Skip to content

Phase 3: DSL query completed #5

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
rcholic merged 1 commit into from
Dec 22, 2025
Merged

Phase 3: DSL query completed #5

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
187 changes: 184 additions & 3 deletions sentience/query.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,33 +16,121 @@ def parse_selector(selector: str) -> Dict[str, Any]:
"role=textbox name~'email'"
"clickable=true role=link"
"role!=link"
"importance>500"
"text^='Sign'"
"text$='in'"
"""
query: Dict[str, Any] = {}

# Match patterns like: key=value, key~'value', key!="value"
# This regex matches: key, operator (=, ~, !=), and value (quoted or unquoted)
pattern = r'(\w+)([=~!]+)((?:\'[^\']+\'|\"[^\"]+\"|[^\s]+))'
# Match patterns like: key=value, key~'value', key!="value", key>123, key^='prefix', key$='suffix'
# Updated regex to support: =, !=, ~, ^=, $=, >, >=, <, <=
# Supports dot notation: attr.id, css.color
# Note: Handle ^= and $= first (before single char operators) to avoid regex conflicts
# Pattern matches: key, operator (including ^= and $=), and value (quoted or unquoted)
pattern = r'([\w.]+)(\^=|\$=|>=|<=|!=|[=~<>])((?:\'[^\']+\'|\"[^\"]+\"|[^\s]+))'
matches = re.findall(pattern, selector)

for key, op, value in matches:
# Remove quotes from value
value = value.strip().strip('"\'')

# Handle numeric comparisons
is_numeric = False
try:
numeric_value = float(value)
is_numeric = True
except ValueError:
pass

if op == '!=':
if key == "role":
query["role_exclude"] = value
elif key == "clickable":
query["clickable"] = False
elif key == "visible":
query["visible"] = False
elif op == '~':
# Substring match (case-insensitive)
if key == "text" or key == "name":
query["text_contains"] = value
elif op == '^=':
# Prefix match
if key == "text" or key == "name":
query["text_prefix"] = value
elif op == '$=':
# Suffix match
if key == "text" or key == "name":
query["text_suffix"] = value
elif op == '>':
# Greater than
if is_numeric:
if key == "importance":
query["importance_min"] = numeric_value + 0.0001 # Exclusive
elif key.startswith("bbox."):
query[f"{key}_min"] = numeric_value + 0.0001
elif key == "z_index":
query["z_index_min"] = numeric_value + 0.0001
elif key.startswith("attr.") or key.startswith("css."):
query[f"{key}_gt"] = value
elif op == '>=':
# Greater than or equal
if is_numeric:
if key == "importance":
query["importance_min"] = numeric_value
elif key.startswith("bbox."):
query[f"{key}_min"] = numeric_value
elif key == "z_index":
query["z_index_min"] = numeric_value
elif key.startswith("attr.") or key.startswith("css."):
query[f"{key}_gte"] = value
elif op == '<':
# Less than
if is_numeric:
if key == "importance":
query["importance_max"] = numeric_value - 0.0001 # Exclusive
elif key.startswith("bbox."):
query[f"{key}_max"] = numeric_value - 0.0001
elif key == "z_index":
query["z_index_max"] = numeric_value - 0.0001
elif key.startswith("attr.") or key.startswith("css."):
query[f"{key}_lt"] = value
elif op == '<=':
# Less than or equal
if is_numeric:
if key == "importance":
query["importance_max"] = numeric_value
elif key.startswith("bbox."):
query[f"{key}_max"] = numeric_value
elif key == "z_index":
query["z_index_max"] = numeric_value
elif key.startswith("attr.") or key.startswith("css."):
query[f"{key}_lte"] = value
elif op == '=':
# Exact match
if key == "role":
query["role"] = value
elif key == "clickable":
query["clickable"] = value.lower() == "true"
elif key == "visible":
query["visible"] = value.lower() == "true"
elif key == "tag":
query["tag"] = value
elif key == "name" or key == "text":
query["text"] = value
elif key == "importance" and is_numeric:
query["importance"] = numeric_value
elif key.startswith("attr."):
# Dot notation for attributes: attr.id="submit-btn"
attr_key = key[5:] # Remove "attr." prefix
if "attr" not in query:
query["attr"] = {}
query["attr"][attr_key] = value
elif key.startswith("css."):
# Dot notation for CSS: css.color="red"
css_key = key[4:] # Remove "css." prefix
if "css" not in query:
query["css"] = {}
query["css"][css_key] = value

return query

Expand All @@ -65,6 +153,18 @@ def match_element(element: Element, query: Dict[str, Any]) -> bool:
if element.visual_cues.is_clickable != query["clickable"]:
return False

# Visible (using in_viewport and !is_occluded)
if "visible" in query:
is_visible = element.in_viewport and not element.is_occluded
if is_visible != query["visible"]:
return False

# Tag (not yet in Element model, but prepare for future)
if "tag" in query:
# For now, this will always fail since tag is not in Element model
# This is a placeholder for future implementation
pass

# Text exact match
if "text" in query:
if not element.text or element.text != query["text"]:
Expand All @@ -77,6 +177,87 @@ def match_element(element: Element, query: Dict[str, Any]) -> bool:
if query["text_contains"].lower() not in element.text.lower():
return False

# Text prefix match
if "text_prefix" in query:
if not element.text:
return False
if not element.text.lower().startswith(query["text_prefix"].lower()):
return False

# Text suffix match
if "text_suffix" in query:
if not element.text:
return False
if not element.text.lower().endswith(query["text_suffix"].lower()):
return False

# Importance filtering
if "importance" in query:
if element.importance != query["importance"]:
return False
if "importance_min" in query:
if element.importance < query["importance_min"]:
return False
if "importance_max" in query:
if element.importance > query["importance_max"]:
return False

# BBox filtering (spatial)
if "bbox.x_min" in query:
if element.bbox.x < query["bbox.x_min"]:
return False
if "bbox.x_max" in query:
if element.bbox.x > query["bbox.x_max"]:
return False
if "bbox.y_min" in query:
if element.bbox.y < query["bbox.y_min"]:
return False
if "bbox.y_max" in query:
if element.bbox.y > query["bbox.y_max"]:
return False
if "bbox.width_min" in query:
if element.bbox.width < query["bbox.width_min"]:
return False
if "bbox.width_max" in query:
if element.bbox.width > query["bbox.width_max"]:
return False
if "bbox.height_min" in query:
if element.bbox.height < query["bbox.height_min"]:
return False
if "bbox.height_max" in query:
if element.bbox.height > query["bbox.height_max"]:
return False

# Z-index filtering
if "z_index_min" in query:
if element.z_index < query["z_index_min"]:
return False
if "z_index_max" in query:
if element.z_index > query["z_index_max"]:
return False

# In viewport filtering
if "in_viewport" in query:
if element.in_viewport != query["in_viewport"]:
return False

# Occlusion filtering
if "is_occluded" in query:
if element.is_occluded != query["is_occluded"]:
return False

# Attribute filtering (dot notation: attr.id="submit-btn")
if "attr" in query:
# This requires DOM access, which is not available in the Element model
# This is a placeholder for future implementation when we add DOM access
pass

# CSS property filtering (dot notation: css.color="red")
if "css" in query:
# This requires DOM access, which is not available in the Element model
# This is a placeholder for future implementation when we add DOM access
pass

return True


Expand Down
Binary file removed tests/__pycache__/__init__.cpython-311.pyc
View file
Open in desktop
Binary file not shown.
Binary file removed tests/__pycache__/test_actions.cpython-311-pytest-7.4.4.pyc
View file
Open in desktop
Binary file not shown.
Binary file removed tests/__pycache__/test_query.cpython-311-pytest-7.4.4.pyc
View file
Open in desktop
Binary file not shown.
Binary file removed tests/__pycache__/test_snapshot.cpython-311-pytest-7.4.4.pyc
View file
Open in desktop
Binary file not shown.
Binary file removed tests/__pycache__/test_spec_validation.cpython-311-pytest-7.4.4.pyc
View file
Open in desktop
Binary file not shown.
Binary file removed tests/__pycache__/test_wait.cpython-311-pytest-7.4.4.pyc
View file
Open in desktop
Binary file not shown.
Loading
Loading