Update pdf_reports.py

pdf_reports.py

@@ -1,186 +1,196 @@
-# pdf_reports.py
 import os
-import uuid
-from datetime import datetime
 from reportlab.lib.pagesizes import A4
 from reportlab.lib.styles import ParagraphStyle, getSampleStyleSheet
-from reportlab.lib.enums import TA_CENTER, TA_RIGHT, TA_JUSTIFY, TA_LEFT
+from reportlab.lib.enums import TA_CENTER, TA_RIGHT, TA_LEFT, TA_JUSTIFY
 from reportlab.lib import colors
-from reportlab.platypus import (
-    BaseDocTemplate, PageTemplate, Frame, Paragraph, Spacer, Table, TableStyle
-)
+from reportlab.platypus import BaseDocTemplate, PageTemplate, Frame, Paragraph, Spacer, Table, TableStyle, KeepTogether
 from reportlab.pdfbase.ttfonts import TTFont
 from reportlab.pdfbase import pdfmetrics
+from datetime import datetime
 
-# Try to load a nicer serif; fallback to Times
+# Ensure output directory exists
+out_dir = "/mnt/data"
+os.makedirs(out_dir, exist_ok=True)
+
+# Register a serif-like font if available (fallback to Times)
+body_font = 'Times-Roman'
 try:
-    pdfmetrics.registerFont(TTFont('DejaVuSerif', '/usr/share/fonts/truetype/dejavu/DejaVuSerif.ttf'))
-    BODY_FONT = 'DejaVuSerif'
+    font_path = '/usr/share/fonts/truetype/dejavu/DejaVuSerif.ttf'
+    if os.path.exists(font_path):
+        pdfmetrics.registerFont(TTFont('DejaVuSerif', font_path))
+        body_font = 'DejaVuSerif'
+    else:
+        # try common alternative (mac)
+        font_path2 = '/Library/Fonts/DejaVuSerif.ttf'
+        if os.path.exists(font_path2):
+            pdfmetrics.registerFont(TTFont('DejaVuSerif', font_path2))
+            body_font = 'DejaVuSerif'
 except Exception:
-    BODY_FONT = 'Times-Roman'
-
-
-def _build_doc(filepath, title_text, tiles, counts, sections, matched_sources, footer_text):
-    PAGE_WIDTH, PAGE_HEIGHT = A4
-    MARGIN = 36
-    usable_width = PAGE_WIDTH - 2 * MARGIN
-
-    styles = getSampleStyleSheet()
-    styles.add(ParagraphStyle(name='ReportTitle', fontName=BODY_FONT, fontSize=18, alignment=TA_CENTER, leading=22))
-    styles.add(ParagraphStyle(name='SmallRight', fontName=BODY_FONT, fontSize=9, alignment=TA_RIGHT, textColor=colors.HexColor("#555555")))
-    styles.add(ParagraphStyle(name='TileBig', fontName=BODY_FONT, fontSize=30, alignment=TA_CENTER, leading=32))
-    styles.add(ParagraphStyle(name='TileLabel', fontName=BODY_FONT, fontSize=10, alignment=TA_CENTER, textColor=colors.HexColor("#666666")))
-    styles.add(ParagraphStyle(name='SectionHeading', fontName=BODY_FONT, fontSize=13, spaceBefore=8, spaceAfter=4, leading=15))
-    styles.add(ParagraphStyle(name='Body', fontName=BODY_FONT, fontSize=11, leading=15, alignment=TA_JUSTIFY))
-    styles.add(ParagraphStyle(name='HighlightYellow', fontName=BODY_FONT, fontSize=11, leading=15, backColor=colors.HexColor("#fff3b0"), alignment=TA_JUSTIFY))
-    styles.add(ParagraphStyle(name='HighlightRed', fontName=BODY_FONT, fontSize=11, leading=15, backColor=colors.HexColor("#ffd6d6"), alignment=TA_JUSTIFY))
-    styles.add(ParagraphStyle(name='Footer', fontName=BODY_FONT, fontSize=9, alignment=TA_RIGHT, textColor=colors.HexColor("#666666")))
-    styles.add(ParagraphStyle(name='MatchedHeader', fontName=BODY_FONT, fontSize=12, leading=14, alignment=TA_LEFT, spaceBefore=6, spaceAfter=6))
-
-    def header_footer(canvas, doc):
-        canvas.saveState()
-        date_str = datetime.now().strftime("%d %B %Y, %H:%M")
-        canvas.setFont(BODY_FONT, 9)
-        canvas.setFillColor(colors.HexColor("#555555"))
-        canvas.drawString(MARGIN, PAGE_HEIGHT - MARGIN + 8, f"Date: {date_str}")
-        canvas.setFont(BODY_FONT, 16)
-        canvas.setFillColor(colors.black)
-        canvas.drawCentredString(PAGE_WIDTH / 2.0, PAGE_HEIGHT - MARGIN + 4, title_text)
-        canvas.setFont(BODY_FONT, 9)
-        canvas.setFillColor(colors.HexColor("#666666"))
-        canvas.drawRightString(PAGE_WIDTH - MARGIN, MARGIN - 10, f"Page {doc.page}")
-        canvas.restoreState()
-
-    doc = BaseDocTemplate(filepath, pagesize=A4,
-                          leftMargin=MARGIN, rightMargin=MARGIN,
-                          topMargin=MARGIN, bottomMargin=MARGIN)
-    frame = Frame(MARGIN, MARGIN, usable_width, PAGE_HEIGHT - 2 * MARGIN, id='normal')
-    template = PageTemplate(id='report', frames=[frame], onPage=header_footer)
-    doc.addPageTemplates([template])
-
-    story = []
-
-    # Tiles (4 small summary tiles)
-    tile_values = tiles  # list of 4 dicts: {'value': '12%', 'label': 'Plagiarism'}
-    tiles_data = [
-        [Paragraph(f"<b>{tile_values[i]['value']}</b>", styles['TileBig']) for i in range(4)],
-        [Paragraph(tile_values[i]['label'], styles['TileLabel']) for i in range(4)]
-    ]
-    tiles_table = Table(tiles_data, colWidths=[usable_width / 4.0] * 4, rowHeights=[46, 18])
-    tiles_table.setStyle(TableStyle([
-        ('BACKGROUND', (0, 0), (-1, 0), colors.HexColor("#f7f7f9")),
-        ('ALIGN', (0, 0), (-1, -1), 'CENTER'),
-        ('VALIGN', (0, 0), (-1, -1), 'MIDDLE'),
-        ('BOX', (0, 0), (-1, -1), 0.6, colors.HexColor("#dddddd")),
-    ]))
-    story.append(tiles_table)
-    story.append(Spacer(1, 12))
-
-    # Counts row
-    if counts:
-        counts_table = Table([list(counts.keys()), list(counts.values())],
-                             colWidths=[usable_width / len(counts)] * len(counts))
-        counts_table.setStyle(TableStyle([
-            ('BACKGROUND', (0, 0), (-1, 0), colors.HexColor("#f4f6f7")),
-            ('ALIGN', (0, 0), (-1, -1), 'CENTER'),
-            ('BOX', (0, 0), (-1, -1), 0.5, colors.HexColor("#e6e6e6")),
-        ]))
-        story.append(counts_table)
-        story.append(Spacer(1, 12))
-
-    # Sections + highlighting: sections is list of dicts: {'heading': 'Abstract', 'paragraphs': [p1, p2...]}
-    for sec in sections or []:
-        if sec.get('heading'):
-            story.append(Paragraph(sec['heading'], styles['SectionHeading']))
-        for para in sec.get('paragraphs', []):
-            # para may be dict {'text': '...', 'highlight':'yellow'/'red'/None}
-            if isinstance(para, dict):
-                text = para.get('text', '')
-                hl = para.get('highlight')
-                if hl == 'yellow':
-                    story.append(Paragraph(text, styles['HighlightYellow']))
-                elif hl == 'red':
-                    story.append(Paragraph(text, styles['HighlightRed']))
-                else:
-                    story.append(Paragraph(text, styles['Body']))
-            else:
-                story.append(Paragraph(para, styles['Body']))
-            story.append(Spacer(1, 6))
-
-    story.append(Spacer(1, 10))
-
-    # Matched Sources table (if any)
-    if matched_sources:
-        story.append(Paragraph("Matched Sources", styles['MatchedHeader']))
-        ms_table_data = [["#", "Source Title", "URL", "Similarity"]]
-        for i, ms in enumerate(matched_sources, start=1):
-            title_par = Paragraph(ms.get('title', ''), styles['Body'])
-            url_par = Paragraph(f'<link href="{ms.get("url", "")}">{ms.get("url", "")}</link>', styles['Body'])
-            ms_table_data.append([str(i), title_par, url_par, ms.get('similarity', '')])
-        ms_table = Table(ms_table_data, colWidths=[30, usable_width * 0.35, usable_width * 0.45, usable_width * 0.15])
-        ms_table.setStyle(TableStyle([
-            ('BACKGROUND', (0, 0), (-1, 0), colors.HexColor("#f2f4f5")),
-            ('TEXTCOLOR', (0, 0), (-1, 0), colors.HexColor("#333333")),
-            ('ALIGN', (0, 0), (-1, 0), 'CENTER'),
-            ('VALIGN', (0, 0), (-1, -1), 'MIDDLE'),
-            ('BOX', (0, 0), (-1, -1), 0.6, colors.HexColor("#e0e0e0")),
-            ('INNERGRID', (0, 0), (-1, -1), 0.4, colors.HexColor("#efefef")),
-            ('LEFTPADDING', (1, 1), (1, -1), 6),
-            ('LEFTPADDING', (2, 1), (2, -1), 6),
-        ]))
-        story.append(ms_table)
-        story.append(Spacer(1, 14))
-
-    # Matched Source Overview block (footer_text)
-    if footer_text:
-        matched_table = Table(
-            [[Paragraph("<b>Matched Source Overview</b>", styles['Body'])],
-             [Paragraph(footer_text, styles['Body'])]],
-            colWidths=[usable_width]
-        )
-        matched_table.setStyle(TableStyle([
-            ('BACKGROUND', (0, 0), (-1, 0), colors.HexColor("#f7fafb")),
-            ('BOX', (0, 0), (-1, -1), 0.5, colors.HexColor("#e6e6e6")),
-            ('LEFTPADDING', (0, 0), (-1, -1), 8),
-            ('RIGHTPADDING', (0, 0), (-1, -1), 8),
-            ('TOPPADDING', (0, 0), (-1, -1), 6),
-            ('BOTTOMPADDING', (0, 0), (-1, -1), 6),
-        ]))
-        story.append(matched_table)
-        story.append(Spacer(1, 24))
-
-    story.append(Paragraph("Generated by TrueWrite Scan • https://gopalkrushnamahapatra-truewrite-scan.static.hf.space", styles['Footer']))
-
-    doc.build(story)
-
-
-def generate_report(report_type: str, out_dir: str = "/tmp", **kwargs) -> str:
-    """
-    report_type: "ai" | "grammar" | "plagiarism"
-    kwargs expected:
-      - title_text: str
-      - tiles: list of 4 dicts [{'value': '12%', 'label': 'Plagiarism'}, ...]
-      - counts: dict {'Words': 950, ...}
-      - sections: list [{'heading':'','paragraphs':[...]}]
-      - matched_sources: list [{'title','url','similarity'}]
-      - footer_text: str
-    Returns: path to generated PDF
-    """
-    os.makedirs(out_dir, exist_ok=True)
-    filename = f"{report_type}_report_{uuid.uuid4().hex[:8]}.pdf"
-    filepath = os.path.join(out_dir, filename)
-
-    title_text = kwargs.get('title_text', "Report")
-    tiles = kwargs.get('tiles') or [
-        {'value': '0%', 'label': 'Plagiarism'},
-        {'value': '0%', 'label': 'Exact Match'},
-        {'value': '0%', 'label': 'Partial Match'},
-        {'value': '100%', 'label': 'Unique'},
-    ]
-    counts = kwargs.get('counts') or {}
-    sections = kwargs.get('sections') or []
-    matched_sources = kwargs.get('matched_sources') or []
-    footer_text = kwargs.get('footer_text') or ''
-
-    _build_doc(filepath, title_text, tiles, counts, sections, matched_sources, footer_text)
-    return filepath
+    body_font = 'Times-Roman'
+
+pdf_path = os.path.join(out_dir, "TrueWriteScan_Duplichecker_PixelPerfect_Report.pdf")
+PAGE_WIDTH, PAGE_HEIGHT = A4
+MARGIN = 36
+usable_width = PAGE_WIDTH - 2*MARGIN
+
+styles = getSampleStyleSheet()
+styles.add(ParagraphStyle(name='ReportTitle', fontName=body_font, fontSize=18, alignment=TA_CENTER, leading=22))
+styles.add(ParagraphStyle(name='SmallRight', fontName=body_font, fontSize=9, alignment=TA_RIGHT, textColor=colors.HexColor("#555555")))
+styles.add(ParagraphStyle(name='TileBig', fontName=body_font, fontSize=30, alignment=TA_CENTER, leading=32))
+styles.add(ParagraphStyle(name='TileLabel', fontName=body_font, fontSize=10, alignment=TA_CENTER, textColor=colors.HexColor("#666666")))
+styles.add(ParagraphStyle(name='SectionHeading', fontName=body_font, fontSize=13, spaceBefore=8, spaceAfter=4, leading=15))
+styles.add(ParagraphStyle(name='Body', fontName=body_font, fontSize=11, leading=15, alignment=TA_JUSTIFY))
+styles.add(ParagraphStyle(name='HighlightYellow', fontName=body_font, fontSize=11, leading=15, backColor=colors.HexColor("#fff3b0"), alignment=TA_JUSTIFY))
+styles.add(ParagraphStyle(name='HighlightRed', fontName=body_font, fontSize=11, leading=15, backColor=colors.HexColor("#ffd6d6"), alignment=TA_JUSTIFY))
+styles.add(ParagraphStyle(name='Footer', fontName=body_font, fontSize=9, alignment=TA_RIGHT, textColor=colors.HexColor("#666666")))
+styles.add(ParagraphStyle(name='MatchedHeader', fontName=body_font, fontSize=12, leading=14, alignment=TA_LEFT, spaceBefore=6, spaceAfter=6))
+
+def header_footer(canvas, doc):
+    canvas.saveState()
+    date_str = datetime.now().strftime("%d %B %Y, %H:%M")
+    canvas.setFont(body_font, 9)
+    canvas.setFillColor(colors.HexColor("#555555"))
+    canvas.drawString(MARGIN, PAGE_HEIGHT - MARGIN + 8, f"Date: {date_str}")
+    canvas.setFont(body_font, 16)
+    canvas.setFillColor(colors.black)
+    canvas.drawCentredString(PAGE_WIDTH/2.0, PAGE_HEIGHT - MARGIN + 4, "Plagiarism Scan Report")
+    canvas.setFont(body_font, 9)
+    canvas.setFillColor(colors.HexColor("#666666"))
+    canvas.drawRightString(PAGE_WIDTH - MARGIN, MARGIN - 10, f"Page {doc.page}")
+    canvas.restoreState()
+
+doc = BaseDocTemplate(pdf_path, pagesize=A4,
+                      leftMargin=MARGIN, rightMargin=MARGIN,
+                      topMargin=MARGIN, bottomMargin=MARGIN)
+
+frame = Frame(MARGIN, MARGIN, usable_width, PAGE_HEIGHT - 2*MARGIN, id='normal')
+template = PageTemplate(id='report', frames=[frame], onPage=header_footer)
+doc.addPageTemplates([template])
+
+# Optionally set metadata
+doc.title = "TrueWrite Scan — Plagiarism Report"
+doc.author = "TrueWrite Scan"
+
+story = []
+
+# Tiles (colors matched to Duplichecker-like look)
+tiles_data = [
+    [Paragraph("<b>12%</b>", styles['TileBig']), Paragraph("<b>4%</b>", styles['TileBig']), Paragraph("<b>8%</b>", styles['TileBig']), Paragraph("<b>88%</b>", styles['TileBig'])],
+    [Paragraph("Plagiarism", styles['TileLabel']), Paragraph("Exact Match", styles['TileLabel']), Paragraph("Partial Match", styles['TileLabel']), Paragraph("Unique", styles['TileLabel'])]
+]
+tiles_table = Table(tiles_data, colWidths=[usable_width/4.0]*4, rowHeights=[46, 18])
+tiles_table.setStyle(TableStyle([
+    ('BACKGROUND', (0,0), (-1,0), colors.HexColor("#f7f7f9")),
+    ('ALIGN', (0,0), (-1,-1), 'CENTER'),
+    ('VALIGN', (0,0), (-1,-1), 'MIDDLE'),
+    ('BOX', (0,0), (-1,-1), 0.6, colors.HexColor("#dddddd")),
+]))
+story.append(tiles_table)
+story.append(Spacer(1, 12))
+
+# Counts row
+counts = [
+    ['Words', 'Characters', 'Sentences', 'Paragraphs', 'Read Time'],
+    ['950', '7138', '43', '16', '5 minute(s)']
+]
+counts_table = Table(counts, colWidths=[usable_width/5.0]*5)
+counts_table.setStyle(TableStyle([
+    ('BACKGROUND', (0,0), (-1,0), colors.HexColor("#f4f6f7")),
+    ('ALIGN', (0,0), (-1,-1), 'CENTER'),
+    ('BOX', (0,0), (-1,-1), 0.5, colors.HexColor("#e6e6e6")),
+]))
+story.append(counts_table)
+story.append(Spacer(1, 12))
+
+# Sample content with highlighted spans (we'll highlight whole sentences as blocks)
+story.append(Paragraph("Abstract—", styles['SectionHeading']))
+
+para1 = ("Using AI, machine learning, and advanced computing together opens up a lot of possibilities to effectively tackle "
+         "major issues such as climate resilience and health equity.")
+para2 = ("The research looks at the ways in which AI can be a social developmental tool by aiding first responders during calamities "
+         "as well as healthcare personalization.")
+para3 = ("In particular, it mentions the use of deep learning and NLP for better prediction, efficient resource management, and improved accessibility of services.")
+para4 = ("Moreover, the paper points to the issues that revolve around the ethics of the technology and the need for transparent models.")
+
+story.append(Paragraph(para1, styles['Body']))
+story.append(Spacer(1, 6))
+story.append(Paragraph(para2, styles['HighlightYellow']))
+story.append(Spacer(1, 6))
+story.append(Paragraph(para3, styles['HighlightRed']))
+story.append(Spacer(1, 6))
+story.append(Paragraph(para4, styles['Body']))
+story.append(Spacer(1, 10))
+
+story.append(Paragraph("I. Introduction", styles['SectionHeading']))
+for _ in range(3):
+    story.append(Paragraph(para1 + " " + para2, styles['Body']))
+    story.append(Spacer(1, 6))
+
+story.append(Paragraph("II. Literature Review", styles['SectionHeading']))
+for i in range(4):
+    # Insert occasional highlighted sentences
+    story.append(Paragraph(para2, styles['Body']))
+    story.append(Paragraph(para3, styles['HighlightYellow'] if i % 2 == 0 else styles['Body']))
+    story.append(Spacer(1, 6))
+
+story.append(Paragraph("III. Methodology", styles['SectionHeading']))
+for _ in range(4):
+    story.append(Paragraph(para3, styles['Body']))
+    story.append(Spacer(1, 6))
+
+story.append(Spacer(1, 12))
+
+# Matched sources header
+story.append(Paragraph("Matched Sources", styles['MatchedHeader']))
+
+matched_sources = [
+    {"title": "AI for Social Good", "url": "https://example.com/ai-social-good", "similarity": "42%"},
+    {"title": "Deep Learning Predictions", "url": "https://example.org/deep-learning", "similarity": "67%"},
+    {"title": "NLP and Resource Management", "url": "https://sample.org/nlp-resource-management", "similarity": "18%"}
+]
+
+ms_table_data = [["#", "Source Title", "URL", "Similarity"]]
+for i, ms in enumerate(matched_sources, start=1):
+    title_par = Paragraph(ms['title'], styles['Body'])
+    # Use <a href="..."> for clickable link
+    url_par = Paragraph(f'<a href="{ms["url"]}">{ms["url"]}</a>', styles['Body'])
+    ms_table_data.append([str(i), title_par, url_par, ms['similarity']])
+
+ms_table = Table(ms_table_data, colWidths=[30, usable_width*0.35, usable_width*0.45, usable_width*0.15])
+ms_table.setStyle(TableStyle([
+    ('BACKGROUND', (0,0), (-1,0), colors.HexColor("#f2f4f5")),
+    ('TEXTCOLOR', (0,0), (-1,0), colors.HexColor("#333333")),
+    ('ALIGN', (0,0), (-1,0), 'CENTER'),
+    ('VALIGN', (0,0), (-1,-1), 'MIDDLE'),
+    ('BOX', (0,0), (-1,-1), 0.6, colors.HexColor("#e0e0e0")),
+    ('INNERGRID', (0,0), (-1,-1), 0.4, colors.HexColor("#efefef")),
+    ('LEFTPADDING', (1,1), (1,-1), 6),
+    ('LEFTPADDING', (2,1), (2,-1), 6),
+    ('WORDWRAP', (2,1), (2,-1), 'CJK')  # helps long URLs wrap
+]))
+story.append(ms_table)
+story.append(Spacer(1, 14))
+
+# Matched Source final block
+matched_table = Table(
+    [[Paragraph("<b>Matched Source Overview</b>", styles['Body'])],
+     [Paragraph("Congratulations! Sections highlighted indicate similarity with external sources. Please review matched entries for exact references.", styles['Body'])]],
+    colWidths=[usable_width]
+)
+matched_table.setStyle(TableStyle([
+    ('BACKGROUND', (0,0), (-1,0), colors.HexColor("#f7fafb")),
+    ('BOX', (0,0), (-1,-1), 0.5, colors.HexColor("#e6e6e6")),
+    ('LEFTPADDING', (0,0), (-1,-1), 8),
+    ('RIGHTPADDING', (0,0), (-1,-1), 8),
+    ('TOPPADDING', (0,0), (-1,-1), 6),
+    ('BOTTOMPADDING', (0,0), (-1,-1), 6),
+]))
+story.append(matched_table)
+story.append(Spacer(1, 24))
+
+story.append(Paragraph("Generated by TrueWrite Scan • https://gopalkrushnamahapatra-truewrite-scan.static.hf.space", styles['Footer']))
+
+# Build PDF
+doc.build(story)
+
+print("PDF written to:", pdf_path)