from pypdf import PdfReader

def extract_text(pdf_path):
    reader = PdfReader(pdf_path)
    text = ""
    for page in reader.pages:
        text += page.extract_text()
    return text

text = extract_text("document.pdf")
print(text)

from pypdf import PdfMerger

def merge_pdfs(files, output):
    merger = PdfMerger()
    for f in files:
        merger.append(f)
    merger.write(output)
    merger.close()

merge_pdfs(["a.pdf", "b.pdf"], "merged.pdf")

from reportlab.lib import colors
from reportlab.platypus import SimpleDocTemplate, Table, TableStyle

def create_table_pdf():
    doc = SimpleDocTemplate("table.pdf")
    data = [
        ['产品', 'Q1', 'Q2', 'Q3', 'Q4'],
        ['产品 A', '12500', '15800', '18200', '22100'],
        ['产品 B', '8900', '10200', '12500', '14800'],
    ]
    table = Table(data)
    table.setStyle(TableStyle([
        ('BACKGROUND', (0,0), (-1,0), colors.blue),
        ('TEXTCOLOR', (0,0), (-1,0), colors.white),
        ('GRID', (0,0), (-1,-1), 1, colors.grey),
    ]))
    doc.build([table])

create_table_pdf()

from reportlab.graphics.shapes import Drawing
from reportlab.graphics.charts.barcharts import VerticalBarChart

def create_chart():
    drawing = Drawing(400, 200)
    chart = VerticalBarChart()
    chart.data = [(12500, 15800, 18200, 22100)]
    chart.categoryAxis.categoryNames = ['Q1', 'Q2', 'Q3', 'Q4']
    chart.bars.fillColor = colors.blue
    drawing.add(chart)
    return drawing

# 安装
pip install mermaid-py

# 基本使用
from mermaid import Mermaid, Graph, SequenceDiagram, ClassDiagram

# 创建流程图
graph = Graph('TD')
graph.add_node('A', '开始')
graph.add_node('B', '处理')
graph.add_node('C', '结束')
graph.add_edge('A', 'B')
graph.add_edge('B', 'C')

# 渲染为 SVG
mermaid_code = str(graph)
print(mermaid_code)

# 保存为文件
with open('diagram.mmd', 'w') as f:
    f.write(mermaid_code)

# 转换为 PDF（需要 mermaid-cli）
# mmdc -i diagram.mmd -o diagram.pdf

# 序列图示例
seq = SequenceDiagram()
seq.participant('用户')
seq.participant('系统')
seq.message('用户', '系统', '请求数据')
seq.message('系统', '用户', '返回结果')

# 类图示例
class_diag = ClassDiagram()
class_diag.add_class('User', ['name', 'email'], ['login()', 'logout()'])
class_diag.add_class('Order', ['id', 'total'], ['create()', 'cancel()'])
class_diag.add_relation('User', 'Order', 'places', '1..*')

# 完整 PDF 报告示例
from reportlab.lib.pagesizes import A4
from reportlab.platypus import SimpleDocTemplate, Paragraph, Image, PageBreak
from reportlab.lib.styles import getSampleStyleSheet

def create_mermaid_pdf():
    doc = SimpleDocTemplate("mermaid_report.pdf", pagesize=A4)
    elements = []
    styles = getSampleStyleSheet()
    
    # 添加标题
    elements.append(Paragraph("Mermaid 流程图报告", styles['Title']))
    elements.append(Paragraph("使用 Mermaid-Py 生成", styles['Normal']))
    
    # 添加 Mermaid 图表（需要先生成 SVG/PNG）
    # 1. 使用 mermaid-cli 转换
    # 2. 或使用在线 API 渲染
    
    # 示例：添加流程图描述
    elements.append(Paragraph("流程图说明:", styles['Heading2']))
    elements.append(Paragraph("开始 → 处理 → 结束", styles['Normal']))
    
    doc.build(elements)

create_mermaid_pdf()

# 安装
pip install pyecharts

# 基本使用 - 柱状图
from pyecharts import options as opts
from pyecharts.charts import Bar

bar = Bar()
bar.add_xaxis(["衬衫", "羊毛衫", "雪纺衫", "裤子", "高跟鞋", "袜子"])
bar.add_yaxis("商家 A", [120, 200, 150, 80, 70, 110])
bar.add_yaxis("商家 B", [80, 150, 110, 100, 90, 130])
bar.set_global_opts(title_opts=opts.TitleOpts(title="销售数据"))

# 渲染为 HTML
bar.render("bar_chart.html")

# 渲染为 PNG（需要 snapshot-selenium）
# pip install snapshot-selenium
from snapshot_selenium import snapshot
bar.render_snapshot("bar_chart.png")

# 饼图
from pyecharts.charts import Pie

pie = Pie()
pie.add(
    "销售占比",
    [
        ("衬衫", 120),
        ("羊毛衫", 200),
        ("雪纺衫", 150),
        ("裤子", 80),
        ("高跟鞋", 70),
        ("袜子", 110)
    ]
)
pie.set_global_opts(title_opts=opts.TitleOpts(title="饼图示例"))
pie.render("pie_chart.html")

# 折线图
from pyecharts.charts import Line

line = Line()
line.add_xaxis(["1 月", "2 月", "3 月", "4 月", "5 月", "6 月"])
line.add_yaxis("销售额", [220, 182, 191, 234, 290, 330])
line.set_global_opts(title_opts=opts.TitleOpts(title="月度趋势"))
line.render("line_chart.html")

# 完整 PDF 报告示例
from reportlab.lib.pagesizes import A4
from reportlab.platypus import SimpleDocTemplate, Paragraph, Image, PageBreak
from reportlab.lib.styles import getSampleStyleSheet
from pyecharts.charts import Bar
import pyecharts.options as opts

def create_pyecharts_pdf():
    # 1. 创建图表并保存为图片
    bar = Bar()
    bar.add_xaxis(["产品 A", "产品 B", "产品 C", "产品 D"])
    bar.add_yaxis("销量", [120, 200, 150, 80])
    bar.set_global_opts(title_opts=opts.TitleOpts(title="销售对比"))
    bar.render_snapshot("chart.png")
    
    # 2. 创建 PDF 报告
    doc = SimpleDocTemplate("pyecharts_report.pdf", pagesize=A4)
    elements = []
    styles = getSampleStyleSheet()
    
    # 添加标题
    elements.append(Paragraph("Pyecharts 可视化报告", styles['Title']))
    elements.append(Paragraph("使用 Python 生成 ECharts 图表", styles['Normal']))
    elements.append(PageBreak())
    
    # 添加图表
    elements.append(Paragraph("销售数据对比", styles['Heading2']))
    elements.append(Image("chart.png", width=400, height=300))
    
    doc.build(elements)

create_pyecharts_pdf()

go get github.com/unidoc/unipdf/v3
go get github.com/jung-kurt/gofpdf

package main

import (
    "fmt"
    "github.com/unidoc/unipdf/v3/model"
)

func extractText(path string) error {
    f, err := model.NewPdfReaderFromFile(path, nil)
    if err != nil { return err }
    
    numPages, _ := f.GetNumPages()
    for i := 1; i <= numPages; i++ {
        page, _ := f.GetPage(i)
        ex, _ := extractor.New(page)
        text, _ := ex.ExtractText()
        fmt.Println(text)
    }
    return nil
}

package main

import "github.com/jung-kurt/gofpdf"

func createPDF() {
    pdf := gofpdf.New("P", "mm", "A4", "")
    pdf.AddPage()
    pdf.SetFont("Arial", "B", 16)
    pdf.Cell(40, 10, "Hello World!")
    pdf.OutputFileAndClose("hello.pdf")
}

import * as pdfjsLib from 'pdfjs-dist';

async function extractText(url) {
    const pdf = await pdfjsLib.getDocument(url).promise;
    let text = '';
    for (let i = 1; i <= pdf.numPages; i++) {
        const page = await pdf.getPage(i);
        const content = await page.getTextContent();
        text += content.items.map(item => item.str).join(' ');
    }
    return text;
}

import { PDFDocument, rgb } from 'pdf-lib';

async function createPDF() {
    const doc = await PDFDocument.create();
    const page = doc.addPage([600, 400]);
    page.drawText('Hello, PDF!', {
        x: 50, y: 350, size: 30,
        color: rgb(0, 0.4, 0.8)
    });
    const bytes = await doc.save();
    // 保存或下载 bytes
}

const PDFDocument = require('pdfkit');
const doc = new PDFDocument();
doc.pipe(require('fs').createWriteStream('table.pdf'));

const data = [['产品', 'Q1', 'Q2'], ['A', '100', '150']];
let y = 50, x = 50;

// 表头
doc.font('Helvetica-Bold').fontSize(12);
data[0].forEach((h, i) => doc.text(h, x + i * 100, y));

// 数据行
doc.font('Helvetica').fontSize(10);
data[1].forEach((d, i) => doc.text(d, x + i * 100, y + 20));

doc.end();

🌐 HTML 转 PDF 工具

将 HTML/CSS 页面转换为 PDF 文档的常用工具