Pull request #30: Feature/table

Merge in FRAG/50ohm.de from feature/table to main

* commit '5fe7510819e85412ea0dbc80bfa6c23789249343':
  feature: caption for tables in LaTeX
  feat: parse table captions
  fix: parse column alignment with regex
  refactor: format file with ruff
  refactor: simplify rendering of table cells
  feat: improve table tests
  refactor: format file with ruff
  fix: include new dependencies in requirements.txt
  refactor: inline row_helper
  feature: render latex table
  feature: html table implemented
  fix: fixed bug in morse code
  feature: first table implementation

renderer/fifty_ohm_html_renderer.py

@@ -12,10 +12,13 @@ from .picture import Picture
 from .question import Question
 from .quote import Quote
 from .references import References
+from .table import Table
 from .tag import Tag
 from .underline import Underline
 from .unit import Unit
 
+table_alignment = {"l": "left", "c": "center", "r": "right"}
+
 
 class FiftyOhmHtmlRenderer(HtmlRenderer):
     margin_anchor_id = 0
@@ -38,7 +41,8 @@ class FiftyOhmHtmlRenderer(HtmlRenderer):
             References,
             Question,
             Picture,
-            Photo
+            Photo,
+            Table,
         )
         self.question_renderer = question_renderer
 
@@ -182,7 +186,7 @@ class FiftyOhmHtmlRenderer(HtmlRenderer):
                 </figure>
             """
 
-    def render_picture(self, token) :
+    def render_picture(self, token):
         return self.render_picture_helper(token.id, token.ref, token.text, token.number)
 
     @staticmethod
@@ -194,5 +198,28 @@ class FiftyOhmHtmlRenderer(HtmlRenderer):
                 </figure>
             """
 
-    def render_photo(self, token) :
+    def render_photo(self, token):
-        return self.render_photo_helper(token.id, token.ref, token.text, token.number)
+        return self.render_photo_helper(token.id, token.ref, token.text, token.number)
+
+    @staticmethod
+    def render_cell_helper(cell, alignment, type):
+        style = ""
+        if alignment in table_alignment:
+            style = f' style="text-align: {table_alignment[alignment]};"'
+        return f"<{type}{style}>{cell}</{type}>\n"
+
+    def render_table(self, token):
+        alignment = token.alignment
+        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"))
+            table += f"<tr>\n{content}</tr>\n"
+
+        if token.caption != "":
+            table += f"<caption>{token.caption}</caption>"
+        table += "</table>"
+
+        return table

renderer/fifty_ohm_latex_renderer.py

@@ -8,6 +8,7 @@ from .photo import Photo
 from .picture import Picture
 from .question import Question
 from .quote import Quote
+from .table import Table
 from .tag import Tag
 from .underline import Underline
 
@@ -25,7 +26,8 @@ class FiftyOhmLaTeXRenderer(LaTeXRenderer):
             NonbreakingSpacesDots,
             Question,
             Picture,
-            Photo
+            Photo,
+            Table
         )
         self.question_renderer = question_renderer
 
@@ -113,4 +115,26 @@ class FiftyOhmLaTeXRenderer(LaTeXRenderer):
 \label{{{ref}}}"""
 
     def render_photo(self, token) :
-        return self.render_photo_helper(token.id, token.ref, token.text, token.number)
+        return self.render_photo_helper(token.id, token.ref, token.text, token.number)
+
+    def render_table(self, token):
+        alignments = token.alignment
+        align = ""
+        for alignment in alignments :
+            align += alignment
+
+        table = f"\\begin{{DARCtabular}}{{{align}}}\n"
+        
+        for row in token.children:
+            for j, cell in enumerate(row.children):
+                table += self.render_inner(cell)
+                if j < len(row.children) - 1:
+                    table += " & "
+            table += r"\\"+ "\n"
+
+        table += "\\end{DARCtabular}"
+        if token.caption != "" :
+            table += f"\\captionof{{figure}}{{{token.caption}}}\n"
+            table += f"\\label{{{token.name}}}\n"
+
+        return table

renderer/morse.py

@@ -87,4 +87,4 @@ class Morse(SpanToken):
         return result
 
     parse_inner = False
-    pattern = re.compile(r'"\[morse:\s*(.+)\]"')
+    pattern = re.compile(r'\[morse:\s*(.+)\]')

renderer/table.py

@@ -0,0 +1,78 @@
+import re
+
+from mistletoe import span_token
+from mistletoe.block_token import BlockToken
+
+
+class Table(BlockToken):
+    @staticmethod
+    def start(line):
+        return "|" in line
+
+    @classmethod
+    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
+        columns = [col.strip() for col in line.split("|")]  # Split by pipe |
+
+        alignment = []
+        header = "| "
+
+        for column in columns:
+            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)]
+
+        # Read table until the end: No more column definitions, caption, or empty 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
+            elif "|" not in next_line:
+                break
+            else:
+                line_buffer.append(next(lines))
+                next_line = lines.peek()
+
+        rows.extend([TableRow(line) for line in line_buffer])
+
+        return alignment, rows, name, caption
+
+    def __init__(self, match):
+        self.alignment, self.children, self.name, self.caption = match
+
+
+class TableRow(BlockToken):
+    @classmethod
+    def parse_row(cls, line):
+        line = line.strip().strip("|")
+        columns = [col.strip() for col in line.split("|")]
+
+        data = []
+
+        for column in columns:
+            data.append(TableCell(column))
+
+        return data
+
+    def __init__(self, match):
+        self.children = self.parse_row(match)
+
+
+class TableCell(BlockToken):
+    def __init__(self, match):
+        super().__init__(match, span_token.tokenize_inner)

requirements.txt

@@ -1,7 +1,12 @@
+beautifulsoup4==4.13.4
 coverage==7.6.9
 iniconfig==2.0.0
+jinja2==3.1.6
+markupsafe==3.0.2
 mistletoe==1.4.0
 packaging==24.2
 pluggy==1.5.0
 pytest==8.3.4
 pytest-cov==6.0.0
+soupsieve==2.7
+typing-extensions==4.13.2

test/acceptanceTest.md

@@ -79,7 +79,7 @@ CQ CQ hier ist DL1PZ mit einem allgemeinen Anruf, hier ist DL1PZ und hört.
 </indepth>
 
 <webmargin>
-| c | l | c | l | c | l |
+| c: | l: | c: | l: | c: | l: |
 |  |  |  |  |  |  |
 | A | [morse:a] | K | [morse:k] | U | [morse:u] |
 | B | [morse:b] | L | [morse:l] | V | [morse:v] |

test/test_html.py

@@ -8,59 +8,56 @@ from renderer.fifty_ohm_html_renderer import FiftyOhmHtmlRenderer
 
 
 def parse_katalog():
-    with open('data/fragenkatalog3b.json') as fragenkatalog_file:
+    with open("data/fragenkatalog3b.json") as fragenkatalog_file:
         fragenkatalog = json.load(fragenkatalog_file)
 
         questions = {}
 
-        for exampart in fragenkatalog["sections"] : 
+        for exampart in fragenkatalog["sections"]:
-            for chapter in exampart["sections"] :
+            for chapter in exampart["sections"]:
-                if "questions" in chapter :
+                if "questions" in chapter:
-                    for question in chapter["questions"] :
+                    for question in chapter["questions"]:
                         questions[question["number"]] = question
-                if "sections" in chapter :
+                if "sections" in chapter:
-                    for section in chapter["sections"] :
+                    for section in chapter["sections"]:
-                        for question in section["questions"] :
+                        for question in section["questions"]:
                             questions[question["number"]] = question
-        
+
         return questions
 
-env = Environment(loader=FileSystemLoader('templates/html'))
+
-question_template = env.get_template('question.html')
+env = Environment(loader=FileSystemLoader("templates/html"))
+question_template = env.get_template("question.html")
 questions = parse_katalog()
 
+
 def question_stub(input):
-    """ Combines the original question dataset from BNetzA with our internal metadata """
+    """Combines the original question dataset from BNetzA with our internal metadata"""
-    with open('data/metadata.json') as metadata_file :
+    with open("data/metadata.json") as metadata_file:
         metadata = json.load(metadata_file)
-        number = metadata[f"{input}"]["number"] # Fragennummer z.B. AB123
+        number = metadata[f"{input}"]["number"]  # Fragennummer z.B. AB123
 
         question = questions[number]
         metadata = metadata[f"{input}"]
 
-        if "answer_a" in question :
+        if "answer_a" in question:
-            answers = [
+            answers = [question["answer_a"], question["answer_b"], question["answer_c"], question["answer_d"]]
-                    question["answer_a"],
+        else:
-                    question["answer_b"],
-                    question["answer_c"],
-                    question["answer_d"]
-            ]
-        else :
             answers = []
 
-        if metadata["picture_a"] != "" :
+        if metadata["picture_a"] != "":
             answer_pictures = [
-                    metadata["picture_a"],
+                metadata["picture_a"],
-                    metadata["picture_b"],
+                metadata["picture_b"],
-                    metadata["picture_c"],
+                metadata["picture_c"],
-                    metadata["picture_d"]
+                metadata["picture_d"],
-                ]
+            ]
-        else :
+        else:
             answer_pictures = []
 
-        if "picture_question" in question :
+        if "picture_question" in question:
             picture_question = metadata["picture_question"]
-        else :
+        else:
             picture_question = ""
 
         return question_template.render(
@@ -69,9 +66,10 @@ def question_stub(input):
             layout=metadata["layout"],
             picture_question=picture_question,
             answers=answers,
-            answer_pictures=answer_pictures
+            answer_pictures=answer_pictures,
         )
 
+
 def test_html(capsys):
     with capsys.disabled():
         with open("test/acceptanceTest.md") as file:
@@ -79,5 +77,5 @@ def test_html(capsys):
             with FiftyOhmHtmlRenderer(question_stub) as renderer:
                 output = renderer.render(Document(content))
                 with open("test/acceptanceTest.html", "w") as output_file:
-                    pretty = BeautifulSoup(output, 'html.parser').prettify()
+                    pretty = BeautifulSoup(output, "html.parser").prettify()
                     output_file.write(pretty)

test/test_morse.py

@@ -8,11 +8,11 @@ from test.util import paragraph
 @pytest.mark.html
 def test_morse_html():
     assertions = {
-        '"[morse:a]"': FiftyOhmHtmlRenderer.render_morse_helper([[1, 2]]),
+        '[morse:a]': FiftyOhmHtmlRenderer.render_morse_helper([[1, 2]]),
-        '"[morse: a]"': FiftyOhmHtmlRenderer.render_morse_helper([[1, 2]]),
+        '[morse: a]': FiftyOhmHtmlRenderer.render_morse_helper([[1, 2]]),
-        '"[morse: aaa]"': FiftyOhmHtmlRenderer.render_morse_helper([[1, 2], [1, 2], [1, 2]]),
+        '[morse: aaa]': FiftyOhmHtmlRenderer.render_morse_helper([[1, 2], [1, 2], [1, 2]]),
-        '"[morse: a a]"': FiftyOhmHtmlRenderer.render_morse_helper([[1, 2], [3], [1, 2]]),
+        '[morse: a a]': FiftyOhmHtmlRenderer.render_morse_helper([[1, 2], [3], [1, 2]]),
-        '"[morse: db0]"': FiftyOhmHtmlRenderer.render_morse_helper([[2, 1, 1], [2, 1, 1, 1], [2, 2, 2, 2, 2]]),
+        '[morse: db0]': FiftyOhmHtmlRenderer.render_morse_helper([[2, 1, 1], [2, 1, 1, 1], [2, 2, 2, 2, 2]]),
     }
 
     for assertion in assertions:

test/test_table.py

@@ -0,0 +1,39 @@
+import pytest
+from mistletoe import Document
+
+from renderer.fifty_ohm_html_renderer import FiftyOhmHtmlRenderer
+from renderer.fifty_ohm_latex_renderer import FiftyOhmLaTeXRenderer
+
+
+@pytest.mark.html
+def test_table_html():
+    assertions = {
+        "| l: a |": '<table>\n<tr>\n<th style="text-align: left;">a</th>\n</tr>\n</table>\n',
+        "| r: a |": '<table>\n<tr>\n<th style="text-align: right;">a</th>\n</tr>\n</table>\n',
+        "| c: a |": '<table>\n<tr>\n<th style="text-align: center;">a</th>\n</tr>\n</table>\n',
+        "| l: a |\n| b |": '<table>\n<tr>\n<th style="text-align: left;">a</th>\n</tr>\n<tr>\n<td style="text-align: left;">b</td>\n</tr>\n</table>\n',
+        "| l: a |\n| b |\n| c |": '<table>\n<tr>\n<th style="text-align: left;">a</th>\n</tr>\n<tr>\n<td style="text-align: left;">b</td>\n</tr>\n<tr>\n<td style="text-align: left;">c</td>\n</tr>\n</table>\n',
+        "| a |\n| b |\n| c |": "<table>\n<tr>\n<th>a</th>\n</tr>\n<tr>\n<td>b</td>\n</tr>\n<tr>\n<td>c</td>\n</tr>\n</table>\n",
+        "| l: a |\n| *b* |": '<table>\n<tr>\n<th style="text-align: left;">a</th>\n</tr>\n<tr>\n<td style="text-align: left;"><em>b</em></td>\n</tr>\n</table>\n',
+        "| l: a |\n| b |\n[table:n_ab:A und B]": '<table>\n<tr>\n<th style="text-align: left;">a</th>\n</tr>\n<tr>\n<td style="text-align: left;">b</td>\n</tr>\n<caption>A und B</caption></table>\n',
+        "| l: *a* |": '<table>\n<tr>\n<th style="text-align: left;"><em>a</em></th>\n</tr>\n</table>\n',
+        "| l: a |\n| [morse:a] |": '<table>\n<tr>\n<th style="text-align: left;">a</th>\n</tr>\n<tr>\n<td style="text-align: left;"><span class="morse"><span class="morse_char">\n<span class="morse_char">\n▄</span>\n<span class="morse_char">\n▄▄▄</span>\n</span>\n</span></td>\n</tr>\n</table>\n',
+        "| l: abc | r: abc | X: abc |\n| 123 | 456 | 789 | \n | 987 | 654 | 321 | ": '<table>\n<tr>\n<th style="text-align: left;">abc</th>\n<th style="text-align: right;">abc</th>\n<th>abc</th>\n</tr>\n<tr>\n<td style="text-align: left;">123</td>\n<td style="text-align: right;">456</td>\n<td>789</td>\n</tr>\n<tr>\n<td style="text-align: left;">987</td>\n<td style="text-align: right;">654</td>\n<td>321</td>\n</tr>\n</table>\n',
+    }
+
+    with FiftyOhmHtmlRenderer() as renderer:
+        for assertion in assertions:
+            assert renderer.render(Document(assertion)) == assertions[assertion]
+
+
+@pytest.mark.latex
+def test_thematic_break_latex():
+    assertions = {
+        "| l: a |": "\\begin{DARCtabular}{l}\na\\\\\n\\end{DARCtabular}",
+        "| l: a |\n| *b* |": "\\begin{DARCtabular}{l}\na\\\\\n\\emph{b}\\\\\n\\end{DARCtabular}",
+        "| l: a | X: b |\n| c | d |": "\\begin{DARCtabular}{lX}\na & b\\\\\nc & d\\\\\n\\end{DARCtabular}",
+    }
+
+    with FiftyOhmLaTeXRenderer() as renderer:
+        for assertion in assertions:
+            assert renderer.render(Document(assertion)) == assertions[assertion]