Skip to content

⚡ Bolt: Optimize regex and caching in ATSGenerator #281

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

Open
anchapin wants to merge 2 commits into
base: main
Choose a base branch
from
Open

⚡ Bolt: Optimize regex and caching in ATSGenerator #281

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
912efb4
⚡ Bolt: Optimize regex and caching in ATSGenerator
google-labs-jules[bot] May 3, 2026
8b8c5bf
⚡ Bolt: Optimize regex and caching in ATSGenerator and fix formatting
google-labs-jules[bot] May 3, 2026
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
64 changes: 39 additions & 25 deletions cli/generators/ats_generator.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,6 +37,33 @@

console = Console()

# Pre-compiled regular expressions and constants for performance optimization
_TABLE_PATTERN = re.compile(r"\|[^\n]+\|")
_SPECIAL_CHARS_PATTERN = re.compile(r"[^a-zA-Z0-9\s\-\.\,\@\(\)\#\/]")
_EMAIL_PATTERN = re.compile(r"^[^@]+@[^@]+\.[^@]+$")
_PHONE_PATTERN = re.compile(r"\d")
_QUANTIFIABLE_PATTERN = re.compile(
r"\d+%|\$\d+|\d+\s*(?:users|customers|projects)", flags=re.IGNORECASE
)
_ACRONYM_PATTERN = re.compile(r"\b[A-Z]{2,4}\b")
_JSON_ARRAY_PATTERN = re.compile(r"\[.*\]", flags=re.DOTALL)
Copy link
Copy Markdown

@sourcery-ai sourcery-ai Bot May 3, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

suggestion: Make JSON array extraction regex non-greedy to avoid overmatching

With DOTALL, r"\[.*\]" will match from the first [ to the last ] in the entire response. If the model returns multiple bracketed sections or other text containing [/], json_match.group(0) may be much larger than intended and break json.loads or parse the wrong data. A non-greedy pattern like r"\[.*?\]" limits the match to the smallest bracketed JSON array, which is likely what we want.

Suggested change
_JSON_ARRAY_PATTERN = re.compile(r"\[.*\]", flags=re.DOTALL)
_JSON_ARRAY_PATTERN = re.compile(r"\[.*?\]", flags=re.DOTALL)

_TECH_TERM_PATTERN = re.compile(r"\b[a-z]+(?:\s+[a-z]+)?\b")
_SUMMARY_TERM_PATTERN = re.compile(r"\b[a-z]{2,}\b")

_ACTION_VERBS = [
"developed",
"implemented",
"built",
"created",
"designed",
"managed",
"led",
"increased",
"decreased",
"improved",
"achieved",
]


@dataclass
class ATSCategoryScore:
Expand Down Expand Up @@ -214,8 +241,8 @@ def _check_format_parsing(self, resume_data: Dict[str, Any]) -> ATSCategoryScore

# Check for complex formatting indicators
all_text = self._get_all_text(resume_data)
has_tables = bool(re.search(r"\|[^\n]+\|", all_text))
has_special_chars = len(re.findall(r"[^a-zA-Z0-9\s\-\.\,\@\(\)\#\/]", all_text))
has_tables = bool(_TABLE_PATTERN.search(all_text))
has_special_chars = len(_SPECIAL_CHARS_PATTERN.findall(all_text))

if not has_tables:
details.append("No tables detected (ATS-friendly)")
Expand Down Expand Up @@ -349,15 +376,15 @@ def _check_contact_info(self, resume_data: Dict[str, Any]) -> ATSCategoryScore:

# Check required contact fields
contact_fields = {
"email": (contact.get("email"), 5, r"^[^@]+@[^@]+\.[^@]+$"),
"phone": (contact.get("phone"), 5, r"\d"),
"email": (contact.get("email"), 5, _EMAIL_PATTERN),
"phone": (contact.get("phone"), 5, _PHONE_PATTERN),
"location": (contact.get("location"), 5, None), # Just presence check
}

for field_name, (field_value, field_points, pattern) in contact_fields.items():
if field_value:
if pattern:
if re.search(pattern, field_value):
if pattern.search(field_value):
points += field_points
details.append(f"✓ {field_name.capitalize()} present and valid")
else:
Expand Down Expand Up @@ -392,22 +419,10 @@ def _check_readability(self, resume_data: Dict[str, Any]) -> ATSCategoryScore:
suggestions = []

all_text = self._get_all_text(resume_data)
all_text_lower = all_text.lower()

# Check for action verbs in experience bullets
action_verbs = [
"developed",
"implemented",
"built",
"created",
"designed",
"managed",
"led",
"increased",
"decreased",
"improved",
"achieved",
]
action_verb_count = sum(1 for verb in action_verbs if verb in all_text.lower())
action_verb_count = sum(1 for verb in _ACTION_VERBS if verb in all_text_lower)

if action_verb_count >= 3:
details.append(f"✓ Uses action verbs ({action_verb_count} found)")
Expand All @@ -416,7 +431,7 @@ def _check_readability(self, resume_data: Dict[str, Any]) -> ATSCategoryScore:
suggestions.append("Use more action verbs (e.g., developed, implemented)")

# Check for quantifiable achievements
has_numbers = bool(re.search(r"\d+%|\$\d+|\d+\s*(users|customers|projects)", all_text))
has_numbers = bool(_QUANTIFIABLE_PATTERN.search(all_text))
if has_numbers:
details.append("✓ Includes quantifiable achievements")
else:
Expand All @@ -425,8 +440,7 @@ def _check_readability(self, resume_data: Dict[str, Any]) -> ATSCategoryScore:

# Check for acronyms (should be minimal or defined)
# This is a simple heuristic
acronym_pattern = r"\b[A-Z]{2,4}\b"
acronyms = re.findall(acronym_pattern, all_text)
acronyms = _ACRONYM_PATTERN.findall(all_text)
if len(acronyms) < 10:
details.append(f"✓ Minimal acronyms ({len(acronyms)} found)")
else:
Expand Down Expand Up @@ -505,7 +519,7 @@ def _extract_job_keywords(self, job_description: str) -> List[str]:
response = self._call_openai(prompt)

# Parse JSON from response
json_match = re.search(r"\[.*\]", response, re.DOTALL)
json_match = _JSON_ARRAY_PATTERN.search(response)
if json_match:
keywords = json.loads(json_match.group(0))
if isinstance(keywords, list):
Expand Down Expand Up @@ -547,12 +561,12 @@ def _extract_resume_keywords(self, resume_data: Dict[str, Any]) -> List[str]:
text = bullet.get("text", "").lower()
# Extract common tech terms from text
# This is a simple heuristic - AI could do better
keywords.extend(re.findall(r"\b[a-z]+(?:\s+[a-z]+)?\b", text))
keywords.extend(_TECH_TERM_PATTERN.findall(text))

# Extract from summary
summary = resume_data.get("summary", "")
if summary:
keywords.extend(re.findall(r"\b[a-z]{2,}\b", summary.lower()))
keywords.extend(_SUMMARY_TERM_PATTERN.findall(summary.lower()))

return list(set(k.strip() for k in keywords if len(k) > 2))

Expand Down
Loading