viso-sdk-python 1.1.17__tar.gz → 1.1.19__tar.gz

{viso_sdk_python-1.1.17 → viso_sdk_python-1.1.19}/CHANGELOG.md

@@ -1,6 +1,14 @@
 # Changelog
 
 <!--next-version-placeholder-->
+## v1.1.19
+### Fix
+- Fix timestamp error
+- Add missing test files
+
+## v1.1.18
+### Fix
+missing nms script
 
 ## v1.1.17
 

{viso_sdk_python-1.1.17 → viso_sdk_python-1.1.19}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: viso-sdk-python
-Version: 1.1.17
+Version: 1.1.19
 Summary: VisoSDK: A Python SDK for use in Viso containers
 Home-page: https://gitlab.com/TopKamera/03_edge/Base/viso-sdk-python-public.git
 Author: support@viso.ai

{viso_sdk_python-1.1.17 → viso_sdk_python-1.1.19}/tests/unit/test_model_fetch.py

@@ -2,7 +2,6 @@
 """Tests for viso_sdk.model.fetch"""
 import io
 import tarfile
-import tempfile
 from unittest.mock import patch
 
 import pytest

viso_sdk_python-1.1.19/tests/unit/test_nms.py

@@ -0,0 +1,231 @@
+"""Unit tests for NMS (Non-Maximum Suppression) utilities."""
+from viso_sdk.detection.nms import non_max_suppression_cv, non_max_suppression_cv_per_class
+
+
+class TestNonMaxSuppressionCV:
+    """Tests for standard NMS."""
+
+    def test_empty_input_returns_empty(self):
+        """Test that empty boxes list returns empty result."""
+        result = non_max_suppression_cv([], [], score_threshold=0.5, iou_threshold=0.5)
+        assert result == []
+
+    def test_single_box_is_kept(self):
+        """Test that a single box is always kept."""
+        boxes = [[10, 10, 50, 50]]
+        scores = [0.9]
+        result = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.5)
+        assert result == [0]
+
+    def test_non_overlapping_boxes_all_kept(self):
+        """Test that non-overlapping boxes are all kept."""
+        boxes = [
+            [10, 10, 30, 30],    # Box 0
+            [100, 100, 30, 30],  # Box 1 - far away
+            [200, 200, 30, 30],  # Box 2 - far away
+        ]
+        scores = [0.9, 0.85, 0.8]
+        result = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.5)
+        assert len(result) == 3
+        assert set(result) == {0, 1, 2}
+
+    def test_overlapping_boxes_highest_score_kept(self):
+        """Test that overlapping boxes keep only the highest score."""
+        boxes = [
+            [10, 10, 50, 50],   # Box 0 - score 0.9
+            [15, 15, 50, 50],   # Box 1 - score 0.8 (overlaps with 0)
+        ]
+        scores = [0.9, 0.8]
+        result = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.3)
+        assert result == [0]  # Box 0 kept, box 1 suppressed
+
+    def test_score_threshold_filters_low_scores(self):
+        """Test that boxes below score_threshold are filtered out."""
+        boxes = [
+            [10, 10, 50, 50],   # High score
+            [100, 100, 30, 30],  # Low score
+        ]
+        scores = [0.9, 0.3]
+        result = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.5)
+        assert result == [0]
+
+    def test_high_iou_threshold_keeps_more_boxes(self):
+        """Test that higher IoU threshold allows more overlapping boxes."""
+        boxes = [
+            [10, 10, 50, 50],
+            [20, 20, 50, 50],  # Moderate overlap
+        ]
+        scores = [0.9, 0.85]
+        
+        # With low IoU threshold, should suppress
+        result_low = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.1)
+        assert len(result_low) == 1
+        
+        # With high IoU threshold, should keep both
+        result_high = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.9)
+        assert len(result_high) == 2
+
+    def test_three_overlapping_boxes_keeps_best(self):
+        """Test NMS with three overlapping boxes."""
+        boxes = [
+            [10, 10, 50, 50],   # score 0.95
+            [12, 12, 50, 50],   # score 0.9 (high overlap with 0)
+            [14, 14, 50, 50],   # score 0.85 (high overlap with 0 and 1)
+        ]
+        scores = [0.95, 0.9, 0.85]
+        result = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.3)
+        assert result == [0]  # Only highest score kept
+
+    def test_float_boxes_and_scores(self):
+        """Test that float boxes and scores are handled correctly."""
+        boxes = [[10.5, 10.5, 50.3, 50.7]]
+        scores = [0.934]
+        result = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.5)
+        assert result == [0]
+
+    def test_mixed_overlapping_and_non_overlapping(self):
+        """Test mix of overlapping and non-overlapping boxes."""
+        boxes = [
+            [10, 10, 50, 50],    # Group 1 - score 0.9
+            [15, 15, 50, 50],    # Group 1 - score 0.8 (overlaps with 0)
+            [200, 200, 30, 30],  # Group 2 - score 0.95 (isolated)
+        ]
+        scores = [0.9, 0.8, 0.95]
+        result = non_max_suppression_cv(boxes, scores, score_threshold=0.5, iou_threshold=0.3)
+        assert len(result) == 2
+        assert 2 in result  # Isolated box kept
+        assert 0 in result  # Highest score from group 1 kept
+        assert 1 not in result  # Suppressed by box 0
+
+
+class TestNonMaxSuppressionCVPerClass:
+    """Tests for per-class NMS."""
+
+    def test_empty_input_returns_empty(self):
+        """Test that empty inputs return empty result."""
+        result = non_max_suppression_cv_per_class(
+            [], [], [], score_threshold=0.5, iou_threshold=0.5
+        )
+        assert result == []
+
+    def test_single_box_is_kept(self):
+        """Test that a single box is always kept."""
+        boxes = [[10, 10, 50, 50]]
+        scores = [0.9]
+        class_ids = [1]
+        result = non_max_suppression_cv_per_class(
+            boxes, scores, class_ids, score_threshold=0.5, iou_threshold=0.5
+        )
+        assert result == [0]
+
+    def test_different_classes_overlap_both_kept(self):
+        """Test that overlapping boxes of different classes are both kept."""
+        boxes = [
+            [10, 10, 50, 50],   # Class 1
+            [15, 15, 50, 50],   # Class 2 (overlaps with class 1)
+        ]
+        scores = [0.9, 0.85]
+        class_ids = [1, 2]
+        result = non_max_suppression_cv_per_class(
+            boxes, scores, class_ids, score_threshold=0.5, iou_threshold=0.3
+        )
+        assert len(result) == 2
+        assert set(result) == {0, 1}
+
+    def test_same_class_overlap_highest_kept(self):
+        """Test that overlapping boxes of same class keep highest score."""
+        boxes = [
+            [10, 10, 50, 50],   # Class 1 - score 0.9
+            [15, 15, 50, 50],   # Class 1 - score 0.8 (overlaps)
+        ]
+        scores = [0.9, 0.8]
+        class_ids = [1, 1]
+        result = non_max_suppression_cv_per_class(
+            boxes, scores, class_ids, score_threshold=0.5, iou_threshold=0.3
+        )
+        assert result == [0]
+
+    def test_multiple_classes_with_overlap(self):
+        """Test complex scenario with multiple classes and overlaps."""
+        boxes = [
+            [10, 10, 50, 50],   # Class 1 - score 0.9
+            [15, 15, 50, 50],   # Class 1 - score 0.8 (overlaps with 0)
+            [12, 12, 48, 48],   # Class 2 - score 0.85 (overlaps with 0 and 1)
+            [200, 200, 30, 30], # Class 1 - score 0.95 (isolated)
+        ]
+        scores = [0.9, 0.8, 0.85, 0.95]
+        class_ids = [1, 1, 2, 1]
+        result = non_max_suppression_cv_per_class(
+            boxes, scores, class_ids, score_threshold=0.5, iou_threshold=0.3
+        )
+        
+        assert len(result) == 3
+        assert 0 in result  # Class 1, highest in overlapping group
+        assert 2 in result  # Class 2, only box
+        assert 3 in result  # Class 1, isolated
+        assert 1 not in result  # Class 1, suppressed by box 0
+
+    def test_string_class_ids(self):
+        """Test that string class IDs work correctly."""
+        boxes = [
+            [10, 10, 50, 50],   # Class 'person'
+            [15, 15, 50, 50],   # Class 'car' (overlaps)
+        ]
+        scores = [0.9, 0.85]
+        class_ids = ['person', 'car']
+        result = non_max_suppression_cv_per_class(
+            boxes, scores, class_ids, score_threshold=0.5, iou_threshold=0.3
+        )
+        assert len(result) == 2
+        assert set(result) == {0, 1}
+
+    def test_three_classes_no_suppression(self):
+        """Test that different classes at same location are all kept."""
+        boxes = [
+            [10, 10, 50, 50],   # Class 1
+            [10, 10, 50, 50],   # Class 2 (exact same location)
+            [10, 10, 50, 50],   # Class 3 (exact same location)
+        ]
+        scores = [0.9, 0.85, 0.8]
+        class_ids = [1, 2, 3]
+        result = non_max_suppression_cv_per_class(
+            boxes, scores, class_ids, score_threshold=0.5, iou_threshold=0.3
+        )
+        assert len(result) == 3
+        assert set(result) == {0, 1, 2}
+
+    def test_score_threshold_filters_per_class(self):
+        """Test that score threshold is applied per class."""
+        boxes = [
+            [10, 10, 50, 50],    # Class 1 - high score
+            [100, 100, 30, 30],  # Class 1 - low score
+            [200, 200, 30, 30],  # Class 2 - low score
+        ]
+        scores = [0.9, 0.3, 0.4]
+        class_ids = [1, 1, 2]
+        result = non_max_suppression_cv_per_class(
+            boxes, scores, class_ids, score_threshold=0.5, iou_threshold=0.5
+        )
+        assert result == [0]  # Only box 0 passes score threshold
+
+    def test_per_class_nms_order_independence(self):
+        """Test that result is independent of input order."""
+        boxes = [
+            [10, 10, 50, 50],   # Class 1 - score 0.8
+            [15, 15, 50, 50],   # Class 2 - score 0.9
+            [12, 12, 48, 48],   # Class 1 - score 0.95 (overlaps with box 0)
+        ]
+        scores = [0.8, 0.9, 0.95]
+        class_ids = [1, 2, 1]
+        
+        result = non_max_suppression_cv_per_class(
+            boxes, scores, class_ids, score_threshold=0.5, iou_threshold=0.3
+        )
+        
+        # Box 2 (class 1, score 0.95) should be kept
+        # Box 0 (class 1, score 0.8) should be suppressed
+        # Box 1 (class 2, score 0.9) should be kept
+        assert len(result) == 2
+        assert 1 in result  # Class 2
+        assert 2 in result  # Class 1 with higher score
+        assert 0 not in result  # Class 1 with lower score suppressed

viso_sdk_python-1.1.19/tests/unit/test_object.py

@@ -0,0 +1,189 @@
+"""Unit tests for Detection class and object detection utilities."""
+import numpy as np
+import pytest
+
+from viso_sdk.detection.object import Detection, filter_iou_threshold, filter_labels
+from viso_sdk.constants import KEY
+
+
+class TestDetectionInit:
+    def test_init_with_direct_params(self):
+        det = Detection(class_id=1, label="car", score=0.95, tlwh=[0.1, 0.2, 0.3, 0.4])
+        assert det.class_id == 1
+        assert det.label == "car"
+        assert det.score == 0.95
+        assert np.allclose(det.tlwh, [0.1, 0.2, 0.3, 0.4])
+
+    def test_init_with_dict(self):
+        det_dict = {
+            KEY.CLASS_ID: 2,
+            KEY.LABEL: "person",
+            KEY.SCORE: 0.88,
+            KEY.TLWH: [0.2, 0.3, 0.4, 0.5],
+        }
+        det = Detection(det=det_dict)
+        assert det.class_id == 2
+        assert det.label == "person"
+        assert det.score == 0.88
+        assert np.allclose(det.tlwh, [0.2, 0.3, 0.4, 0.5])
+
+    def test_init_with_roi_id_and_name(self):
+        det = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.2, 0.3, 0.4],
+                       roi_id=[1, 2], roi_name=["zone_a", "zone_b"])
+        assert det.roi_id == [1, 2]
+        assert det.roi_name == ["zone_a", "zone_b"]
+
+    def test_init_converts_single_roi_to_list(self):
+        det = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.2, 0.3, 0.4],
+                       roi_id=5, roi_name="parking")
+        assert det.roi_id == [5]
+        assert det.roi_name == ["parking"]
+
+    def test_init_defaults_none_roi_to_empty_list(self):
+        det = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.2, 0.3, 0.4])
+        assert det.roi_id == []
+        assert det.roi_name == []
+
+    def test_init_with_default_values(self):
+        det = Detection(tlwh=[0.1, 0.2, 0.3, 0.4])
+        assert det.class_id == -1
+        assert det.label == ""
+        assert det.score == 0.0
+
+
+class TestDetectionCoordinateConversion:
+    def test_normalized_coordinates_stay_normalized(self):
+        # When tlwh has values < 1.0, treat as normalized
+        det = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.2, 0.3, 0.4])
+        assert np.allclose(det.tlwh, [0.1, 0.2, 0.3, 0.4])
+
+    def test_pixel_coordinates_converted_to_normalized(self):
+        # When tlwh has values >= 1.0, treat as pixels and normalize
+        det = Detection(class_id=1, label="car", score=0.9, 
+                       tlwh=[100, 200, 300, 400], frame_sz=(1000, 1000))
+        expected = [0.1, 0.2, 0.3, 0.4]
+        assert np.allclose(det.tlwh, expected)
+
+    def test_pixel_coords_clamped_to_frame_bounds(self):
+        # Coordinates exceeding frame bounds should be clamped
+        det = Detection(class_id=1, label="car", score=0.9,
+                       tlwh=[900, 900, 200, 200], frame_sz=(1000, 1000))
+        # x + w = 1100 > 1000, so xmax should be clamped to 1.0
+        # Final normalized: x=0.9, y=0.9, w=0.1, h=0.1
+        assert det.tlwh[0] == 0.9
+        assert det.tlwh[1] == 0.9
+        assert det.tlwh[2] <= 0.1
+        assert det.tlwh[3] <= 0.1
+
+    def test_negative_coordinates_clamped_to_zero(self):
+        det = Detection(class_id=1, label="car", score=0.9,
+                       tlwh=[-10, -20, 100, 100], frame_sz=(1000, 1000))
+        # Negative coords should be clamped to 0
+        assert det.tlwh[0] >= 0.0
+        assert det.tlwh[1] >= 0.0
+
+
+class TestDetectionConversionMethods:
+    def test_to_tlbr(self):
+        det = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.2, 0.3, 0.4])
+        tlbr = det.to_tlbr()
+        # tlbr = [xmin, ymin, xmax, ymax] = [0.1, 0.2, 0.4, 0.6]
+        assert np.allclose(tlbr, [0.1, 0.2, 0.4, 0.6])
+
+    def test_to_xyah(self):
+        det = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.2, 0.4, 0.2])
+        xyah = det.to_xyah()
+        # center_x = 0.1 + 0.4/2 = 0.3
+        # center_y = 0.2 + 0.2/2 = 0.3
+        # aspect = 0.4 / 0.2 = 2.0
+        # height = 0.2
+        assert np.allclose(xyah, [0.3, 0.3, 2.0, 0.2])
+
+    def test_to_dict(self):
+        det = Detection(class_id=3, label="truck", score=0.876, 
+                       tlwh=[0.123, 0.456, 0.789, 0.234],
+                       roi_id=[1], roi_name=["zone1"])
+        d = det.to_dict()
+        assert d[KEY.CLASS_ID] == 3
+        assert d[KEY.LABEL] == "truck"
+        assert d[KEY.SCORE] == 0.88  # Rounded to 2 decimals
+        assert d[KEY.ROI_ID] == [1]
+        assert d[KEY.ROI_NAME] == ["zone1"]
+        # TLWH rounded to 3 decimals
+        assert len(d[KEY.TLWH]) == 4
+        assert isinstance(d[KEY.TLWH], list)
+
+
+class TestFilterIouThreshold:
+    def test_filters_overlapping_detections(self):
+        # Create overlapping detections with different scores
+        det1 = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.1, 0.3, 0.3])
+        det2 = Detection(class_id=1, label="car", score=0.7, tlwh=[0.15, 0.15, 0.3, 0.3])  # Overlaps with det1
+        det3 = Detection(class_id=1, label="car", score=0.95, tlwh=[0.6, 0.6, 0.2, 0.2])  # Separate
+        
+        detections = [det1, det2, det3]
+        filtered = filter_iou_threshold(detections, score_threshold=0.5, iou_threshold=0.3, img_sz=(1000, 1000))
+        
+        # Should keep det1 (higher score than det2) and det3 (non-overlapping)
+        assert len(filtered) >= 1
+        assert det3 in filtered  # Non-overlapping detection should remain
+
+    def test_filters_by_score_threshold(self):
+        det1 = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.1, 0.3, 0.3])
+        det2 = Detection(class_id=1, label="car", score=0.4, tlwh=[0.6, 0.6, 0.2, 0.2])  # Below threshold
+        
+        detections = [det1, det2]
+        filtered = filter_iou_threshold(detections, score_threshold=0.5, iou_threshold=0.5, img_sz=(1000, 1000))
+        
+        # Only det1 should pass score threshold
+        assert len(filtered) == 1
+        assert filtered[0].score == 0.9
+
+    def test_empty_detections_returns_empty(self):
+        filtered = filter_iou_threshold([], score_threshold=0.5, iou_threshold=0.5, img_sz=(1000, 1000))
+        assert filtered == []
+
+
+class TestFilterLabels:
+    def test_filters_by_label_list(self):
+        det1 = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.1, 0.3, 0.3])
+        det2 = Detection(class_id=2, label="person", score=0.8, tlwh=[0.5, 0.5, 0.2, 0.2])
+        det3 = Detection(class_id=3, label="truck", score=0.85, tlwh=[0.7, 0.7, 0.2, 0.2])
+        
+        detections = [det1, det2, det3]
+        filtered = filter_labels(detections, labels_to_detect=["car", "truck"])
+        
+        assert len(filtered) == 2
+        labels = [d.label for d in filtered]
+        assert "car" in labels
+        assert "truck" in labels
+        assert "person" not in labels
+
+    def test_none_labels_returns_all(self):
+        det1 = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.1, 0.3, 0.3])
+        det2 = Detection(class_id=2, label="person", score=0.8, tlwh=[0.5, 0.5, 0.2, 0.2])
+        
+        detections = [det1, det2]
+        filtered = filter_labels(detections, labels_to_detect=None)
+        
+        assert len(filtered) == 2
+        assert filtered == detections
+
+    def test_empty_labels_returns_all(self):
+        det1 = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.1, 0.3, 0.3])
+        det2 = Detection(class_id=2, label="person", score=0.8, tlwh=[0.5, 0.5, 0.2, 0.2])
+        
+        detections = [det1, det2]
+        filtered = filter_labels(detections, labels_to_detect=[])
+        
+        assert len(filtered) == 2
+        assert filtered == detections
+
+    def test_no_matching_labels_returns_empty(self):
+        det1 = Detection(class_id=1, label="car", score=0.9, tlwh=[0.1, 0.1, 0.3, 0.3])
+        det2 = Detection(class_id=2, label="person", score=0.8, tlwh=[0.5, 0.5, 0.2, 0.2])
+        
+        detections = [det1, det2]
+        filtered = filter_labels(detections, labels_to_detect=["truck", "bike"])
+        
+        assert len(filtered) == 0

{viso_sdk_python-1.1.17 → viso_sdk_python-1.1.19}/tests/unit/test_stream_cleanup.py

@@ -1,8 +1,6 @@
 # -*- coding: utf-8 -*-
 """Tests for viso_sdk.redis.stream_cleanup"""
-from unittest.mock import MagicMock, patch
-
-import pytest
+from unittest.mock import MagicMock
 
 from viso_sdk.redis.stream_cleanup import StreamCleanupManager, clear_stream_redis_keys
 

{viso_sdk_python-1.1.17 → viso_sdk_python-1.1.19}/tests/unit/test_trace_malloc.py

@@ -1,5 +1,4 @@
 """Unit tests for TraceMallocManager."""
-import pytest
 from viso_sdk.debug import TraceMallocManager
 
 

{viso_sdk_python-1.1.17 → viso_sdk_python-1.1.19}/tests/unit/test_viso_data.py

@@ -2,8 +2,6 @@
 """Tests for viso_sdk.redis.viso_data"""
 import time
 
-import pytest
-
 from viso_sdk.redis.viso_data import Meta, MetaFrameFormat, MetaModule, MetaSource, VisoData
 
 NODE = {"id": "node-1", "type": "object-detection", "name": "OD Node"}
@@ -60,12 +58,15 @@ class TestMetaModule:
 
 
 class TestMeta:
-    def test_to_dict_uses_stored_ts_not_fresh_time(self):
-        # Regression: to_dict() was calling time.time() instead of using self.ts
+    def test_to_dict_uses_current_time(self):
+        # to_dict() should use time.time() to get fresh timestamp
         meta = Meta(cur_node=NODE, feed_idx=0, src_node=NODE, src_port=0, src_name=None)
-        stored_ts = meta.ts
+        time.sleep(0.01)  # Small delay to ensure timestamps differ
+        t_before = time.time()
         d = meta.to_dict()
-        assert d["ts"] == stored_ts
+        t_after = time.time()
+        assert t_before <= d["ts"] <= t_after
+        assert d["ts"] != meta.ts  # Should be different from stored ts
 
     def test_to_dict_has_required_keys(self):
         meta = Meta(cur_node=NODE, feed_idx=0, src_node=NODE, src_port=0, src_name=None)
@@ -79,6 +80,22 @@ class TestMeta:
         t_after = time.time()
         assert t_before <= meta.ts <= t_after
 
+    def test_update_source(self):
+        meta = Meta(cur_node=NODE, feed_idx=0, src_node=NODE, src_port=0, src_name=None)
+        meta.update_source({"frame_id": 100})
+        assert meta.source.frame_id == 100
+
+    def test_update_module(self):
+        meta = Meta(cur_node=NODE, feed_idx=0, src_node=NODE, src_port=0, src_name=None)
+        meta.update_module({"custom_field": "test"})
+        assert meta.module.custom_field == "test"
+
+    def test_update_frame_format(self):
+        meta = Meta(cur_node=NODE, feed_idx=0, src_node=NODE, src_port=0, src_name=None)
+        meta.update_frame_format({"encoder": ".jpg", "with_base64": True})
+        assert meta.frame_format.encoder == ".jpg"
+        assert meta.frame_format.with_base64 is True
+
 
 class TestVisoData:
     def test_create_without_frame_returns_none_frame(self):

{viso_sdk_python-1.1.17 → viso_sdk_python-1.1.19}/viso_sdk/_version.c

@@ -1994,7 +1994,7 @@ static __pyx_mstatetype * const __pyx_mstate_global = &__pyx_mstate_global_stati
 #endif
 /* #### Code section: constant_name_defines ### */
 #define __pyx_kp_u_ __pyx_string_tab[0]
-#define __pyx_kp_u_1_1_17 __pyx_string_tab[1]
+#define __pyx_kp_u_1_1_19 __pyx_string_tab[1]
 #define __pyx_kp_u_Copyright_2020 __pyx_string_tab[2]
 #define __pyx_kp_u_Viso_ai_team __pyx_string_tab[3]
 #define __pyx_kp_u__2 __pyx_string_tab[4]
@@ -2605,16 +2605,16 @@ __Pyx_RefNannySetupContext("PyInit__version", 0);
  * __license__ = "LGPL3"
  * __title__ = "viso-sdk-python"             # <<<<<<<<<<<<<<
  * 
- * __version__ = "1.1.17"
+ * __version__ = "1.1.19"
 */
   if (PyDict_SetItem(__pyx_mstate_global->__pyx_d, __pyx_mstate_global->__pyx_n_u_title, __pyx_mstate_global->__pyx_kp_u_viso_sdk_python) < (0)) __PYX_ERR(0, 15, __pyx_L1_error)
 
   /* "viso_sdk/_version.py":17
  * __title__ = "viso-sdk-python"
  * 
- * __version__ = "1.1.17"             # <<<<<<<<<<<<<<
+ * __version__ = "1.1.19"             # <<<<<<<<<<<<<<
 */
-  if (PyDict_SetItem(__pyx_mstate_global->__pyx_d, __pyx_mstate_global->__pyx_n_u_version, __pyx_mstate_global->__pyx_kp_u_1_1_17) < (0)) __PYX_ERR(0, 17, __pyx_L1_error)
+  if (PyDict_SetItem(__pyx_mstate_global->__pyx_d, __pyx_mstate_global->__pyx_n_u_version, __pyx_mstate_global->__pyx_kp_u_1_1_19) < (0)) __PYX_ERR(0, 17, __pyx_L1_error)
 
   /* "viso_sdk/_version.py":1
  * """             # <<<<<<<<<<<<<<
@@ -2689,7 +2689,7 @@ static int __Pyx_InitConstants(__pyx_mstatetype *__pyx_mstate) {
   {
     const struct { const unsigned int length: 5; } index[] = {{1},{6},{15},{12},{1},{18},{26},{15},{5},{7},{10},{18},{13},{8},{8},{7},{9},{8},{11},{8},{10},{8},{3},{2},{4},{12},{10},{8},{9},{11},{4}};
     #if (CYTHON_COMPRESS_STRINGS) == 2 /* compression: bz2 (242 bytes) */
-const char* const cstring = "BZh91AY&SYI\035\357\304\000\000\007\337\200@\000@\003x\205\350\204A\000\277\357\377\2400\000\315[\014\246\251\261\032\217\005<\240\r\007\244\323@DM6\223LM2\000\000\001\240\324\3656\251\246\241\232\020\304\310\3204\321\241B}\317\023\177kg\260 \035\340 Z\376\rG[']N7\363d\253B\010\313\310Z\267(\347m\323\314\021\232\010\010\024@c\311w\270\016\003\344b\354\037\236\203J\027\241\211a\270\276(\315:U\024$b\337\332]'j\372\314\025\256\316\316e`Tm\0023)E\244\027\351\362\025\rH\213YH\275X\250Gz{\240\367\332\304\326\244\302\251\344\341_\034a\254\230\376\276!0f\271\314\347\255\367\004GrU*\004\201\212\020\326\026\203r\226\221\215=K\030Y%%\037\225\001,<\037\342\356H\247\n\022\t#\275\370\200";
+const char* const cstring = "BZh91AY&SYo\247\030\321\000\000\007\337\200@\000@\003x%\350\204A\000\277\357\377\2400\000\315[\014\246\251\261\032\217\005<\240\r\007\244\323@DM6\223LM2\000\000\001\240\324\3656\251\246\241\232\020\304\310\3204\321\241B}\317\023\177kg\260 \035\340 Z\376\rG[']N7\363d\253B\010\313\310Z\267(\347m\323\314\021\232\010\010\024@c\311w\270\016\003\344b\354\037\236\203J\027\241\211a\270\276(\315:U\024$b\337\332]'j\372\314\025\256\316\316e`Tm\0023)E\244\027\351\362\025\rH\213YH\275X\250Gz{\240\367\332\304\326\244\302\251\344\341_\034a\254\230\376\276!0f\271\314\347\255\367\004GrU*\004\201\212\020\326\026\203r\226\221\215=K\030Y%%\037\225\001,<\037\342\356H\247\n\022\r\364\343\032 ";
     PyObject *data = __Pyx_DecompressString(cstring, 242, 2);
     if (unlikely(!data)) __PYX_ERR(0, 1, __pyx_L1_error)
     const char* const bytes = __Pyx_PyBytes_AsString(data);
@@ -2697,7 +2697,7 @@ const char* const cstring = "BZh91AY&SYI\035\357\304\000\000\007\337\200@\000@\0
     if (likely(bytes)); else { Py_DECREF(data); __PYX_ERR(0, 1, __pyx_L1_error) }
     #endif
     #elif (CYTHON_COMPRESS_STRINGS) != 0 /* compression: zlib (196 bytes) */
-const char* const cstring = "x\3325\216\275n\3030\014\204_\305/`#M\207.A\177\320!K\206N]\tF\242c\"2\351JT\002\277}\3518\006\207;\334\007\362\330\275t>o\337:\315\231/\2035\373\335~\327\376r\321\016\2711\302\361\343\322\005\315\016Q>o\317|\323\257cs`\351u\003\357\213\266%^\333i\266A\345t\3749\275\342\271Lh\003\000V\3172@H,\004,`\031\003\2351\\=\333\036\000\3675C\344\034\321\310x$\267\202#\001\320\210\234\234C\317\211\026M\034H\312\303:\222\207j\254+\\wD\357\372\254\377\253\230\326\260\220E\352\261&o3*K'\030\333\272w\243\\X\375\330L\230\377\0019@m\355";
+const char* const cstring = "x\3325\216\275n\3030\014\204_\305/`#M\247\002A\177\320!K\206N]\tF\242c\"2\351JT\002\277}\3518\006\207;\334\007\362\330\275t>o\337:\315\231/\2035\373\335~\327\376r\321\016\2711\302\361\343\322\005\315\016Q>o\317|\323\257cs`\351u\003\357\213\266%^\333i\266A\345t\3749\275\342\271Lh\003\000V\3172@H,\004,`\031\003\2351\\=\333\036\000\3675C\344\034\321\310x$\267\202#\001\320\210\234\234C\317\211\026M\034H\312\303:\222\207j\254+\\wD\357\372\254\377\253\230\326\260\220E\352\261&o3*K'\030\333\272w\243\\X\375\330L\230\377\001;rm\357";
     PyObject *data = __Pyx_DecompressString(cstring, 196, 1);
     if (unlikely(!data)) __PYX_ERR(0, 1, __pyx_L1_error)
     const char* const bytes = __Pyx_PyBytes_AsString(data);
@@ -2705,7 +2705,7 @@ const char* const cstring = "x\3325\216\275n\3030\014\204_\305/`#M\207.A\177\320
     if (likely(bytes)); else { Py_DECREF(data); __PYX_ERR(0, 1, __pyx_L1_error) }
     #endif
     #else /* compression: none (287 bytes) */
-const char* const bytes = ".1.1.17Copyright 2020-Viso.ai team?g.corrigan@viso.ai viso.ai AG <info@viso.ai>viso-sdk-pythonLGPL3abspath__author__cline_in_traceback__copyright___cur_dirdatetimedirname__email____file____license____main____module____name__nowospath__qualname__setdefault__test____title____version__year";
+const char* const bytes = ".1.1.19Copyright 2020-Viso.ai team?g.corrigan@viso.ai viso.ai AG <info@viso.ai>viso-sdk-pythonLGPL3abspath__author__cline_in_traceback__copyright___cur_dirdatetimedirname__email____file____license____main____module____name__nowospath__qualname__setdefault__test____title____version__year";
     PyObject *data = NULL;
     CYTHON_UNUSED_VAR(__Pyx_DecompressString);
     #endif

{viso_sdk_python-1.1.17 → viso_sdk_python-1.1.19}/viso_sdk/_version.py

@@ -14,4 +14,4 @@ __email__ = "g.corrigan@viso.ai"
 __license__ = "LGPL3"
 __title__ = "viso-sdk-python"
 
-__version__ = "1.1.17"
+__version__ = "1.1.19"