diff --git a/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/pdf.bin b/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/pdf.bin new file mode 100644 index 00000000..05592c58 Binary files /dev/null and b/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/pdf.bin differ diff --git a/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/stderr.bin b/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/stderr.bin new file mode 100644 index 00000000..61f78d82 --- /dev/null +++ b/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/stderr.bin @@ -0,0 +1 @@ +Tesseract Open Source OCR Engine v4.0.0 with Leptonica diff --git a/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/stdout.bin b/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/stdout.bin new file mode 100644 index 00000000..e69de29b diff --git a/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/txt.bin b/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/txt.bin new file mode 100644 index 00000000..080288f6 --- /dev/null +++ b/tests/cache/francais/__-l__deu__000001.ocr.png__000001.text__pdf__txt/txt.bin @@ -0,0 +1,13 @@ +Portez ce vieux whisky au juge +blond qui fume sur son ile +interieure, a cöte de l'alcöve +ovolde, oU les büches se +consument dans l'ätre, ce qui +lui permet de penser ä la +cgnogenese de l'&tre dont il +est question dans la cause +ambigu6 entendue ä Moy, dans +un capharnaüÜm qui, pense-t-il, +diminue ca et 13 la qualite de son +ceuvre. + \ No newline at end of file diff --git a/tests/cache/manifest.jsonl b/tests/cache/manifest.jsonl index afa91683..6ec2a17e 100644 --- a/tests/cache/manifest.jsonl +++ b/tests/cache/manifest.jsonl @@ -44,3 +44,4 @@ {"tesseract_version": "tesseract 4.0.0 leptonica-1.77.0 libjpeg 9c : libpng 1.6.35 : libtiff 4.0.10 : zlib 1.2.11 : libopenjp2 2.3.0 Found AVX2 Found AVX Found SSE ", "platform": "Darwin-18.0.0-x86_64-i386-64bit", "python": "3.7.1", "argv_slug": "__-l__eng__000004.ocr.png__000004__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000004.ocr.png", "$TMPDIR/000004", "hocr", "txt"]} {"tesseract_version": "tesseract 4.0.0 leptonica-1.77.0 libjpeg 9c : libpng 1.6.35 : libtiff 4.0.10 : zlib 1.2.11 : libopenjp2 2.3.0 Found AVX2 Found AVX Found SSE ", "platform": "Darwin-18.0.0-x86_64-i386-64bit", "python": "3.7.1", "argv_slug": "__-l__eng__000003.ocr.png__000003__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000003.ocr.png", "$TMPDIR/000003", "hocr", "txt"]} {"tesseract_version": "tesseract 4.0.0 leptonica-1.77.0 libjpeg 9c : libpng 1.6.35 : libtiff 4.0.10 : zlib 1.2.11 : libopenjp2 2.3.0 Found AVX2 Found AVX Found SSE ", "platform": "Darwin-18.0.0-x86_64-i386-64bit", "python": "3.7.1", "argv_slug": "__-l__eng__000002.ocr.png__000002__hocr__txt", "sourcefile": "resources/cardinal.pdf", "args": ["-l", "eng", "$TMPDIR/000002.ocr.png", "$TMPDIR/000002", "hocr", "txt"]} +{"tesseract_version": "tesseract 4.0.0 leptonica-1.77.0 libgif 5.1.4 : libjpeg 9c : libpng 1.6.36 : libtiff 4.0.10 : zlib 1.2.11 : libwebp 1.0.2 : libopenjp2 2.3.0 Found AVX2 Found AVX Found SSE ", "platform": "Darwin-18.2.0-x86_64-i386-64bit", "python": "3.7.2", "argv_slug": "__-l__deu__000001.ocr.png__000001.text__pdf__txt", "sourcefile": "resources/francais.pdf", "args": ["-l", "deu", "-c", "textonly_pdf=1", "$TMPDIR/000001.ocr.png", "$TMPDIR/000001.text", "pdf", "txt"]}