Mercurial > hgrepos > Python2 > PyMuPDF

comparison mupdf-source/thirdparty/tesseract/src/training/wordlist2dawg.cpp @ 2:b50eed0cc0ef upstream

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
ADD: MuPDF v1.26.7: the MuPDF source as downloaded by a default build of PyMuPDF 1.26.4. The directory name has changed: no version number in the expanded directory now.
author Franz Glasner <fzglas.hg@dom66.de>
date Mon, 15 Sep 2025 11:43:07 +0200
parents
children
comparison
equal deleted inserted replaced
1:1d09e1dec1d9 2:b50eed0cc0ef
1 ///////////////////////////////////////////////////////////////////////
2 // File: wordlist2dawg.cpp
3 // Description: Program to generate a DAWG from a word list file
4 // Author: Thomas Kielbus
5 //
6 // (C) Copyright 2006, Google Inc.
7 // Licensed under the Apache License, Version 2.0 (the "License");
8 // you may not use this file except in compliance with the License.
9 // You may obtain a copy of the License at
10 // http://www.apache.org/licenses/LICENSE-2.0
11 // Unless required by applicable law or agreed to in writing, software
12 // distributed under the License is distributed on an "AS IS" BASIS,
13 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 // See the License for the specific language governing permissions and
15 // limitations under the License.
16 //
17 ///////////////////////////////////////////////////////////////////////
18
19 // Given a file that contains a list of words (one word per line) this program
20 // generates the corresponding squished DAWG file.
21
22 #include "classify.h"
23 #include "commontraining.h" // CheckSharedLibraryVersion
24 #include "dawg.h"
25 #include "dict.h"
26 #include "helpers.h"
27 #include "serialis.h"
28 #include "trie.h"
29 #include "unicharset.h"
30
31 using namespace tesseract;
32
33 int main(int argc, char **argv) {
34 tesseract::CheckSharedLibraryVersion();
35
36 if (argc > 1 && (!strcmp(argv[1], "-v") || !strcmp(argv[1], "--version"))) {
37 printf("%s\n", tesseract::TessBaseAPI::Version());
38 return EXIT_SUCCESS;
39 } else if (!(argc == 4 || (argc == 5 && strcmp(argv[1], "-t") == 0) ||
40 (argc == 6 && strcmp(argv[1], "-r") == 0))) {
41 printf(
42 "Usage: %s -v | --version |\n"
43 " %s [-t | -r [reverse policy] ] word_list_file"
44 " dawg_file unicharset_file\n",
45 argv[0], argv[0]);
46 return EXIT_FAILURE;
47 }
48 tesseract::Classify classify;
49 int argv_index = 0;
50 if (argc == 5) {
51 ++argv_index;
52 }
53 tesseract::Trie::RTLReversePolicy reverse_policy = tesseract::Trie::RRP_DO_NO_REVERSE;
54 if (argc == 6) {
55 ++argv_index;
56 int tmp_int;
57 sscanf(argv[++argv_index], "%d", &tmp_int);
58 reverse_policy = static_cast<tesseract::Trie::RTLReversePolicy>(tmp_int);
59 tprintf("Set reverse_policy to %s\n", tesseract::Trie::get_reverse_policy_name(reverse_policy));
60 }
61 const char *wordlist_filename = argv[++argv_index];
62 const char *dawg_filename = argv[++argv_index];
63 const char *unicharset_file = argv[++argv_index];
64 tprintf("Loading unicharset from '%s'\n", unicharset_file);
65 if (!classify.getDict().getUnicharset().load_from_file(unicharset_file)) {
66 tprintf("Failed to load unicharset from '%s'\n", unicharset_file);
67 return EXIT_FAILURE;
68 }
69 const UNICHARSET &unicharset = classify.getDict().getUnicharset();
70 if (argc == 4 || argc == 6) {
71 tesseract::Trie trie(
72 // the first 3 arguments are not used in this case
73 tesseract::DAWG_TYPE_WORD, "", SYSTEM_DAWG_PERM, unicharset.size(),
74 classify.getDict().dawg_debug_level);
75 tprintf("Reading word list from '%s'\n", wordlist_filename);
76 if (!trie.read_and_add_word_list(wordlist_filename, unicharset, reverse_policy)) {
77 tprintf("Failed to add word list from '%s'\n", wordlist_filename);
78 return EXIT_FAILURE;
79 }
80 tprintf("Reducing Trie to SquishedDawg\n");
81 std::unique_ptr<tesseract::SquishedDawg> dawg(trie.trie_to_dawg());
82 if (dawg && dawg->NumEdges() > 0) {
83 tprintf("Writing squished DAWG to '%s'\n", dawg_filename);
84 dawg->write_squished_dawg(dawg_filename);
85 } else {
86 tprintf("Dawg is empty, skip producing the output file\n");
87 }
88 } else if (argc == 5) {
89 tprintf("Loading dawg DAWG from '%s'\n", dawg_filename);
90 tesseract::SquishedDawg words(dawg_filename,
91 // these 3 arguments are not used in this case
92 tesseract::DAWG_TYPE_WORD, "", SYSTEM_DAWG_PERM,
93 classify.getDict().dawg_debug_level);
94 tprintf("Checking word list from '%s'\n", wordlist_filename);
95 words.check_for_words(wordlist_filename, unicharset, true);
96 } else { // should never get here
97 tprintf("Invalid command-line options\n");
98 return EXIT_FAILURE;
99 }
100 return EXIT_SUCCESS;
101 }