parse existing tables from markdown

files/doc_regular_table/00-main/table.md

@@ -0,0 +1,14 @@
+# A basic table
+
+Some text before the table
+
+|           | cat A [unit] | cat 2 [unitB] | num ex [-] |
+|:----------|-------------:|--------------:|-----------:|
+| example1  |         4000 |          1.13 |          6 |
+| example4  |         4000 |          2.15 |          6 |
+| example9  |         4000 |          4.04 |          6 |
+| example10 |         4500 |             2 |          6 |
+
+Table: A basic table {#tbl:a-basic-table}
+
+And some text after

files/doc_regular_table/metadata.yaml

@@ -0,0 +1,5 @@
+lang: en-GB
+linkReferences: true
+nameInLink: true
+figPrefix: "Figure"
+tblPrefix: "Table"

src/paradoc/cli_app.py

@@ -11,7 +11,7 @@ def main(
     source_dir: str,
     report_name: str,
     auto_open: bool = False,
-    work_dir: str = None,
+    work_dir: str = "temp",
     export_format: ExportFormats = ExportFormats.DOCX,
 ):
     one = OneDoc(source_dir, work_dir=work_dir)

src/paradoc/common.py

@@ -41,6 +41,7 @@ class Table:
     add_link: bool = True
     md_instances: List[MarkDownFile] = field(default_factory=list)
     docx_instances: List[object] = field(default_factory=list)
+    link_name_override: str = None
 
     def __post_init__(self):
         if self.df is None:
@@ -61,9 +62,40 @@ def to_markdown(self, include_name_in_cell=False, flags=None):
             return tbl_str
         tbl_str += f"\n\nTable: {self.caption}"
         if self.add_link:
-            tbl_str += f" {{#tbl:{self.name}}}"
+            if self.link_name_override is None:
+                link_name = self.name
+            else:
+                link_name = self.link_name_override
+
+            tbl_str += f" {{#tbl:{link_name}}}"
         return tbl_str
 
+    @staticmethod
+    def from_markdown_str(table_str: str) -> Table:
+        """Parse a markdown table string and return a Table instance"""
+        lines = table_str.splitlines()
+        header = [x.strip() for x in lines[0].split("|")[1:-1]]
+        data = []
+        table_caption_str = None
+        for line in lines[2:]:
+            if line == "":
+                continue
+            if line.strip().startswith("Table:"):
+                table_caption_str = line.strip()
+                break
+            data.append([x.strip() for x in line.split("|")[1:-1]])
+
+        caption = table_caption_str.split("Table:")[1].strip()
+        caption = caption.split('{')[0].strip()
+        # Create a pandas DataFrame using the extracted header and data rows
+        df = pd.DataFrame(data, columns=header)
+        name = str(df.values[0][0])
+        tbl_ref = re.search(r"{#tbl:(.*?)}", table_str)
+        link_override = None
+        if tbl_ref is not None:
+            link_override = tbl_ref.group(1)
+        return Table(name=name, df=df, caption=caption, link_name_override=link_override)
+
 
 @dataclass
 class Figure:
@@ -110,6 +142,10 @@ def get_figures(self):
         regx = re.compile(r'<img src="(?P<file_path>.*?)" alt="(?P<caption>.*?)"\s*(?:width="(?P<width>.*?)"|)\/>')
         yield from regx.finditer(self.read_original_file())
 
+    def get_tables(self):
+        regx = re.compile(r'(\|.*?\nTable:.*?$)', re.MULTILINE | re.DOTALL)
+        yield from regx.finditer(self.read_original_file())
+
 
 class ExportFormats(str, Enum):
     DOCX = "docx"

src/paradoc/document.py

@@ -61,16 +61,16 @@ class OneDoc:
     FORMATS = ExportFormats
 
     def __init__(
-        self,
-        source_dir=None,
-        main_prefix="00-main",
-        app_prefix="01-app",
-        clean_build_dir=True,
-        create_dirs=False,
-        output_dir=None,
-        work_dir="temp",
-        use_default_html_style=True,
-        **kwargs,
+            self,
+            source_dir=None,
+            main_prefix="00-main",
+            app_prefix="01-app",
+            clean_build_dir=True,
+            create_dirs=False,
+            output_dir=None,
+            work_dir="temp",
+            use_default_html_style=True,
+            **kwargs,
     ):
         self.source_dir = pathlib.Path().resolve().absolute() if source_dir is None else pathlib.Path(source_dir)
         self.work_dir = pathlib.Path(work_dir).resolve().absolute()
@@ -138,7 +138,7 @@ def _setup(self, create_dirs, clean_build_dir):
                 # Check if the figure is commented out
                 # Get first newline right before regex search found start and till the end (capture entire line)
                 start = fig.string[: fig.start()].rfind("\n") + 1
-                end = fig.string[fig.start() :].find("\n") + fig.start()
+                end = fig.string[fig.start():].find("\n") + fig.start()
                 line = fig.string[start:end]
                 if line.startswith("[//]: #"):
                     continue
@@ -155,6 +155,10 @@ def _setup(self, create_dirs, clean_build_dir):
                     )
                 self.figures[caption] = Figure(name, caption, ref, file_path, md_instance=md_file)
 
+            for re_table in md_file.get_tables():
+                table = Table.from_markdown_str(re_table.group(1))
+                self.tables[table.name] = table
+
         if clean_build_dir is True:
             shutil.rmtree(self.build_dir, ignore_errors=True)
 

tests/tables/test_tables.py

@@ -16,3 +16,10 @@ def test_table(files_dir, test_dir):
     one.add_table("my_table_5", df, "No Space 3")
 
     one.compile("TableDoc")
+
+
+def test_regular_table(files_dir, test_dir):
+    report_dir = files_dir / "doc_regular_table"
+    one = OneDoc(report_dir, work_dir=test_dir / "doc_regular_table")
+
+    one.compile("TableDoc", export_format="docx")