|
1 /** |
|
2 * Relational pipes |
|
3 * Copyright © 2023 František Kučera (Frantovo.cz, GlobalCode.info) |
|
4 * |
|
5 * This program is free software: you can redistribute it and/or modify |
|
6 * it under the terms of the GNU General Public License as published by |
|
7 * the Free Software Foundation, version 3 of the License. |
|
8 * |
|
9 * This program is distributed in the hope that it will be useful, |
|
10 * but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
11 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
12 * GNU General Public License for more details. |
|
13 * |
|
14 * You should have received a copy of the GNU General Public License |
|
15 * along with this program. If not, see <http://www.gnu.org/licenses/>. |
|
16 */ |
|
17 #pragma once |
|
18 |
|
19 #include <cstdlib> |
|
20 #include <iostream> |
|
21 #include <string> |
|
22 #include <sstream> |
|
23 #include <vector> |
|
24 #include <algorithm> |
|
25 #include <exception> |
|
26 #include <regex> |
|
27 |
|
28 #include <libxml++-2.6/libxml++/libxml++.h> |
|
29 |
|
30 #include <relpipe/writer/typedefs.h> |
|
31 |
|
32 namespace relpipe { |
|
33 namespace in { |
|
34 namespace ods { |
|
35 |
|
36 using namespace relpipe::writer; |
|
37 |
|
38 class ODSCommand { |
|
39 private: |
|
40 std::wstring_convert<codecvt_utf8<wchar_t>> convertor; |
|
41 |
|
42 xmlpp::Node::PrefixNsMap ns; |
|
43 std::shared_ptr<RelationalWriter> writer; |
|
44 |
|
45 string_t xpath(xmlpp::Node* node, std::string xpath) { |
|
46 return convertor.from_bytes(node->eval_to_string(xpath, ns)); |
|
47 } |
|
48 |
|
49 void processCell(xmlpp::Node* c, AttributeMetadata& am) { |
|
50 string_t value = xpath(c, "@o:value"); |
|
51 if (value.size()) writer->writeAttribute(value); |
|
52 else writer->writeAttribute(xpath(c, "tx:p")); |
|
53 // TODO: be aware of the current data types |
|
54 } |
|
55 |
|
56 void processRow(xmlpp::Node* r, std::vector<AttributeMetadata>& am) { |
|
57 for (int i = 0; i < am.size(); i++) { |
|
58 auto xpe = std::string("t:table-cell[") |
|
59 + std::to_string(i + 1) |
|
60 + "]"; |
|
61 auto cells = r->find(xpe, ns); |
|
62 if (cells.size() == 1) { |
|
63 processCell(cells[0], am[i]); |
|
64 } else { |
|
65 writer->writeAttribute(L""); |
|
66 // TODO: support also other data types |
|
67 } |
|
68 } |
|
69 |
|
70 // FIXME: support sparse data / missing values: |
|
71 // <table:table-row table:number-rows-repeated="2"> |
|
72 // <table:table-cell table:number-columns-repeated="3"/> |
|
73 |
|
74 } |
|
75 |
|
76 void processTable(xmlpp::Node* t) { |
|
77 auto relation = xpath(t, "@t:name"); |
|
78 std::vector<AttributeMetadata> metadata; |
|
79 |
|
80 for (xmlpp::Node* c : t->find("t:table-row[1]/t:table-cell", ns)) { |
|
81 auto name = xpath(c, "tx:p"); |
|
82 if (name.size()) { |
|
83 metadata.push_back({name, TypeId::STRING}); |
|
84 // TODO: detect and support other data types |
|
85 } |
|
86 } |
|
87 |
|
88 writer->startRelation(relation, metadata, true); |
|
89 |
|
90 int i = 0; |
|
91 for (xmlpp::Node* r : t->find("t:table-row", ns)) { |
|
92 i++; |
|
93 if (i == 1) continue; // skip header row |
|
94 processRow(r, metadata); |
|
95 } |
|
96 } |
|
97 |
|
98 public: |
|
99 |
|
100 ODSCommand(std::shared_ptr<RelationalWriter> writer) : writer(writer) { |
|
101 ns["o"] = "urn:oasis:names:tc:opendocument:xmlns:office:1.0"; |
|
102 ns["t"] = "urn:oasis:names:tc:opendocument:xmlns:table:1.0"; |
|
103 ns["tx"] = "urn:oasis:names:tc:opendocument:xmlns:text:1.0"; |
|
104 } |
|
105 |
|
106 void process(std::istream & input) { |
|
107 xmlpp::DomParser parser; |
|
108 parser.parse_stream(input); |
|
109 |
|
110 xmlpp::Element* root = parser.get_document()->get_root_node(); |
|
111 auto spreadsheets = root->find("/o:document/o:body/o:spreadsheet", ns); |
|
112 |
|
113 if (spreadsheets.size() == 1) { |
|
114 for (xmlpp::Node* table : spreadsheets[0]->find("t:table", ns)) { |
|
115 processTable(table); |
|
116 } |
|
117 } else { |
|
118 throw RelpipeWriterException(L"Invalid XML structure. " |
|
119 "Expecting OpenDocument spreadsheet."); |
|
120 } |
|
121 |
|
122 |
|
123 } |
|
124 }; |
|
125 |
|
126 } |
|
127 } |
|
128 } |