xapian-applications/omega/xlsxparser.cc

   1 /** @file
   2  * @brief Extract fields from XLSX sheet*.xml.
   3  */
   4 /* Copyright (C) 2012,2013,2021 Olly Betts
   5  *
   6  * This program is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (at your option) any later version.
  10  *
  11  * This program is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14  * GNU General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License
  17  * along with this program; if not, write to the Free Software
  18  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301 USA
  19  */
  20
  21 #include <config.h>
  22
  23 #include "xlsxparser.h"
  24
  25 #include <cstdlib>
  26 #include <cstring>
  27 #include <ctime>
  28
  29 using namespace std;
  30
  31 bool
  32 XlsxParser::opening_tag(const string &tag)
  33 {
  34     if (tag == "c") {
  35         // We need to distinguish <v> tags which are inside <c t="s">, as these
  36         // are numeric references to shared strings.
  37         string type;
  38         if (get_attribute("t", type) && type == "s") {
  39             mode = MODE_C_STRING;
  40         } else {
  41             mode = MODE_C_LITERAL;
  42             if (get_attribute("s", type)) {
  43                 unsigned long style_id = strtoul(type.c_str(), NULL, 10);
  44                 if (date_style.find(style_id) != date_style.end()) {
  45                     mode = MODE_C_DATE;
  46                 }
  47             }
  48         }
  49     } else if (tag == "v") {
  50         if (mode == MODE_C_LITERAL) {
  51             mode = MODE_V_LITERAL;
  52         } else if (mode == MODE_C_STRING) {
  53             mode = MODE_V_STRING;
  54         } else if (mode == MODE_C_DATE) {
  55             mode = MODE_V_DATE;
  56         }
  57     } else if (tag == "si") {
  58         mode = MODE_SI;
  59     } else if (tag == "sst") {
  60         string unique_count;
  61         if (get_attribute("uniqueCount", unique_count)) {
  62             unsigned long c = strtoul(unique_count.c_str(), NULL, 10);
  63             // This reserving is just a performance tweak, so don't go
  64             // reserving ludicrous amounts of space just because an XML
  65             // attribute told us to.
  66             sst.reserve(std::min(c, 1000000ul));
  67         }
  68     } else if (tag == "workbookPr") {
  69         string v;
  70         if (get_attribute("date1904", v)) {
  71             date1904 = (v == "true" || v == "1");
  72         }
  73     } else if (tag == "numFmt") {
  74         string formatcode;
  75         if (get_attribute("formatCode", formatcode)) {
  76             // Heuristic for "date format" (FIXME: implement properly)
  77             if (strchr(formatcode.c_str(), 'd') &&
  78                 strchr(formatcode.c_str(), 'm') &&
  79                 strchr(formatcode.c_str(), 'y')) {
  80                 string v;
  81                 if (get_attribute("numFmtId", v)) {
  82                     unsigned long id = strtoul(v.c_str(), NULL, 10);
  83                     date_format.insert(id);
  84                 }
  85             }
  86         }
  87     } else if (tag == "cellXfs") {
  88         mode = MODE_CELLXFS;
  89     } else if (tag == "xf") {
  90         if (mode == MODE_CELLXFS) {
  91             string v;
  92             if (get_attribute("numFmtId", v)) {
  93                 unsigned long id = strtoul(v.c_str(), NULL, 10);
  94                 if ((id >= 14 && id <= 17) ||
  95                     date_format.find(id) != date_format.end()) {
  96                     date_style.insert(style_index);
  97                 }
  98             }
  99             ++style_index;
 100         }
 101     }
 102     return true;
 103 }
 104
 105 void
 106 XlsxParser::process_content(const string& content)
 107 {
 108     switch (mode) {
 109         case MODE_V_DATE: {
 110             // Date field.
 111             unsigned long c = strtoul(content.c_str(), NULL, 10);
 112             if (date1904) {
 113                 c -= 24107;
 114             } else {
 115                 // The spec insists we treat 1900 as a leap year!
 116                 if (c > 60) --c;
 117                 c -= 25568;
 118             }
 119             time_t t = c * 86400 + 43200;
 120             struct tm * tm = gmtime(&t);
 121             if (tm) {
 122                 char buf[32];
 123                 size_t res = strftime(buf, sizeof(buf), "%Y-%m-%d", tm);
 124                 if (res)
 125                     append_field(string(buf, res));
 126             }
 127             mode = MODE_NONE;
 128             return;
 129         }
 130         case MODE_V_STRING: {
 131             // Shared string use.
 132             unsigned long c = strtoul(content.c_str(), NULL, 10);
 133             if (c < sst.size()) {
 134                 append_field(sst[c]);
 135             }
 136             mode = MODE_NONE;
 137             return;
 138         }
 139         case MODE_V_LITERAL:
 140             // Literal (possibly calculated) field value.
 141             append_field(content);
 142             mode = MODE_NONE;
 143             return;
 144         case MODE_SI:
 145             // Shared string definition.
 146             sst.push_back(content);
 147             mode = MODE_NONE;
 148             return;
 149         default:
 150             return;
 151     }
 152 }