp++; // get past the "/"
|
|
78 item->file_as = (char*)xmalloc(strlen(temp)+1);
|
|
79 strcpy(item->file_as, temp);
|
|
80 }
|
|
81 fprintf(stderr, "item->file_as = '%s'.\n", item->file_as);
|
|
82
|
|
83 // setup head file_ll
|
|
84 head = (struct file_ll*) malloc(sizeof(struct file_ll));
|
|
85 memset(head, 0, sizeof(struct file_ll));
|
|
86 head->email_count = 0;
|
|
87 head->skip_count = 0;
|
|
88 head->next = NULL;
|
|
89 head->name = "mbox";
|
|
90 head->dname = (char*) malloc(strlen(item->file_as)+1);
|
|
91 strcpy(head->dname, item->file_as);
|
|
92 head->type = item->type;
|
|
93 DEBUG_MAIN(("head @ %p: name = '%s', dname = '%s', next = %p.\n", head, head->name, head->dname, head->next));
|
|
94
|
|
95 if ((d_ptr = pst_getTopOfFolders(&pstfile, item)) == NULL) {
|
|
96 DIE(("Top of folders record not found. Cannot continue\n"));
|
|
97 }
|
|
98 DEBUG_MAIN(("d_ptr(TOF) = %p.\n", d_ptr));
|
|
99
|
|
100 if (item){
|
|
101 _pst_freeItem(item);
|
|
102 item = NULL;
|
|
103 }
|
|
104
|
|
105 d_ptr = d_ptr->child; // do the children of TOPF
|
|
106 DEBUG_MAIN(("d_ptr(TOF->child) = %p.\n", d_ptr));
|
|
107
|
|
108 DEBUG_MAIN(("main: About to do email stuff\n"));
|
|
109 while (d_ptr != NULL) {
|
|
110 // Process d_ptr {{{2
|
|
111 DEBUG_MAIN(("main: New item record, d_ptr = %p.\n", d_ptr));
|
|
112 if (d_ptr->desc == NULL) {
|
|
113 DEBUG_WARN(("main: ERROR ?? item's desc record is NULL\n"));
|
|
114 f->skip_count++;
|
|
115 goto check_parent;
|
|
116 }
|
|
117 DEBUG_MAIN(("main: Desc Email ID %x [d_ptr->id = %x]\n", d_ptr->desc->id, d_ptr->id));
|
|
118
|
|
119 item = _pst_parse_item(&pstfile, d_ptr);
|
|
120 DEBUG_MAIN(("main: About to process item @ %p.\n", item));
|
|
121 if (item != NULL) {
|
|
122
|
|
123 // there should only be one message_store, and we have already
|
|
124 // done it
|
|
125 if (item->message_store != NULL) {
|
|
126 DIE(("ERROR(main): A second message_store has been found.\n"));
|
|
127 }
|
|
128
|
|
129 if (item->folder != NULL) {
|
|
130 // Process Folder item {{{3
|
|
131 // if this is a folder, we want to recurse into it
|
|
132 printf("Folder");
|
|
133 if (item->file_as != NULL)
|
|
134 printf("\t%s/", item->file_as);
|
|
135 printf("\n");
|
|
136
|
|
137 DEBUG_MAIN(("main: I think I may try to go into folder \"%s\"\n", item->file_as));
|
|
138 f = (struct file_ll*) malloc(sizeof(struct file_ll));
|
|
139 memset(f, 0, sizeof(struct file_ll));
|
|
140 f->next = head;
|
|
141 f->email_count = 0;
|
|
142 f->type = item->type;
|
|
143 f->stored_count = item->folder->email_count;
|
|
144 head = f;
|
|
145 f->name = "mbox";
|
|
146 f->dname = (char*) xmalloc(strlen(item->file_as)+1);
|
|
147 strcpy(f->dname, item->file_as);
|
|
148
|
|
149 DEBUG_MAIN(("main: f->name = %s\nitem->folder_name = %s\n", f->name, item->file_as));
|
|
150 canonicalize_filename(f->name);
|
|
151
|
|
152 if (d_ptr->child != NULL) {
|
|
153 d_ptr = d_ptr->child;
|
|
154 skip_child = 1;
|
|
155 } else {
|
|
156 DEBUG_MAIN(("main: Folder has NO children. Creating directory, and closing again\n"));
|
|
157 // printf("\tNo items to process in folder \"%s\", should have been %i\n", f->dname, f->stored_count);
|
|
158 head = f->next;
|
|
159 if (f->output != NULL)
|
|
160 fclose(f->output);
|
|
161 free(f->dname);
|
|
162 free(f->name);
|
|
163 free(f);
|
|
164
|
|
165 f = head;
|
|
166 }
|
|
167 _pst_freeItem(item);
|
|
168 item = NULL; // just for the odd situations!
|
|
169 goto check_parent;
|
|
170 // }}}3
|
|
171 } else if (item->contact != NULL) {
|
|
172 // Process Contact item {{{3
|
|
173 if (f->type != PST_TYPE_CONTACT) {
|
|
174 DEBUG_MAIN(("main: I have a contact, but the folder isn't a contacts folder. "
|
|
175 "Will process anyway\n"));
|
|
176 }
|
|
177 if (item->type != PST_TYPE_CONTACT) {
|
|
178 DEBUG_MAIN(("main: I have an item that has contact info, but doesn't say that"
|
|
179 " it is a contact. Type is \"%s\"\n", item->ascii_type));
|
|
180 DEBUG_MAIN(("main: Processing anyway\n"));
|
|
181 }
|
|
182
|
|
183 printf("Contact");
|
|
184 if (item->contact->fullname != NULL)
|
|
185 printf("\t%s", rfc2426_escape(item->contact->fullname));
|
|
186 printf("\n");
|
|
187 // }}}3
|
|
188 } else if (item->email != NULL &&
|
|
189 (item->type == PST_TYPE_NOTE || item->type == PST_TYPE_REPORT)) {
|
|
190 // Process Email item {{{3
|
|
191 printf("Email");
|
|
192 if (item->email->outlook_sender_name != NULL)
|
|
193 printf("\tFrom: %s", item->email->outlook_sender_name);
|
|
194 if (item->email->subject->subj != NULL)
|
|
195 printf("\tSubject: %s", item->email->subject->subj);
|
|
196 printf("\n");
|
|
197 // }}}3
|
|
198 } else if (item->type == PST_TYPE_JOURNAL) {
|
|
199 // Process Journal item {{{3
|
|
200 if (f->type != PST_TYPE_JOURNAL) {
|
|
201 DEBUG_MAIN(("main: I have a journal entry, but folder isn't specified as a journal type. Processing...\n"));
|
|
202 }
|
|
203
|
|
204 printf("Journal\t%s\n", rfc2426_escape(item->email->subject->subj));
|
|
205 // }}}3
|
|
206 } else if (item->type == PST_TYPE_APPOINTMENT) {
|
|
207 // Process Calendar Appointment item {{{3
|
|
208 // deal with Calendar appointments
|
|
209
|
|
210 DEBUG_MAIN(("main: Processing Appointment Entry\n"));
|
|
211 if (f->type != PST_TYPE_APPOINTMENT) {
|
|
212 DEBUG_MAIN(("main: I have an appointment, but folder isn't specified as an appointment type. Processing...\n"));
|
|
213 }
|
|
214
|
|
215 printf("Appointment");
|
|
216 if (item->email != NULL && item->email->subject != NULL)
|
|
217 printf("\tSUMMARY: %s", rfc2426_escape(item->email->subject->subj));
|
|
218 if (item->appointment != NULL && item->appointment->start != NULL)
|
|
219 printf("\tSTART: %s", rfc2445_datetime_format(item->appointment->start));
|
|
220 printf("\n");
|
|
221
|
|
222 // }}}3
|
|
223 } else {
|
|
224 f->skip_count++;
|
|
225 DEBUG_MAIN(("main: Unknown item type. %i. Ascii1=\"%s\"\n", \
|
|
226 item->type, item->ascii_type));
|
|
227 }
|
|
228 } else {
|
|
229 f->skip_count++;
|
|
230 DEBUG_MAIN(("main: A NULL item was seen\n"));
|
|
231 }
|
|
232
|
|
233 check_parent:
|
|
234 // _pst_freeItem(item);
|
|
235 while (!skip_child && d_ptr->next == NULL && d_ptr->parent != NULL) {
|
|
236 DEBUG_MAIN(("main: Going to Parent\n"));
|
|
237 head = f->next;
|
|
238 if (f->output != NULL)
|
|
239 fclose(f->output);
|
|
240 DEBUG_MAIN(("main: Email Count for folder %s is %i\n", f->dname, f->email_count));
|
|
241 /*
|
|
242 printf("\t\"%s\" - %i items done, skipped %i, should have been %i\n", \
|
|
243 f->dname, f->email_count, f->skip_count, f->stored_count);
|
|
244 */
|
|
245
|
|
246 free(f->name);
|
|
247 free(f->dname);
|
|
248 free(f);
|
|
249 f = head;
|
|
250 if (head == NULL) { //we can't go higher. Must be at start?
|
|
251 DEBUG_MAIN(("main: We are now trying to go above the highest level. We must be finished\n"));
|
|
252 break; //from main while loop
|
|
253 }
|
|
254 d_ptr = d_ptr->parent;
|
|
255 skip_child = 0;
|
|
256 }
|
|
257
|
|
258 if (item != NULL) {
|
|
259 DEBUG_MAIN(("main: Freeing memory used by item\n"));
|
|
260 _pst_freeItem(item);
|
|
261 item = NULL;
|
|
262 }
|
|
263
|
|
264 if (!skip_child)
|
|
265 d_ptr = d_ptr->next;
|
|
266 else
|
|
267 skip_child = 0;
|
|
268
|
|
269 if (d_ptr == NULL) { DEBUG_MAIN(("main: d_ptr is now NULL\n")); }
|
|
270
|
|
271 // }}}2
|
|
272 } // end while(d_ptr != NULL)
|
|
273 DEBUG_MAIN(("main: Finished.\n"));
|
|
274
|
|
275 // Cleanup {{{2
|
|
276 pst_close(&pstfile);
|
|
277 while (f != NULL) {
|
|
278 if (f->output != NULL)
|
|
279 fclose(f->output);
|
|
280 free(f->name);
|
|
281 free(f->dname);
|
|
282
|
|
283 head = f->next;
|
|
284 free(f);
|
|
285 f = head;
|
|
286 }
|
|
287 DEBUG_RET();
|
|
288 // }}}2
|
|
289
|
|
290 return 0;
|
|
291 }
|
|
292 // }}}1
|
|
293 // void canonicalize_filename(char *fname) {{{1
|
|
294 // This function will make sure that a filename is in cannonical form. That
|
|
295 // is, it will replace any slashes, backslashes, or colons with underscores.
|
|
296 void canonicalize_filename(char *fname) {
|
|
297 DEBUG_ENT("canonicalize_filename");
|
|
298 if (fname == NULL) {
|
|
299 DEBUG_RET();
|
|
300 return;
|
|
301 }
|
|
302 while ((fname = strpbrk(fname, "/\\:")) != NULL)
|
|
303 *fname = '_';
|
|
304 DEBUG_RET();
|
|
305 }
|
|
306 // }}}1
|
|
307 // int chr_count(char *str, char x) {{{1
|
|
308 int chr_count(char *str, char x) {
|
|
309 int r = 0;
|
|
310 if (str == NULL) return 0;
|
|
311 while (*str != '\0') {
|
|
312 if (*str == x)
|
|
313 r++;
|
|
314 str++;
|
|
315 }
|
|
316 return r;
|
|
317 }
|
|
318 // }}}1
|
|
319 // void debug_print(char *fmt, ...) {{{1
|
|
320 void debug_print(char *fmt, ...) {
|
|
321 // shamlessly stolen from minprintf() in K&R pg. 156
|
|
322 va_list ap;
|
|
323 char *p, *sval;
|
|
324 void *pval;
|
|
325 int ival;
|
|
326 double dval;
|
|
327 FILE *fp = stderr;
|
|
328
|
|
329 va_start(ap, fmt);
|
|
330 for(p = fmt; *p; p++) {
|
|
331 if (*p != '%') {
|
|
332 fputc(*p, fp);
|
|
333 continue;
|
|
334 }
|
|
335 switch (tolower(*++p)) {
|
|
336 case 'd': case 'i':
|
|
337 ival = va_arg(ap, int);
|
|
338 fprintf(fp, "%d", ival);
|
|
339 break;
|
|
340 case 'f':
|
|
341 dval = va_arg(ap, double);
|
|
342 fprintf(fp, "%f", dval);
|
|
343 break;
|
|
344 case 's':
|
|
345 for (sval = va_arg(ap, char *); *sval; ++sval)
|
|
346 fputc(*sval, fp);
|
|
347 break;
|
|
348 case 'p':
|
|
349 pval = va_arg(ap, void *);
|
|
350 fprintf(fp, "%p", pval);
|
|
351 break;
|
|
352 case 'x':
|
|
353 ival = va_arg(ap, int);
|
|
354 fprintf(fp, "%#010x", ival);
|
|
355 break;
|
|
356 default:
|
|
357 fputc(*p, fp);
|
|
358 break;
|
|
359 }
|
|
360 }
|
|
361 va_end(ap);
|
|
362 }
|
|
363 // }}}1
|
|
364 // char *rfc2426_escape(char *str) {{{1
|
|
365 char *rfc2426_escape(char *str) {
|
|
366 static char *buf = NULL;
|
|
367 char *a, *b;
|
|
368 int y, z;
|
|
369
|
|
370 DEBUG_ENT("rfc2426_escape");
|
|
371 if (str == NULL) {
|
|
372 DEBUG_RET();
|
|
373 return NULL;
|
|
374 }
|
|
375
|
|
376 // calculate space required to escape all the commas, semi-colons, backslashes, and newlines
|
|
377 y = chr_count(str, ',') + chr_count(str, '\\') + chr_count(str, ';') + chr_count(str, '\n');
|
|
378 // count how many carriage-returns we have to skip
|
|
379 z = chr_count(str, '\r');
|
|
380
|
|
381 if (y == 0 && z == 0) {
|
|
382 // there isn't any work required
|
|
383 DEBUG_RET();
|
|
384 return str;
|
|
385 }
|
|
386
|
|
387 buf = (char *) realloc( buf, strlen(str) + y - z + 1 );
|
|
388 for (a = str, b = buf; *a != '\0'; ++a, ++b)
|
|
389 switch (*a) {
|
|
390 case ',' : case '\\': case ';' : case '\n':
|
|
391 // insert backslash to escape
|
|
392 *(b++) = '\\';
|
|
393 *b = *a;
|
|
394 break;
|
|
395 case '\r':
|
|
396 // skip
|
|
397 break;
|
|
398 default:
|
|
399 *b = *a;
|
|
400 }
|
|
401 *b = '\0'; // NUL-terminate the string
|
|
402
|
|
403 DEBUG_RET();
|
|
404 return buf;
|
|
405 }
|
|
406 // }}}1
|
|
407 // char *rfc2445_datetime_format(FILETIME *ft) {{{1
|
|
408 char *rfc2445_datetime_format(FILETIME *ft) {
|
|
409 static char* buffer = NULL;
|
|
410 struct tm *stm = NULL;
|
|
411 DEBUG_ENT("rfc2445_datetime_format");
|
|
412 if (buffer == NULL)
|
|
413 buffer = malloc(30); // should be enough
|
|
414 stm = fileTimeToStructTM(ft);
|
|
415 if (strftime(buffer, 30, "%Y%m%dT%H%M%SZ", stm)==0) {
|
|
416 DEBUG_INFO(("Problem occured formatting date\n"));
|
|
417 }
|
|
418 DEBUG_RET();
|
|
419 return buffer;
|
|
420 }
|
|
421 // }}}1
|
|
422
|
|
423 // vim:sw=4 ts=4:
|
|
424 // vim600: set foldlevel=0 foldmethod=marker:
|