version 0.5.5

Avoid crashes when passing unknown dicts to the -u flag

Fix #87

.clang-format

@@ -15,7 +15,7 @@ BreakBeforeBinaryOperators: true
 BreakBeforeTernaryOperators: true
 BreakConstructorInitializersBeforeComma: true
 BinPackParameters: true
-ColumnLimit:     0
+ColumnLimit:     120
 ConstructorInitializerAllOnOneLineOrOnePerLine: false
 DerivePointerAlignment: false
 ExperimentalAutoDetectBinPacking: false

.github/workflows/main.yml

@@ -20,12 +20,13 @@ jobs:
       fail-fast: true
 
       matrix:
-        os: [ubuntu-latest]
+        os: [ubuntu-20.04, ubuntu-latest]
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
         with:
           submodules: 'recursive'
       - uses: jwlawson/actions-setup-cmake@v1.4
+        if: matrix.os != 'ubuntu-latest'
         with:
           cmake-version: '3.5.1'
           github-api-token: ${{ secrets.GITHUB_TOKEN }}

CMakeLists.txt

@@ -3,6 +3,10 @@ project(sdcv)
 cmake_minimum_required(VERSION 3.5 FATAL_ERROR)
 cmake_policy(VERSION 3.5)
 
+set(CMAKE_CXX_STANDARD 11)
+set(CMAKE_CXX_STANDARD_REQUIRED True)
+set(CMAKE_CXX_EXTENSIONS False)
+
 include("${CMAKE_CURRENT_SOURCE_DIR}/cmake/compiler.cmake")
 
 set(ZLIB_FIND_REQUIRED True)
@@ -91,7 +95,7 @@ set(CPACK_PACKAGE_VENDOR "Evgeniy Dushistov <dushistov@mail.ru>")
 set(CPACK_PACKAGE_DESCRIPTION_FILE "${CMAKE_CURRENT_SOURCE_DIR}/README.org")
 set(CPACK_PACKAGE_VERSION_MAJOR "0")
 set(CPACK_PACKAGE_VERSION_MINOR "5")
-set(CPACK_PACKAGE_VERSION_PATCH "3")
+set(CPACK_PACKAGE_VERSION_PATCH "5")
 
 set(sdcv_VERSION
 	"${CPACK_PACKAGE_VERSION_MAJOR}.${CPACK_PACKAGE_VERSION_MINOR}.${CPACK_PACKAGE_VERSION_PATCH}")
@@ -144,5 +148,6 @@ if (BUILD_TESTS)
   add_sdcv_shell_test(t_datadir)
   add_sdcv_shell_test(t_return_code)
   add_sdcv_shell_test(t_multiple_results)
+  add_sdcv_shell_test(t_newlines_in_ifo)
 
 endif (BUILD_TESTS)

NEWS

@@ -1,3 +1,13 @@
+Version 0.5.5
+    - Avoid crashes when passing unknown dicts to the -u flag (by NiLuJe)
+    - Use off_t for stuff mainly assigned to a stat.st_size value
+Version 0.5.4
+    - Use binary search for synonyms
+    - Various improvments in work with synonyms
+    - Added --json (same as --json-output) to match man
+    - Show all matched result
+    - More robust parsing of ifo file
+    - Prevent crash if file size of files not matched expecting one for .oft files
 Version 0.5.3
   - Use single quotes around JSON data to reduce need for escaping
   - Store integer magic in cache file

README.org

@@ -1,6 +1,9 @@
 #+OPTIONS: ^:nil
 [[https://github.com/Dushistov/sdcv/actions?query=workflow%3ACI+branch%3Amaster][https://github.com/Dushistov/sdcv/workflows/CI/badge.svg]]
 [[https://github.com/Dushistov/sdcv/blob/master/LICENSE][https://img.shields.io/badge/license-GPL%202-brightgreen.svg]]
+
+* sdcv
+*sdcv* is a simple, cross-platform, text-based utility for working with dictionaries in [[http://stardict-4.sourceforge.net/][StarDict]] format.
 * How to compile and install
 #+BEGIN_SRC sh
 mkdir /tmp/build-sdcv

cmake/compiler.cmake

@@ -16,19 +16,6 @@ if (NOT DEFINED SDCV_COMPILER_IS_GCC_COMPATIBLE)
   endif()
 endif()
 
-if (MSVC AND (MSVC_VERSION LESS 1900))
-  message(FATAL_ERROR "MSVC version ${MSVC_VERSION} have no full c++11 support")
-elseif (MSVC)
-  add_definitions(-DNOMINMAX)
-elseif (NOT MSVC)
-  check_cxx_compiler_flag("-std=c++11" CXX_SUPPORTS_CXX11)
-  if (CXX_SUPPORTS_CXX11)
-    append("-std=c++11" CMAKE_CXX_FLAGS)
-  else ()
-    message(FATAL_ERROR "sdcv requires C++11 support but the '-std=c++11' flag isn't supported.")
-  endif()
-endif ()
-
 if (SDCV_COMPILER_IS_GCC_COMPATIBLE)
   append("-Wall" "-Wextra" "-Wformat-security" "-Wcast-align" "-Werror=format" "-Wcast-qual" CMAKE_C_FLAGS)
   append("-Wall" "-pedantic" "-Wextra" "-Wformat-security" "-Wcast-align" "-Werror=format" "-Wcast-qual" CMAKE_CXX_FLAGS)

src/dictziplib.hpp

@@ -27,7 +27,7 @@ public:
 private:
     const char *start; /* start of mmap'd area */
     const char *end; /* end of mmap'd area */
-    unsigned long size; /* size of mmap */
+    off_t size; /* size of mmap */
 
     int type;
     z_stream zStream;
@@ -47,7 +47,7 @@ private:
     std::string origFilename;
     std::string comment;
     unsigned long crc;
-    unsigned long length;
+    off_t length;
     unsigned long compressedLength;
     DictCache cache[DICT_CACHE_SIZE];
     MapFile mapfile;

src/mapfile.hpp

@@ -7,6 +7,7 @@
 #ifdef HAVE_MMAP
 #include <fcntl.h>
 #include <sys/mman.h>
+#include <sys/stat.h>
 #include <sys/types.h>
 #endif
 #ifdef _WIN32
@@ -21,13 +22,13 @@ public:
     ~MapFile();
     MapFile(const MapFile &) = delete;
     MapFile &operator=(const MapFile &) = delete;
-    bool open(const char *file_name, unsigned long file_size);
+    bool open(const char *file_name, off_t file_size);
     gchar *begin() { return data; }
 
 private:
     char *data = nullptr;
-    unsigned long size = 0ul;
 #ifdef HAVE_MMAP
+    size_t size = 0u;
     int mmap_fd = -1;
 #elif defined(_WIN32)
     HANDLE hFile = 0;
@@ -35,25 +36,31 @@ private:
 #endif
 };
 
-inline bool MapFile::open(const char *file_name, unsigned long file_size)
+inline bool MapFile::open(const char *file_name, off_t file_size)
 {
-    size = file_size;
 #ifdef HAVE_MMAP
     if ((mmap_fd = ::open(file_name, O_RDONLY)) < 0) {
         // g_print("Open file %s failed!\n",fullfilename);
         return false;
     }
-    data = (gchar *)mmap(nullptr, file_size, PROT_READ, MAP_SHARED, mmap_fd, 0);
+    struct stat st;
+    if (fstat(mmap_fd, &st) == -1 || st.st_size < 0 || (st.st_size == 0 && S_ISREG(st.st_mode))
+        || st.st_size != file_size) {
+        close(mmap_fd);
+        return false;
+    }
+
+    size = static_cast<size_t>(st.st_size);
+    data = (gchar *)mmap(nullptr, size, PROT_READ, MAP_SHARED, mmap_fd, 0);
     if ((void *)data == (void *)(-1)) {
         // g_print("mmap file %s failed!\n",idxfilename);
+        size = 0u;
         data = nullptr;
         return false;
     }
 #elif defined(_WIN32)
-    hFile = CreateFile(file_name, GENERIC_READ, 0, nullptr, OPEN_ALWAYS,
-                       FILE_ATTRIBUTE_NORMAL, 0);
-    hFileMap = CreateFileMapping(hFile, nullptr, PAGE_READONLY, 0,
-                                 file_size, nullptr);
+    hFile = CreateFile(file_name, GENERIC_READ, 0, nullptr, OPEN_ALWAYS, FILE_ATTRIBUTE_NORMAL, 0);
+    hFileMap = CreateFileMapping(hFile, nullptr, PAGE_READONLY, 0, file_size, nullptr);
     data = (gchar *)MapViewOfFile(hFileMap, FILE_MAP_READ, 0, 0, file_size);
 #else
     gsize read_len;

src/sdcv.cpp

@@ -186,10 +186,13 @@ try {
         }
 
         // add bookname to list
-        gchar **p = get_impl(use_dict_list);
-        while (*p) {
-            order_list.push_back(bookname_to_ifo.at(*p));
-            ++p;
+        for (gchar **p = get_impl(use_dict_list); *p != nullptr; ++p) {
+            auto it = bookname_to_ifo.find(*p);
+            if (it != bookname_to_ifo.end()) {
+                order_list.push_back(it->second);
+            } else {
+                fprintf(stderr, _("Unknown dictionary: %s\n"), *p);
+            }
         }
     } else {
         std::string ordering_cfg_file = std::string(g_get_user_config_dir()) + G_DIR_SEPARATOR_S "sdcv_ordering";
@@ -201,7 +204,12 @@ try {
         if (ordering_file != nullptr) {
             std::string line;
             while (stdio_getline(ordering_file, line)) {
-                order_list.push_back(bookname_to_ifo.at(line));
+                auto it = bookname_to_ifo.find(line);
+                if (it != bookname_to_ifo.end()) {
+                    order_list.push_back(it->second);
+                } else {
+                    fprintf(stderr, _("Unknown dictionary: %s\n"), line.c_str());
+                }
             }
             fclose(ordering_file);
         }

src/stardict_lib.cpp

@@ -5,6 +5,7 @@
 #include <algorithm>
 #include <cctype>
 #include <cstring>
+#include <map>
 #include <stdexcept>
 
 #include <glib/gstdio.h>
@@ -78,108 +79,93 @@ bool DictInfo::load_from_ifo_file(const std::string &ifofilename,
 {
     ifo_file_name = ifofilename;
     glib::CharStr buffer;
-    if (!g_file_get_contents(ifofilename.c_str(), get_addr(buffer), nullptr, nullptr))
+    gsize length = 0;
+    if (!g_file_get_contents(ifofilename.c_str(), get_addr(buffer), &length, nullptr)) {
+        fprintf(stderr, "Can not read from %s\n", ifofilename.c_str());
         return false;
+    }
 
     static const char TREEDICT_MAGIC_DATA[] = "StarDict's treedict ifo file";
     static const char DICT_MAGIC_DATA[] = "StarDict's dict ifo file";
 
     const gchar *magic_data = istreedict ? TREEDICT_MAGIC_DATA : DICT_MAGIC_DATA;
-    static const unsigned char utf8_bom[] = { 0xEF, 0xBB, 0xBF, '\0' };
-    if (!g_str_has_prefix(
-            g_str_has_prefix(get_impl(buffer), (const gchar *)(utf8_bom)) ? get_impl(buffer) + 3 : get_impl(buffer),
-            magic_data)) {
+    static const gchar utf8_bom[] = { (gchar)0xEF, (gchar)0xBB, (gchar)0xBF, '\0' };
+
+    const gchar *p = get_impl(buffer);
+    const gchar *end = p + length;
+
+    if (g_str_has_prefix(p, utf8_bom)) {
+        p += strlen(utf8_bom);
+    }
+    if (!g_str_has_prefix(p, magic_data)) {
+        fprintf(stderr, "No magic header(%s) in ifo file\n", magic_data);
         return false;
     }
+    p += strlen(magic_data);
 
-    gchar *p1 = get_impl(buffer) + strlen(magic_data) - 1;
-
-    gchar *p2 = strstr(p1, "\nwordcount=");
-    if (p2 == nullptr)
+    std::map<std::string, std::string> key_value_map;
+    while (p != end) {
+        auto key_it = std::find_if(p, end, [](gchar ch) { return !g_ascii_isspace(ch); });
+        if (key_it == end) {
+            break;
+        }
+        auto eq_it = std::find(key_it, end, gchar('='));
+        if (eq_it == end) {
+            fprintf(stderr, "Invalid part of ifo (no '=') here: %s\n", key_it);
             return false;
+        }
+        auto val_it = std::find_if(eq_it + 1, end, [](gchar ch) { return !g_ascii_isspace(ch); });
+        if (val_it == end) {
+            key_value_map.insert(std::make_pair(std::string(key_it, eq_it), std::string()));
+            break;
+        }
 
-    gchar *p3 = strchr(p2 + sizeof("\nwordcount=") - 1, '\n');
+        auto line_end_it = std::find_if(val_it, end, [](gchar ch) { return ch == '\r' || ch == '\n'; });
+        key_value_map.insert(std::make_pair(std::string(key_it, eq_it), std::string(val_it, line_end_it)));
+        if (line_end_it == end)
+            break;
+        p = line_end_it + 1;
+    }
 
-    wordcount = atol(std::string(p2 + sizeof("\nwordcount=") - 1, p3 - (p2 + sizeof("\nwordcount=") - 1)).c_str());
+    std::map<std::string, std::string>::const_iterator it;
+#define FIND_KEY(_key_)                                            \
+    it = key_value_map.find(_key_);                                \
+    if (it == key_value_map.end()) {                               \
+        fprintf(stderr, "Can not find '%s' in ifo file\n", _key_); \
+        return false;                                              \
+    }
+
+    FIND_KEY("wordcount")
+    wordcount = atol(it->second.c_str());
 
     if (istreedict) {
-        p2 = strstr(p1, "\ntdxfilesize=");
-        if (p2 == nullptr)
-            return false;
-
-        p3 = strchr(p2 + sizeof("\ntdxfilesize=") - 1, '\n');
-
-        index_file_size = atol(std::string(p2 + sizeof("\ntdxfilesize=") - 1, p3 - (p2 + sizeof("\ntdxfilesize=") - 1)).c_str());
-
+        FIND_KEY("tdxfilesize")
+        index_file_size = atol(it->second.c_str());
     } else {
+        FIND_KEY("idxfilesize")
+        index_file_size = atol(it->second.c_str());
+    }
+    FIND_KEY("bookname")
+    bookname = it->second;
 
-        p2 = strstr(p1, "\nidxfilesize=");
-        if (p2 == nullptr)
-            return false;
-
-        p3 = strchr(p2 + sizeof("\nidxfilesize=") - 1, '\n');
-        index_file_size = atol(std::string(p2 + sizeof("\nidxfilesize=") - 1, p3 - (p2 + sizeof("\nidxfilesize=") - 1)).c_str());
+#define SET_IF_EXISTS(_key_)         \
+    it = key_value_map.find(#_key_); \
+    if (it != key_value_map.end()) { \
+        _key_ = it->second;          \
     }
 
-    p2 = strstr(p1, "\nbookname=");
-
-    if (p2 == nullptr)
-        return false;
-
-    p2 = p2 + sizeof("\nbookname=") - 1;
-    p3 = strchr(p2, '\n');
-    bookname.assign(p2, p3 - p2);
-
-    p2 = strstr(p1, "\nauthor=");
-    if (p2) {
-        p2 = p2 + sizeof("\nauthor=") - 1;
-        p3 = strchr(p2, '\n');
-        author.assign(p2, p3 - p2);
-    }
-
-    p2 = strstr(p1, "\nemail=");
-    if (p2) {
-        p2 = p2 + sizeof("\nemail=") - 1;
-        p3 = strchr(p2, '\n');
-        email.assign(p2, p3 - p2);
-    }
-
-    p2 = strstr(p1, "\nwebsite=");
-    if (p2) {
-        p2 = p2 + sizeof("\nwebsite=") - 1;
-        p3 = strchr(p2, '\n');
-        website.assign(p2, p3 - p2);
-    }
-
-    p2 = strstr(p1, "\ndate=");
-    if (p2) {
-        p2 = p2 + sizeof("\ndate=") - 1;
-        p3 = strchr(p2, '\n');
-        date.assign(p2, p3 - p2);
-    }
-
-    p2 = strstr(p1, "\ndescription=");
-    if (p2) {
-        p2 = p2 + sizeof("\ndescription=") - 1;
-        p3 = strchr(p2, '\n');
-        description.assign(p2, p3 - p2);
-    }
-
-    p2 = strstr(p1, "\nsametypesequence=");
-    if (p2) {
-        p2 += sizeof("\nsametypesequence=") - 1;
-        p3 = strchr(p2, '\n');
-        sametypesequence.assign(p2, p3 - p2);
-    }
-
-    p2 = strstr(p1, "\nsynwordcount=");
+    SET_IF_EXISTS(author)
+    SET_IF_EXISTS(email)
+    SET_IF_EXISTS(website)
+    SET_IF_EXISTS(date)
+    SET_IF_EXISTS(description)
+    SET_IF_EXISTS(sametypesequence)
     syn_wordcount = 0;
-    if (p2) {
-        p2 += sizeof("\nsynwordcount=") - 1;
-        p3 = strchr(p2, '\n');
-        syn_wordcount = atol(std::string(p2, p3 - p2).c_str());
-    }
-
+    it = key_value_map.find("synwordcount");
+    if (it != key_value_map.end())
+        syn_wordcount = atol(it->second.c_str());
+#undef FIND_KEY
+#undef SET_IF_EXISTS
     return true;
 }
 
@@ -443,7 +429,7 @@ public:
         if (idxfile)
             fclose(idxfile);
     }
-    bool load(const std::string &url, gulong wc, gulong fsize, bool verbose) override;
+    bool load(const std::string &url, gulong wc, off_t fsize, bool verbose) override;
     const gchar *get_key(glong idx) override;
     void get_data(glong idx) override { get_key(idx); }
     const gchar *get_key_and_data(glong idx) override
@@ -503,7 +489,7 @@ public:
     {
     }
     ~WordListIndex() { g_free(idxdatabuf); }
-    bool load(const std::string &url, gulong wc, gulong fsize, bool verbose) override;
+    bool load(const std::string &url, gulong wc, off_t fsize, bool verbose) override;
     const gchar *get_key(glong idx) override { return wordlist[idx]; }
     void get_data(glong idx) override;
     const gchar *get_key_and_data(glong idx) override
@@ -629,7 +615,7 @@ bool OffsetIndex::save_cache(const std::string &url, bool verbose)
     return false;
 }
 
-bool OffsetIndex::load(const std::string &url, gulong wc, gulong fsize, bool verbose)
+bool OffsetIndex::load(const std::string &url, gulong wc, off_t fsize, bool verbose)
 {
     wordcount = wc;
     gulong npages = (wc - 1) / ENTR_PER_PAGE + 2;
@@ -772,7 +758,7 @@ bool OffsetIndex::lookup(const char *str, std::set<glong> &idxs, glong &next_idx
     return bFound;
 }
 
-bool WordListIndex::load(const std::string &url, gulong wc, gulong fsize, bool)
+bool WordListIndex::load(const std::string &url, gulong wc, off_t fsize, bool)
 {
     gzFile in = gzopen(url.c_str(), "rb");
     if (in == nullptr)
@@ -785,7 +771,7 @@ bool WordListIndex::load(const std::string &url, gulong wc, gulong fsize, bool)
     if (len < 0)
         return false;
 
-    if (gulong(len) != fsize)
+    if (static_cast<off_t>(len) != fsize)
         return false;
 
     wordlist.resize(wc + 1);
@@ -934,7 +920,7 @@ bool Dict::Lookup(const char *str, std::set<glong> &idxs, glong &next_idx)
 
 bool Dict::load(const std::string &ifofilename, bool verbose)
 {
-    gulong idxfilesize;
+    off_t idxfilesize;
     if (!load_ifofile(ifofilename, idxfilesize))
         return false;
 
@@ -978,7 +964,7 @@ bool Dict::load(const std::string &ifofilename, bool verbose)
     return true;
 }
 
-bool Dict::load_ifofile(const std::string &ifofilename, gulong &idxfilesize)
+bool Dict::load_ifofile(const std::string &ifofilename, off_t &idxfilesize)
 {
     DictInfo dict_info;
     if (!dict_info.load_from_ifo_file(ifofilename, false))

src/stardict_lib.hpp

@@ -1,10 +1,8 @@
 #pragma once
 
-#include <cstdio>
 #include <cstring>
 #include <functional>
 #include <list>
-#include <map>
 #include <memory>
 #include <set>
 #include <string>
@@ -79,8 +77,8 @@ struct DictInfo {
     std::string website;
     std::string date;
     std::string description;
-    guint32 index_file_size;
-    guint32 syn_file_size;
+    off_t index_file_size;
+    off_t syn_file_size;
     std::string sametypesequence;
 
     bool load_from_ifo_file(const std::string &ifofilename, bool istreedict);
@@ -93,12 +91,13 @@ public:
     guint32 wordentry_size;
 
     virtual ~IIndexFile() {}
-    virtual bool load(const std::string &url, gulong wc, gulong fsize, bool verbose) = 0;
+    virtual bool load(const std::string &url, gulong wc, off_t fsize, bool verbose) = 0;
     virtual const gchar *get_key(glong idx) = 0;
     virtual void get_data(glong idx) = 0;
     virtual const gchar *get_key_and_data(glong idx) = 0;
     virtual bool lookup(const char *str, std::set<glong> &idxs, glong &next_idx) = 0;
-    virtual bool lookup(const char *str, std::set<glong> &idxs) {
+    virtual bool lookup(const char *str, std::set<glong> &idxs)
+    {
         glong unused_next_idx;
         return lookup(str, idxs, unused_next_idx);
     };
@@ -144,7 +143,8 @@ public:
         *size = idx_file->wordentry_size;
     }
     bool Lookup(const char *str, std::set<glong> &idxs, glong &next_idx);
-    bool Lookup(const char *str, std::set<glong> &idxs) {
+    bool Lookup(const char *str, std::set<glong> &idxs)
+    {
         glong unused_next_idx;
         return Lookup(str, idxs, unused_next_idx);
     }
@@ -160,7 +160,7 @@ private:
     std::unique_ptr<IIndexFile> idx_file;
     std::unique_ptr<SynFile> syn_file;
 
-    bool load_ifofile(const std::string &ifofilename, gulong &idxfilesize);
+    bool load_ifofile(const std::string &ifofilename, off_t &idxfilesize);
 };
 
 class Libs

tests/not-unix-newlines-ifo/russian/russian.ifo

@@ -0,0 +1,9 @@
+StarDict's dict ifo file
+version=3.0.0
+bookname=Russian-English Dictionary (ru-en)
+wordcount=415144
+idxfilesize=12344255
+sametypesequence=h
+synwordcount=1277580
+author=Vuizur
+description=

tests/t_json

@@ -18,7 +18,8 @@ test_json() {
     fi
 }
 
-test_json '[{"name": "Test synonyms", "wordcount": "2"},
+test_json '[{"name": "Russian-English Dictionary (ru-en)", "wordcount": "415144"},
+            {"name": "Test synonyms", "wordcount": "2"},
             {"name": "Test multiple results", "wordcount": "246"},
             {"name": "Sample 1 test dictionary", "wordcount": "1"},
             {"name": "test_dict", "wordcount": "1"}]' -x -j -l -n --data-dir "$TEST_DIR"

tests/t_newlines_in_ifo

@@ -0,0 +1,18 @@
+#!/bin/sh
+
+set -e
+
+PATH_TO_SDCV="$1"
+TEST_DIR="$2"
+
+unset SDCV_PAGER
+unset STARDICT_DATA_DIR
+
+RES=$("$PATH_TO_SDCV" -n -x --data-dir="$TEST_DIR/not-unix-newlines-ifo" -l | tail -n 1)
+
+if [ "$RES" = "Russian-English Dictionary (ru-en)    415144" ]; then
+  exit 0
+else
+  echo "test failed, unexpected result: $RES" >&2
+  exit 1
+fi