Add cover pages to transcript/summary .docx with AI-generated descriptions; include combined .docx when both requested

2026-06-14 21:33:15 +00:00
parent efb34dd9ff
commit 7adca3d921
2 changed files with 194 additions and 48 deletions
@@ -284,25 +284,13 @@ def send_email(
        raise EmailError(f"Failed to send email: {e}")


-def create_transcript_docx(text: str, filename: str):
-    """
-    Create a .docx transcript with:
-      - 1.5" left margin, 1" right margin
-      - 12pt Courier
-      - Continuous line numbering on the left
-      - Speaker names capitalized and indented; spoken text further indented
-      - No section headings; use bold/underline only.
-    """
-    doc = Document()
-
-    # Set margins via section properties
+def _setup_docx_style(doc):
    section = doc.sections[0]
    section.left_margin = Inches(1.5)
    section.right_margin = Inches(1.0)
    section.top_margin = Inches(1.0)
    section.bottom_margin = Inches(1.0)

-    # Enable continuous line numbering on the left
    sectPr = section._sectPr
    lnNumType = sectPr.find(qn("w:lnNumType"))
    if lnNumType is None:
@@ -311,85 +299,82 @@ def create_transcript_docx(text: str, filename: str):
    lnNumType.set(qn("w:start"), "continuous")
    lnNumType.set(qn("w:countBy"), "1")

-    # Default font
    style = doc.styles["Normal"]
    font = style.font
    font.name = "Courier"
    font.size = Pt(12)

-    # Parse lines
+
+def _add_cover_page(doc, doc_type, date, description):
+    p_type = doc.add_paragraph()
+    run_type = p_type.add_run(doc_type)
+    run_type.bold = True
+    run_type.font.name = "Courier"
+    run_type.font.size = Pt(12)
+
+    p_date = doc.add_paragraph()
+    run_date = p_date.add_run(date)
+    run_date.font.name = "Courier"
+    run_date.font.size = Pt(12)
+
+    for _ in range(3):
+        doc.add_paragraph()
+
+    p_desc = doc.add_paragraph()
+    run_desc = p_desc.add_run(description)
+    run_desc.font.name = "Courier"
+    run_desc.font.size = Pt(12)
+
+    doc.add_page_break()
+
+
+def _add_transcript_content(doc, text):
    lines = text.strip().split("\n")
    for line in lines:
        line = line.strip()
        if not line:
            continue

-        # Try to parse: [00:00] SPEAKER: text
        m = re.match(r"\[(\d+:\d+(?::\d+)?)\]\s*(.+?):\s*(.*)", line)
        if m:
            ts, speaker, content = m.groups()
-            # Single paragraph: [timestamp] SPEAKER NAME underlined, then text inline
            p = doc.add_paragraph()
            p.paragraph_format.left_indent = Inches(0.25)

-            # Timestamp + speaker name (underline only, not bold)
            run_label = p.add_run(f"[{ts}] {speaker.upper()}:")
            run_label.bold = False
            run_label.underline = True
            run_label.font.name = "Courier"
            run_label.font.size = Pt(12)

-            # Space after colon (no underline)
            run_space = p.add_run(" ")
            run_space.bold = False
            run_space.underline = False
            run_space.font.name = "Courier"
            run_space.font.size = Pt(12)

-            # Spoken text (no underline, no bold)
            run_txt = p.add_run(content.strip())
            run_txt.bold = False
            run_txt.underline = False
            run_txt.font.name = "Courier"
            run_txt.font.size = Pt(12)
        else:
-            # Fallback for non-standard lines
            p = doc.add_paragraph()
            run = p.add_run(line)
            run.font.name = "Courier"
            run.font.size = Pt(12)

-    doc.save(filename)
-
-
-def create_summary_docx(text: str, filename: str):
-    """
-    Create a .docx summary with consistent font.
-    Translates markdown headings into WYSIWYG formatting:
-      - First heading: bold
-      - Second heading: italic
-      - Third heading: underline
-      - Fourth heading: italic + underline
-    No section headings; use bold/underline only.
-    """
-    doc = Document()
-    style = doc.styles["Normal"]
-    font = style.font
-    font.name = "Courier"
-    font.size = Pt(12)
-
-    heading_count = 0  # track headings in order

+def _add_summary_content(doc, text):
+    heading_count = 0
    for line in text.splitlines():
        stripped = line.strip()
        if not stripped:
            continue

-        # Detect markdown-style headings: #, ##, ###, ####
        m = re.match(r"^(#{1,4})\s+(.*)", stripped)
        if m:
            heading_count += 1
-            level = len(m.group(1))  # 1..4
            content = m.group(2).strip()

            p = doc.add_paragraph()
@@ -399,7 +384,6 @@ def create_summary_docx(text: str, filename: str):
            run.font.name = "Courier"
            run.font.size = Pt(12)

-            # Apply formatting based on this heading's ordinal position
            if heading_count == 1:
                run.bold = True
            elif heading_count == 2:
@@ -410,8 +394,84 @@ def create_summary_docx(text: str, filename: str):
                run.italic = True
                run.underline = True
        else:
-            # Normal text line
            p = doc.add_paragraph(stripped)
            p.paragraph_format.space_after = Pt(4)

+
+def create_transcript_docx(
+    text: str,
+    filename: str,
+    include_cover: bool = False,
+    cover_date: str = "",
+    cover_desc: str = "",
+):
+    """
+    Create a .docx transcript with:
+      - 1.5" left margin, 1" right margin
+      - 12pt Courier
+      - Continuous line numbering on the left
+      - Optional cover page with type, date, and AI-generated description.
+    """
+    doc = Document()
+    _setup_docx_style(doc)
+
+    if include_cover and cover_date and cover_desc:
+        _add_cover_page(doc, "TRANSCRIPT", cover_date, cover_desc)
+
+    _add_transcript_content(doc, text)
+    doc.save(filename)
+
+
+def create_summary_docx(
+    text: str,
+    filename: str,
+    include_cover: bool = False,
+    cover_date: str = "",
+    cover_desc: str = "",
+):
+    """
+    Create a .docx summary with consistent font and heading styles.
+    Optional cover page with type, date, and AI-generated description.
+    """
+    doc = Document()
+    _setup_docx_style(doc)
+
+    if include_cover and cover_date and cover_desc:
+        _add_cover_page(doc, "SUMMARY", cover_date, cover_desc)
+
+    _add_summary_content(doc, text)
+    doc.save(filename)
+
+
+def create_combined_docx(
+    transcript_text: str,
+    summary_text: str,
+    filename: str,
+    transcript_cover_date: str,
+    transcript_cover_desc: str,
+    summary_cover_date: str,
+    summary_cover_desc: str,
+):
+    """
+    Create a combined .docx with:
+      - Transcript cover page
+      - Transcript content
+      - Page break
+      - Summary cover page
+      - Summary content
+    """
+    doc = Document()
+    _setup_docx_style(doc)
+
+    # Transcript cover page
+    _add_cover_page(doc, "TRANSCRIPT", transcript_cover_date, transcript_cover_desc)
+    _add_transcript_content(doc, transcript_text)
+
+    # Page break before summary
+    doc.add_page_break()
+
+    # Summary cover page
+    _add_cover_page(doc, "SUMMARY", summary_cover_date, summary_cover_desc)
+    _add_summary_content(doc, summary_text)
+
    doc.save(filename)
@@ -14,6 +14,7 @@ from .summarizer import SummarizerClient, SummarizerError
 from .misc import setup_logging
 from .email_sender import send_email, EmailError, load_template
 from .email_sender import create_transcript_docx, create_summary_docx
+from .email_sender import create_combined_docx

 logger = logging.getLogger("scraibe.tasks")

@@ -405,6 +406,58 @@ def process_transcription_task(
                )
                speaker_map = {}

+        # 3c) Generate short cover-page descriptions using summarizer
+        transcript_cover_desc = ""
+        summary_cover_desc = ""
+        today_str = datetime.utcnow().strftime("%Y-%m-%d")
+
+        try:
+            scraibe._ensure_summarizer()
+            summarizer = scraibe._summarizer
+
+            # Transcript description
+            prompt = (
+                "In one short sentence (max 25 words), describe the content of this transcript "
+                "for use on a cover page. Do not include speaker names. Output only the sentence.\n\n"
+                + transcript_text
+            )
+            response = summarizer._chat_completion(
+                messages=[{"role": "user", "content": prompt}],
+                temperature=0.3,
+                max_tokens=60,
+            )
+            transcript_cover_desc = (
+                (response or {}).get("choices", [{}])[0].get("message", {}).get("content", "")
+            ).strip()
+            if not transcript_cover_desc:
+                transcript_cover_desc = "Transcript of the recorded conversation."
+            logger.info("Transcript cover description: %s", transcript_cover_desc)
+
+            # Summary description (if summary requested)
+            if include_summary:
+                prompt = (
+                    "In one short sentence (max 25 words), describe the content of this summary "
+                    "for use on a cover page. Output only the sentence.\n\n"
+                    + summary_text
+                )
+                response = summarizer._chat_completion(
+                    messages=[{"role": "user", "content": prompt}],
+                    temperature=0.3,
+                    max_tokens=60,
+                )
+                summary_cover_desc = (
+                    (response or {}).get("choices", [{}])[0].get("message", {}).get("content", "")
+                ).strip()
+                if not summary_cover_desc:
+                    summary_cover_desc = "Summary of the recorded conversation."
+                logger.info("Summary cover description: %s", summary_cover_desc)
+
+        except (SummarizerError, Exception) as e:
+            logger.warning("Failed to generate cover descriptions: %s", e)
+            transcript_cover_desc = transcript_cover_desc or "Transcript of the recorded conversation."
+            if include_summary:
+                summary_cover_desc = summary_cover_desc or "Summary of the recorded conversation."
+
        # 4) Prepare files

        # Transcript .md
@@ -414,9 +467,15 @@ def process_transcription_task(
            f.write(transcript_text)
        temp_files.append(md_transcript_path)

-        # Transcript .docx
+        # Transcript .docx (standalone, with cover page)
        docx_transcript_path = _safe_filename("TRANSCRIPT", local, date_tag, ".docx")
-        create_transcript_docx(transcript_text, docx_transcript_path)
+        create_transcript_docx(
+            transcript_text,
+            docx_transcript_path,
+            include_cover=True,
+            cover_date=today_str,
+            cover_desc=transcript_cover_desc,
+        )
        temp_files.append(docx_transcript_path)

        # JSON as SOURCE
@@ -440,17 +499,42 @@ def process_transcription_task(
        temp_files.append(json_path)

        # Summary files (if present)
+        md_summary_path = None
+        docx_summary_path = None
+        docx_combined_path = None
+
        if summary_text:
+            # Summary .md
            md_summary_path = _safe_filename("SUMMARY", local, date_tag, ".md")
            with open(md_summary_path, "w", encoding="utf-8") as f:
                f.write("# Summary\n\n")
                f.write(summary_text)
            temp_files.append(md_summary_path)

+            # Summary .docx (standalone, with cover page)
            docx_summary_path = _safe_filename("SUMMARY", local, date_tag, ".docx")
-            create_summary_docx(summary_text, docx_summary_path)
+            create_summary_docx(
+                summary_text,
+                docx_summary_path,
+                include_cover=True,
+                cover_date=today_str,
+                cover_desc=summary_cover_desc,
+            )
            temp_files.append(docx_summary_path)

+            # Combined .docx: Transcript + Summary with cover pages
+            docx_combined_path = _safe_filename("TRANSCRIPT_AND_SUMMARY", local, date_tag, ".docx")
+            create_combined_docx(
+                transcript_text=transcript_text,
+                summary_text=summary_text,
+                filename=docx_combined_path,
+                transcript_cover_date=today_str,
+                transcript_cover_desc=transcript_cover_desc,
+                summary_cover_date=today_str,
+                summary_cover_desc=summary_cover_desc,
+            )
+            temp_files.append(docx_combined_path)
+
        attachments = [
            md_transcript_path,
            docx_transcript_path,
@@ -458,6 +542,8 @@ def process_transcription_task(
        ]
        if summary_text:
            attachments += [md_summary_path, docx_summary_path]
+            if docx_combined_path:
+                attachments.append(docx_combined_path)

        # 5) Send success email
        send_success_email(