feat: parse table captions

renderer/fifty_ohm_html_renderer.py

@@ -210,13 +210,16 @@ class FiftyOhmHtmlRenderer(HtmlRenderer):
 
     def render_table(self, token):
         alignment = token.alignment
-        table = "<table>"
+        table = "<table>\n"
-        
+
         for i, row in enumerate(token.children):
             content = ""
             for j, cell in enumerate(row.children):
-                content += self.render_cell_helper(self.render_inner(cell), alignment[j], ("th" if i==0 else "td"))
+                content += self.render_cell_helper(self.render_inner(cell), alignment[j], ("th" if i == 0 else "td"))
-            table += f"<tr>{content}</tr>" 
+            table += f"<tr>\n{content}</tr>\n"
+
+        if token.caption != "":
+            table += f"<caption>{token.caption}</caption>"
         table += "</table>"
 
-        return table
+        return table

renderer/table.py

@@ -1,3 +1,4 @@
+import re
 
 from mistletoe import span_token
 from mistletoe.block_token import BlockToken
@@ -6,13 +7,13 @@ from mistletoe.block_token import BlockToken
 class Table(BlockToken):
     @staticmethod
     def start(line):
-        return '|' in line
+        return "|" in line
 
     @classmethod
-    def parse_header(cls, line): 
+    def parse_header(cls, line):
         # Parse header, extract alignment and make an ordinary row out of it
-        line = line.strip().strip('|') # Remove leading/trailing whitespace and pipes
+        line = line.strip().strip("|")  # Remove leading/trailing whitespace and pipes
-        columns = [col.strip() for col in line.split('|')] # Split by pipe |
+        columns = [col.strip() for col in line.split("|")]  # Split by pipe |
 
         alignment = []
         header = "| "
@@ -21,32 +22,45 @@ class Table(BlockToken):
             attr, val = re.match(r"^ ?(?:([lcrX]):)?(.*)", column).group(1, 2)
             alignment.append(attr)
             header += val + " | "
-        
+
         return header, alignment
 
     @classmethod
     def read(cls, lines):
         header = next(lines)
         header, alignment = cls.parse_header(header)
+        name = ""
+        caption = ""
         rows = [TableRow(header)]
 
-        for line in lines:
+        # Read table until the end: No more column definitions, caption, or empty line.
-            if "|" not in line :
+        line_buffer = []
+        next_line = lines.peek()
+        while next_line is not None and next_line.strip() != "":
+            maybe_caption = re.match(r"\[table:([^:\]]+):([^:\]]+)\]", next_line)
+            if maybe_caption is not None:
+                name, caption = maybe_caption.group(1, 2)
+                next(lines)
                 break
-            else :
+            elif "|" not in next_line:
-                rows.append(TableRow(line))
+                break
+            else:
+                line_buffer.append(next(lines))
+                next_line = lines.peek()
 
-        return alignment, rows
+        rows.extend([TableRow(line) for line in line_buffer])
+
+        return alignment, rows, name, caption
 
     def __init__(self, match):
-        self.alignment, self.children = match
+        self.alignment, self.children, self.name, self.caption = match
 
 
 class TableRow(BlockToken):
     @classmethod
     def parse_row(cls, line):
-        line = line.strip().strip('|')
+        line = line.strip().strip("|")
-        columns = [col.strip() for col in line.split('|')]
+        columns = [col.strip() for col in line.split("|")]
 
         data = []
 
@@ -58,6 +72,7 @@ class TableRow(BlockToken):
     def __init__(self, match):
         self.children = self.parse_row(match)
 
-class TableCell(BlockToken) :
+
+class TableCell(BlockToken):
     def __init__(self, match):
         super().__init__(match, span_token.tokenize_inner)