Class: Pubid::Ieee::Parser

Inherits:
Parslet::Parser
Object
Parslet::Parser
Pubid::Ieee::Parser
show all
Defined in:: lib/pubid/ieee/parser.rb
Overview

Parser class for IEEE identifiers Single Responsibility: Parsing IEEE identifier syntax Note: IEEE is extremely complex with many edge cases
Class Method Summary collapse

.parse(string) ⇒ Object
Class Method Details

.parse(string) ⇒ `Object`

# File 'lib/pubid/ieee/parser.rb', line 797

def self.parse(string)
  # Strip .pdf extension if present (Pattern 3: File Extensions)
  cleaned = string.sub(/\.pdf$/i, "")

  # Note: IEC and ANSI identifiers are NOT filtered here because they can have
  # IEEE co-publication or adoption. The Base.parse method handles determining
  # which standards are actually IEEE-related.
  # ISO-only standards are still filtered as they have separate handling.

  # Pattern 3: Replace underscore before ISO stage codes with slash
  # These are joint development drafts that use underscore instead of slash
  cleaned = cleaned.gsub(/_(FDIS|CDV|CD|DIS|WD|PWI|NP)/, '/\1')

  # NEW: Normalize multiple spaces to single space
  # No valid IEEE identifier pattern needs more than 1 space
  cleaned = cleaned.gsub(/\s+/, " ")

  # NEW Session 171: CONSERVATIVE data quality fixes for TODO.IEEE-MUST-DO.txt
  # Only fix clear typos: space before dash + 4-digit year, OR dash + space + 4-digit year
  # Do NOT touch " - " (space-dash-space) which is valid formatting
  cleaned = cleaned.gsub(/(\d)\s+-(\d{4})\b/, '\1-\2')  # "C37.101 -2006" → "C37.101-2006"
  cleaned = cleaned.gsub(/(\d)-\s+(\d{4})\b/, '\1-\2')  # "C62.35- 2010" → "C62.35-2010"

  # NEW Session 171: HTML entity for en dash (&#x2013;)
  # ONLY convert if not already followed by a dash (avoid creating --)
  cleaned = cleaned.gsub(/&#x2013;(?!-)/, "-")  # En dash → regular hyphen (if not followed by dash)
  cleaned = cleaned.gsub("&#x2013;-", "-")      # En-dash-dash → single dash

  # NEW Session 171: Remove wrong ! prefix
  cleaned = cleaned.gsub(/^!IEEE /, "IEEE ")

  # NEW Session 171: Fix "IEEE/ ASTM" spacing (extra space after slash)
  cleaned = cleaned.gsub("IEEE/ ASTM", "IEEE/ASTM")

  # NEW Phase 1: Handle HTML entities comprehensively
  cleaned = cleaned.gsub("&#x2122;", "™") # Trademark symbol
  cleaned = cleaned.gsub("&#x2019;", "'") # Smart apostrophe
  cleaned = cleaned.gsub("&amp;amp;", "&")   # Double-encoded ampersand
  cleaned = cleaned.gsub("&amp;", "&")       # Single-encoded ampersand

  # NEW: Wrap P&V notation in parentheses (Paper & Video, etc.)
  # Pattern: "IEEE Std 500-1984 P&V" → "IEEE Std 500-1984 (P&V)"
  cleaned = cleaned.gsub(/\s+(P&V)\s*$/, ' (\1)')

  # NEW Phase 1: Fix number spacing issues (e.g., "C57.1 2.25" → "C57.12.25")
  # This handles cases where a space appears in the middle of a number
  cleaned = cleaned.gsub(/(\d+\.\d+)\s+(\d+\.)/, '\1\2')

  # NEW Phase 1: Fix year spacing issues (e.g., "1 996" → "1996")
  # Remove spaces within 4-digit years
  cleaned = cleaned.gsub(/\b(1|2)\s+(\d{3})\b/, '\1\2')

  # NEW: Fix month+year spacing (e.g., "March2016" → "March 2016")
  # Add space between month name and 4-digit year when they're concatenated
  cleaned = cleaned.gsub(
    /\b(January|February|March|April|May|June|July|August|September|October|November|December)(\d{4})\b/, '\1 \2'
  )
  # Also handle abbreviated months
  cleaned = cleaned.gsub(
    /\b(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Sept|Oct|Nov|Dec)(\d{4})\b/, '\1 \2'
  )

  # NEW: Convert IEC/IEEE space-separated to semicolon format
  # Pattern: "IEC 61523-3 First edition 2004-09; IEEE 1497" → already semicolon
  # Pattern: "IEC 62539 First Edition 2007-07 IEEE 930" → needs semicolon
  # Match: IEC identifier (with edition) + space + IEEE identifier
  # Be conservative: only convert if IEC has "First edition" or similar and followed by IEEE
  cleaned = cleaned.gsub(
    /(IEC\s+\d+(?:-\d+)?(?:\s+First?\s+Edition\s+\d{4}-\d{2})?)\s+(IEEE\s+\S+)/, '\1; \2'
  )

  # NEW Phase 1 (Session 141): Remove literal trademark symbol
  # "C57.110™-2018" → "C57.110-2018"
  cleaned = cleaned.gsub(/™/, "")

  # NEW Phase 1 (Session 141): Fix specific year typo
  # "19969" → "1969" (very specific pattern, won't affect other text)
  cleaned = cleaned.gsub(/\b19969\b/, "1969")

  # NEW Session 169: Fix comma typo in 802.3 series numbers
  # "802.3ch-2020,802.3ca-2020" → "802.3ch-2020, 802.3ca-2020"
  # Very specific: 4 digits, comma, 3 digits (likely 802.3xx typo)
  cleaned = cleaned.gsub(/(\d{4}),(\d{3})/, '\1, \2')

  # NEW Session 169: Fix /lNT typo (lowercase L as 1)
  # "1003.1/2003.l/lNT" → "1003.1/2003.1/INT"
  cleaned = cleaned.gsub(/\/lNT\b/, "/INT")
  cleaned = cleaned.gsub(".l/", ".1/") # Also fix .l/ -> .1/

  # NEW Session 169: Fix I99O typo (letter I and O instead of digits)
  # "IEEE 1076-CONC-I99O" → "IEEE 1076-CONC-1990"
  cleaned = cleaned.gsub(/\bI99O\b/, "1990")

  # NEW: Fix common typos (Category 9)
  cleaned = cleaned.gsub(/^EEE /, "IEEE ")

  # NEW Session 170: Additional safe typo fixes
  # Fix "I EEE" (space between I and EEE)
  cleaned = cleaned.gsub(/^I EEE /, "IEEE ")

  # Fix "lEEE" (lowercase L instead of I)
  cleaned = cleaned.gsub(/^lEEE /, "IEEE ")

  # Fix missing closing parenthesis at end only (very conservative)
  # Only if there's exactly one more opening than closing paren
  open_count = cleaned.count("(")
  close_count = cleaned.count(")")
  if open_count == close_count + 1 && !cleaned.end_with?(")")
    cleaned = "#{cleaned})"
  end

  # NEW Phase 1: Remove trailing commas/colons and text
  cleaned = cleaned.gsub(/,\s*Standard\s*$/, "") # ", Standard" at end
  cleaned = cleaned.gsub(/[,:]\s*$/, "") # Trailing comma/colon
  cleaned = cleaned.gsub(/,\s+and\s+IEEE\s+Std\s/, " and ") # Handle "IEEE Std and Std" case

  # Enhanced: Fix unbalanced parentheses comprehensively
  # Handle three cases: missing closing, extra opening, nested unbalanced
  open_count = cleaned.count("(")
  close_count = cleaned.count(")")

  if open_count > close_count
    # More opening than closing - add closing parens at end
    # This handles both simple missing and nested unbalanced cases
    missing = open_count - close_count
    cleaned = cleaned + (")" * missing)
  elsif close_count > open_count
    # More closing than opening - remove extra closing from end
    # Very conservative: only remove trailing excess closing parens
    extra = close_count - open_count
    cleaned = cleaned.sub(/\){#{extra}}$/, "")
  end

  # === SESSION 173: TODO.IEEE-MUST-DO.txt Preprocessing Enhancements ===

  # Part A: Simple Normalizations (Lines 13, 16, 32-35, 36, 39-41 from TODO)

  # 1. Missing dash before year: "802.16g 2007" → "802.16g-2007"
  # But be careful not to affect month names (already have space)
  # Only apply if: digit + space + 4-digit year (and not after a month name)
  cleaned = cleaned.gsub(/(\d)\s+(\d{4})(?=\s*\(|\s*$)/, '\1-\2')

  # 2. Space-dash-space before year: "802.1ag - 2007" → "802.1ag-2007"
  # This is distinct from " - " in titles, targets space-dash-space-year pattern
  cleaned = cleaned.gsub(/\s+-\s+(\d{4})\b/, '-\1')

  # 3. Add missing "Std" after IEEE: "IEEE 1070-1995" → "IEEE Std 1070-1995"
  # Only at start of string, IEEE + space + digit
  cleaned = cleaned.gsub(/^IEEE\s+(?!Std\b)(\d)/, 'IEEE Std \1')

  # 3.5. Convert "IEEE No." to "IEEE Std": "IEEE No. 264-1968" → "IEEE Std 264-1968"
  # NOTE: Do NOT convert AIEE No - AIEE uses "No" as standard format
  cleaned = cleaned.gsub(/^IEEE\s+No\.\s*/, "IEEE Std ")
  cleaned = cleaned.gsub(/^IEEE\s+No\s/, "IEEE Std ")
  # Skip AIEE No conversion - AIEE preserves "No" format

  # 4. Space before slash in dual published: "262-1973 /ANSI" → "262-1973/ANSI"
  cleaned = cleaned.gsub(/\s+\//, "/")

  # 5. Comma before Edition: ", 1998 Edition" → "-1998"
  # Normalize to standard year format for parser
  cleaned = cleaned.gsub(/,\s+(\d{4})\s+Edition/, '-\1')

  # 6. ISO/IEC spacing: "ISO/IEC15802" → "ISO/IEC 15802"
  # Add space between publisher prefix and number
  cleaned = cleaned.gsub(/(ISO\/IEC)(\d)/, '\1 \2')

  # Part B: Publisher Order (Line 38 from TODO)

  # Fix wrong publisher order: "IEEE Std ANSI/IEEE" → "ANSI/IEEE Std"
  # This handles cases where IEEE Std appears before ANSI/IEEE publisher
  cleaned = cleaned.gsub(/^IEEE\s+Std\s+(ANSI\/IEEE)/, '\1 Std')

  # Part C: Dual Published Formats (Lines 8, 19 from TODO)

  # 1. Semicolon to parenthetical for dual published (MultiLabeledIdentifier)
  # "IEEE Std 120-1955; ASME PTC 19.6-1955" → "IEEE Std 120-1955 (ASME PTC 19.6-1955)"
  # Only if semicolon + space + organization abbreviation (capital letters)
  if cleaned.match?(/;\s+[A-Z]{2,}/)
    cleaned = cleaned.sub(/;\s+([A-Z][^;]+)$/, ' (\1)')
  end

  # === SESSION 174: Additional TODO.IEEE-MUST-DO.txt Preprocessing ===

  # Part A: Edition Abbreviation Normalization (Lines 10-11)
  # Pattern: ", 1999 Edn. (Reaff 2003)" → "-1999 (R2003)"
  # Normalize both the Edition abbreviation and the Reaffirmed format
  cleaned = cleaned.gsub(/,\s+(\d{4})\s+Edn\.\s+\(Reaff\s+(\d{4})\)/,
                         '-\1 (R\2)')
  # Also handle without initial comma (might occur in relationships)
  cleaned = cleaned.gsub(/(\d{4})\s+Edn\.\s+\(Reaff\s+(\d{4})\)/,
                         '\1 (R\2)')

  # Part B: IRE Parenthetical Split (Line 9)
  # Pattern: "(Reaffirmed 1980, 56 IRE 28.S2)" → "(R1980) (56 IRE 28.S2)"
  # Split nested reaffirmation + IRE reference into two parentheticals
  cleaned = cleaned.gsub(/\(Reaffirmed\s+(\d{4}),\s+(\d+\s+IRE[^)]+)\)/,
                         '(R\1) (\2)')

  # Part C: Slash to Parenthetical (Line 37)
  # Pattern: "number-year/ANSI identifier" → "number-year (ANSI identifier)"
  # Only convert if slash is followed by ANSI and NOT a relationship keyword
  # Look ahead to ensure we're at end of main identifier (before paren or end of string)
  cleaned = cleaned.gsub(%r{(\d{4})/ANSI\s+([^(]+)(?=\s*\(|$)},
                         '\1 (ANSI \2)')

  # Part D: ISO/IEC TR Spacing (Line 40)
  # Pattern: "ISO/IEC TR11802" → "ISO/IEC TR 11802"
  # Add space after TR when directly followed by digit
  cleaned = cleaned.gsub(/(ISO\/IEC\s+TR)(\d)/, '\1 \2')
  # === SESSION 178: AIEE Dual Numbers Expansion (Line 45) ===

  # Part E: AIEE "Nos X and Y" Expansion
  # Pattern: "AIEE Nos 72 and 73 - 1932" → "AIEE No 72-1932 and AIEE No 73-1932"
  # Expands dual AIEE numbers to separate identifiers with shared year
  if cleaned.match?(/AIEE\s+Nos\s+(\d+)\s+and\s+(\d+)\s+-\s+(\d{4})/)
    cleaned = cleaned.sub(/AIEE\s+Nos\s+(\d+)\s+and\s+(\d+)\s+-\s+(\d{4})/) do
      first_num = $1
      second_num = $2
      year = $3
      "AIEE No #{first_num}-#{year} and AIEE No #{second_num}-#{year}"
    end
  end

  # === SESSION 222: TODO.IEEE-MUST-FIX-IDs.txt Comprehensive Fixes ===

  # Part A: Typo Fixes
  # 1. "Stad" -> "Std" (typo)
  cleaned = cleaned.gsub(/\bStad\b/, "Std")

  # 2. Lowercase "std" -> "Std" when after IEEE/ANSI publishers
  cleaned = cleaned.gsub(/\b(IEEE|ANSI|AIEE)\s+std\b/, '\1 Std')

  # Part B: Symbol Normalization
  # 3. Additional (TM) patterns - strip them out
  cleaned = cleaned.gsub("(TM)", "")

  # Part C: Year-first format normalization
  # 4. Pattern "62704-4/D4, 2020" -> "IEEE P62704-4/D4, 2020"
  # Only if starts with digits-dash-digits/D pattern
  if cleaned.match?(/^(\d+[-.]\d+)\/D\d+/)
    cleaned = "IEEE P#{cleaned}"
  end

  # Part D: Suffix Normalization
  # 5. "/Preprint" -> remove (data quality - not standard suffix)
  cleaned = cleaned.gsub(/\/Preprint\b/, "")

  # Part E: Relationship Text Normalization
  # 6. "Proposed Revision of" -> "Revision of"
  cleaned = cleaned.gsub("Proposed Revision of", "Revision of")

  # 7. "ammended" typo -> "amended"
  cleaned = cleaned.gsub(/\bammended\b/i, "amended")

  # Part F: Trailing Characters After Special Patterns
  # 8. Remove trailing periods after /INT, /Cor, etc.
  cleaned = cleaned.gsub(/(\/INT|\/Cor\s+\d+-\d{4})\./, '\1')

  # Part G: Conformance Pattern Spacing
  # 9. Fix spacing in "/Conformance" patterns WITHOUT year (malformed only)
  # "1904.1(TM)/Conformance02" -> "1904.1 /Conformance02" (space before slash)
  # BUT: DO NOT touch valid patterns like "802.16/Conformance01-2003" (with year)
  # Use positive check for year suffix to exclude valid patterns
  # Actually, this preprocessing is breaking valid patterns - just remove it entirely
  # The parser can handle both "6/Conformance01-2003" and "6 /Conformance02" formats

  # Part H: Edition Text After /INT
  # 10. Handle ", Month YYYY Edition" after /INT by converting to month-year format
  # "1003.1/INT, March 1994 Edition" -> "1003.1/INT, March 1994"
  cleaned = cleaned.gsub(/(\/INT),\s+([A-Z][a-z]+)\s+(\d{4})\s+Edition/,
                         '\1, \2 \3')

  # Part I: Handle "Ed." abbreviation
  # 11. "Dec. 1994 Ed." -> "Dec. 1994"
  cleaned = cleaned.gsub(/\s+Ed\.\s*$/, "")

  # === PHASE 2: High-impact preprocessing for fixture failures ===

  # Quick wins from SESSION 224 (must come before more complex fixes)

  # Remove period after "Std": "IEEE Std." -> "IEEE Std"
  cleaned = cleaned.gsub(/\bStd\.\s+/, "Std ")

  # Redline Suffix Removal: " - Redline" at end
  cleaned = cleaned.gsub(/\s+-\s+Redline\b.*$/, "")

  # Title portion removal after year: "YYYY - IEEE Standard for..."
  cleaned = cleaned.gsub(
    /(\d{4})(\s+\([^)]+\))?\s+-\s+IEEE\s+Standard\s+for.*$/, '\1\2'
  )

  # Fix 2A: "IEEE PC" prefix -> "IEEE Std PC" or "IEEE P" treatment
  # "IEEE PC37.20.9/D7.3A" -> needs to parse as IEEE project draft
  # Strategy: Add "Std" after "IEEE" when followed by "PC" to route to standard pattern
  # Actually, the issue is the number rule consumes "PC37" as P + C37.
  # Better: normalize "IEEE PC" to "IEEE Std PC" so it hits the standard identifier path
  cleaned = cleaned.gsub(/^IEEE\s+PC(\d)/, 'IEEE Std PC\1')
  cleaned = cleaned.gsub(/^IEEE\s+Unapproved\s+Draft\s+Std\s+PC(\d)/,
                         'IEEE Unapproved Draft Std PC\1')

  # Fix 2B: "IEEE P" without "Std"/"Draft" prefix
  # ieee_p_identifier rule handles these directly - no preprocessing needed
  # Only handle "IEEE P" followed by "and ASHRAE" (copub case)
  cleaned = cleaned.gsub(/^IEEE\s+P(\d+)\s+and\s+ASHRAE/,
                         'IEEE Std P\1 and ASHRAE')

  # Fix 2C: "ISO/IEC XXXX-YYYY: Title" -> strip title after colon for ISO/IEC published standards
  # These are ISO-format identifiers with IEEE adoption, strip the title
  cleaned = cleaned.gsub(/^(ISO\/IEC \d+[-.]\d+-\d{4}):.*$/, '\1')
  cleaned = cleaned.gsub(/^(ISO\/IEC \d+-\d{4}):.*$/, '\1')

  # Fix 2D: "ISO/IEC XXXX : YYYY" -> normalize spacing around colon
  cleaned = cleaned.gsub(/^(ISO\/IEC \d+[-.]\d*)\s*:\s*(\d{4})/, '\1:\2')
  cleaned = cleaned.gsub(/^(ISO\/IEC \d+)\s*:\s*(\d{4})/, '\1:\2')

  # Fix 2G: "IEC/IEEE PXXX_D5" -> underscore to slash
  cleaned = cleaned.gsub(/^(IEC\/IEEE P[\w.-]+)_D/, '\1/D')

  # Fix 2H: "IEC XXXX First edition YYYY-MM; IEEE NNNN" -> normalize semicolon
  # Already handled by earlier semicolon normalization

  # Fix 2I: "IEEE/ISO/IEC PXXX/DIS" -> normalize to "ISO/IEC/IEEE PXXX/DIS"
  cleaned = cleaned.gsub(/^IEEE\/ISO\/IEC\s+(P[\w.-]+)/,
                         'ISO/IEC/IEEE \1')
  cleaned = cleaned.gsub(/^IEEE\/IEC\/ISO\s+(P[\w.-]+)/,
                         'IEC/ISO/IEEE \1')

  # Fix 2J: "IEEE/IEC PXXX D5" -> normalize space to slash before D
  cleaned = cleaned.gsub(/^(IEEE\/IEC P[\w.-]+)\s+D(\d)/, '\1/D\2')
  cleaned = cleaned.gsub(
    /^(IEEE\/IEC P[\w.-]+)\s+(CDV|FDIS|CD|DIS|ED\d)/, '\1/\2'
  )

  # Fix 2K: "ISO /IEC/IEEE" -> fix space before slash
  cleaned = cleaned.gsub(/^ISO\s+\/IEC\/IEEE/, "ISO/IEC/IEEE")
  cleaned = cleaned.gsub(/^ISO\s+\/IEC/, "ISO/IEC")

  # Fix 2L: "IS0" typo (letter O instead of digit 0)
  cleaned = cleaned.gsub(/^IS0\//, "ISO/")

  # Fix 2M: "IEEE-P15026-3-DIS-January 2015" -> dash-separated format
  # Normalize to "ISO/IEC/IEEE P15026-3/DIS, January 2015"
  cleaned = cleaned.gsub(/^IEEE-P(\d+)-(\d+)-DIS-(.*)/,
                         'ISO/IEC/IEEE P\1-\2/DIS, \3')

  # Fix 2N: "IEEE/CSA P844.1/293.1/D2" -> normalize CSA dual numbering
  cleaned = cleaned.gsub(/^IEEE\/CSA\s+(P[\d.]+)\/([\d.]+)\/D(\d+)/,
                         'IEEE/CSA \1/D\3')

  # Fix 2O: "IEEE Approved Draft Std P" -> normalize spacing
  cleaned = cleaned.gsub(/^IEEE\s+Approved\s+Draft\s+Std\s+(P\d)/,
                         'IEEE Approved Draft Std \1')
  # Fix: "IEEE Approved Draft Std P1234 / D12" -> remove space before slash
  cleaned = cleaned.gsub(/^(IEEE Approved Draft Std P[\w.-]+)\s+\/\s*D/,
                         '\1/D')

  # Fix 2P: "IEEE/EIA" -> normalize (parser handles IEEE/EIA via copublisher)
  # Already works - no fix needed

  # Fix 2Q: AIEE format variations
  # "AIEE No.1C-1954" -> "AIEE No. 1C-1954" (add space after No.)
  cleaned = cleaned.gsub(/^AIEE\s+No\.\s*(\d)/, 'AIEE No. \1')
  # "AIEE no 700-1945" -> "AIEE No 700-1945" (capitalize)
  cleaned = cleaned.gsub(/^AIEE\s+no\s/, "AIEE No ")
  # "AIEE Std No. 800" -> "AIEE Standard No 800" (normalize type word)
  cleaned = cleaned.gsub(/^AIEE\s+Std\s+No\.\s*/, "AIEE Standard No ")
  # "AIEE No 750.1-1960" -> handled by AIEE parser if decimal support added

  # Fix 2R: "IEEE PSI 10/D2" -> normalize to "IEEE/ASTM PSI 10/D2"
  cleaned = cleaned.gsub(/^IEEE\s+PSI\s+(\d)/, 'IEEE/ASTM PSI \1')

  # Fix 2S: "IEEE/IEC P62271-111/PC37.60_D5" -> normalize
  cleaned = cleaned.gsub(/^(IEEE\/IEC P[\d.-]+\/PC[\d.]+)_D/, '\1/D')

  # Fix 2T: "IEC P62271-111/IEEE PC37.60_D5" -> normalize to IEC/IEEE format
  cleaned = cleaned.gsub(/^IEC\s+(P[\d.-]+)\/IEEE\s+(PC[\d.]+)_D/,
                         'IEC/IEEE \2/D')

  # Fix 2U: "IEC/IEC P" -> "IEC/IEEE P" (typo)
  cleaned = cleaned.gsub(/^IEC\/IEC\s+(P\d)/, 'IEC/IEEE \1')

  # Fix 2V: "NACE SPXXXX-YYYY/IEEE Std NNNN-YYYY" -> normalize slash to parenthetical
  cleaned = cleaned.gsub(/^(NACE\s+SP\d+-\d+)\/(IEEE\s+Std\s+\d+-\d+)$/,
                         '\1 (\2)')

  # Fix 2W: "IEEE Std 802.11g-2003 (Amendment to IEEE Std 802.11, 1999 Edn. (Reaff 2003) as amended by"
  # This is a complex relationship - strip the parenthetical if too complex
  # Let the parser handle it but fix "Edn." to "Edition"
  cleaned = cleaned.gsub("Edn.", "Edition")

  # Fix 2X: "IEEE-P15026-3-DIS" format -> normalize
  # Already handled by Fix 2M

  # Fix 2Y: "P1635/D10/ASHARE 21/D10" -> fix ASHARE typo to ASHRAE
  cleaned = cleaned.gsub("ASHARE", "ASHRAE")

  # Fix 2Z: "PC37.30.2/D043 Rev 18" -> normalize draft version with Rev
  # "PC57-15 D2.0" -> normalize to "P57-15/D2.0"
  cleaned = cleaned.gsub(/^PC(\d)/, 'P\1')

  # Fix 2AA: "IEEE/ISO/IEC 8802-1Q-2020/Amd31-2021" -> normalize
  cleaned = cleaned.gsub(/^IEEE\/ISO\/IEC\s+(8802[\w.-]+)/,
                         'ISO/IEC/IEEE \1')

  # Fix 2AB: "IEEE C57.139/D14June 2010" -> add missing space
  cleaned = cleaned.gsub(
    /^(IEEE\s+C?\d[\d.]*\/D\d+)([A-Z][a-z]+\s+\d{4})/, '\1, \2'
  )

  # Fix 2AC: "IEEE Std: Title" -> strip colon and title (ANSI/IEEE Std: )
  cleaned = cleaned.gsub(/^(ANSI\/IEEE Std):\s+.*$/, '\1')

  # Fix 2AD: "IEEE 1076 IEC 61691-1-1 First edition 2004-10" -> semicolon format
  cleaned = cleaned.gsub(
    /^(IEEE\s+[\d.]+)\s+(IEC\s+\d+[-\d]*\s+.*edition\s+\d{4}-\d{2})$/i, '\1; \2'
  )

  # Fix 2AE: "IEEE No 29-1941 / ASA C77.1-1943" -> normalize to IEEE Std format
  cleaned = cleaned.gsub(/^IEEE\s+No\s+(\d+-\d+)\s+\/\s+ASA\s+(.*)/,
                         'IEEE Std \1 (ASA \2)')

  # Fix 2AF: "IEEE Std 1003.1/2003.l/lNT" -> fix typos
  # .l -> .1 and lNT -> INT handled by existing fixes

  new.parse(cleaned)
end
Class: Pubid::Ieee::Parser

Overview

Class Method Summary collapse

Class Method Details

.parse(string) ⇒ Object

.parse(string) ⇒ `Object`
