summaryrefslogtreecommitdiff
path: root/helpcompiler/source/HelpIndexer.cxx
blob: 25ff6ef0b09ec113c71f08a0e3b5fab0b2484b64 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
/*
 * This file is part of the LibreOffice project.
 *
 * This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at http://mozilla.org/MPL/2.0/.
 */

#include <helpcompiler/HelpIndexer.hxx>

#include <rtl/string.hxx>
#include <rtl/uri.hxx>
#include <rtl/ustrbuf.hxx>
#include <o3tl/runtimetooustring.hxx>
#include <osl/file.hxx>
#include <osl/thread.h>
#include <algorithm>
#include <memory>

#include "LuceneHelper.hxx"

using namespace lucene::document;

HelpIndexer::HelpIndexer(OUString const &lang, OUString const &module,
    OUString const &srcDir, OUString const &outDir)
    : d_lang(lang), d_module(module)
{
    d_indexDir = outDir + OUStringLiteral1('/') + module + ".idxl";
    d_captionDir = srcDir + "/caption";
    d_contentDir = srcDir + "/content";
}

bool HelpIndexer::indexDocuments()
{
    if (!scanForFiles())
        return false;

    try
    {
        OUString sLang = d_lang.getToken(0, '-');
        bool bUseCJK = sLang == "ja" || sLang == "ko" || sLang == "zh";

        // Construct the analyzer appropriate for the given language
        std::unique_ptr<lucene::analysis::Analyzer> analyzer;
        if (bUseCJK)
            analyzer.reset(new lucene::analysis::LanguageBasedAnalyzer(L"cjk"));
        else
            analyzer.reset(new lucene::analysis::standard::StandardAnalyzer());

        OUString ustrSystemPath;
        osl::File::getSystemPathFromFileURL(d_indexDir, ustrSystemPath);

        OString indexDirStr = OUStringToOString(ustrSystemPath, osl_getThreadTextEncoding());
        lucene::index::IndexWriter writer(indexDirStr.getStr(), analyzer.get(), true);
        //Double limit of tokens allowed, otherwise we'll get a too-many-tokens
        //exception for ja help. Could alternative ignore the exception and get
        //truncated results as per java-Lucene apparently
        writer.setMaxFieldLength(lucene::index::IndexWriter::DEFAULT_MAX_FIELD_LENGTH*2);

        // Index the identified help files
        Document doc;
        for (auto const& elem : d_files)
        {
            helpDocument(elem, &doc);
            writer.addDocument(&doc);
            doc.clear();
        }
        writer.optimize();

        // Optimize the index
        writer.optimize();
    }
    catch (CLuceneError &e)
    {
        d_error = o3tl::runtimeToOUString(e.what());
        return false;
    }

    return true;
}


bool HelpIndexer::scanForFiles() {
    if (!scanForFiles(d_contentDir)) {
        return false;
    }
    if (!scanForFiles(d_captionDir)) {
        return false;
    }
    return true;
}

bool HelpIndexer::scanForFiles(OUString const & path) {

    osl::Directory dir(path);
    if (osl::FileBase::E_None != dir.open()) {
        d_error = "Error reading directory " + path;
        return false;
    }

    osl::DirectoryItem item;
    osl::FileStatus fileStatus(osl_FileStatus_Mask_FileName | osl_FileStatus_Mask_Type);
    while (dir.getNextItem(item) == osl::FileBase::E_None) {
        item.getFileStatus(fileStatus);
        if (fileStatus.getFileType() == osl::FileStatus::Regular) {
            d_files.insert(fileStatus.getFileName());
        }
    }

    return true;
}

void HelpIndexer::helpDocument(OUString const & fileName, Document *doc) const {
    // Add the help path as an indexed, untokenized field.

    OUString path = "#HLP#" + d_module + "/" + fileName;
    std::vector<TCHAR> aPath(OUStringToTCHARVec(path));
    doc->add(*_CLNEW Field(_T("path"), aPath.data(), Field::STORE_YES | Field::INDEX_UNTOKENIZED));

    OUString sEscapedFileName =
        rtl::Uri::encode(fileName,
        rtl_UriCharClassUric, rtl_UriEncodeIgnoreEscapes, RTL_TEXTENCODING_UTF8);

    // Add the caption as a field.
    OUString captionPath = d_captionDir + "/" + sEscapedFileName;
    doc->add(*_CLNEW Field(_T("caption"), helpFileReader(captionPath), Field::STORE_NO | Field::INDEX_TOKENIZED));

    // Add the content as a field.
    OUString contentPath = d_contentDir + "/" + sEscapedFileName;
    doc->add(*_CLNEW Field(_T("content"), helpFileReader(contentPath), Field::STORE_NO | Field::INDEX_TOKENIZED));
}

lucene::util::Reader *HelpIndexer::helpFileReader(OUString const & path) {
    osl::File file(path);
    if (osl::FileBase::E_None == file.open(osl_File_OpenFlag_Read)) {
        file.close();
        OUString ustrSystemPath;
        osl::File::getSystemPathFromFileURL(path, ustrSystemPath);
        OString pathStr = OUStringToOString(ustrSystemPath, osl_getThreadTextEncoding());
        return _CLNEW lucene::util::FileReader(pathStr.getStr(), "UTF-8");
    } else {
        return _CLNEW lucene::util::StringReader(L"");
    }
}

/* vim:set shiftwidth=4 softtabstop=4 expandtab: */