Mercurial > hgrepos > Python2 > PyMuPDF

comparison mupdf-source/thirdparty/tesseract/src/api/renderer.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: renderer.cpp
3 // Description: Rendering interface to inject into TessBaseAPI
4 //
5 // (C) Copyright 2011, Google Inc.
6 // Licensed under the Apache License, Version 2.0 (the "License");
7 // you may not use this file except in compliance with the License.
8 // You may obtain a copy of the License at
9 // http://www.apache.org/licenses/LICENSE-2.0
10 // Unless required by applicable law or agreed to in writing, software
11 // distributed under the License is distributed on an "AS IS" BASIS,
12 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 // See the License for the specific language governing permissions and
14 // limitations under the License.
15 //
16 ///////////////////////////////////////////////////////////////////////
17
18 #ifdef HAVE_CONFIG_H
19 # include "config_auto.h"
20 #endif
21 #include <tesseract/baseapi.h>
22 #include <tesseract/renderer.h>
23 #include <cstring>
24 #include <memory> // std::unique_ptr
25 #include <string> // std::string
26 #include "serialis.h" // Serialize
27
28 namespace tesseract {
29
30 /**********************************************************************
31 * Base Renderer interface implementation
32 **********************************************************************/
33 TessResultRenderer::TessResultRenderer(const char *outputbase, const char *extension)
34 : next_(nullptr)
35 , fout_(stdout)
36 , file_extension_(extension)
37 , title_("")
38 , imagenum_(-1)
39 , happy_(true) {
40 if (strcmp(outputbase, "-") && strcmp(outputbase, "stdout")) {
41 std::string outfile = std::string(outputbase) + "." + extension;
42 fout_ = fopen(outfile.c_str(), "wb");
43 if (fout_ == nullptr) {
44 happy_ = false;
45 }
46 }
47 }
48
49 TessResultRenderer::~TessResultRenderer() {
50 if (fout_ != nullptr) {
51 if (fout_ != stdout) {
52 fclose(fout_);
53 } else {
54 clearerr(fout_);
55 }
56 }
57 delete next_;
58 }
59
60 void TessResultRenderer::insert(TessResultRenderer *next) {
61 if (next == nullptr) {
62 return;
63 }
64
65 TessResultRenderer *remainder = next_;
66 next_ = next;
67 if (remainder) {
68 while (next->next_ != nullptr) {
69 next = next->next_;
70 }
71 next->next_ = remainder;
72 }
73 }
74
75 bool TessResultRenderer::BeginDocument(const char *title) {
76 if (!happy_) {
77 return false;
78 }
79 title_ = title;
80 imagenum_ = -1;
81 bool ok = BeginDocumentHandler();
82 if (next_) {
83 ok = next_->BeginDocument(title) && ok;
84 }
85 return ok;
86 }
87
88 bool TessResultRenderer::AddImage(TessBaseAPI *api) {
89 if (!happy_) {
90 return false;
91 }
92 ++imagenum_;
93 bool ok = AddImageHandler(api);
94 if (next_) {
95 ok = next_->AddImage(api) && ok;
96 }
97 return ok;
98 }
99
100 bool TessResultRenderer::EndDocument() {
101 if (!happy_) {
102 return false;
103 }
104 bool ok = EndDocumentHandler();
105 if (next_) {
106 ok = next_->EndDocument() && ok;
107 }
108 return ok;
109 }
110
111 void TessResultRenderer::AppendString(const char *s) {
112 if (s == nullptr) {
113 return;
114 }
115 AppendData(s, strlen(s));
116 }
117
118 void TessResultRenderer::AppendData(const char *s, int len) {
119 if (!tesseract::Serialize(fout_, s, len)) {
120 happy_ = false;
121 }
122 fflush(fout_);
123 }
124
125 bool TessResultRenderer::BeginDocumentHandler() {
126 return happy_;
127 }
128
129 bool TessResultRenderer::EndDocumentHandler() {
130 return happy_;
131 }
132
133 /**********************************************************************
134 * UTF8 Text Renderer interface implementation
135 **********************************************************************/
136 TessTextRenderer::TessTextRenderer(const char *outputbase)
137 : TessResultRenderer(outputbase, "txt") {}
138
139 bool TessTextRenderer::AddImageHandler(TessBaseAPI *api) {
140 const std::unique_ptr<const char[]> utf8(api->GetUTF8Text());
141 if (utf8 == nullptr) {
142 return false;
143 }
144
145 const char *pageSeparator = api->GetStringVariable("page_separator");
146 if (pageSeparator != nullptr && *pageSeparator != '\0' && imagenum() > 0) {
147 AppendString(pageSeparator);
148 }
149
150 AppendString(utf8.get());
151
152 return true;
153 }
154
155 /**********************************************************************
156 * TSV Text Renderer interface implementation
157 **********************************************************************/
158 TessTsvRenderer::TessTsvRenderer(const char *outputbase) : TessResultRenderer(outputbase, "tsv") {
159 font_info_ = false;
160 }
161
162 TessTsvRenderer::TessTsvRenderer(const char *outputbase, bool font_info)
163 : TessResultRenderer(outputbase, "tsv") {
164 font_info_ = font_info;
165 }
166
167 bool TessTsvRenderer::BeginDocumentHandler() {
168 // Output TSV column headings
169 AppendString(
170 "level\tpage_num\tblock_num\tpar_num\tline_num\tword_"
171 "num\tleft\ttop\twidth\theight\tconf\ttext\n");
172 return true;
173 }
174
175 bool TessTsvRenderer::EndDocumentHandler() {
176 return true;
177 }
178
179 bool TessTsvRenderer::AddImageHandler(TessBaseAPI *api) {
180 const std::unique_ptr<const char[]> tsv(api->GetTSVText(imagenum()));
181 if (tsv == nullptr) {
182 return false;
183 }
184
185 AppendString(tsv.get());
186
187 return true;
188 }
189
190 /**********************************************************************
191 * UNLV Text Renderer interface implementation
192 **********************************************************************/
193 TessUnlvRenderer::TessUnlvRenderer(const char *outputbase)
194 : TessResultRenderer(outputbase, "unlv") {}
195
196 bool TessUnlvRenderer::AddImageHandler(TessBaseAPI *api) {
197 const std::unique_ptr<const char[]> unlv(api->GetUNLVText());
198 if (unlv == nullptr) {
199 return false;
200 }
201
202 AppendString(unlv.get());
203
204 return true;
205 }
206
207 /**********************************************************************
208 * BoxText Renderer interface implementation
209 **********************************************************************/
210 TessBoxTextRenderer::TessBoxTextRenderer(const char *outputbase)
211 : TessResultRenderer(outputbase, "box") {}
212
213 bool TessBoxTextRenderer::AddImageHandler(TessBaseAPI *api) {
214 const std::unique_ptr<const char[]> text(api->GetBoxText(imagenum()));
215 if (text == nullptr) {
216 return false;
217 }
218
219 AppendString(text.get());
220
221 return true;
222 }
223
224 #ifndef DISABLED_LEGACY_ENGINE
225
226 /**********************************************************************
227 * Osd Text Renderer interface implementation
228 **********************************************************************/
229 TessOsdRenderer::TessOsdRenderer(const char *outputbase) : TessResultRenderer(outputbase, "osd") {}
230
231 bool TessOsdRenderer::AddImageHandler(TessBaseAPI *api) {
232 const std::unique_ptr<const char[]> osd(api->GetOsdText(imagenum()));
233 if (osd == nullptr) {
234 return false;
235 }
236
237 AppendString(osd.get());
238
239 return true;
240 }
241
242 #endif // ndef DISABLED_LEGACY_ENGINE
243
244 } // namespace tesseract