Initial RAA parser poster copy

src/vk_parser_app/text_utils.py

@@ -0,0 +1,107 @@
+from __future__ import annotations
+
+import re
+
+
+def normalize_hash_tag(value: str, fallback: str = "source") -> str:
+    tag = (value or "").strip().lstrip("#").lower()
+    tag = re.sub(r"\s+", "_", tag)
+    tag = re.sub(r"[^\wа-яё_]+", "_", tag, flags=re.IGNORECASE)
+    tag = re.sub(r"_+", "_", tag).strip("_")
+    return tag or fallback
+
+
+def strip_trailing_hashtag_line(text: str) -> str:
+    lines = str(text or "").rstrip().splitlines()
+    while lines and not lines[-1].strip():
+        lines.pop()
+    if not lines:
+        return ""
+    parts = lines[-1].split()
+    if parts and all(part.startswith("#") and normalize_hash_tag(part, "") for part in parts):
+        lines.pop()
+    return "\n".join(lines).rstrip()
+
+
+def publication_hashtags(category_tag: str, source_tag: str) -> str:
+    tags = [
+        normalize_hash_tag(category_tag, "category"),
+        normalize_hash_tag(source_tag, "source"),
+    ]
+    return " ".join(f"#{tag}" for tag in tags if tag)
+
+
+import html
+
+
+def build_publication_text(
+    text: str,
+    category_tag: str,
+    source_tag: str,
+    format_title: bool = False,
+    parse_mode: str | None = None,
+) -> str:
+    base = strip_trailing_hashtag_line(text)
+    lines = base.splitlines()
+    sep_pattern = re.compile(r"^\s*[━—─\-=\*\#_]{2,}\s*$")
+    
+    cleaned_lines = []
+    for line in lines:
+        if sep_pattern.match(line):
+            cleaned_lines.append("")
+        else:
+            cleaned_lines.append(line)
+            
+    title_idx = None
+    for idx, line in enumerate(cleaned_lines):
+        if line.strip():
+            title_idx = idx
+            break
+
+    if title_idx is None:
+        return publication_hashtags(category_tag, source_tag)
+
+    formatted_lines = []
+    for idx, line in enumerate(cleaned_lines):
+        if idx == title_idx and format_title:
+            if parse_mode == "html":
+                escaped_title = html.escape(line.strip())
+                formatted_lines.append(f"<b>{escaped_title}</b>")
+            elif parse_mode == "markdown":
+                formatted_lines.append(f"**{line.strip()}**")
+            else:
+                formatted_lines.append(line.strip())
+            
+            # Ensure a newline break (blank line) after the bold title
+            if idx + 1 < len(cleaned_lines) and cleaned_lines[idx + 1].strip() != "":
+                formatted_lines.append("")
+        else:
+            if parse_mode == "html":
+                formatted_lines.append(html.escape(line))
+            else:
+                formatted_lines.append(line)
+
+    raw_body = "\n".join(formatted_lines)
+    # Collapse multiple consecutive blank lines to at most two newlines (\n\n)
+    normalized_body = re.sub(r"\n{3,}", "\n\n", raw_body).strip()
+    hashtags = publication_hashtags(category_tag, source_tag)
+    if hashtags:
+        return f"{normalized_body}\n\n{hashtags}" if normalized_body else hashtags
+    return normalized_body
+
+
+def parse_categories(value: object) -> list[str]:
+    if isinstance(value, list):
+        raw_items = [str(item) for item in value]
+    else:
+        text = str(value or "")
+        raw_items = re.split(r"[\n,|]+", text)
+    categories: list[str] = []
+    seen: set[str] = set()
+    for item in raw_items:
+        category = item.strip().strip("#")
+        key = category.lower()
+        if category and key not in seen:
+            categories.append(category)
+            seen.add(key)
+    return categories