scripts/generate_analysis_report.py

from reportlab.lib import colors
from reportlab.lib.pagesizes import A4
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
from reportlab.lib.units import inch, cm
from reportlab.platypus import (
    SimpleDocTemplate,
    Paragraph,
    Spacer,
    Table,
    TableStyle,
    PageBreak,
    HRFlowable,
)
from reportlab.lib.enums import TA_CENTER, TA_LEFT, TA_JUSTIFY
from reportlab.pdfgen import canvas
from datetime import datetime
import os


class AnalysisReport:
    def __init__(self, output_path):
        self.output_path = output_path
        self.doc = SimpleDocTemplate(
            output_path,
            pagesize=A4,
            rightMargin=2 * cm,
            leftMargin=2 * cm,
            topMargin=2 * cm,
            bottomMargin=2 * cm,
        )
        self.styles = getSampleStyleSheet()
        self.story = []
        self._setup_styles()

    def _setup_styles(self):
        self.title_style = ParagraphStyle(
            "CustomTitle",
            parent=self.styles["Title"],
            fontSize=24,
            textColor=colors.HexColor("#0B0F1A"),
            spaceAfter=30,
            alignment=TA_CENTER,
        )
        self.heading1_style = ParagraphStyle(
            "CustomHeading1",
            parent=self.styles["Heading1"],
            fontSize=16,
            textColor=colors.HexColor("#1E3A5F"),
            spaceBefore=20,
            spaceAfter=12,
            borderPadding=5,
        )
        self.heading2_style = ParagraphStyle(
            "CustomHeading2",
            parent=self.styles["Heading2"],
            fontSize=13,
            textColor=colors.HexColor("#2563EB"),
            spaceBefore=15,
            spaceAfter=8,
        )
        self.body_style = ParagraphStyle(
            "CustomBody",
            parent=self.styles["Normal"],
            fontSize=10,
            leading=14,
            alignment=TA_JUSTIFY,
            spaceAfter=8,
        )
        self.bullet_style = ParagraphStyle(
            "BulletStyle",
            parent=self.styles["Normal"],
            fontSize=10,
            leading=14,
            leftIndent=20,
            spaceAfter=4,
        )

    def add_header(self):
        self.story.append(Paragraph("CitationEdge", self.title_style))
        self.story.append(
            Paragraph(
                "Agentic Architecture Analysis Report",
                ParagraphStyle(
                    "Subtitle",
                    parent=self.styles["Normal"],
                    fontSize=14,
                    textColor=colors.HexColor("#64748B"),
                    alignment=TA_CENTER,
                    spaceAfter=5,
                ),
            )
        )
        self.story.append(
            Paragraph(
                f"Generated: {datetime.now().strftime('%B %d, %Y')}",
                ParagraphStyle(
                    "Date",
                    parent=self.styles["Normal"],
                    fontSize=10,
                    textColor=colors.HexColor("#94A3B8"),
                    alignment=TA_CENTER,
                    spaceAfter=30,
                ),
            )
        )
        self.story.append(
            HRFlowable(
                width="100%",
                thickness=2,
                color=colors.HexColor("#2563EB"),
                spaceAfter=30,
            )
        )

    def add_summary_table(self):
        self.story.append(Paragraph("1. Executive Summary", self.heading1_style))

        data = [
            ["Metric", "Value"],
            ["Total Papers Analyzed", "13"],
            ["Success Rate", "100% (all completed)"],
            ["Avg Processing Time", "~40 seconds"],
            ["Papers with Complete Data", "10/13 (77%)"],
            ["Papers with Partial Data", "3/13 (23%)"],
        ]

        table = Table(data, colWidths=[5 * cm, 10 * cm])
        table.setStyle(
            TableStyle(
                [
                    ("BACKGROUND", (0, 0), (-1, 0), colors.HexColor("#1E3A5F")),
                    ("TEXTCOLOR", (0, 0), (-1, 0), colors.white),
                    ("ALIGN", (0, 0), (-1, -1), "CENTER"),
                    ("FONTNAME", (0, 0), (-1, 0), "Helvetica-Bold"),
                    ("FONTSIZE", (0, 0), (-1, 0), 11),
                    ("BOTTOMPADDING", (0, 0), (-1, 0), 12),
                    ("TOPPADDING", (0, 0), (-1, 0), 12),
                    ("BACKGROUND", (0, 1), (-1, -1), colors.HexColor("#F8FAFC")),
                    ("GRID", (0, 0), (-1, -1), 1, colors.HexColor("#E2E8F0")),
                    ("FONTNAME", (0, 1), (-1, -1), "Helvetica"),
                    ("FONTSIZE", (0, 1), (-1, -1), 10),
                    ("BOTTOMPADDING", (0, 1), (-1, -1), 8),
                    ("TOPPADDING", (0, 1), (-1, -1), 8),
                ]
            )
        )
        self.story.append(table)
        self.story.append(Spacer(1, 20))

    def add_strengths(self):
        self.story.append(Paragraph("2. Strengths", self.heading1_style))

        strengths = [
            "<b>Fast Processing:</b> 30-140s per paper is excellent for multi-agent analysis",
            "<b>Keyword Extraction:</b> Captures domain-specific terms accurately (LSTM, Self-Attention, reinforcement learning)",
            "<b>Citation Detection:</b> Identifies real citations and cross-references well",
            "<b>Claim Classification:</b> Distinguishes between method/result/finding claims effectively",
            "<b>Multi-Agent Pipeline:</b> Good separation of concerns between agents (parser, keyword, claim, citation, evidence, argumentation, scoring)",
        ]

        for s in strengths:
            self.story.append(Paragraph(f"• {s}", self.bullet_style))
        self.story.append(Spacer(1, 15))

    def add_weaknesses(self):
        self.story.append(Paragraph("3. Weaknesses & Gaps", self.heading1_style))

        self.story.append(
            Paragraph("3.1 Low Citation Gap Detection", self.heading2_style)
        )
        self.story.append(
            Paragraph(
                "Most reports show only 0-2 citation gaps, which appears significantly under-detected. "
                "Academic papers typically cite 15-40+ references. For a thorough analysis, expected citation "
                "gaps should be in the range of 5-15 per paper, depending on the paper length and field.",
                self.body_style,
            )
        )

        self.story.append(Paragraph("3.2 Weak Evidence Grounding", self.heading2_style))
        self.story.append(
            Paragraph(
                "Only 1-5% of claims are grounded with citations. For example, Paper 4 with 50 claims "
                "shows only 3 grounded claims (0.06%). A well-structured research paper should have 30-50% "
                "of claims properly grounded with citations to source materials.",
                self.body_style,
            )
        )

        self.story.append(
            Paragraph("3.3 Incomplete Claim Detection", self.heading2_style)
        )
        self.story.append(
            Paragraph(
                "Average 20-50 claims per paper seems low for dense ML papers. Papers 1, 6, and 9 show "
                "only 15-18 claims, which likely indicates under-detection. High-impact ML papers often "
                "contain 80-150+ distinct claims that should be identified.",
                self.body_style,
            )
        )

        self.story.append(Paragraph("3.4 Citation Format Issues", self.heading2_style))
        self.story.append(
            Paragraph(
                "Some citations show 'Unknown Title' or missing metadata. Author names are occasionally "
                "wrong or truncated. The citation extraction should leverage reference section parsing "
                "more aggressively.",
                self.body_style,
            )
        )

        self.story.append(Paragraph("3.5 No Limitation Analysis", self.heading2_style))
        self.story.append(
            Paragraph(
                "Papers mention 'Limitations' in text but no agent captures this systematically. "
                "A dedicated Limitations agent should identify: (a) scope constraints, (b) generalizability "
                "issues, (c) comparison baselines, (d) reproducibility concerns.",
                self.body_style,
            )
        )

        self.story.append(
            Paragraph("3.6 Inconsistent Argumentation Mapping", self.heading2_style)
        )
        self.story.append(
            Paragraph(
                "While 'Related To' chains exist in reports, they are sparse. Papers with 50+ claims "
                "show minimal argument mapping. The argumentation agent may be failing silently or not "
                "finding sufficient claim relationships to map.",
                self.body_style,
            )
        )

    def add_improvements(self):
        self.story.append(PageBreak())
        self.story.append(Paragraph("4. Priority Improvements", self.heading1_style))

        improvements = [
            (
                "HIGH",
                "Evidence Grounding Agent",
                "Increase grounding from 1-5% to 30-50%. Add sentence-level citation matching. "
                "Use coreference resolution to link pronouns to entities.",
            ),
            (
                "HIGH",
                "Citation Gap Detection",
                "Improve to 5-15 gaps per paper. Implement cross-reference analysis (when A cites B, "
                "but B's claims aren't addressed). Add methodology comparison gaps.",
            ),
            (
                "HIGH",
                "Claim Detection",
                "Target 80-150 claims per high-impact paper. Add hedging language detection "
                "(may, might, suggests). Separate major from minor claims.",
            ),
            (
                "MEDIUM",
                "Citation Extraction",
                "Parse reference section more aggressively. Extract full author lists, venue, year. "
                "Handle multiple citation formats (APA, MLA, IEEE, Nature).",
            ),
            (
                "MEDIUM",
                "Limitation Analysis Agent",
                "Add new agent or extend claim agent to identify limitations. Classify: scope, "
                "generalizability, baseline, reproducibility. Rate severity.",
            ),
            (
                "MEDIUM",
                "Argumentation Mapping",
                "Ensure all claims with relationships are mapped. Connect claims across sections "
                "(introduction → method → result → conclusion).",
            ),
            (
                "LOW",
                "Multi-modal Support",
                "Add support for tables, figures, equations. Extract claims from figure captions. "
                "Link table data to supporting claims.",
            ),
            (
                "LOW",
                "Real-time Streaming",
                "Provide incremental agent results as jobs run. Better UX for long papers.",
            ),
        ]

        for priority, title, desc in improvements:
            color = (
                colors.HexColor("#DC2626")
                if priority == "HIGH"
                else (
                    colors.HexColor("#CA8A04")
                    if priority == "MEDIUM"
                    else colors.HexColor("#64748B")
                )
            )

            data = [
                [
                    Paragraph(
                        f"<b>[{priority}]</b>",
                        ParagraphStyle("P", textColor=color, fontSize=9),
                    ),
                    Paragraph(f"<b>{title}</b><br/>{desc}", self.body_style),
                ]
            ]

            table = Table(data, colWidths=[2 * cm, 13 * cm])
            table.setStyle(
                TableStyle(
                    [
                        ("VALIGN", (0, 0), (-1, -1), "TOP"),
                        ("BACKGROUND", (0, 0), (-1, -1), colors.HexColor("#F8FAFC")),
                        ("BOX", (0, 0), (-1, -1), 1, colors.HexColor("#E2E8F0")),
                        ("LEFTPADDING", (0, 0), (-1, -1), 8),
                        ("RIGHTPADDING", (0, 0), (-1, -1), 8),
                        ("TOPPADDING", (0, 0), (-1, -1), 8),
                        ("BOTTOMPADDING", (0, 0), (-1, -1), 8),
                    ]
                )
            )
            self.story.append(table)
            self.story.append(Spacer(1, 8))

    def add_agent_breakdown(self):
        self.story.append(Spacer(1, 20))
        self.story.append(Paragraph("5. Agent Pipeline Analysis", self.heading1_style))

        agents = [
            ["Agent", "Purpose", "Current Status", "Rating"],
            ["Parser Agent", "Extract text & metadata", "Working well", "Good"],
            ["Keyword Agent", "Key concept extraction", "Working well", "Good"],
            ["Claim Agent", "Research claim detection", "Under-detecting", "Fair"],
            ["Citation Gap Agent", "Gap identification", "Too few gaps", "Poor"],
            [
                "Evidence Grounding Agent",
                "Grounding in source",
                "Very low grounding",
                "Poor",
            ],
            ["Argumentation Agent", "Structure mapping", "Sparse mapping", "Fair"],
            ["Scoring Agent", "Quality assessment", "Working", "Good"],
        ]

        table = Table(agents, colWidths=[4 * cm, 4 * cm, 4 * cm, 2 * cm])
        table.setStyle(
            TableStyle(
                [
                    ("BACKGROUND", (0, 0), (-1, 0), colors.HexColor("#1E3A5F")),
                    ("TEXTCOLOR", (0, 0), (-1, 0), colors.white),
                    ("ALIGN", (0, 0), (-1, -1), "CENTER"),
                    ("FONTNAME", (0, 0), (-1, 0), "Helvetica-Bold"),
                    ("FONTSIZE", (0, 0), (-1, -1), 9),
                    ("BOTTOMPADDING", (0, 0), (-1, -1), 8),
                    ("TOPPADDING", (0, 0), (-1, -1), 8),
                    ("BACKGROUND", (0, 1), (-1, 1), colors.HexColor("#DCFCE7")),
                    ("BACKGROUND", (0, 2), (-1, 2), colors.HexColor("#DCFCE7")),
                    ("BACKGROUND", (0, 3), (-1, 3), colors.HexColor("#FEF3C7")),
                    ("BACKGROUND", (0, 4), (-1, 4), colors.HexColor("#FEE2E2")),
                    ("BACKGROUND", (0, 5), (-1, 5), colors.HexColor("#FEE2E2")),
                    ("BACKGROUND", (0, 6), (-1, 6), colors.HexColor("#FEF3C7")),
                    ("BACKGROUND", (0, 7), (-1, 7), colors.HexColor("#DCFCE7")),
                    ("GRID", (0, 0), (-1, -1), 1, colors.HexColor("#E2E8F0")),
                ]
            )
        )
        self.story.append(table)

    def add_recommendations(self):
        self.story.append(PageBreak())
        self.story.append(Paragraph("6. Recommendations", self.heading1_style))

        short_term = [
            "Improve evidence grounding to target 30%+ coverage",
            "Enhance citation gap detection with cross-reference analysis",
            "Add hedging language detection for claim classification",
            "Parse reference sections more aggressively",
        ]

        medium_term = [
            "Add dedicated Limitations analysis agent",
            "Implement full argumentation chain mapping",
            "Add multi-modal support for tables and figures",
            "Improve real-time progress streaming",
        ]

        long_term = [
            "Fine-tune models on research paper corpus",
            "Add comparison with arXiv/PubMed knowledge base",
            "Implement claim verification against source claims",
            "Add novelty/innovation scoring",
        ]

        self.story.append(Paragraph("Short-term (1-2 weeks)", self.heading2_style))
        for item in short_term:
            self.story.append(Paragraph(f"• {item}", self.bullet_style))

        self.story.append(Paragraph("Medium-term (1-2 months)", self.heading2_style))
        for item in medium_term:
            self.story.append(Paragraph(f"• {item}", self.bullet_style))

        self.story.append(Paragraph("Long-term (3-6 months)", self.heading2_style))
        for item in long_term:
            self.story.append(Paragraph(f"• {item}", self.bullet_style))

    def add_footer(self):
        self.story.append(Spacer(1, 40))
        self.story.append(
            HRFlowable(
                width="100%",
                thickness=1,
                color=colors.HexColor("#E2E8F0"),
                spaceAfter=20,
            )
        )
        self.story.append(
            Paragraph(
                "Report generated by CitationEdge Analysis Platform | IIT Patna Research Project",
                ParagraphStyle(
                    "Footer",
                    fontSize=8,
                    textColor=colors.HexColor("#94A3B8"),
                    alignment=TA_CENTER,
                ),
            )
        )

    def build(self):
        self.add_header()
        self.add_summary_table()
        self.add_strengths()
        self.add_weaknesses()
        self.add_improvements()
        self.add_agent_breakdown()
        self.add_recommendations()
        self.add_footer()
        self.doc.build(self.story)
        return self.output_path


if __name__ == "__main__":
    output_path = os.path.join(
        os.path.dirname(__file__), "..", "reports", "architecture_analysis_report.pdf"
    )
    report = AnalysisReport(output_path)
    report.build()
    print(f"Report saved: {output_path}")