Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.11/site-packages/PIL/PdfImagePlugin.py: 10%

1# 

2# The Python Imaging Library. 

3# $Id$ 

4# 

5# PDF (Acrobat) file handling 

6# 

7# History: 

8# 1996-07-16 fl Created 

9# 1997-01-18 fl Fixed header 

10# 2004-02-21 fl Fixes for 1/L/CMYK images, etc. 

11# 2004-02-24 fl Fixes for 1 and P images. 

12# 

13# Copyright (c) 1997-2004 by Secret Labs AB. All rights reserved. 

14# Copyright (c) 1996-1997 by Fredrik Lundh. 

15# 

16# See the README file for information on usage and redistribution. 

17# 

18 

19## 

20# Image plugin for PDF images (output only). 

21## 

22from __future__ import annotations 

23 

24import io 

25import math 

26import os 

27import time 

28from typing import IO, Any 

29 

30from . import Image, ImageFile, ImageSequence, PdfParser, __version__, features 

31 

32# 

33# -------------------------------------------------------------------- 

34 

35# object ids: 

36# 1. catalogue 

37# 2. pages 

38# 3. image 

39# 4. page 

40# 5. page contents 

41 

42 

43def _save_all(im: Image.Image, fp: IO[bytes], filename: str | bytes) -> None: 

44 _save(im, fp, filename, save_all=True) 

45 

46 

47## 

48# (Internal) Image save plugin for the PDF format. 

49 

50 

51def _write_image( 

52 im: Image.Image, 

53 filename: str | bytes, 

54 existing_pdf: PdfParser.PdfParser, 

55 image_refs: list[PdfParser.IndirectReference], 

56) -> tuple[PdfParser.IndirectReference, str]: 

57 # FIXME: Should replace ASCIIHexDecode with RunLengthDecode 

58 # (packbits) or LZWDecode (tiff/lzw compression). Note that 

59 # PDF 1.2 also supports Flatedecode (zip compression). 

60 

61 params = None 

62 decode = None 

63 

64 # 

65 # Get image characteristics 

66 

67 width, height = im.size 

68 

69 dict_obj: dict[str, Any] = {"BitsPerComponent": 8} 

70 if im.mode == "1": 

71 if features.check("libtiff"): 

72 decode_filter = "CCITTFaxDecode" 

73 dict_obj["BitsPerComponent"] = 1 

74 params = PdfParser.PdfArray( 

75 [ 

76 PdfParser.PdfDict( 

77 { 

78 "K": -1, 

79 "BlackIs1": True, 

80 "Columns": width, 

81 "Rows": height, 

82 } 

83 ) 

84 ] 

85 ) 

86 else: 

87 decode_filter = "DCTDecode" 

88 dict_obj["ColorSpace"] = PdfParser.PdfName("DeviceGray") 

89 procset = "ImageB" # grayscale 

90 elif im.mode == "L": 

91 decode_filter = "DCTDecode" 

92 # params = f"<< /Predictor 15 /Columns {width-2} >>" 

93 dict_obj["ColorSpace"] = PdfParser.PdfName("DeviceGray") 

94 procset = "ImageB" # grayscale 

95 elif im.mode == "LA": 

96 decode_filter = "JPXDecode" 

97 # params = f"<< /Predictor 15 /Columns {width-2} >>" 

98 procset = "ImageB" # grayscale 

99 dict_obj["SMaskInData"] = 1 

100 elif im.mode == "P": 

101 decode_filter = "ASCIIHexDecode" 

102 palette = im.getpalette() 

103 assert palette is not None 

104 dict_obj["ColorSpace"] = [ 

105 PdfParser.PdfName("Indexed"), 

106 PdfParser.PdfName("DeviceRGB"), 

107 len(palette) // 3 - 1, 

108 PdfParser.PdfBinary(palette), 

109 ] 

110 procset = "ImageI" # indexed color 

111 

112 if "transparency" in im.info: 

113 smask = im.convert("LA").getchannel("A") 

114 smask.encoderinfo = {} 

115 

116 image_ref = _write_image(smask, filename, existing_pdf, image_refs)[0] 

117 dict_obj["SMask"] = image_ref 

118 elif im.mode == "RGB": 

119 decode_filter = "DCTDecode" 

120 dict_obj["ColorSpace"] = PdfParser.PdfName("DeviceRGB") 

121 procset = "ImageC" # color images 

122 elif im.mode == "RGBA": 

123 decode_filter = "JPXDecode" 

124 procset = "ImageC" # color images 

125 dict_obj["SMaskInData"] = 1 

126 elif im.mode == "CMYK": 

127 decode_filter = "DCTDecode" 

128 dict_obj["ColorSpace"] = PdfParser.PdfName("DeviceCMYK") 

129 procset = "ImageC" # color images 

130 decode = [1, 0, 1, 0, 1, 0, 1, 0] 

131 else: 

132 msg = f"cannot save mode {im.mode}" 

133 raise ValueError(msg) 

134 

135 # 

136 # image 

137 

138 op = io.BytesIO() 

139 

140 if decode_filter == "ASCIIHexDecode": 

141 ImageFile._save(im, op, [ImageFile._Tile("hex", (0, 0) + im.size, 0, im.mode)]) 

142 elif decode_filter == "CCITTFaxDecode": 

143 im.save( 

144 op, 

145 "TIFF", 

146 compression="group4", 

147 # use a single strip 

148 strip_size=math.ceil(width / 8) * height, 

149 ) 

150 elif decode_filter == "DCTDecode": 

151 Image.SAVE["JPEG"](im, op, filename) 

152 elif decode_filter == "JPXDecode": 

153 del dict_obj["BitsPerComponent"] 

154 Image.SAVE["JPEG2000"](im, op, filename) 

155 else: 

156 msg = f"unsupported PDF filter ({decode_filter})" 

157 raise ValueError(msg) 

158 

159 stream = op.getvalue() 

160 filter: PdfParser.PdfArray | PdfParser.PdfName 

161 if decode_filter == "CCITTFaxDecode": 

162 stream = stream[8:] 

163 filter = PdfParser.PdfArray([PdfParser.PdfName(decode_filter)]) 

164 else: 

165 filter = PdfParser.PdfName(decode_filter) 

166 

167 image_ref = image_refs.pop(0) 

168 existing_pdf.write_obj( 

169 image_ref, 

170 stream=stream, 

171 Type=PdfParser.PdfName("XObject"), 

172 Subtype=PdfParser.PdfName("Image"), 

173 Width=width, # * 72.0 / x_resolution, 

174 Height=height, # * 72.0 / y_resolution, 

175 Filter=filter, 

176 Decode=decode, 

177 DecodeParms=params, 

178 **dict_obj, 

179 ) 

180 

181 return image_ref, procset 

182 

183 

184def _save( 

185 im: Image.Image, fp: IO[bytes], filename: str | bytes, save_all: bool = False 

186) -> None: 

187 is_appending = im.encoderinfo.get("append", False) 

188 filename_str = filename.decode() if isinstance(filename, bytes) else filename 

189 if is_appending: 

190 existing_pdf = PdfParser.PdfParser(f=fp, filename=filename_str, mode="r+b") 

191 else: 

192 existing_pdf = PdfParser.PdfParser(f=fp, filename=filename_str, mode="w+b") 

193 

194 dpi = im.encoderinfo.get("dpi") 

195 if dpi: 

196 x_resolution = dpi[0] 

197 y_resolution = dpi[1] 

198 else: 

199 x_resolution = y_resolution = im.encoderinfo.get("resolution", 72.0) 

200 

201 info = { 

202 "title": ( 

203 None if is_appending else os.path.splitext(os.path.basename(filename))[0] 

204 ), 

205 "author": None, 

206 "subject": None, 

207 "keywords": None, 

208 "creator": None, 

209 "producer": None, 

210 "creationDate": None if is_appending else time.gmtime(), 

211 "modDate": None if is_appending else time.gmtime(), 

212 } 

213 for k, default in info.items(): 

214 v = im.encoderinfo.get(k) if k in im.encoderinfo else default 

215 if v: 

216 existing_pdf.info[k[0].upper() + k[1:]] = v 

217 

218 # 

219 # make sure image data is available 

220 im.load() 

221 

222 existing_pdf.start_writing() 

223 existing_pdf.write_header() 

224 existing_pdf.write_comment(f"created by Pillow {__version__} PDF driver") 

225 

226 # 

227 # pages 

228 ims = [im] 

229 if save_all: 

230 append_images = im.encoderinfo.get("append_images", []) 

231 for append_im in append_images: 

232 append_im.encoderinfo = im.encoderinfo.copy() 

233 ims.append(append_im) 

234 number_of_pages = 0 

235 image_refs = [] 

236 page_refs = [] 

237 contents_refs = [] 

238 for im in ims: 

239 im_number_of_pages = 1 

240 if save_all: 

241 im_number_of_pages = getattr(im, "n_frames", 1) 

242 number_of_pages += im_number_of_pages 

243 for i in range(im_number_of_pages): 

244 image_refs.append(existing_pdf.next_object_id(0)) 

245 if im.mode == "P" and "transparency" in im.info: 

246 image_refs.append(existing_pdf.next_object_id(0)) 

247 

248 page_refs.append(existing_pdf.next_object_id(0)) 

249 contents_refs.append(existing_pdf.next_object_id(0)) 

250 existing_pdf.pages.append(page_refs[-1]) 

251 

252 # 

253 # catalog and list of pages 

254 existing_pdf.write_catalog() 

255 

256 page_number = 0 

257 for im_sequence in ims: 

258 im_pages: ImageSequence.Iterator | list[Image.Image] = ( 

259 ImageSequence.Iterator(im_sequence) if save_all else [im_sequence] 

260 ) 

261 for im in im_pages: 

262 image_ref, procset = _write_image(im, filename, existing_pdf, image_refs) 

263 

264 # 

265 # page 

266 

267 existing_pdf.write_page( 

268 page_refs[page_number], 

269 Resources=PdfParser.PdfDict( 

270 ProcSet=[PdfParser.PdfName("PDF"), PdfParser.PdfName(procset)], 

271 XObject=PdfParser.PdfDict(image=image_ref), 

272 ), 

273 MediaBox=[ 

274 0, 

275 0, 

276 im.width * 72.0 / x_resolution, 

277 im.height * 72.0 / y_resolution, 

278 ], 

279 Contents=contents_refs[page_number], 

280 ) 

281 

282 # 

283 # page contents 

284 

285 page_contents = b"q %f 0 0 %f 0 0 cm /image Do Q\n" % ( 

286 im.width * 72.0 / x_resolution, 

287 im.height * 72.0 / y_resolution, 

288 ) 

289 

290 existing_pdf.write_obj(contents_refs[page_number], stream=page_contents) 

291 

292 page_number += 1 

293 

294 # 

295 # trailer 

296 existing_pdf.write_xref_and_trailer() 

297 if hasattr(fp, "flush"): 

298 fp.flush() 

299 existing_pdf.close() 

300 

301 

302# 

303# -------------------------------------------------------------------- 

304 

305 

306Image.register_save("PDF", _save) 

307Image.register_save_all("PDF", _save_all) 

308 

309Image.register_extension("PDF", ".pdf") 

310 

311Image.register_mime("PDF", "application/pdf")