Mercurial > libpst
view src/pst2dii.cpp.in @ 386:f1f9920cc7b1
Add AM_GNU_GETTEXT macros
AM_ICONV relies on config.rpath and autopoint/gettextize will
only copy these in when the AM_GNU_GETTEXT macros are present.
This is needed for the next commit that deletes config.rpath since
it is cruft that should be copied in by autotools not embedded.
Run autopoint to copy in config.rpath and
leave it to automake to copy into the tarball.
author | Paul Wise <pabs3@bonedaddy.net> |
---|---|
date | Sat, 21 Dec 2019 21:25:44 +0800 |
parents | cb67b335afcc |
children |
line wrap: on
line source
/* Copyright (c) 2008 Carl Byington - 510 Software Group, released under the GPL version 2 or any later version at your choice available at http://www.fsf.org/licenses/gpl.txt Based on readpst.c by David Smith */ #include <iostream> #include <string> #include <vector> using namespace std; extern "C" { #include "define.h" #include "lzfu.h" } struct file_ll { string name; int32_t stored_count; int32_t email_count; int32_t skip_count; int32_t type; file_ll() { stored_count = 0; email_count = 0; skip_count = 0; type = 0; }; }; // global settings const char* convert = "@CONVERT@"; // fully qualified path of the convert program from image magick const char* prog_name = NULL; // our arg0 name const char* bates_prefix = ""; // string to prefix bates numbers int bates_index = 0; // current bates sequence const char* output_directory = "."; const char* output_file = "load.dii"; char* font_file = NULL; int bates_color = 0xff0000; // color of bates header stamp int email_sequence = 0; // current pdf sequence number char* pdf_name = NULL; // current pdf file name FILE* dii_file = NULL; // the output dii load file pst_file pstfile; // the input pst file // pdf writer globals bool pdf_open = false; // is pdf writer started char* pst_folder; // current folder name int page_sequence; // current page number string conversion; // conversion command vector<string> png_names; // png writer globals bool png_open = false; // is current page open int line_height; // in pixels int char_width; // in pixels int col_number, col_max; // in characters int line_number, line_max; // lines per page int x_position, y_position; // in pixels int black, red; // text colors gdImagePtr image; // current gd image const int DPI = 300; const double sz = 10.0; const int margin = DPI/2; const int LINE_SIZE = 2000; const int PAGE_WIDTH = DPI*17/2; const int PAGE_HEIGHT = DPI*11; // max size of the c_time char*. It will store the date of the email #define C_TIME_SIZE 500 static void open_png(); static void close_png(); static void version(); static void version() { printf("pst2dii v%s\n", VERSION); #if BYTE_ORDER == BIG_ENDIAN printf("Big Endian implementation being used.\n"); #elif BYTE_ORDER == LITTLE_ENDIAN printf("Little Endian implementation being used.\n"); #else # error "Byte order not supported by this library" #endif } static void usage(); static void usage() { version(); printf("Usage: %s -f ttf-font-file [OPTIONS] {PST FILENAME}\n", prog_name); printf("\t-f ttf-font-file \t- Set the font file\n"); printf("OPTIONS:\n"); printf("\t-B bates-prefix \t- Set the bates prefix string\n"); printf("\t-O dii-output-file\t- Set the dii load file output filename\n"); printf("\t-V \t- Version. Display program version\n"); printf("\t-b bates-number \t- Set the starting bates sequence number\n"); printf("\t-c bates-color \t- Specify the color of the bates stamps as 6 digit hex\n"); printf("\t-d filename \t- Debug to file.\n"); printf("\t-h \t- Help. This screen\n"); printf("\t-o dirname \t- Output directory to write files to.\n"); } static char *removeCR (char *c); static char *removeCR (char *c) { // converts /r/n to /n char *a, *b; DEBUG_ENT("removeCR"); a = b = c; while (*a != '\0') { *b = *a; if (*a != '\r') b++; a++; } *b = '\0'; DEBUG_RET(); return c; } // The sole purpose of this function is to bypass the pseudo-header prologue // that Microsoft Outlook inserts at the beginning of the internet email // headers for emails stored in their "Personal Folders" files. static char *skip_header_prologue(char *headers); static char *skip_header_prologue(char *headers) { const char *bad = "Microsoft Mail Internet Headers"; if (strncmp(headers, bad, strlen(bad)) == 0) { // Found the offensive header prologue char *pc = strchr(headers, '\n'); return pc + 1; } return headers; } static void check_filename(string &fname); static void check_filename(string &fname) { char *t = strdup(fname.c_str()); DEBUG_ENT("check_filename"); if (!t) { DEBUG_RET(); return; } char *tt = t; bool fixed = false; while ((t = strpbrk(t, " /\\:"))) { // while there are characters in the second string that we don't want *t = '_'; //replace them with an underscore fixed = true; } if (fixed) fname = string(tt); free(tt); DEBUG_RET(); } static string write_separate_attachment(string fname, pst_item_attach* current_attach, int attach_num, pst_file* pst); static string write_separate_attachment(string fname, pst_item_attach* current_attach, int attach_num, pst_file* pst) { FILE *fp = NULL; int x = 0; char *temp = NULL; // If there is a long filename (filename2) use that, otherwise // use the 8.3 filename (filename1) char *attach_filename = (current_attach->filename2.str) ? current_attach->filename2.str : current_attach->filename1.str; DEBUG_ENT("write_separate_attachment"); check_filename(fname); const char* f_name = fname.c_str(); DEBUG_INFO(("dirname=%s, pathname=%s, filename=%s\n", output_directory, f_name, attach_filename)); int len = strlen(output_directory) + 1 + strlen(f_name) + 15; if (!attach_filename) { // generate our own (dummy) filename for the attachment temp = (char*)pst_malloc(len); sprintf(temp, "%s/%s_attach%i", output_directory, f_name, attach_num); } else { // have an attachment name, make sure it's unique temp = (char*)pst_malloc(len+strlen(attach_filename)); do { if (fp) fclose(fp); if (x == 0) sprintf(temp, "%s/%s_%s", output_directory, f_name, attach_filename); else sprintf(temp, "%s/%s_%s-%i", output_directory, f_name, attach_filename, x); } while ((fp = fopen(temp, "r")) && ++x < 99999999); if (x > 99999999) { DIE(("error finding attachment name. exhausted possibilities to %s\n", temp)); } } DEBUG_INFO(("Saving attachment to %s\n", temp)); if (!(fp = fopen(temp, "wb"))) { DEBUG_WARN(("write_separate_attachment: Cannot open attachment save file \"%s\"\n", temp)); } else { (void)pst_attach_to_file(pst, current_attach, fp); fclose(fp); } string rc(temp); if (temp) free(temp); DEBUG_RET(); return rc; } static void print_pdf_short(const char *line, int len, int color); static void print_pdf_short(const char *line, int len, int color) { if (line_number >= line_max) { close_png(); open_png(); } int brect[8]; gdFTStringExtra strex; strex.flags = gdFTEX_RESOLUTION; strex.linespacing = 1.20; strex.charmap = 0; strex.hdpi = DPI; strex.vdpi = DPI; char xline[len+1]; memcpy(xline, line, len); xline[len] = '\0'; char *p; char *l = xline; while ((p = strchr(l, '&'))) { *p = '\0'; char *err = gdImageStringFTEx(image, &brect[0], color, font_file, sz, 0.0, x_position, y_position, l, &strex); if (err) printf("%s", err); x_position += (brect[2]-brect[6]); l = p+1; err = gdImageStringFTEx(image, &brect[0], color, font_file, sz, 0.0, x_position, y_position, (char*)"&", &strex); if (err) printf("%s", err); x_position += (brect[2]-brect[6]); } char *err = gdImageStringFTEx(image, &brect[0], color, font_file, sz, 0.0, x_position, y_position, l, &strex); if (err) printf("%s", err); x_position += (brect[2]-brect[6]); col_number += len; } static void new_line(); static void new_line() { y_position += line_height; line_number += 1; x_position = margin; col_number = 0; } static void print_pdf_single(const char *line, int color); static void print_pdf_single(const char *line, int color) { while (*line == '\t') { char blanks[5]; memset(blanks, ' ', 5); print_pdf_short(blanks, 4, color); line++; if (col_number >= col_max) new_line(); } int n = strlen(line); while (n) { int m = col_max - col_number; // number of chars that will fit on this line m = (n > m) ? m : n; print_pdf_short(line, m, color); line += m; n -= m; if (n) new_line(); } } static void print_pdf_only(char *line, int color); static void print_pdf_only(char *line, int color) { char *p; while ((p = strchr(line, '\n'))) { *p = '\0'; print_pdf_single(line, color); *p = '\n'; line = p+1; new_line(); } print_pdf_single(line, color); } static void print_pdf(char *line); static void print_pdf(char *line) { pst_fwrite(line, 1, strlen(line), dii_file); print_pdf_only(line, black); } static void open_png() { if (!png_open) { png_open = true; int brect[8]; image = gdImageCreate(PAGE_WIDTH, PAGE_HEIGHT); gdImageColorAllocate(image, 255, 255, 255); // background color first one allocated black = gdImageColorAllocate(image, 0, 0, 0); int r = (bates_color & 0xff0000) >> 16; int g = (bates_color & 0x00ff00) >> 8; int b = (bates_color & 0x0000ff); red = gdImageColorAllocate(image, r, g, b); gdFTStringExtra strex; strex.flags = gdFTEX_RESOLUTION; strex.linespacing = 1.20; strex.charmap = 0; strex.hdpi = DPI; strex.vdpi = DPI; char line[LINE_SIZE]; char *err = gdImageStringFTEx(NULL, &brect[0], black, font_file, sz, 0.0, margin, margin, (char*)"LMgqQ", &strex); if (err) printf("%s", err); line_height = (brect[3]-brect[7]) * 12/10; char_width = (brect[2]-brect[6]) / 5; col_number = 0; col_max = (PAGE_WIDTH - margin*2) / char_width; line_number = 0; line_max = (PAGE_HEIGHT - margin*2) / line_height; x_position = margin; y_position = margin + line_height; snprintf(line, sizeof(line), "%s%06d\n", bates_prefix, bates_index++); print_pdf_only(line, red); print_pdf_only(pst_folder, red); } } static void close_png() { if (png_open) { png_open = false; int len = 4 + 11 + 4 +1; char *fn = (char*)pst_malloc(len); snprintf(fn, len, "page%d.png", ++page_sequence); FILE *pngout = fopen(fn, "wb"); if (pngout) { gdImagePng(image, pngout); fclose(pngout); } gdImageDestroy(image); // free memory png_names.push_back(fn); conversion += string(" ") + fn; free(fn); } } static void open_pdf(char *line); static void open_pdf(char *line) { pst_folder = line; page_sequence = 0; conversion = string(convert); png_names.clear(); open_png(); /* Note; allocating the largest string length to pdf_name */ int len = strlen(output_directory) + 4 + 6 + 4 + 1; pdf_name = (char*)pst_malloc(len); snprintf(pdf_name, 3 + 6 + 1, "dii%06d", ++email_sequence); fprintf(dii_file, "\n@T %s\n", pdf_name); snprintf(pdf_name, len, "%s/dii%06d.pdf", output_directory, email_sequence); } static void close_pdf(); static void close_pdf() { close_png(); conversion += string(" ") + pdf_name; (void)system(conversion.c_str()); for (vector<string>::iterator i=png_names.begin(); i!=png_names.end(); i++) { remove((*i).c_str()); } fprintf(dii_file, "@D %s\n", pdf_name); free(pdf_name); } static void write_simple(const char *tag, const char *value); static void write_simple(const char *tag, const char *value) { if (value) fprintf(dii_file, "@%s %s\n", tag, value); } static void write_simple(const char *tag, string value); static void write_simple(const char *tag, string value) { fprintf(dii_file, "@%s %s\n", tag, value.c_str()); } static void write_simple(const char *tag, const char *value, const char *value2); static void write_simple(const char *tag, const char *value, const char *value2) { if (value) { if (value2) fprintf(dii_file, "@%s \"%s\" <%s>\n", tag, value, value2); else fprintf(dii_file, "@%s \"%s\"\n", tag, value); } } static string extract_header(char *headers, const char *field); static string extract_header(char *headers, const char *field) { string rc; int len = strlen(field) + 4; char f[len]; snprintf(f, len, "\n%s: ", field); char *p = strstr(headers, f); if (p) { p += strlen(f); char *n = strchr(p, '\n'); if (n) { *n = '\0'; rc = string(p); *n = '\n'; } else { rc = string(p); } } return rc; } static void write_normal_email(file_ll &f, pst_item* item, pst_file* pst); static void write_normal_email(file_ll &f, pst_item* item, pst_file* pst) { DEBUG_ENT("write_normal_email"); char *soh = NULL; // real start of headers. if (item->email->header.str) { // some of the headers we get from the file are not properly defined. // they can contain some email stuff too. We will cut off the header // when we see a \n\n or \r\n\r\n removeCR(item->email->header.str); char *temp = strstr(item->email->header.str, "\n\n"); if (temp) { DEBUG_INFO(("Found body text in header\n")); temp[1] = '\0'; // stop after first \n } soh = skip_header_prologue(item->email->header.str); } char folder_line[LINE_SIZE]; char line[LINE_SIZE]; // reset pdf writer to new file int bates = bates_index; // save starting index snprintf(folder_line, sizeof(folder_line), "pst folder = %s\n", f.name.c_str()); open_pdf(folder_line); // start printing this email fprintf(dii_file, "@FOLDERNAME %s\n", f.name.c_str()); string myfrom = extract_header(soh, "From"); string myto = extract_header(soh, "To"); string mycc = extract_header(soh, "Cc"); string mybcc = extract_header(soh, "Bcc"); if (myfrom.empty()) write_simple("FROM", item->email->outlook_sender_name.str, item->email->sender_address.str); else write_simple("FROM", myfrom); if (myto.empty()) write_simple("TO", item->email->sentto_address.str, item->email->recip_address.str); else write_simple("TO", myto); if (mycc.empty()) write_simple("CC", item->email->cc_address.str); else write_simple("CC", mycc); if (mybcc.empty()) write_simple("BCC", item->email->bcc_address.str); else write_simple("BCC", mybcc); if (item->email->sent_date) { time_t t = pst_fileTimeToUnixTime(item->email->sent_date); char c_time[C_TIME_SIZE]; strftime(c_time, C_TIME_SIZE, "%F", gmtime(&t)); write_simple("DATESENT", c_time); strftime(c_time, C_TIME_SIZE, "%T+0000", gmtime(&t)); write_simple("TIMESENT", c_time); } if (item->email->arrival_date) { time_t t = pst_fileTimeToUnixTime(item->email->arrival_date); char c_time[C_TIME_SIZE]; strftime(c_time, C_TIME_SIZE, "%F", gmtime(&t)); write_simple("DATERCVD", c_time); strftime(c_time, C_TIME_SIZE, "%T+0000", gmtime(&t)); write_simple("TIMERCVD", c_time); } if (item->subject.str) { write_simple("SUBJECT", item->subject.str); } write_simple("MSGID", item->email->messageid.str); write_simple("READ", (item->flags & 1) ? "Y" : "N"); DEBUG_INFO(("About to print Header\n")); fprintf(dii_file, "@HEADER\n"); if (item && item->subject.str) { DEBUG_INFO(("item->subject = %s\n", item->subject.str)); } if (soh) { // Now, write out the header... print_pdf(soh); int len = strlen(soh); if (!len || (soh[len-1] != '\n')) { snprintf(line, sizeof(line), "\n"); print_pdf(line); } } else { //make up our own headers const char *temp = item->email->outlook_sender.str; if (!temp) temp = ""; snprintf(line, sizeof(line), "From: \"%s\" <%s>\n", item->email->outlook_sender_name.str, temp); print_pdf(line); if (item->subject.str) { snprintf(line, sizeof(line), "Subject: %s\n", item->subject.str); } else { snprintf(line, sizeof(line), "Subject: \n"); } print_pdf(line); snprintf(line, sizeof(line), "To: %s\n", item->email->sentto_address.str); print_pdf(line); if (item->email->cc_address.str) { snprintf(line, sizeof(line), "Cc: %s\n", item->email->cc_address.str); print_pdf(line); } if (item->email->sent_date) { time_t em_time = pst_fileTimeToUnixTime(item->email->sent_date); char c_time[C_TIME_SIZE]; strftime(c_time, C_TIME_SIZE, "%a, %d %b %Y %H:%M:%S %z", gmtime(&em_time)); snprintf(line, sizeof(line), "Date: %s\n", c_time); print_pdf(line); } } snprintf(line, sizeof(line), "\n"); print_pdf_only(line, black); fprintf(dii_file, "@HEADER-END\n"); DEBUG_INFO(("About to print Body\n")); fprintf(dii_file, "@EMAIL-BODY\n"); if (item->body.str) { removeCR(item->body.str); print_pdf(item->body.str); } else if (item->email->htmlbody.str) { removeCR(item->email->htmlbody.str); print_pdf(item->email->htmlbody.str); } else if (item->email->encrypted_body.data || item->email->encrypted_htmlbody.data) { char ln[LINE_SIZE]; snprintf(ln, sizeof(ln), "%s", "The body of this email is encrypted. This isn't supported yet, but the body is now an attachment\n"); print_pdf(ln); } fprintf(dii_file, "@EMAIL-END\n"); int attach_num = 0; for (pst_item_attach* attach = item->attach; attach; attach = attach->next) { pst_convert_utf8_null(item, &attach->filename1); pst_convert_utf8_null(item, &attach->filename2); pst_convert_utf8_null(item, &attach->mimetype); DEBUG_INFO(("Attempting Attachment encoding\n")); if (attach->data.data || attach->i_id) { string an = write_separate_attachment(f.name, attach, ++attach_num, pst); fprintf(dii_file, "@EATTACH %s\n", an.c_str()); } } close_pdf(); fprintf(dii_file, "@BATESBEG %d\n", bates); fprintf(dii_file, "@BATESEND %d\n", bates_index-1); DEBUG_RET(); } static void create_enter_dir(file_ll &f, file_ll *parent, pst_item *item); static void create_enter_dir(file_ll &f, file_ll *parent, pst_item *item) { pst_convert_utf8(item, &item->file_as); f.type = item->type; f.stored_count = (item->folder) ? item->folder->item_count : 0; f.name = ((parent) ? parent->name + "/" : "") + string(item->file_as.str); } static void close_enter_dir(file_ll &f); static void close_enter_dir(file_ll &f) { } static void process(pst_item *outeritem, file_ll *parent, pst_desc_tree *d_ptr); static void process(pst_item *outeritem, file_ll *parent, pst_desc_tree *d_ptr) { file_ll ff; pst_item *item = NULL; DEBUG_ENT("process"); create_enter_dir(ff, parent, outeritem); for (; d_ptr; d_ptr = d_ptr->next) { if (d_ptr->desc) { item = pst_parse_item(&pstfile, d_ptr, NULL); DEBUG_INFO(("item pointer is %p\n", item)); if (item) { if (item->folder && item->file_as.str && d_ptr->child ) { //if this is a non-empty folder, we want to recurse into it fprintf(stderr, "entering folder %s\n", item->file_as.str); process(item, &ff, d_ptr->child); } else if (item->email && (item->type == PST_TYPE_NOTE || item->type == PST_TYPE_SCHEDULE || item->type == PST_TYPE_REPORT)) { ff.email_count++; write_normal_email(ff, item, &pstfile); } else { ff.skip_count++; // other mapi objects } pst_freeItem(item); } else { ff.skip_count++; DEBUG_INFO(("A NULL item was seen\n")); } } } close_enter_dir(ff); DEBUG_RET(); } int main(int argc, char* const* argv) { pst_desc_tree *d_ptr; char *fname = NULL; char c; char *d_log = NULL; prog_name = argv[0]; pst_item *item = NULL; while ((c = getopt(argc, argv, "B:b:c:d:f:o:O:Vh"))!= -1) { switch (c) { case 'B': bates_prefix = optarg; break; case 'b': bates_index = atoi(optarg); break; case 'c': bates_color = (int)strtol(optarg, (char**)NULL, 16); break; case 'f': font_file = optarg; break; case 'o': output_directory = optarg; break; case 'O': output_file = optarg; break; case 'd': d_log = optarg; break; case 'h': usage(); exit(0); break; case 'V': version(); exit(0); break; default: usage(); exit(1); break; } } if (!font_file) { usage(); exit(1); } if (argc > optind) { fname = argv[optind]; } else { usage(); exit(2); } #ifdef DEBUG_ALL // force a log file if (!d_log) d_log = "pst2dii.log"; #endif DEBUG_INIT(d_log, NULL); DEBUG_ENT("main"); RET_DERROR(pst_open(&pstfile, fname, NULL), 1, ("Error opening File\n")); RET_DERROR(pst_load_index(&pstfile), 2, ("Index Error\n")); pst_load_extended_attributes(&pstfile); d_ptr = pstfile.d_head; // first record is main record item = (pst_item*)pst_parse_item(&pstfile, d_ptr, NULL); if (!item || !item->message_store) { DEBUG_RET(); DIE(("Could not get root record\n")); } d_ptr = pst_getTopOfFolders(&pstfile, item); if (!d_ptr) { DEBUG_RET(); DIE(("Top of folders record not found. Cannot continue\n")); } dii_file = fopen(output_file, "wb"); if (dii_file) { process(item, NULL, d_ptr->child); // do the children of TOPF pst_freeItem(item); pst_close(&pstfile); fclose(dii_file); } DEBUG_RET(); return 0; }