python-substack 0.1.21__tar.gz → 0.1.23__tar.gz

{python_substack-0.1.21 → python_substack-0.1.23}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: python-substack
-Version: 0.1.21
+Version: 0.1.23
 Summary: A Python wrapper around the Substack API.
 License: MIT
 License-File: LICENSE
@@ -180,6 +180,21 @@ This is a paragraph with **bold** and *italic* text.
 """
 post.from_markdown(markdown_content, api=api)
 
+# Markdown footnotes are supported too. References become inline anchors and
+# definitions become footnote blocks, numbered by order of first appearance.
+# Labels can be numbers or names (e.g. [^1] or [^source]).
+footnote_markdown = """
+A claim that needs support.[^1] Another, with a named label.[^source]
+
+[^1]: The supporting detail, with a [link](https://example.com).
+[^source]: Author, *Title* (2025).
+"""
+post.from_markdown(footnote_markdown, api=api)
+
+# Or build footnotes manually:
+post.paragraph(content=[{"content": "Some claim."}]).footnote_anchor(1)
+post.footnote(1, "The note text, with **formatting** allowed.")
+
 draft = api.post_draft(post.get_draft())
 
 # set section (can only be done after first posting the draft)

{python_substack-0.1.21 → python_substack-0.1.23}/README.md

@@ -156,6 +156,21 @@ This is a paragraph with **bold** and *italic* text.
 """
 post.from_markdown(markdown_content, api=api)
 
+# Markdown footnotes are supported too. References become inline anchors and
+# definitions become footnote blocks, numbered by order of first appearance.
+# Labels can be numbers or names (e.g. [^1] or [^source]).
+footnote_markdown = """
+A claim that needs support.[^1] Another, with a named label.[^source]
+
+[^1]: The supporting detail, with a [link](https://example.com).
+[^source]: Author, *Title* (2025).
+"""
+post.from_markdown(footnote_markdown, api=api)
+
+# Or build footnotes manually:
+post.paragraph(content=[{"content": "Some claim."}]).footnote_anchor(1)
+post.footnote(1, "The note text, with **formatting** allowed.")
+
 draft = api.post_draft(post.get_draft())
 
 # set section (can only be done after first posting the draft)

{python_substack-0.1.21 → python_substack-0.1.23}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "python-substack"
-version = "0.1.21"
+version = "0.1.23"
 description = "A Python wrapper around the Substack API."
 authors = ["Paolo Mazza <mazzapaolo2019@gmail.com>"]
 license = "MIT"

{python_substack-0.1.21 → python_substack-0.1.23}/substack/__init__.py

@@ -3,7 +3,7 @@
 __author__ = "Paolo Mazza"
 __email__ = "mazzapaolo2019@gmail.com"
 __license__ = "MIT License"
-__version__ = "0.1.21"
+__version__ = "0.1.23"
 __url__ = "https://github.com/ma2za/python-substack"
 __download_url__ = "https://pypi.python.org/pypi/python-substack"
 __description__ = "A Python wrapper around the Substack API"

{python_substack-0.1.21 → python_substack-0.1.23}/substack/post.py

@@ -8,10 +8,32 @@ import json
 import re
 from typing import Dict, List
 
-__all__ = ["Post", "parse_inline"]
+__all__ = ["Post", "parse_inline", "tokens_to_text_nodes"]
 
 from substack.exceptions import SectionNotExistsException
 
+# Markdown footnotes: ``text.[^label]`` references and ``[^label]: definition`` lines.
+FOOTNOTE_REFERENCE_PATTERN = re.compile(r"\[\^([^\]]+)\]")
+FOOTNOTE_DEFINITION_PATTERN = re.compile(r"^\[\^([^\]]+)\]:\s?(.*)$")
+
+
+def tokens_to_text_nodes(tokens: List[Dict]) -> List[Dict]:
+    """Convert parse_inline() tokens to ProseMirror text nodes.
+
+    parse_inline() returns {"content": "text", "marks": [...]}.
+    ProseMirror expects {"type": "text", "text": "text", "marks": [...]}.
+    """
+    nodes = []
+    for token in tokens:
+        if not token or not token.get("content"):
+            continue
+        node = {"type": "text", "text": token["content"]}
+        marks = token.get("marks")
+        if marks:
+            node["marks"] = marks
+        nodes.append(node)
+    return nodes
+
 
 def parse_inline(text: str) -> List[Dict]:
     """
@@ -19,8 +41,11 @@ def parse_inline(text: str) -> List[Dict]:
     for use in the post content.
 
     Supported formatting:
+      - `code`: Text wrapped in backticks.
       - **Bold**: Text wrapped in double asterisks.
       - *Italic*: Text wrapped in single asterisks.
+      - ***Bold+Italic***: Text wrapped in triple asterisks.
+      - ~~Strikethrough~~: Text wrapped in double tildes.
       - [Links]: Text wrapped in square brackets followed by URL in parentheses.
 
     Args:
@@ -37,33 +62,50 @@ def parse_inline(text: str) -> List[Dict]:
         return []
 
     tokens = []
-    # Process text character by character to handle nested formatting
-    # We'll use regex to find all markdown patterns, then process them in order
 
-    # Find all markdown patterns: links, bold, italic
-    # Pattern order: links first (to avoid conflicts), then bold, then italic
+    # Pattern order matters: code > links > bold+italic > bold > italic > strikethrough
+    code_pattern = r'`([^`]+)`'
     link_pattern = r'\[([^\]]+)\]\(([^)]+)\)'
+    bold_italic_pattern = r'\*\*\*([^*]+)\*\*\*'
     bold_pattern = r'\*\*([^*]+)\*\*'
     italic_pattern = r'(?<!\*)\*([^*]+)\*(?!\*)'  # Not preceded or followed by *
+    strikethrough_pattern = r'~~([^~]+)~~'
 
     # Find all matches with their positions
     matches = []
+
+    # Inline code FIRST -- content inside backticks must not be parsed for other formatting
+    for match in re.finditer(code_pattern, text):
+        matches.append((match.start(), match.end(), "code", match.group(1), None))
+
+    # Links
     for match in re.finditer(link_pattern, text):
         # Skip if it's an image link (starts with ![)
         # But do NOT skip normal links at position 0.
         if match.start() == 0 or text[match.start()-1:match.start()+1] != "![":
-            matches.append((match.start(), match.end(), "link", match.group(1), match.group(2)))
+            if not any(start <= match.start() < end for start, end, _, _, _ in matches):
+                matches.append((match.start(), match.end(), "link", match.group(1), match.group(2)))
+
+    # Bold+italic combo
+    for match in re.finditer(bold_italic_pattern, text):
+        if not any(start <= match.start() < end for start, end, _, _, _ in matches):
+            matches.append((match.start(), match.end(), "bold_italic", match.group(1), None))
 
+    # Bold
     for match in re.finditer(bold_pattern, text):
-        # Check if this range is already covered by a link
         if not any(start <= match.start() < end for start, end, _, _, _ in matches):
             matches.append((match.start(), match.end(), "bold", match.group(1), None))
 
+    # Italic
     for match in re.finditer(italic_pattern, text):
-        # Check if this range is already covered by a link or bold
         if not any(start <= match.start() < end for start, end, _, _, _ in matches):
             matches.append((match.start(), match.end(), "italic", match.group(1), None))
 
+    # Strikethrough
+    for match in re.finditer(strikethrough_pattern, text):
+        if not any(start <= match.start() < end for start, end, _, _, _ in matches):
+            matches.append((match.start(), match.end(), "strikethrough", match.group(1), None))
+
     # Sort matches by position
     matches.sort(key=lambda x: x[0])
 
@@ -75,11 +117,21 @@ def parse_inline(text: str) -> List[Dict]:
             tokens.append({"content": text[last_pos:start]})
 
         # Add the formatted content
-        if match_type == "link":
+        if match_type == "code":
+            tokens.append({
+                "content": content,
+                "marks": [{"type": "code"}]
+            })
+        elif match_type == "link":
             tokens.append({
                 "content": content,
                 "marks": [{"type": "link", "attrs": {"href": url}}]
             })
+        elif match_type == "bold_italic":
+            tokens.append({
+                "content": content,
+                "marks": [{"type": "strong"}, {"type": "em"}]
+            })
         elif match_type == "bold":
             tokens.append({
                 "content": content,
@@ -90,6 +142,11 @@ def parse_inline(text: str) -> List[Dict]:
                 "content": content,
                 "marks": [{"type": "em"}]
             })
+        elif match_type == "strikethrough":
+            tokens.append({
+                "content": content,
+                "marks": [{"type": "strikethrough"}]
+            })
 
         last_pos = end
 
@@ -490,6 +547,186 @@ class Post:
 
         return self
 
+    def footnote_anchor(self, number: int):
+        """
+
+        Add an inline footnote reference (the superscript marker) to the last block.
+
+        Args:
+            number: The footnote number this anchor points to.
+
+        Returns:
+            Self for method chaining.
+
+        """
+        content = self.draft_body["content"][-1].get("content", [])
+        content += [{"type": "footnoteAnchor", "attrs": {"number": number}}]
+        self.draft_body["content"][-1]["content"] = content
+        return self
+
+    def footnote(self, number: int, content=None):
+        """
+
+        Append a footnote block (the note shown at the foot of the post).
+
+        Args:
+            number: The footnote number, matching a footnote_anchor.
+            content: Text string or list of inline token dicts. A plain string is
+                parsed for inline Markdown and may contain blank-line-separated
+                paragraphs; a parse_inline() token list or a list of ready text
+                nodes is also accepted (single paragraph).
+
+        Returns:
+            Self for method chaining.
+
+        """
+        paragraphs: List[Dict] = []
+        if isinstance(content, str):
+            # Blank lines separate paragraphs within the footnote.
+            for chunk in re.split(r"\n\s*\n", content):
+                chunk = chunk.strip()
+                if chunk:
+                    paragraphs.append(
+                        {"type": "paragraph", "content": tokens_to_text_nodes(parse_inline(chunk))}
+                    )
+        elif isinstance(content, list):
+            # Accept either parse_inline tokens ({"content": ...}) or text nodes.
+            if content and content[0].get("type") == "text":
+                text_nodes = content
+            else:
+                text_nodes = tokens_to_text_nodes(content)
+            paragraphs.append({"type": "paragraph", "content": text_nodes})
+
+        if not paragraphs:
+            paragraphs = [{"type": "paragraph", "content": []}]
+
+        node: Dict = {
+            "type": "footnote",
+            "attrs": {"number": number},
+            "content": paragraphs,
+        }
+        self.draft_body["content"] = self.draft_body.get("content", []) + [node]
+        return self
+
+    @staticmethod
+    def _extract_footnote_definitions(markdown_content: str):
+        """
+
+        Pull ``[^label]: definition`` lines out of the Markdown.
+
+        Definitions may wrap onto indented continuation lines and may contain
+        multiple paragraphs (blank line followed by an indented block). Returns
+        the body with definitions removed plus a {label: definition_text} mapping,
+        where paragraphs are separated by a blank line.
+
+        """
+        lines = markdown_content.split("\n")
+        body_lines: List[str] = []
+        definitions: Dict[str, str] = {}
+        in_code_fence = False
+        i = 0
+        while i < len(lines):
+            # Track fenced code blocks so footnote-like lines inside them are
+            # left untouched.
+            if lines[i].lstrip().startswith("```"):
+                in_code_fence = not in_code_fence
+                body_lines.append(lines[i])
+                i += 1
+                continue
+            match = None if in_code_fence else FOOTNOTE_DEFINITION_PATTERN.match(lines[i])
+            if match:
+                label, first = match.group(1), match.group(2)
+                paragraphs: List[str] = []
+                current = [first.strip()] if first.strip() else []
+                i += 1
+                while i < len(lines):
+                    line = lines[i]
+                    if line.strip() == "":
+                        # A blank line stays in the footnote only if the next
+                        # non-empty line is indented (a further paragraph).
+                        nxt = i + 1
+                        if (
+                            nxt < len(lines)
+                            and lines[nxt].strip()
+                            and lines[nxt][:1] in (" ", "\t")
+                        ):
+                            if current:
+                                paragraphs.append(" ".join(current))
+                                current = []
+                            i += 1
+                            continue
+                        break
+                    if line[:1] in (" ", "\t"):
+                        current.append(line.strip())
+                        i += 1
+                    else:
+                        break
+                if current:
+                    paragraphs.append(" ".join(current))
+                definitions[label] = "\n\n".join(paragraphs)
+            else:
+                body_lines.append(lines[i])
+                i += 1
+        return "\n".join(body_lines), definitions
+
+    @staticmethod
+    def _number_footnotes(markdown_content: str, definitions: Dict[str, str]):
+        """Number footnotes by order of first inline reference in the body."""
+        order: List[str] = []
+        for match in FOOTNOTE_REFERENCE_PATTERN.finditer(markdown_content):
+            label = match.group(1)
+            if label in definitions and label not in order:
+                order.append(label)
+        # Defined-but-unreferenced footnotes go last, in definition order.
+        for label in definitions:
+            if label not in order:
+                order.append(label)
+        return {label: index + 1 for index, label in enumerate(order)}
+
+    def _inject_footnote_anchors(self, node: Dict, numbers_by_label: Dict[str, int]):
+        """Recursively replace ``[^label]`` in text nodes with footnoteAnchor nodes."""
+        # Never rewrite the contents of a code block.
+        if node.get("type") == "codeBlock":
+            return
+        content = node.get("content")
+        if not isinstance(content, list):
+            return
+        new_content: List[Dict] = []
+        for child in content:
+            text = child.get("text", "")
+            has_code_mark = any(
+                mark.get("type") == "code" for mark in (child.get("marks") or [])
+            )
+            if (
+                child.get("type") == "text"
+                and not has_code_mark
+                and FOOTNOTE_REFERENCE_PATTERN.search(text)
+            ):
+                marks = child.get("marks")
+                last = 0
+                for match in FOOTNOTE_REFERENCE_PATTERN.finditer(text):
+                    label = match.group(1)
+                    if label not in numbers_by_label:
+                        continue  # Unknown label: leave the literal text in place.
+                    if match.start() > last:
+                        segment = {"type": "text", "text": text[last:match.start()]}
+                        if marks:
+                            segment["marks"] = marks
+                        new_content.append(segment)
+                    new_content.append(
+                        {"type": "footnoteAnchor", "attrs": {"number": numbers_by_label[label]}}
+                    )
+                    last = match.end()
+                if last < len(text):
+                    segment = {"type": "text", "text": text[last:]}
+                    if marks:
+                        segment["marks"] = marks
+                    new_content.append(segment)
+            else:
+                self._inject_footnote_anchors(child, numbers_by_label)
+                new_content.append(child)
+        node["content"] = new_content
+
     def from_markdown(self, markdown_content: str, api=None):
         """
         Parse Markdown content and add it to the post.
@@ -503,7 +740,13 @@ class Post:
           - Blockquotes: Lines starting with '>' (consecutive lines grouped)
           - Paragraphs: Regular text blocks
           - Bullet lists: Lines starting with '*' or '-'
-          - Inline formatting: **bold** and *italic* within paragraphs
+          - Ordered lists: Lines starting with '1.', '2.', etc.
+          - Horizontal rules: Lines with ---, ***, or ___
+          - Inline formatting: **bold**, *italic*, ***bold+italic***, `code`, ~~strikethrough~~
+          - Footnotes: ``text.[^label]`` references plus ``[^label]: definition``
+            lines. References become inline anchors and definitions become
+            footnote blocks, numbered by order of first appearance. Labels may be
+            numbers or names (e.g. ``[^1]`` or ``[^agi-book]``).
 
         Args:
             markdown_content: Markdown string to parse and add to the post.
@@ -517,6 +760,13 @@ class Post:
             >>> post = Post("Title", "Subtitle", user_id)
             >>> post.from_markdown("# Heading\\n\\nThis is **bold** text with [a link](https://example.com).")
         """
+        # Footnotes: extract ``[^label]: ...`` definitions and number them by
+        # order of first reference before parsing the rest of the body.
+        markdown_content, footnote_definitions = self._extract_footnote_definitions(
+            markdown_content
+        )
+        footnote_numbers = self._number_footnotes(markdown_content, footnote_definitions)
+
         lines = markdown_content.split("\n")
         blocks = []
         current_block: List[str] = []
@@ -593,6 +843,11 @@ class Post:
                 if not text_content:
                     continue
 
+                # Check for horizontal rule: ---, ***, ___
+                if re.match(r'^(\*{3,}|-{3,}|_{3,})\s*$', text_content):
+                    self.horizontal_rule()
+                    continue
+
                 # Process headings (lines starting with '#' characters)
                 if text_content.startswith("#"):
                     level = len(text_content) - len(text_content.lstrip("#"))
@@ -648,14 +903,15 @@ class Post:
 
                             self.add({"type": "captionedImage", "src": image_url})
 
-                # Process paragraphs, bullet lists, or blockquotes
+                # Process paragraphs, bullet lists, ordered lists, or blockquotes
                 else:
                     if "\n" in text_content:
-                        # Process each line, grouping consecutive bullets
-                        # into a single bullet_list node and consecutive
-                        # blockquote lines into a single blockquote node.
+                        # Process each line, grouping consecutive bullets/ordered items
+                        # into list nodes and consecutive blockquote lines into a
+                        # single blockquote node.
                         pending_bullets: List[List[Dict]] = []
                         pending_quotes: List[str] = []
+                        pending_ordered: List[List[Dict]] = []
 
                         def flush_bullets():
                             if not pending_bullets:
@@ -677,10 +933,7 @@ class Post:
                             paragraphs: List[Dict] = []
                             for quote_line in pending_quotes:
                                 tokens = parse_inline(quote_line)
-                                text_nodes = [
-                                    {"type": "text", "text": t["content"]}
-                                    for t in tokens if t
-                                ]
+                                text_nodes = tokens_to_text_nodes(tokens)
                                 if text_nodes:
                                     paragraphs.append({"type": "paragraph", "content": text_nodes})
                             node: Dict = {"type": "blockquote"}
@@ -689,20 +942,48 @@ class Post:
                             self.draft_body["content"].append(node)
                             pending_quotes.clear()
 
+                        def flush_ordered():
+                            if not pending_ordered:
+                                return
+                            list_items = []
+                            for item_nodes in pending_ordered:
+                                list_items.append({
+                                    "type": "list_item",
+                                    "content": [{"type": "paragraph", "content": item_nodes}],
+                                })
+                            self.draft_body["content"].append(
+                                {"type": "ordered_list", "content": list_items}
+                            )
+                            pending_ordered.clear()
+
                         for line in text_content.split("\n"):
                             line = line.strip()
                             if not line:
                                 flush_bullets()
+                                flush_ordered()
                                 flush_quotes()
                                 continue
 
                             # Check for blockquote marker
                             if line.startswith("> ") or line == ">":
                                 flush_bullets()
+                                flush_ordered()
                                 quote_text = line[2:] if line.startswith("> ") else ""
                                 pending_quotes.append(quote_text)
                                 continue
 
+                            # Check for ordered list marker
+                            ordered_match = re.match(r'^(\d+)\.\s+(.*)', line)
+                            if ordered_match:
+                                flush_bullets()
+                                flush_quotes()
+                                item_text = ordered_match.group(2).strip()
+                                tokens = parse_inline(item_text)
+                                text_nodes = tokens_to_text_nodes(tokens)
+                                if text_nodes:
+                                    pending_ordered.append(text_nodes)
+                                continue
+
                             # Check for bullet marker
                             bullet_text = None
                             if line.startswith("* "):
@@ -713,33 +994,56 @@ class Post:
                                 bullet_text = line[1:].strip()
 
                             if bullet_text is not None:
+                                flush_ordered()
                                 flush_quotes()
                                 tokens = parse_inline(bullet_text)
-                                if tokens:
-                                    pending_bullets.append(tokens)
+                                text_nodes = tokens_to_text_nodes(tokens)
+                                if text_nodes:
+                                    pending_bullets.append(text_nodes)
                             else:
                                 flush_bullets()
+                                flush_ordered()
                                 flush_quotes()
                                 tokens = parse_inline(line)
                                 self.add({"type": "paragraph", "content": tokens})
 
                         flush_bullets()
+                        flush_ordered()
                         flush_quotes()
                     else:
-                        # Single line — could be a blockquote or paragraph
+                        # Single line — blockquote, ordered list, or paragraph
                         if text_content.startswith("> ") or text_content == ">":
                             quote_text = text_content[2:] if text_content.startswith("> ") else ""
                             tokens = parse_inline(quote_text)
-                            text_nodes = [
-                                {"type": "text", "text": t["content"]}
-                                for t in tokens if t
-                            ]
+                            text_nodes = tokens_to_text_nodes(tokens)
                             para = {"type": "paragraph", "content": text_nodes} if text_nodes else {"type": "paragraph"}
                             self.draft_body["content"] = self.draft_body.get("content", []) + [
                                 {"type": "blockquote", "content": [para]}
                             ]
+
+                        elif re.match(r'^(\d+)\.\s+(.*)', text_content):
+                            ordered_match = re.match(r'^(\d+)\.\s+(.*)', text_content)
+                            item_text = ordered_match.group(2).strip()
+                            tokens = parse_inline(item_text)
+                            text_nodes = tokens_to_text_nodes(tokens)
+                            if text_nodes:
+                                list_item = {
+                                    "type": "list_item",
+                                    "content": [{"type": "paragraph", "content": text_nodes}],
+                                }
+                                self.draft_body["content"].append(
+                                    {"type": "ordered_list", "content": [list_item]}
+                                )
+
                         else:
                             tokens = parse_inline(text_content)
                             self.add({"type": "paragraph", "content": tokens})
 
+        # Footnotes: turn ``[^label]`` references into inline anchors, then append
+        # the footnote blocks in numbered order.
+        if footnote_numbers:
+            self._inject_footnote_anchors(self.draft_body, footnote_numbers)
+            for label, number in sorted(footnote_numbers.items(), key=lambda item: item[1]):
+                self.footnote(number, footnote_definitions[label])
+
         return self