rolfedh-doc-utils 0.1.29__py3-none-any.whl → 0.1.31__py3-none-any.whl

callout_lib/table_parser.py

@@ -38,6 +38,7 @@ class AsciiDocTable:
     end_line: int
     attributes: str  # Table attributes like [cols="1,3"]
     rows: List[TableRow]
+    title: str = ""  # Block title like ".Table description"
 
 
 class TableParser:
@@ -47,8 +48,8 @@ class TableParser:
     TABLE_START = re.compile(r'^\[.*?\]$')
     TABLE_DELIMITER = re.compile(r'^\|===\s*$')
 
-    # Pattern for table cell separator
-    CELL_SEPARATOR = re.compile(r'^\|')
+    # Pattern for table cell separator (| or cell type specifier like a|, s|, etc.)
+    CELL_SEPARATOR = re.compile(r'^(\||[ashdmev]\|)')
 
     # Pattern for conditional directives
     IFDEF_PATTERN = re.compile(r'^(ifdef::|ifndef::).+\[\]\s*$')
@@ -56,6 +57,64 @@ class TableParser:
 
     # Pattern for callout number (used for callout table detection)
     CALLOUT_NUMBER = re.compile(r'^<(\d+)>\s*$')
+    PLAIN_NUMBER = re.compile(r'^(\d+)\s*$')
+
+    def _is_callout_or_number(self, text: str) -> tuple[bool, int]:
+        """
+        Check if text is a callout number (<1>) or plain number (1).
+        Returns (is_match, number) or (False, 0) if no match.
+        """
+        text = text.strip()
+
+        # Try callout format first: <1>
+        match = self.CALLOUT_NUMBER.match(text)
+        if match:
+            return (True, int(match.group(1)))
+
+        # Try plain number format: 1
+        match = self.PLAIN_NUMBER.match(text)
+        if match:
+            return (True, int(match.group(1)))
+
+        return (False, 0)
+
+    def _finalize_row_if_complete(self, current_row_cells, conditionals_before_row,
+                                   conditionals_after_row, expected_columns, rows):
+        """
+        Check if we have enough cells for a complete row, and if so, save it.
+
+        Returns: (new_current_row_cells, new_conditionals_before, new_conditionals_after)
+        """
+        if expected_columns > 0 and len(current_row_cells) >= expected_columns:
+            # Row is complete - save it
+            rows.append(TableRow(
+                cells=current_row_cells.copy(),
+                conditionals_before=conditionals_before_row.copy(),
+                conditionals_after=conditionals_after_row.copy()
+            ))
+            return [], [], []  # Reset for next row
+
+        # Row not complete yet
+        return current_row_cells, conditionals_before_row, conditionals_after_row
+
+    def _parse_column_count(self, attributes: str) -> int:
+        """
+        Parse the cols attribute to determine number of columns.
+
+        Example: '[cols="1,7a"]' returns 2
+                 '[cols="1,2,3"]' returns 3
+        """
+        import re
+        # Match cols="..." or cols='...'
+        match = re.search(r'cols=["\']([^"\']+)["\']', attributes)
+        if not match:
+            return 0  # Unknown column count
+
+        cols_spec = match.group(1)
+        # Count comma-separated values
+        # Handle formats like: "1,2", "1a,2a", "1,2,3", etc.
+        columns = cols_spec.split(',')
+        return len(columns)
 
     def find_tables(self, lines: List[str]) -> List[AsciiDocTable]:
         """Find all tables in the document."""
@@ -65,16 +124,27 @@ class TableParser:
         while i < len(lines):
             # Look for table delimiter
             if self.TABLE_DELIMITER.match(lines[i]):
-                # Check if there are attributes on the line before
+                # Check for attributes and title before the table
                 attributes = ""
+                title = ""
                 start_line = i
 
+                # Check line before delimiter for attributes [cols="..."]
                 if i > 0 and self.TABLE_START.match(lines[i - 1]):
                     attributes = lines[i - 1]
                     start_line = i - 1
 
+                    # Check line before attributes for title .Title
+                    if i > 1 and lines[i - 2].strip().startswith('.') and not lines[i - 2].strip().startswith('..'):
+                        title = lines[i - 2].strip()
+                        start_line = i - 2
+                elif i > 0 and lines[i - 1].strip().startswith('.') and not lines[i - 1].strip().startswith('..'):
+                    # Title directly before delimiter (no attributes)
+                    title = lines[i - 1].strip()
+                    start_line = i - 1
+
                 # Parse table content
-                table = self._parse_table(lines, start_line, i)
+                table = self._parse_table(lines, start_line, i, title)
                 if table:
                     tables.append(table)
                     i = table.end_line + 1
@@ -83,11 +153,13 @@ class TableParser:
 
         return tables
 
-    def _parse_table(self, lines: List[str], start_line: int, delimiter_line: int) -> Optional[AsciiDocTable]:
+    def _parse_table(self, lines: List[str], start_line: int, delimiter_line: int, title: str = "") -> Optional[AsciiDocTable]:
         """
         Parse a single table starting at the delimiter.
 
         AsciiDoc table format:
+        .Optional title
+        [optional attributes]
         |===
         |Cell1
         |Cell2
@@ -96,12 +168,28 @@ class TableParser:
         |Cell4
         |===
         """
+        # Get attributes and parse column count
+        attributes = ""
+        if start_line < delimiter_line:
+            # Check if start line is title or attributes
+            start_content = lines[start_line].strip()
+            if start_content.startswith('.') and not start_content.startswith('..'):
+                # Start line is title, attributes might be on next line
+                if start_line + 1 < delimiter_line:
+                    attributes = lines[start_line + 1]
+            else:
+                # Start line is attributes
+                attributes = lines[start_line]
+
+        expected_columns = self._parse_column_count(attributes)
+
         i = delimiter_line + 1
         rows = []
         current_row_cells = []
         current_cell_lines = []
         conditionals_before_row = []
         conditionals_after_row = []
+        in_asciidoc_cell = False  # Track if we're in an a| (AsciiDoc) cell
 
         while i < len(lines):
             line = lines[i]
@@ -124,36 +212,40 @@ class TableParser:
                         conditionals_after=conditionals_after_row.copy()
                     ))
 
-                # Get attributes if present
-                attributes = ""
-                if start_line < delimiter_line:
-                    attributes = lines[start_line]
-
+                # Get attributes if present (already extracted above)
                 return AsciiDocTable(
                     start_line=start_line,
                     end_line=i,
                     attributes=attributes,
-                    rows=rows
+                    rows=rows,
+                    title=title
                 )
 
             # Check for conditional directives
             if self.IFDEF_PATTERN.match(line) or self.ENDIF_PATTERN.match(line):
-                if not current_row_cells:
+                # If we're building a cell (current_cell_lines is not empty) OR
+                # we're in an AsciiDoc cell, add conditional to cell content
+                if current_cell_lines or in_asciidoc_cell:
+                    # Inside a cell - conditional is part of cell content
+                    current_cell_lines.append(line)
+                elif current_row_cells:
+                    # Between cells in the same row
+                    conditionals_after_row.append(line)
+                else:
                     # Conditional before any cells in this row
                     conditionals_before_row.append(line)
-                else:
-                    # Conditional after cells started - treat as part of current context
-                    if current_cell_lines:
-                        # Inside a cell
-                        current_cell_lines.append(line)
-                    else:
-                        # Between cells in the same row
-                        conditionals_after_row.append(line)
                 i += 1
                 continue
 
-            # Blank line separates rows
+            # Blank line handling
             if not line.strip():
+                # In AsciiDoc cells (a|), blank lines are part of cell content
+                if in_asciidoc_cell:
+                    current_cell_lines.append(line)
+                    i += 1
+                    continue
+
+                # Otherwise, blank line separates rows
                 # Save pending cell if exists
                 if current_cell_lines:
                     current_row_cells.append(TableCell(
@@ -161,6 +253,7 @@ class TableParser:
                         conditionals=[]
                     ))
                     current_cell_lines = []
+                    in_asciidoc_cell = False
 
                 # Save row if we have cells
                 if current_row_cells:
@@ -176,8 +269,16 @@ class TableParser:
                 i += 1
                 continue
 
-            # Check for cell separator (|)
+            # Check for cell separator (|) or cell type specifier (a|, s|, etc.)
             if self.CELL_SEPARATOR.match(line):
+                # Determine if line starts with | or with a cell type specifier
+                if line.startswith('|'):
+                    # Standard cell separator
+                    cell_content = line[1:]  # Remove leading |
+                else:
+                    # Cell type specifier without leading | (e.g., "a|text")
+                    cell_content = line
+
                 # Save previous cell if exists
                 if current_cell_lines:
                     current_row_cells.append(TableCell(
@@ -185,9 +286,24 @@ class TableParser:
                         conditionals=[]
                     ))
                     current_cell_lines = []
+                    in_asciidoc_cell = False  # Reset for next cell
+
+                    # Check if row is complete (have enough cells based on cols attribute)
+                    current_row_cells, conditionals_before_row, conditionals_after_row = \
+                        self._finalize_row_if_complete(
+                            current_row_cells, conditionals_before_row,
+                            conditionals_after_row, expected_columns, rows
+                        )
 
-                # Extract cell content from this line (text after |)
-                cell_content = line[1:].strip()  # Remove leading |
+                # Check for cell type specifier (a|, s|, etc.)
+                # Type specifiers are single characters followed by |
+                if len(cell_content) > 1 and cell_content[0] in 'ashdmev' and cell_content[1] == '|':
+                    # Track if this is an AsciiDoc cell (a|)
+                    if cell_content[0] == 'a':
+                        in_asciidoc_cell = True
+                    cell_content = cell_content[2:]  # Remove type specifier and |
+
+                cell_content = cell_content.strip()
 
                 # Check if there are multiple cells on the same line (e.g., |Cell1 |Cell2 |Cell3)
                 if '|' in cell_content:
@@ -220,6 +336,39 @@ class TableParser:
                 i += 1
                 continue
 
+            # Check for cell type specifier on its own line (e.g., "a|", "s|", "h|")
+            # This is actually a cell SEPARATOR with type specifier
+            # Example:
+            #   |<1>          ← Cell 1
+            #   a|            ← Start cell 2 with type 'a' (AsciiDoc)
+            #   content...    ← Cell 2 content
+            stripped_line = line.strip()
+            if (len(stripped_line) == 2 and
+                stripped_line[0] in 'ashdmev' and
+                stripped_line[1] == '|' and
+                (current_cell_lines or current_row_cells)):
+                # Save previous cell if we have one
+                if current_cell_lines:
+                    current_row_cells.append(TableCell(
+                        content=current_cell_lines.copy(),
+                        conditionals=[]
+                    ))
+                    current_cell_lines = []
+
+                    # Check if row is complete
+                    current_row_cells, conditionals_before_row, conditionals_after_row = \
+                        self._finalize_row_if_complete(
+                            current_row_cells, conditionals_before_row,
+                            conditionals_after_row, expected_columns, rows
+                        )
+
+                # Set cell type for the NEW cell we're starting
+                if stripped_line[0] == 'a':
+                    in_asciidoc_cell = True
+                # Start collecting content for the new cell (no content on this line)
+                i += 1
+                continue
+
             # Regular content line (continuation of current cell)
             if current_cell_lines or current_row_cells:
                 current_cell_lines.append(line)
@@ -233,6 +382,8 @@ class TableParser:
         """
         Determine if a table is a callout explanation table.
         A callout table has two columns: callout number and explanation.
+        Accepts both callout format (<1>) and plain numbers (1).
+        Skips header rows if present.
         """
         if not table.rows:
             return False
@@ -241,15 +392,23 @@ class TableParser:
         if not all(len(row.cells) == 2 for row in table.rows):
             return False
 
-        # Check if first cell of each row is a callout number
-        for row in table.rows:
+        # Determine if there's a header row and skip it
+        has_header = self._has_header_row(table)
+        data_rows = table.rows[1:] if has_header else table.rows
+
+        if not data_rows:
+            return False
+
+        # Check if first cell of each data row is a callout number (either <1> or 1)
+        for row in data_rows:
             first_cell = row.cells[0]
             if not first_cell.content:
                 return False
 
-            # First line of first cell should be a callout number
+            # First line of first cell should be a callout number or plain number
             first_line = first_cell.content[0].strip()
-            if not self.CALLOUT_NUMBER.match(first_line):
+            is_match, _ = self._is_callout_or_number(first_line)
+            if not is_match:
                 return False
 
         return True
@@ -258,23 +417,36 @@ class TableParser:
         """
         Check if table has a header row.
         Common header patterns: "Item", "Value", "Description", "Column", etc.
+
+        A row is a header if:
+        - It does NOT start with a callout number (<1> or 1)
+        - It contains common header keywords in the cells
         """
         if not table.rows:
             return False
 
         first_row = table.rows[0]
-        if not first_row.cells:
+        if not first_row.cells or len(first_row.cells) < 2:
             return False
 
+        # If first cell is a callout number, this is NOT a header
+        first_cell = first_row.cells[0]
+        if first_cell.content:
+            first_cell_text = first_cell.content[0].strip()
+            is_callout, _ = self._is_callout_or_number(first_cell_text)
+            if is_callout:
+                return False
+
         # Collect text from all cells in first row
         header_text = ' '.join(
             cell.content[0] if cell.content else ''
             for cell in first_row.cells
         ).lower()
 
-        # Check for common header keywords
+        # Check for common header keywords (as whole words)
         header_keywords = ['item', 'description', 'value', 'column', 'parameter', 'field', 'name']
-        return any(keyword in header_text for keyword in header_keywords)
+        import re
+        return any(re.search(r'\b' + re.escape(keyword) + r'\b', header_text) for keyword in header_keywords)
 
     def is_3column_callout_table(self, table: AsciiDocTable) -> bool:
         """
@@ -300,19 +472,52 @@ class TableParser:
         if not all(len(row.cells) == 3 for row in data_rows):
             return False
 
-        # Check if first cell of each data row contains a plain number (1, 2, 3...)
+        # Check if first cell of each data row contains a callout or plain number (1, 2, 3... or <1>, <2>...)
         for row in data_rows:
             first_cell = row.cells[0]
             if not first_cell.content:
                 return False
 
-            # First line of first cell should be a number
+            # First line of first cell should be a callout number or plain number
             first_line = first_cell.content[0].strip()
-            if not first_line.isdigit():
+            is_match, _ = self._is_callout_or_number(first_line)
+            if not is_match:
                 return False
 
         return True
 
+    def get_table_callout_numbers(self, table: AsciiDocTable) -> List[int]:
+        """
+        Extract just the callout numbers from a table (in order, with duplicates).
+        Used for validation and diagnostics.
+
+        Returns:
+            List of callout numbers in the order they appear in the table.
+            Preserves duplicates to help identify table errors.
+        """
+        callout_numbers = []
+
+        # Determine if there's a header row and skip it
+        has_header = self._has_header_row(table)
+        data_rows = table.rows[1:] if has_header else table.rows
+
+        for row in data_rows:
+            # Handle both 2-column and 3-column tables
+            if len(row.cells) < 2:
+                continue
+
+            first_cell = row.cells[0]
+            if not first_cell.content:
+                continue
+
+            # Extract callout number (supports both <1> and 1 formats)
+            first_line = first_cell.content[0].strip()
+            is_match, callout_num = self._is_callout_or_number(first_line)
+            if is_match:
+                callout_numbers.append(callout_num)
+
+        return callout_numbers
+
     def extract_callout_explanations_from_table(self, table: AsciiDocTable) -> Dict[int, Tuple[List[str], List[str]]]:
         """
         Extract callout explanations from a table.
@@ -320,43 +525,46 @@ class TableParser:
 
         The conditionals list includes any ifdef/ifndef/endif statements that should
         be preserved when converting the table to other formats.
+
+        Accepts both callout format (<1>) and plain numbers (1).
+        Skips header rows if present.
+
+        Note: If table contains duplicate callout numbers, the last one wins.
+        Use get_table_callout_numbers() to detect duplicates.
         """
         explanations = {}
 
-        for row in table.rows:
+        # Determine if there's a header row and skip it
+        has_header = self._has_header_row(table)
+        data_rows = table.rows[1:] if has_header else table.rows
+
+        for row in data_rows:
             if len(row.cells) != 2:
                 continue
 
             callout_cell = row.cells[0]
             explanation_cell = row.cells[1]
 
-            # Extract callout number
+            # Extract callout number (supports both <1> and 1 formats)
             first_line = callout_cell.content[0].strip()
-            match = self.CALLOUT_NUMBER.match(first_line)
-            if not match:
+            is_match, callout_num = self._is_callout_or_number(first_line)
+            if not is_match:
                 continue
 
-            callout_num = int(match.group(1))
-
-            # Collect explanation lines
+            # Collect explanation lines, preserving blank lines and conditionals inline
+            # Blank lines will need to become continuation markers (+) in definition lists
             explanation_lines = []
             for line in explanation_cell.content:
-                # Skip conditional directives in explanation (preserve them separately)
-                if not (self.IFDEF_PATTERN.match(line) or self.ENDIF_PATTERN.match(line)):
-                    explanation_lines.append(line)
-
-            # Collect all conditionals for this row
-            all_conditionals = []
-            all_conditionals.extend(row.conditionals_before)
-
-            # Extract conditionals from explanation cell
-            for line in explanation_cell.content:
-                if self.IFDEF_PATTERN.match(line) or self.ENDIF_PATTERN.match(line):
-                    all_conditionals.append(line)
+                # Preserve ALL lines including conditionals and blank lines
+                # Empty lines will be marked as '' which signals need for continuation marker
+                explanation_lines.append(line)
 
-            all_conditionals.extend(row.conditionals_after)
+            # Collect conditionals that appear before/after the row
+            row_conditionals = []
+            row_conditionals.extend(row.conditionals_before)
+            row_conditionals.extend(row.conditionals_after)
 
-            explanations[callout_num] = (explanation_lines, all_conditionals)
+            explanations[callout_num] = (explanation_lines, row_conditionals)
 
         return explanations
 
@@ -366,12 +574,14 @@ class TableParser:
         Returns dict mapping callout number to tuple of (value_lines, description_lines, conditionals).
 
         Format: Item | Value | Description
-        - Item: Number (1, 2, 3...) corresponding to callout number
+        - Item: Number (1, 2, 3...) or callout (<1>, <2>...) corresponding to callout number
         - Value: The code/value being explained
         - Description: Explanation text
 
         The conditionals list includes any ifdef/ifndef/endif statements that should
         be preserved when converting the table to other formats.
+
+        Accepts both callout format (<1>) and plain numbers (1).
         """
         explanations = {}
 
@@ -387,47 +597,31 @@ class TableParser:
             value_cell = row.cells[1]
             desc_cell = row.cells[2]
 
-            # Extract item number (maps to callout number)
+            # Extract item number (maps to callout number) - supports both <1> and 1 formats
             if not item_cell.content:
                 continue
 
             item_num_str = item_cell.content[0].strip()
-            if not item_num_str.isdigit():
+            is_match, callout_num = self._is_callout_or_number(item_num_str)
+            if not is_match:
                 continue
 
-            callout_num = int(item_num_str)
-
-            # Collect value lines (column 2)
+            # Collect value lines (column 2), preserving all content including conditionals
             value_lines = []
             for line in value_cell.content:
-                # Skip conditional directives in value (preserve them separately)
-                if not (self.IFDEF_PATTERN.match(line) or self.ENDIF_PATTERN.match(line)):
-                    value_lines.append(line)
+                value_lines.append(line)
 
-            # Collect description lines (column 3)
+            # Collect description lines (column 3), preserving all content including conditionals
             description_lines = []
             for line in desc_cell.content:
-                # Skip conditional directives in description (preserve them separately)
-                if not (self.IFDEF_PATTERN.match(line) or self.ENDIF_PATTERN.match(line)):
-                    description_lines.append(line)
-
-            # Collect all conditionals for this row
-            all_conditionals = []
-            all_conditionals.extend(row.conditionals_before)
-
-            # Extract conditionals from value cell
-            for line in value_cell.content:
-                if self.IFDEF_PATTERN.match(line) or self.ENDIF_PATTERN.match(line):
-                    all_conditionals.append(line)
-
-            # Extract conditionals from description cell
-            for line in desc_cell.content:
-                if self.IFDEF_PATTERN.match(line) or self.ENDIF_PATTERN.match(line):
-                    all_conditionals.append(line)
+                description_lines.append(line)
 
-            all_conditionals.extend(row.conditionals_after)
+            # Collect conditionals that appear before/after the row
+            row_conditionals = []
+            row_conditionals.extend(row.conditionals_before)
+            row_conditionals.extend(row.conditionals_after)
 
-            explanations[callout_num] = (value_lines, description_lines, all_conditionals)
+            explanations[callout_num] = (value_lines, description_lines, row_conditionals)
 
         return explanations
 
@@ -442,8 +636,12 @@ class TableParser:
         Returns:
             AsciiDocTable if a callout table is found, None otherwise
         """
-        # Skip blank lines and continuation markers after code block
+        # Skip the closing delimiter of the code block (----, ...., etc.)
         i = code_block_end + 1
+        if i < len(lines) and lines[i].strip() in ['----', '....', '====']:
+            i += 1
+
+        # Skip blank lines and continuation markers after code block
         while i < len(lines) and (not lines[i].strip() or lines[i].strip() == '+'):
             i += 1
 
@@ -465,12 +663,26 @@ class TableParser:
 
             # Check for table delimiter
             if self.TABLE_DELIMITER.match(line):
-                # Found a table, parse it
+                # Found a table, extract attributes and title
+                attributes = ""
+                title = ""
                 start_line = j
+
+                # Check line before delimiter for attributes [cols="..."]
                 if j > 0 and self.TABLE_START.match(lines[j - 1]):
+                    attributes = lines[j - 1]
+                    start_line = j - 1
+
+                    # Check line before attributes for title .Title
+                    if j > 1 and lines[j - 2].strip().startswith('.') and not lines[j - 2].strip().startswith('..'):
+                        title = lines[j - 2].strip()
+                        start_line = j - 2
+                elif j > 0 and lines[j - 1].strip().startswith('.') and not lines[j - 1].strip().startswith('..'):
+                    # Title directly before delimiter (no attributes)
+                    title = lines[j - 1].strip()
                     start_line = j - 1
 
-                table = self._parse_table(lines, start_line, j)
+                table = self._parse_table(lines, start_line, j, title)
                 if table and (self.is_callout_table(table) or self.is_3column_callout_table(table)):
                     return table
 

convert_callouts_interactive.py

@@ -307,7 +307,20 @@ class InteractiveCalloutConverter:
             explanations, explanation_end = self.detector.extract_callout_explanations(new_lines, block.end_line)
 
             if not explanations:
-                print_colored(f"Warning: No explanations found for block at line {block.start_line + 1}", Colors.YELLOW)
+                # Get callout numbers for warning message
+                all_callout_nums = []
+                for group in callout_groups:
+                    all_callout_nums.extend(group.callout_numbers)
+
+                warning_msg = (
+                    f"WARNING: {input_file.name} line {block.start_line + 1}: "
+                    f"Code block has callouts {sorted(set(all_callout_nums))} but no explanations found after it. "
+                    f"This may indicate: explanations are shared with another code block, "
+                    f"explanations are in an unexpected location, or documentation error (missing explanations). "
+                    f"Consider reviewing this block manually."
+                )
+                print_colored(warning_msg, Colors.YELLOW)
+                self.warnings.append(warning_msg)
                 continue
 
             # Validate
@@ -332,9 +345,9 @@ class InteractiveCalloutConverter:
             else:
                 converted_content = self.detector.remove_callouts_from_code(block.content)
                 if format_choice == 'bullets':
-                    output_list = BulletListConverter.convert(callout_groups, explanations)
+                    output_list = BulletListConverter.convert(callout_groups, explanations, self.detector.last_table_title)
                 else:  # deflist
-                    output_list = DefListConverter.convert(callout_groups, explanations)
+                    output_list = DefListConverter.convert(callout_groups, explanations, self.detector.last_table_title)
 
             # Replace in document
             has_source_prefix = self.detector.CODE_BLOCK_START.match(new_lines[block.start_line])
@@ -521,6 +534,8 @@ Examples:
         print_colored(f"\n⚠  {len(converter.warnings)} Warning(s):", Colors.YELLOW)
         for warning in converter.warnings:
             print_colored(f"  {warning}", Colors.YELLOW)
+        print()
+        print_colored("Suggestion: Fix the callout mismatches in the files above and rerun this command.", Colors.YELLOW)
 
     if args.dry_run and files_modified > 0:
         print_colored("\nDRY RUN - No files were modified", Colors.YELLOW)