mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-12-10 05:38:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(tesseract): initialize df_osd to avoid uninitialized variable error (#1718)

Browse Source 
* fix: initialize df_osd to avoid uninitialized variable error

Signed-off-by: IoannisMaras <maras2002@gmail.com>

* Fix formatting

Signed-off-by: Christoph Auer <60343111+cau-git@users.noreply.github.com>

* Satisfy mypy, regenerate OCR tests

Signed-off-by: Christoph Auer <cau@zurich.ibm.com>

---------

Signed-off-by: IoannisMaras <maras2002@gmail.com>
Signed-off-by: Christoph Auer <60343111+cau-git@users.noreply.github.com>
Signed-off-by: Christoph Auer <cau@zurich.ibm.com>
Co-authored-by: Christoph Auer <60343111+cau-git@users.noreply.github.com>
Co-authored-by: Christoph Auer <cau@zurich.ibm.com>
This commit is contained in:
Maras Ioannis
2025-06-10 11:57:45 +03:00
committed by GitHub
parent f7f31137f1
commit e979750ce9
20 changed files with 752 additions and 751 deletions
Download Patch File Download Diff File Expand all files Collapse all files

146
tests/data_scanned/groundtruth/docling_v1/ocr_test.pages.json vendored
View File

@@ -40,14 +40,14 @@
"a": 255
},
"rect": {
"r_x0": 70.90211866351085,
"r_y0": 124.83139551297342,
"r_x1": 504.8720079864275,
"r_y1": 124.83139551297342,
"r_x2": 504.8720079864275,
"r_y2": 102.66666671251768,
"r_x3": 70.90211866351085,
"r_y3": 102.66666671251768,
"r_x0": 69.6796630536824,
"r_y0": 124.83139494707741,
"r_x1": 504.8720051760782,
"r_y1": 124.83139494707741,
"r_x2": 504.8720051760782,
"r_y2": 104.00000011573796,
"r_x3": 69.6796630536824,
"r_y3": 104.00000011573796,
"coord_origin": "TOPLEFT"
},
"text": "JSON and Markdown in an easy self contained",
@@ -65,14 +65,14 @@
"a": 255
},
"rect": {
"r_x0": 73.10852522817731,
"r_y0": 152.70503335218433,
"r_x1": 153.04479435252625,
"r_y1": 152.70503335218433,
"r_x2": 153.04479435252625,
"r_y2": 130.00136157890958,
"r_x3": 73.10852522817731,
"r_y3": 130.00136157890958,
"r_x0": 71.84193505100733,
"r_y0": 152.90926970226084,
"r_x1": 153.088934155825,
"r_y1": 152.90926970226084,
"r_x2": 153.088934155825,
"r_y2": 129.797125232046,
"r_x3": 71.84193505100733,
"r_y3": 129.797125232046,
"coord_origin": "TOPLEFT"
},
"text": "package",
@@ -90,10 +90,10 @@
"id": 0,
"label": "text",
"bbox": {
"l": 70.90211866351085,
"l": 69.6796630536824,
"t": 76.99999977896756,
"r": 504.8720079864275,
"b": 152.70503335218433,
"r": 504.8720051760782,
"b": 152.90926970226084,
"coord_origin": "TOPLEFT"
},
"confidence": 0.9715733528137207,
@@ -132,14 +132,14 @@
"a": 255
},
"rect": {
"r_x0": 70.90211866351085,
"r_y0": 124.83139551297342,
"r_x1": 504.8720079864275,
"r_y1": 124.83139551297342,
"r_x2": 504.8720079864275,
"r_y2": 102.66666671251768,
"r_x3": 70.90211866351085,
"r_y3": 102.66666671251768,
"r_x0": 69.6796630536824,
"r_y0": 124.83139494707741,
"r_x1": 504.8720051760782,
"r_y1": 124.83139494707741,
"r_x2": 504.8720051760782,
"r_y2": 104.00000011573796,
"r_x3": 69.6796630536824,
"r_y3": 104.00000011573796,
"coord_origin": "TOPLEFT"
},
"text": "JSON and Markdown in an easy self contained",
@@ -157,14 +157,14 @@
"a": 255
},
"rect": {
"r_x0": 73.10852522817731,
"r_y0": 152.70503335218433,
"r_x1": 153.04479435252625,
"r_y1": 152.70503335218433,
"r_x2": 153.04479435252625,
"r_y2": 130.00136157890958,
"r_x3": 73.10852522817731,
"r_y3": 130.00136157890958,
"r_x0": 71.84193505100733,
"r_y0": 152.90926970226084,
"r_x1": 153.088934155825,
"r_y1": 152.90926970226084,
"r_x2": 153.088934155825,
"r_y2": 129.797125232046,
"r_x3": 71.84193505100733,
"r_y3": 129.797125232046,
"coord_origin": "TOPLEFT"
},
"text": "package",
@@ -195,10 +195,10 @@
"id": 0,
"label": "text",
"bbox": {
"l": 70.90211866351085,
"l": 69.6796630536824,
"t": 76.99999977896756,
"r": 504.8720079864275,
"b": 152.70503335218433,
"r": 504.8720051760782,
"b": 152.90926970226084,
"coord_origin": "TOPLEFT"
},
"confidence": 0.9715733528137207,
@@ -237,14 +237,14 @@
"a": 255
},
"rect": {
"r_x0": 70.90211866351085,
"r_y0": 124.83139551297342,
"r_x1": 504.8720079864275,
"r_y1": 124.83139551297342,
"r_x2": 504.8720079864275,
"r_y2": 102.66666671251768,
"r_x3": 70.90211866351085,
"r_y3": 102.66666671251768,
"r_x0": 69.6796630536824,
"r_y0": 124.83139494707741,
"r_x1": 504.8720051760782,
"r_y1": 124.83139494707741,
"r_x2": 504.8720051760782,
"r_y2": 104.00000011573796,
"r_x3": 69.6796630536824,
"r_y3": 104.00000011573796,
"coord_origin": "TOPLEFT"
},
"text": "JSON and Markdown in an easy self contained",
@@ -262,14 +262,14 @@
"a": 255
},
"rect": {
"r_x0": 73.10852522817731,
"r_y0": 152.70503335218433,
"r_x1": 153.04479435252625,
"r_y1": 152.70503335218433,
"r_x2": 153.04479435252625,
"r_y2": 130.00136157890958,
"r_x3": 73.10852522817731,
"r_y3": 130.00136157890958,
"r_x0": 71.84193505100733,
"r_y0": 152.90926970226084,
"r_x1": 153.088934155825,
"r_y1": 152.90926970226084,
"r_x2": 153.088934155825,
"r_y2": 129.797125232046,
"r_x3": 71.84193505100733,
"r_y3": 129.797125232046,
"coord_origin": "TOPLEFT"
},
"text": "package",
@@ -293,10 +293,10 @@
"id": 0,
"label": "text",
"bbox": {
"l": 70.90211866351085,
"l": 69.6796630536824,
"t": 76.99999977896756,
"r": 504.8720079864275,
"b": 152.70503335218433,
"r": 504.8720051760782,
"b": 152.90926970226084,
"coord_origin": "TOPLEFT"
},
"confidence": 0.9715733528137207,
@@ -335,14 +335,14 @@
"a": 255
},
"rect": {
"r_x0": 70.90211866351085,
"r_y0": 124.83139551297342,
"r_x1": 504.8720079864275,
"r_y1": 124.83139551297342,
"r_x2": 504.8720079864275,
"r_y2": 102.66666671251768,
"r_x3": 70.90211866351085,
"r_y3": 102.66666671251768,
"r_x0": 69.6796630536824,
"r_y0": 124.83139494707741,
"r_x1": 504.8720051760782,
"r_y1": 124.83139494707741,
"r_x2": 504.8720051760782,
"r_y2": 104.00000011573796,
"r_x3": 69.6796630536824,
"r_y3": 104.00000011573796,
"coord_origin": "TOPLEFT"
},
"text": "JSON and Markdown in an easy self contained",
@@ -360,14 +360,14 @@
"a": 255
},
"rect": {
"r_x0": 73.10852522817731,
"r_y0": 152.70503335218433,
"r_x1": 153.04479435252625,
"r_y1": 152.70503335218433,
"r_x2": 153.04479435252625,
"r_y2": 130.00136157890958,
"r_x3": 73.10852522817731,
"r_y3": 130.00136157890958,
"r_x0": 71.84193505100733,
"r_y0": 152.90926970226084,
"r_x1": 153.088934155825,
"r_y1": 152.90926970226084,
"r_x2": 153.088934155825,
"r_y2": 129.797125232046,
"r_x3": 71.84193505100733,
"r_y3": 129.797125232046,
"coord_origin": "TOPLEFT"
},
"text": "package",