⚡ Bolt: Optimize regex and caching in ATSGenerator

[anchapin]

💡 What: Pre-compile multiple regular expressions and a static list of action verbs as module-level constants in cli/generators/ats_generator.py. Furthermore, cache the lowercased version of the entire resume text before invoking the generator expression to parse action verbs.
🎯 Why: Previously, regular expressions were compiled on every invocation, and string lowercasing (all_text.lower()) was triggered inside a generator expression for every single item in the action verbs list, resulting in redundant O(N) string allocations and compilation steps inside the execution paths.
📊 Impact: This significantly reduces CPU cycles and string allocations during ATS score generation, especially on longer resumes. It ensures string manipulation is linear $O(1)$ regarding the loops instead of $O(N)$ with the collection size.
🔬 Measurement: Verified through the existing test suite (using python -m pytest tests/test_ats_generator.py), confirming that no deterministic output logic changed. Running standard profiling against ATSGenerator on a large text blob should display a measurable performance boost.

---

PR created automatically by Jules for task 10442857480046678864 started by @anchapin

Summary by Sourcery

Optimize ATS generator performance by reusing compiled regular expressions and shared constants for common text and keyword checks.

Enhancements:

• Precompile frequently used regular expressions as module-level constants for format, contact, readability, and keyword extraction checks.
• Share a static action verbs list and reuse a cached lowercased resume text when counting action verbs to avoid repeated allocations.

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.

[sourcery-ai]

Reviewer's Guide

Precompiles commonly used regular expressions and the action verb list in ats_generator as module-level constants, and reuses a single lowercased resume text string to reduce redundant allocations and regex compilation while preserving behavior.

Sequence diagram for optimized _check_readability flow in ATSGenerator

sequenceDiagram
    participant ATSGenerator
    participant ResumeData
    participant RegexPatterns

    ATSGenerator->>ResumeData: _get_all_text(resume_data)
    ResumeData-->>ATSGenerator: all_text

    ATSGenerator->>ATSGenerator: all_text_lower = all_text.lower()

    ATSGenerator->>RegexPatterns: access _ACTION_VERBS
    RegexPatterns-->>ATSGenerator: action_verbs
    ATSGenerator->>ATSGenerator: action_verb_count = sum(verb in all_text_lower)

    ATSGenerator->>RegexPatterns: _QUANTIFIABLE_PATTERN.search(all_text)
    RegexPatterns-->>ATSGenerator: has_numbers

    ATSGenerator->>RegexPatterns: _ACRONYM_PATTERN.findall(all_text)
    RegexPatterns-->>ATSGenerator: acronyms

    ATSGenerator-->>ATSGenerator: update details and suggestions based on checks

Loading

Flow diagram for module-level regex caching in ATSGenerator

flowchart TD
    A["ats_generator module import"] --> B["Compile _TABLE_PATTERN, _SPECIAL_CHARS_PATTERN, _EMAIL_PATTERN, _PHONE_PATTERN, _QUANTIFIABLE_PATTERN, _ACRONYM_PATTERN, _JSON_ARRAY_PATTERN, _TECH_TERM_PATTERN, _SUMMARY_TERM_PATTERN once"]
    B --> C["Define _ACTION_VERBS list once"]
    C --> D["Instantiate ATSGenerator"]

    D --> E["_check_format_parsing"]
    E --> E1["Use _TABLE_PATTERN.search(all_text)"]
    E --> E2["Use _SPECIAL_CHARS_PATTERN.findall(all_text)"]

    D --> F["_check_contact_info"]
    F --> F1["Use _EMAIL_PATTERN.search(contact_email)"]
    F --> F2["Use _PHONE_PATTERN.search(contact_phone)"]

    D --> G["_check_readability"]
    G --> G1["all_text = _get_all_text(resume_data)"]
    G1 --> G2["all_text_lower = all_text.lower()"]
    G2 --> G3["count verbs from _ACTION_VERBS in all_text_lower"]
    G1 --> G4["Use _QUANTIFIABLE_PATTERN.search(all_text)"]
    G1 --> G5["Use _ACRONYM_PATTERN.findall(all_text)"]

    D --> H["_extract_job_keywords"]
    H --> H1["Use _JSON_ARRAY_PATTERN.search(response)"]

    D --> I["_extract_resume_keywords"]
    I --> I1["Use _TECH_TERM_PATTERN.findall(text)"]
    I --> I2["Use _SUMMARY_TERM_PATTERN.findall(summary.lower())"]

Loading

File-Level Changes

Change	Details	Files
Precompile regex patterns as module-level constants and replace inline uses with compiled objects.	Introduced compiled regex constants for tables, special characters, email, phone, quantifiable metrics, acronyms, JSON arrays, tech terms, and summary terms at the top of the module. Updated format parsing to use compiled table and special character patterns instead of inline re.search/re.findall calls. Updated contact info validation to pass compiled regex objects in the contact_fields map and call .search() on them. Updated readability checks to use compiled patterns for quantifiable achievements and acronyms. Updated job keyword extraction to use a compiled JSON array pattern for parsing list-like responses from the model. Updated resume keyword extraction to use compiled patterns for tech terms and summary terms.	cli/generators/ats_generator.py
Cache lowercased resume text once per readability check and reuse it for action verb detection.	Compute all_text_lower once from all_text before readability checks. Replace repeated all_text.lower() calls in the action verb counting loop with the cached all_text_lower string.	cli/generators/ats_generator.py
Hoist the static action verb list to a module-level constant and reuse it in readability checks.	Defined _ACTION_VERBS as a module-level list of common action verbs. Replaced the per-call local action_verbs list in _check_readability with the shared _ACTION_VERBS constant when counting action verbs.	cli/generators/ats_generator.py

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[sourcery-ai]

Hey - I've found 1 issue

Prompt for AI Agents

Please address the comments from this code review:

## Individual Comments

### Comment 1
<location path="cli/generators/ats_generator.py" line_range="47" />
<code_context>
+_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)
+_TECH_TERM_PATTERN = re.compile(r"\b[a-z]+(?:\s+[a-z]+)?\b")
+_SUMMARY_TERM_PATTERN = re.compile(r"\b[a-z]{2,}\b")
</code_context>
<issue_to_address>
**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.

```suggestion
_JSON_ARRAY_PATTERN = re.compile(r"\[.*?\]", flags=re.DOTALL)
```
</issue_to_address>

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[sourcery-ai]

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)