rexa 1.0.3

A powerful regex and NLP utility library for text validation, extraction, conversion, formatting, and pattern generation.

🎉 Rexa Usage Guide Welcome to Rexa, a powerful Python library for regex operations and text preprocessing! Whether you're validating emails, extracting URLs, generating fuzzy regex patterns, or cleaning text for NLP pipelines, Rexa has you covered. This guide provides clear examples to help you harness Rexa's full potential.

🚀 Getting Started Install Rexa via pip: pip install rexa

Import and use: from rexa import Rex rex = Rex()

Example: Validate an email

print(rex.validator.Is_Email("[e-mail]")) # True

Example: Generate fuzzy regex

print(rex.mapper.string_to_regex("i'm")) # \b[iIíìîïİ1!|]['‘’][mM]\b

Dependencies: Requires Python 3.10+, regex, dateparser, pydantic, nltk, langdetect. Source: GitHub | Issues

🔍 1. Validator (validation.py) Validate and match common patterns with ease:

Method Description Example

Is_Email(s) ✅ Checks if s is a valid email rex.validator.Is_Email("[e-mail]") → True

Match_Email(s) 🔎 Returns Match object for email, else None rex.validator.Match_Email("bad@") → None

Is_URL(s) ✅ Validates HTTP/HTTPS URLs rex.validator.Is_URL("https://site.io") → True

Match_URL(s) 🔎 Matches URL and captures path/query m = rex.validator.Match_URL("site.com/path")

Is_Date_ISO(s) ✅ Checks YYYY-MM-DD date format rex.validator.Is_Date_ISO("2025-08-02") → True

Match_Date_ISO(s) 🔎 Captures ISO date, if present rex.validator.Match_Date_ISO("02/08/2025") → None

Is_IranianPhone(s) ✅ Validates Iranian mobile numbers rex.validator.Is_IranianPhone("09121234567") → True

Is_UUID4(s) ✅ Checks UUID v4 format rex.validator.Is_UUID4("123e4567-e89b-12d3-a456-426614174000") → True

And more Methods for time, ISBN, hex colors, etc. See validation.py for full list

📥 2. Extractor (extraction.py) Extract specific patterns from text:

Method Extracts… Example

Extract_Emails(text) All email addresses rex.extractor.Extract_Emails("[e-mail] [e-mail]") → ["[e-mail]", "[e-mail]"]

Extract_URLs(text) All web links (http/https/ftp) rex.extractor.Extract_URLs("Go to http://x.com") → ["http://x.com"]

Extract_Dates(text) Dates in ISO/EU formats rex.extractor.Extract_Dates("2021-01-01 or 01/01/2021") → ["2021-01-01", "01/01/2021"]

Extract_Phones(text) Phone numbers (intl & local) rex.extractor.Extract_Phones("[phone], 09121234567") → ["[phone]", "09121234567"]

Extract_IPv4(text) IPv4 addresses rex.extractor.Extract_IPv4("192.168.1.1") → ["192.168.1.1"]

Extract_UUIDs(text) UUID v4 strings rex.extractor.Extract_UUIDs("123e4567-e89b-12d3-a456-426614174000") → ["123e4567-e89b-12d3-a456-426614174000"]

🔄 3. Converter (conversion.py) Normalize and reformat strings:

Method Description Example

Convert_MultipleSpaces(text) Collapses extra spaces rex.converter.Convert_MultipleSpaces("A B") → "A B"

Convert_ThousandSeparatedNumbers(text) Strips commas from large numbers rex.converter.Convert_ThousandSeparatedNumbers("1,000,000") → "1000000"

Convert_DateFormat(s, from, to) Swaps date separators rex.converter.Convert_DateFormat("01.01.2025", ".", "/") → "01/01/2025"

Slugify(text) Generates SEO-friendly URL slugs rex.converter.Slugify("Hello World!") → "hello-world"

✨ 4. Formatter (formatting.py) Clean and standardize text:

Method Description Example

Strip_HTMLTags(s) Removes HTML tags rex.formatter.Strip_HTMLTags("Hi") → "Hi"

Normalize_Spaces(s) Normalizes to single spaces rex.formatter.Normalize_Spaces("A B") → "A B"

Remove_ThousandSeparators(s) Drops commas in numbers rex.formatter.Remove_ThousandSeparators("1,234") → "1234"

Normalize_DateSeparator(s, sep) Standardizes date delimiters rex.formatter.Normalize_DateSeparator("2021/01.01", "-") → "2021-01-01"

🧠 5. Regex Mapper (regex_mapper.py) Generate fuzzy regex patterns and analyze text structure:

Method Description Example

string_to_regex(text, use_anchors=False, use_char_map=True, detailed=False) Generates a fuzzy regex pattern for text. Use use_anchors=True for ^...$, use_char_map=True for fuzzy matching (e.g., i → [iIíìîïİ1! ]), detailed=True for validation info.

string_to_breakdown(text, use_char_map=True, detailed=False) Breaks down text into regex components with metadata (type, pattern, count). detailed=True adds Unicode categories and variant matches. rex.mapper.string_to_breakdown("i'm") → `[{"type": "LETTER", "name": "fuzzy 'i'", "pattern": "[iIíìîïİ1!

get_all_patterns() Returns a dictionary of predefined patterns (email, URL, etc.) rex.mapper.get_all_patterns() → {"email": "^[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+.[A-Za-z]{2,}$", ...}

Example with Detailed Output:

Detailed regex

pattern, details = rex.mapper.string_to_regex("i'm", detailed=True) print(pattern) # \b[iIíìîïİ1!|]['‘’][mM]\b print(details) # {'valid': True, 'score': 0.75, 'sample_matches': ["i'm", "I'm", "1'm"], 'explanation': 'Pattern matches original: True...'}

Detailed breakdown

breakdown = rex.mapper.string_to_breakdown("i'm", detailed=True) print(breakdown[0]) # {'type': 'LETTER', 'name': "fuzzy 'i'", 'pattern': '[iIíìîïİ1!|]', 'count': 1, 'example_chars': ['i'], 'unicode_category': 'Ll', 'variant_matches': ['i', 'I', '1']}

🧹 6. TextTools (texttools.py) Advanced NLP and text cleaning utilities:

Method Description Example

to_lower(s) Lowercases entire string rex.texttools.to_lower("HELLO") → "hello"

to_upper(s) Uppercases entire string rex.texttools.to_upper("hi") → "HI"

remove_emojis(s) Strips Unicode emojis rex.texttools.remove_emojis("I ❤️ you") → "I you"

remove_numbers(s) Removes all digits rex.texttools.remove_numbers("a1b2") → "ab"

remove_usernames(s) Removes @username tokens rex.texttools.remove_usernames("@me hi") → " hi"

remove_punctuation(s) Strips punctuation & symbols rex.texttools.remove_punctuation("Hey!?@") → "Hey"

remove_urls_emails(s) Drops URLs & email addresses rex.texttools.remove_urls_emails("[e-mail] http://x") → " "

remove_stopwords(s) Filters common words (using NLTK) rex.texttools.remove_stopwords("the cat sits") → "cat sits"

lemmatize_text(s) Lemmatizes tokens rex.texttools.lemmatize_text("running") → "running"

stem_text(s) Stems tokens rex.texttools.stem_text("running") → "run"

normalize_whitespace(s) Collapses whitespace rex.texttools.normalize_whitespace(" A B\n") → "A B"

normalize_arabic(s) Persian/Arabic char mapping & diacritics rex.texttools.normalize_arabic("كیف") → "کیف"

count_tokens(s) Counts word tokens rex.texttools.count_tokens("a b c") → 3

remove_short_long_words(s, min, max) Keeps words in length range rex.texttools.remove_short_long_words("a bb ccc", 2, 3) → "bb ccc"

detect_language(s) Auto-detects text language rex.texttools.detect_language("hello") → "en"

clean_text(...kwargs) Pipeline for common cleaning options rex.texttools.clean_text("Hi @you 123 😊", lowercase=True, remove_emoji=True, remove_username=True, remove_urls_emails=True, remove_punct=True) → "hi"

🚀 Quick Tips

Mix & Match: Combine methods for complex workflows, e.g., rex.texttools.clean_text() followed by rex.extractor.Extract_Emails(). Fuzzy Matching: Use string_to_regex with use_char_map=True for robust pattern generation (e.g., matches i'm, I'm, 1'm). Performance: For large texts, batch process with map() or parallelize with multiprocessing. Extensibility: Subclass Rex to add custom patterns or preprocessing steps. Debugging: Enable logging (via utils.Logger) for detailed insights into regex generation and validation.

📚 Example Workflow Clean and validate user input: from rexa import Rex rex = Rex()

text = "Contact: [e-mail], [e-mail] Date: 2025-08-02 😊" cleaned = rex.texttools.clean_text(text, lowercase=True, remove_emoji=True) emails = rex.extractor.Extract_Emails(cleaned) is_valid_date = rex.validator.Is_Date_ISO("2025-08-02") pattern = rex.mapper.string_to_regex("alice", use_char_map=True) print(cleaned) # contact: [e-mail], [e-mail] date: 2025-08-02 print(emails) # ['[e-mail]', '[e-mail]'] print(is_valid_date) # True print(pattern) # \b[aAáàâäãåā@4][lL1|][iIíìîïİ1!|][cCçćč][eEéèêëē3]\b

Happy coding with Rexa! Questions or feedback? Open an issue at https://github.com/arshia82sbn/rexa/issues.