diff --git a/PDF2WORD.md b/PDF2WORD.md new file mode 100644 index 0000000000000000000000000000000000000000..382a0febee57a21ea0b77d95360feadc54ddae60 --- /dev/null +++ b/PDF2WORD.md @@ -0,0 +1,42 @@ +# PDF2WORD + +PDF2WORD是PaddleOCR社区开发者@whj 基于PP-Structure智能文档分析系统实现的PDF转换word应用程序，提供可直接安装的exe，方便windows用户运行 + +
+ +
+ +## 1.使用 + +### 应用程序 + +1. 下载与安装：针对Windows用户，根据[软件下载]()一节下载软件后，运行 `pdf2word.exe` 。若您下载的是lite版本，安装过程中会在线下载环境依赖、模型等必要资源，安装时间较长，请确保网络畅通。serve版本打包了相关依赖，安装时间较短，可按需下载。 + +2. 转换：由于PP-Structure根据中英文数据分别进行适配，在转换相应文件时可**根据文档语言进行相应选择**。 + +### 脚本运行 + +首次运行需要将 + +``` +python pdf2word.py +``` + + + +## 2.自行打包 + +PDF2WORD应用程序通过[QPT](https://github.com/QPT-Family/QPT)工具打包实现，若您修改了界面代码需要重新打包，请在 `ppstructure` 文件夹下运行下方指令 + +``` +cd ./ +mv ./ppstructure/pdf2word .. -r +python GenEXE.py +``` + + + +## 3.软件下载 + +如需获取已打包程序，可以扫描下方二维码，关注公众号填写问卷后，加入PaddleOCR官方交流群免费获取20G OCR学习大礼包，内含OCR场景应用集合（包含数码管、液晶屏、车牌、高精度SVTR模型等7个垂类模型）、《动手学OCR》电子书、课程回放视频、前沿论文等重磅资料 + diff --git a/ScreenShotWidget.py b/ScreenShotWidget.py new file mode 100644 index 0000000000000000000000000000000000000000..87d46198c04dd9975e8adeef5a269b1ccefdd11d --- /dev/null +++ b/ScreenShotWidget.py @@ -0,0 +1,1108 @@ +# coding=utf-8 +# author: fardeas +# from: https://blog.csdn.net/u010501845/article/details/124931326 +import os +import sys +from datetime import datetime + +from qtpy import QtCore, QtGui +from qtpy.QtWidgets import * +from qtpy.QtCore import * +from qtpy.QtGui import * + + +class TextInputWidget(QTextEdit): + '''在截图区域内的文本输入框''' + + def __init__(self, god=None): + super().__init__(god) + self.god = god + # 设置背景透明 + # self.setStyleSheet("QTextEdit{background-color: transparent;}") + palette = self.palette() + palette.setBrush(QtGui.QPalette.ColorRole.Base, self.god.color_transparent) + self.setPalette(palette) + self.setTextColor(self.god.toolbar.curColor()) + self.setCurrentFont(self.god.toolbar.curFont()) + self._doc = self.document() # QTextDocument + self.textChanged.connect(self.adjustSizeByContent) + self.adjustSizeByContent() # 初始化调整高度为一行 + self.hide() + + def adjustSizeByContent(self, margin=30): + '''限制宽度不超出截图区域，根据文本内容调整高度，不会出现滚动条''' + self._doc.setTextWidth(self.viewport().width()) + margins = self.contentsMargins() + h = int(self._doc.size().height() + margins.top() + margins.bottom()) + self.setFixedHeight(h) + + def beginNewInput(self, pos, endPointF): + '''开始新的文本输入''' + self._maxRect = self.god.screenArea.normalizeRectF(pos, endPointF) + self.waitForInput() + + def waitForInput(self): + self.setGeometry(self._maxRect.toRect()) + # self.setGeometry(self._maxRect.adjusted(0, 0, -1, 0)) # 宽度-1 + self.setFocus() + self.show() + + def loadTextInputBy(self, action): + '''载入修改旧的文本 + action:(type, color, font, rectf, txt)''' + self.setTextColor(action[1]) + self.setCurrentFont(action[2]) + self._maxRect = action[3] + self.append(action[4]) + self.god.isDrawing = True + self.waitForInput() + + +class LineWidthAction(QAction): + + '''画笔粗细选择器''' + + def __init__(self, text, parent, lineWidth): + super().__init__(text, parent) + self._lineWidth = lineWidth + self.refresh(QtCore.Qt.GlobalColor.red) + self.triggered.connect(self.onTriggered) + self.setVisible(False) + + def refresh(self, color): + painter = self.parent().god.screenArea._painter + dotRadius = QPointF(self._lineWidth, self._lineWidth) + centerPoint = self.parent().iconPixmapCenter() + pixmap = self.parent().iconPixmapCopy() + painter.begin(pixmap) + painter.setPen(self.parent().god.pen_transparent) + painter.setBrush(color) + painter.drawEllipse(QRectF(centerPoint - dotRadius, centerPoint + dotRadius)) + painter.end() + self.setIcon(QIcon(pixmap)) + + def onTriggered(self): + self.parent()._curLineWidth = self._lineWidth + + +class FontAction(QAction): + + '''字体选择器''' + + def __init__(self, text, parent): + super().__init__(text, parent) + self.setIcon(QIcon(r"img/sys/font.png")) + self._curFont = self.parent().god.font_textInput + self.triggered.connect(self.onTriggered) + self.setVisible(False) + + def onTriggered(self): + font, ok = QFontDialog.getFont(self._curFont, self.parent(), caption='选择字体') + if ok: + self._curFont = font + self.parent().god.textInputWg.setCurrentFont(font) + + +class ColorAction(QAction): + + '''颜色选择器''' + + def __init__(self, text, parent): + super().__init__(text, parent) + self._curColor = QtCore.Qt.GlobalColor.red + self._pixmap = QPixmap(32, 32) + self.refresh(self._curColor) + self.triggered.connect(self.onTriggered) + + def refresh(self, color): + self._curColor = color + self._pixmap.fill(color) + self.setIcon(QIcon(self._pixmap)) + self.parent()._at_line_small.refresh(color) + self.parent()._at_line_normal.refresh(color) + self.parent()._at_line_big.refresh(color) + + def onTriggered(self): + col = QColorDialog.getColor(self._curColor, self.parent(), title='选择颜色') + if col.isValid(): + self.refresh(col) + self.parent().god.textInputWg.setTextColor(col) + + +class ScreenShotToolBar(QToolBar): + '''截图区域工具条''' + + def __init__(self, god): + super().__init__(god) + self.god = god + self.setToolButtonStyle(QtCore.Qt.ToolButtonStyle.ToolButtonTextUnderIcon) + self.setStyleSheet("QToolBar {border-radius: 5px;padding: 3px;background-color: #eeeeef;}") + self._style_normal = "QToolBar QToolButton{color: black;}" + self._style_selected = "QToolBar QToolButton{color: #ff7300;border: 1px solid #BEDAF2;background-color: #D6E4F1}" # 与鼠标悬停样式一样 + self._iconPixmap = QPixmap(32, 32) + self._iconPixmap.fill(self.god.color_transparent) + self._iconPixmapCenter = QPointF(self._iconPixmap.rect().center()) + self._curLineWidth = 3 + self._at_line_small = LineWidthAction('细', self, self._curLineWidth - 2) + self._at_line_normal = LineWidthAction('中', self, self._curLineWidth) + self._at_line_big = LineWidthAction('粗', self, self._curLineWidth + 2) + self._at_font = FontAction('字体', self) + self._at_color = ColorAction('颜色', self) + # self._at_rectangle = QAction(QIcon(r"img/sys/rectangle.png"), '矩形', self, triggered=self.beforeDrawRectangle) + # self._at_ellipse = QAction(QIcon(r"img/sys/ellipse.png"), '椭圆', self, triggered=self.beforeDrawEllipse) + # self._at_graffiti = QAction(QIcon(r"img/sys/graffiti.png"), '涂鸦', self, triggered=self.beforeDrawGraffiti) + # self._at_textInput = QAction(QIcon(r"img/sys/write.png"), '文字', self, triggered=self.beforeDrawText) + # self.addAction(self._at_line_small) + # self.addAction(self._at_line_normal) + # self.addAction(self._at_line_big) + # self.addAction(self._at_font) + # self.addAction(self._at_color) + # self.addSeparator() + # self.addAction(self._at_rectangle) + # self.addAction(self._at_ellipse) + # self.addAction(self._at_graffiti) + # self.addAction(self._at_textInput) + # self.addAction(QAction(QIcon(r"img/sys/undo.png"), '撤销', self, triggered=self.undo)) + # self.addSeparator() + self.addAction(QAction(QApplication.style().standardIcon(40), + '退出', self, triggered=self.god.rejectSlot)) + self.addAction(QAction(QApplication.style().standardIcon(45), + '接受', self, triggered=self.god.acceptSlot)) + # self.addAction(QAction(QIcon(r"img/chat/download.png"), '保存', self, triggered=lambda: self.beforeSave('local'))) + # self.addAction(QAction(QIcon(r"img/chat/sendImg.png"), '复制', self, triggered=lambda: self.beforeSave('clipboard'))) + # self.actionTriggered.connect(self.onActionTriggered) + + def curLineWidth(self): + return self._curLineWidth + + def curFont(self): + return self._at_font._curFont + + def curColor(self): + return self._at_color._curColor + # return QColor(self._at_color._curColor.toRgb()) # 颜色的副本 + + def iconPixmapCopy(self): + return self._iconPixmap.copy() + + def iconPixmapCenter(self): + return self._iconPixmapCenter + + # def onActionTriggered(self, action): + # '''突出显示已选中的画笔粗细、编辑模式''' + # for at in [self._at_line_small, self._at_line_normal, self._at_line_big]: + # if at._lineWidth == self._curLineWidth: + # self.widgetForAction(at).setStyleSheet(self._style_selected) + # else: + # self.widgetForAction(at).setStyleSheet(self._style_normal) + # if self.god.isDrawRectangle: + # self.widgetForAction(self._at_rectangle).setStyleSheet(self._style_selected) + # else: + # self.widgetForAction(self._at_rectangle).setStyleSheet(self._style_normal) + # if self.god.isDrawEllipse: + # self.widgetForAction(self._at_ellipse).setStyleSheet(self._style_selected) + # else: + # self.widgetForAction(self._at_ellipse).setStyleSheet(self._style_normal) + # if self.god.isDrawGraffiti: + # self.widgetForAction(self._at_graffiti).setStyleSheet(self._style_selected) + # else: + # self.widgetForAction(self._at_graffiti).setStyleSheet(self._style_normal) + # if self.god.isDrawText: + # self.widgetForAction(self._at_textInput).setStyleSheet(self._style_selected) + # else: + # self.widgetForAction(self._at_textInput).setStyleSheet(self._style_normal) + + def setLineWidthActionVisible(self, flag): + self._at_line_small.setVisible(flag) + self._at_line_normal.setVisible(flag) + self._at_line_big.setVisible(flag) + + def beforeDrawRectangle(self): + self.god.clearEditFlags() + self.god.isDrawRectangle = True + self.setLineWidthActionVisible(True) + self._at_font.setVisible(False) + + def beforeDrawEllipse(self): + self.god.clearEditFlags() + self.god.isDrawEllipse = True + self.setLineWidthActionVisible(True) + self._at_font.setVisible(False) + + def beforeDrawGraffiti(self): + self.god.clearEditFlags() + self.god.isDrawGraffiti = True + self.setLineWidthActionVisible(True) + self._at_font.setVisible(False) + + def beforeDrawText(self): + self.god.clearEditFlags() + self.god.isDrawText = True + self.setLineWidthActionVisible(False) + self._at_font.setVisible(True) + + def undo(self): + '''撤销上次编辑行为''' + if self.god.screenArea.undoEditAction(): + self.god.update() + + def beforeSave(self, target): + # 若正在编辑文本未保存，先完成编辑 + if self.god.isDrawing and self.god.isDrawText: + self.god.screenArea.saveTextInputAction() + if target == 'local': + self.god.save2Local() + elif target == 'clipboard': + self.god.save2Clipboard() + + def enterEvent(self, event): + self.god.setCursor(QtCore.Qt.CursorShape.ArrowCursor) # 工具条上显示标准箭头cursor + + def leaveEvent(self, event): + self.god.setCursor(QtCore.Qt.CursorShape.CrossCursor) # 十字无箭头 + + +class ScreenArea(QtCore.QObject): + '''屏幕区域（提供各种算法的核心类），划分为9个子区域： + TopLeft，Top，TopRight + Left，Center，Right + BottomLeft，Bottom，BottomRight + 其中Center根据start、end两个QPointF确定 + ''' + + def __init__(self, god): + super().__init__() + self.god = god + self._pt_start = QPointF() # 划定截图区域时鼠标左键按下的位置（topLeft） + self._pt_end = QPointF() # 划定截图区域时鼠标左键松开的位置（bottomRight） + self._rt_toolbar = QRectF() # 工具条的矩形 + self._actions = [] # 在截图区域上的所有编辑行为（矩形、椭圆、涂鸦、文本输入等） + self._pt_startEdit = QPointF() # 在截图区域上绘制矩形、椭圆时鼠标左键按下的位置（topLeft） + self._pt_endEdit = QPointF() # 在截图区域上绘制矩形、椭圆时鼠标左键松开的位置（bottomRight） + self._pointfs = [] # 涂鸦经过的所有点 + self._painter = QPainter() # 独立于ScreenShotWidget之外的画家类 + self._textOption = QtGui.QTextOption(QtCore.Qt.AlignmentFlag.AlignLeft | QtCore.Qt.AlignmentFlag.AlignTop) + self._textOption.setWrapMode(QtGui.QTextOption.WrapMode.WrapAnywhere) # 文本在矩形内自动换行 + # self._textOption.setWrapMode(QtGui.QTextOption.WrapMode.WrapAtWordBoundaryOrAnywhere) + self.captureScreen() + + def captureScreen(self): + '''抓取整个屏幕的截图''' + # screen = QtGui.QGuiApplication.primaryScreen() + self._screenPixmap = QApplication.primaryScreen().grabWindow(0) + self._pixelRatio = self._screenPixmap.devicePixelRatio() # 设备像素比 + self._rt_screen = self.screenLogicalRectF() + self.remakeNightArea() + + def normalizeRectF(self, topLeftPoint, bottomRightPoint): + '''根据起止点生成宽高非负数的QRectF，通常用于bottomRightPoint比topLeftPoint更左更上的情况 + 入参可以是QPoint或QPointF''' + rectf = QRectF(topLeftPoint, bottomRightPoint) + x = rectf.x() + y = rectf.y() + w = rectf.width() + h = rectf.height() + if w < 0: # bottomRightPoint在topLeftPoint左侧时，topLeftPoint往左移动 + x = x + w + w = -w + if h < 0: # bottomRightPoint在topLeftPoint上侧时，topLeftPoint往上移动 + y = y + h + h = -h + return QRectF(x, y, w, h) + + def physicalRectF(self, rectf): + '''计算划定的截图区域的（缩放倍率1.0的）原始矩形（会变大） + rectf：划定的截图区域的矩形。可为QRect或QRectF''' + return QRectF(rectf.x() * self._pixelRatio, rectf.y() * self._pixelRatio, + rectf.width() * self._pixelRatio, rectf.height() * self._pixelRatio) + + def logicalRectF(self, physicalRectF): + '''根据原始矩形计算缩放后的矩形（会变小） + physicalRectF：缩放倍率1.0的原始矩形。可为QRect或QRectF''' + return QRectF(physicalRectF.x() / self._pixelRatio, physicalRectF.y() / self._pixelRatio, + physicalRectF.width() / self._pixelRatio, physicalRectF.height() / self._pixelRatio) + + def physicalPixmap(self, rectf, editAction=False): + '''根据指定区域获取其原始大小的（缩放倍率1.0的）QPixmap + rectf：指定区域。可为QRect或QRectF + editAction:是否带上编辑结果''' + if editAction: + canvasPixmap = self.screenPhysicalPixmapCopy() + self._painter.begin(canvasPixmap) + self.paintEachEditAction(self._painter, textBorder=False) + self._painter.end() + return canvasPixmap.copy(self.physicalRectF(rectf).toRect()) + else: + return self._screenPixmap.copy(self.physicalRectF(rectf).toRect()) + + def screenPhysicalRectF(self): + return QRectF(self._screenPixmap.rect()) + + def screenLogicalRectF(self): + return QRectF(QPointF(0, 0), self.screenLogicalSizeF()) # 即当前屏幕显示的大小 + + def screenPhysicalSizeF(self): + return QSizeF(self._screenPixmap.size()) + + def screenLogicalSizeF(self): + return QSizeF(self._screenPixmap.width() / self._pixelRatio, self._screenPixmap.height() / self._pixelRatio) + + def screenPhysicalPixmapCopy(self): + return self._screenPixmap.copy() + + def screenLogicalPixmapCopy(self): + return self._screenPixmap.scaled(self.screenLogicalSizeF().toSize()) + + def centerPhysicalRectF(self): + return self.physicalRectF(self._rt_center) + + def centerLogicalRectF(self): + '''根据屏幕上的start、end两个QPointF确定''' + return self._rt_center + + def centerPhysicalPixmap(self, editAction=True): + '''截图区域的QPixmap + editAction:是否带上编辑结果''' + return self.physicalPixmap(self._rt_center + QMarginsF(-1, -1, 1, 1), editAction=editAction) + + def centerTopMid(self): + return self._pt_centerTopMid + + def centerBottomMid(self): + return self._pt_centerBottomMid + + def centerLeftMid(self): + return self._pt_centerLeftMid + + def centerRightMid(self): + return self._pt_centerRightMid + + def setStartPoint(self, pointf, remake=False): + self._pt_start = pointf + if remake: + self.remakeNightArea() + + def setEndPoint(self, pointf, remake=False): + self._pt_end = pointf + if remake: + self.remakeNightArea() + + def setCenterArea(self, start, end): + self._pt_start = start + self._pt_end = end + self.remakeNightArea() + + def remakeNightArea(self): + '''重新划分九宫格区域。根据中央截图区域计算出来的其他8个区域、截图区域四个边框中点坐标等都是logical的''' + self._rt_center = self.normalizeRectF(self._pt_start, self._pt_end) + # 中央区域上下左右边框的中点，用于调整大小 + self._pt_centerTopMid = (self._rt_center.topLeft() + self._rt_center.topRight()) / 2 + self._pt_centerBottomMid = (self._rt_center.bottomLeft() + self._rt_center.bottomRight()) / 2 + self._pt_centerLeftMid = (self._rt_center.topLeft() + self._rt_center.bottomLeft()) / 2 + self._pt_centerRightMid = (self._rt_center.topRight() + self._rt_center.bottomRight()) / 2 + # 以截图区域左上、上中、右上、左中、右中、左下、下中、右下为中心的正方形区域，用于调整大小 + self._square_topLeft = self.squareAreaByCenter(self._rt_center.topLeft()) + self._square_topRight = self.squareAreaByCenter(self._rt_center.topRight()) + self._square_bottomLeft = self.squareAreaByCenter(self._rt_center.bottomLeft()) + self._square_bottomRight = self.squareAreaByCenter(self._rt_center.bottomRight()) + self._square_topMid = self.squareAreaByCenter(self._pt_centerTopMid) + self._square_bottomMid = self.squareAreaByCenter(self._pt_centerBottomMid) + self._square_leftMid = self.squareAreaByCenter(self._pt_centerLeftMid) + self._square_rightMid = self.squareAreaByCenter(self._pt_centerRightMid) + # 除中央截图区域外的8个区域 + self._rt_topLeft = QRectF(self._rt_screen.topLeft(), self._rt_center.topLeft()) + self._rt_top = QRectF(QPointF(self._rt_center.topLeft().x(), 0), self._rt_center.topRight()) + self._rt_topRight = QRectF(QPointF(self._rt_center.topRight().x(), 0), QPointF(self._rt_screen.width(), self._rt_center.topRight().y())) + self._rt_left = QRectF(QPointF(0, self._rt_center.topLeft().y()), self._rt_center.bottomLeft()) + self._rt_right = QRectF(self._rt_center.topRight(), QPointF(self._rt_screen.width(), self._rt_center.bottomRight().y())) + self._rt_bottomLeft = QRectF(QPointF(0, self._rt_center.bottomLeft().y()), QPointF(self._rt_center.bottomLeft().x(), self._rt_screen.height())) + self._rt_bottom = QRectF(self._rt_center.bottomLeft(), QPointF(self._rt_center.bottomRight().x(), self._rt_screen.height())) + self._rt_bottomRight = QRectF(self._rt_center.bottomRight(), self._rt_screen.bottomRight()) + + def squareAreaByCenter(self, pointf): + '''以QPointF为中心的正方形QRectF''' + rectf = QRectF(0, 0, 15, 15) + rectf.moveCenter(pointf) + return rectf + + def aroundAreaIn8Direction(self): + '''中央区域周边的8个方向的区域（无交集）''' + return [self._rt_topLeft, self._rt_top, self._rt_topRight, + self._rt_left, self._rt_right, + self._rt_bottomLeft, self._rt_bottom, self._rt_bottomRight] + + def aroundAreaIn4Direction(self): + '''中央区域周边的4个方向的区域（有交集） + 上区域(左上、上、右上)：0, 0, maxX, topRight.y + 下区域(左下、下、右下)：0, bottomLeft.y, maxX, maxY-bottomLeft.y + 左区域(左上、左、左下)：0, 0, bottomLeft.x, maxY + 右区域(右上、右、右下)：topRight.x, 0, maxX - topRight.x, maxY''' + screenSizeF = self.screenLogicalSizeF() + pt_topRight = self._rt_center.topRight() + pt_bottomLeft = self._rt_center.bottomLeft() + return [QRectF(0, 0, screenSizeF.width(), pt_topRight.y()), + QRectF(0, pt_bottomLeft.y(), screenSizeF.width(), screenSizeF.height() - pt_bottomLeft.y()), + QRectF(0, 0, pt_bottomLeft.x(), screenSizeF.height()), + QRectF(pt_topRight.x(), 0, screenSizeF.width() - pt_topRight.x(), screenSizeF.height())] + + def aroundAreaWithoutIntersection(self): + '''中央区域周边的4个方向的区域（无交集） + 上区域(左上、上、右上)：0, 0, maxX, topRight.y + 下区域(左下、下、右下)：0, bottomLeft.y, maxX, maxY-bottomLeft.y + 左区域(左)：0, topRight.y, bottomLeft.x-1, center.height + 右区域(右)：topRight.x+1, topRight.y, maxX - topRight.x, center.height''' + screenSizeF = self.screenLogicalSizeF() + pt_topRight = self._rt_center.topRight() + pt_bottomLeft = self._rt_center.bottomLeft() + centerHeight = pt_bottomLeft.y() - pt_topRight.y() + return [QRectF(0, 0, screenSizeF.width(), pt_topRight.y()), + QRectF(0, pt_bottomLeft.y(), screenSizeF.width(), screenSizeF.height() - pt_bottomLeft.y()), + QRectF(0, pt_topRight.y(), pt_bottomLeft.x() - 1, centerHeight), + QRectF(pt_topRight.x() + 1, pt_topRight.y(), screenSizeF.width() - pt_topRight.x(), centerHeight)] + + def setBeginDragPoint(self, pointf): + '''计算开始拖拽位置距离截图区域左上角的向量''' + self._drag_vector = pointf - self._rt_center.topLeft() + + def getNewPosAfterDrag(self, pointf): + '''计算拖拽后截图区域左上角的新位置''' + return pointf - self._drag_vector + + def moveCenterAreaTo(self, pointf): + '''限制拖拽不能超出屏幕范围''' + self._rt_center.moveTo(self.getNewPosAfterDrag(pointf)) + startPointF = self._rt_center.topLeft() + if startPointF.x() < 0: + self._rt_center.moveTo(0, startPointF.y()) + startPointF = self._rt_center.topLeft() + if startPointF.y() < 0: + self._rt_center.moveTo(startPointF.x(), 0) + screenSizeF = self.screenLogicalSizeF() + endPointF = self._rt_center.bottomRight() + if endPointF.x() > screenSizeF.width(): + self._rt_center.moveBottomRight(QPointF(screenSizeF.width(), endPointF.y())) + endPointF = self._rt_center.bottomRight() + if endPointF.y() > screenSizeF.height(): + self._rt_center.moveBottomRight(QPointF(endPointF.x(), screenSizeF.height())) + self.setCenterArea(self._rt_center.topLeft(), self._rt_center.bottomRight()) + + def setBeginAdjustPoint(self, pointf): + '''判断开始调整截图区域大小时鼠标左键在哪个区（不可能是中央区域），用于判断调整大小的意图方向''' + self._mousePos = self.getMousePosBy(pointf) + + def getMousePosBy(self, pointf): + if self._square_topLeft.contains(pointf): + return 'TL' + elif self._square_topMid.contains(pointf): + return 'T' + elif self._square_topRight.contains(pointf): + return 'TR' + elif self._square_leftMid.contains(pointf): + return 'L' + elif self._rt_center.contains(pointf): + return 'CENTER' + elif self._square_rightMid.contains(pointf): + return 'R' + elif self._square_bottomLeft.contains(pointf): + return 'BL' + elif self._square_bottomMid.contains(pointf): + return 'B' + elif self._square_bottomRight.contains(pointf): + return 'BR' + else: + return 'ERROR' + + def adjustCenterAreaBy(self, pointf): + '''根据开始调整截图区域大小时鼠标左键在哪个区（不可能是中央区域），判断调整大小的意图方向，判定新的开始、结束位置''' + startPointF = self._rt_center.topLeft() + endPointF = self._rt_center.bottomRight() + if self._mousePos == 'TL': + startPointF = pointf + elif self._mousePos == 'T': + startPointF = QPointF(startPointF.x(), pointf.y()) + elif self._mousePos == 'TR': + startPointF = QPointF(startPointF.x(), pointf.y()) + endPointF = QPointF(pointf.x(), endPointF.y()) + elif self._mousePos == 'L': + startPointF = QPointF(pointf.x(), startPointF.y()) + elif self._mousePos == 'R': + endPointF = QPointF(pointf.x(), endPointF.y()) + elif self._mousePos == 'BL': + startPointF = QPointF(pointf.x(), startPointF.y()) + endPointF = QPointF(endPointF.x(), pointf.y()) + elif self._mousePos == 'B': + endPointF = QPointF(endPointF.x(), pointf.y()) + elif self._mousePos == 'BR': + endPointF = pointf + else: # 'ERROR' + return + newRectF = self.normalizeRectF(startPointF, endPointF) + self.setCenterArea(newRectF.topLeft(), newRectF.bottomRight()) + + def getMouseShapeBy(self, pointf): + '''根据鼠标位置返回对应的鼠标样式''' + if self._rt_center.contains(pointf): + if self.god.isDrawRectangle or self.god.isDrawEllipse: + return QtCore.Qt.CursorShape.ArrowCursor + elif self.god.isDrawGraffiti: + return QtCore.Qt.CursorShape.PointingHandCursor # 超链接上的手势 + elif self.god.isDrawText: + return QtCore.Qt.CursorShape.IBeamCursor # 工字 + else: + return QtCore.Qt.CursorShape.SizeAllCursor # 十字有箭头 + # return QtCore.Qt.CursorShape.OpenHandCursor # 打开的手，表示可拖拽 + elif self._square_topLeft.contains(pointf) or self._square_bottomRight.contains(pointf): + return QtCore.Qt.CursorShape.SizeFDiagCursor # ↖↘ + elif self._square_topMid.contains(pointf) or self._square_bottomMid.contains(pointf): + return QtCore.Qt.CursorShape.SizeVerCursor # ↑↓ + elif self._square_topRight.contains(pointf) or self._square_bottomLeft.contains(pointf): + return QtCore.Qt.CursorShape.SizeBDiagCursor # ↙↗ + elif self._square_leftMid.contains(pointf) or self._square_rightMid.contains(pointf): + return QtCore.Qt.CursorShape.SizeHorCursor # ←→ + else: + return QtCore.Qt.CursorShape.CrossCursor # 十字无箭头 + + def isMousePosInCenterRectF(self, pointf): + return self._rt_center.contains(pointf) + + def paintMagnifyingGlassPixmap(self, pos, glassSize): + '''绘制放大镜内的图像(含纵横十字线) + pos:鼠标光标位置 + glassSize:放大镜边框大小''' + pixmapRect = QRect(0, 0, 20, 20) # 以鼠标光标为中心的正方形区域，最好是偶数 + pixmapRect.moveCenter(pos) + glassPixmap = self.physicalPixmap(pixmapRect) + glassPixmap.setDevicePixelRatio(1.0) + glassPixmap = glassPixmap.scaled(glassSize, glassSize, QtCore.Qt.AspectRatioMode.KeepAspectRatio) + # 在放大后的QPixmap上画纵横十字线 + self._painter.begin(glassPixmap) + halfWidth = glassPixmap.width() / 2 + halfHeight = glassPixmap.height() / 2 + self._painter.setPen(self.god.pen_SolidLine_lightBlue) + self._painter.drawLine(QPointF(0, halfHeight), QPointF(glassPixmap.width(), halfHeight)) + self._painter.drawLine(QPointF(halfWidth, 0), QPointF(halfWidth, glassPixmap.height())) + self._painter.end() + return glassPixmap + + def paintEachEditAction(self, painter, textBorder=True): + '''绘制所有已保存的编辑行为。编辑行为超出截图区域也无所谓，保存图像时只截取截图区域内 + textBorder:是否绘制文本边框''' + for action in self.getEditActions(): + if action[0] == 'rectangle': # (type, color, lineWidth, startPoint, endPoint) + self.paintRectangle(painter, action[1], action[2], action[3], action[4]) + elif action[0] == 'ellipse': # (type, color, lineWidth, startPoint, endPoint) + self.paintEllipse(painter, action[1], action[2], action[3], action[4]) + elif action[0] == 'graffiti': # (type, color, lineWidth, points) + self.paintGraffiti(painter, action[1], action[2], action[3]) + elif action[0] == 'text': # (type, color, font, rectf, txt) + self.paintTextInput(painter, action[1], action[2], action[3], action[4], textBorder=textBorder) + + def paintRectangle(self, painter, color, lineWidth, startPoint=None, endPoint=None): + if not startPoint: + startPoint = self._pt_startEdit + if not endPoint: + endPoint = self._pt_endEdit + qrectf = self.normalizeRectF(startPoint, endPoint) + if qrectf.isValid(): + pen = QPen(color) + pen.setWidth(lineWidth) + painter.setPen(pen) + painter.setBrush(self.god.color_transparent) + painter.drawRect(qrectf) + + def paintEllipse(self, painter, color, lineWidth, startPoint=None, endPoint=None): + if not startPoint: + startPoint = self._pt_startEdit + if not endPoint: + endPoint = self._pt_endEdit + qrectf = self.normalizeRectF(startPoint, endPoint) + if qrectf.isValid(): + pen = QPen(color) + pen.setWidth(lineWidth) + painter.setPen(pen) + painter.setBrush(self.god.color_transparent) + painter.drawEllipse(qrectf) + + def paintGraffiti(self, painter, color, lineWidth, pointfs=None): + if not pointfs: + pointfs = self.getGraffitiPointFs() + pen = QPen(color) + pen.setWidth(lineWidth) + painter.setPen(pen) + total = len(pointfs) + if total == 0: + return + elif total == 1: + painter.drawPoint(pointfs[0]) + else: + previousPoint = pointfs[0] + for i in range(1, total): + nextPoint = pointfs[i] + painter.drawLine(previousPoint, nextPoint) + previousPoint = nextPoint + + def paintTextInput(self, painter, color, font, rectf, txt, textBorder=True): + painter.setPen(color) + painter.setFont(font) + painter.drawText(rectf, txt, self._textOption) + if textBorder: + painter.setPen(QtCore.Qt.PenStyle.DotLine) # 点线 + painter.setBrush(self.god.color_transparent) + painter.drawRect(rectf) + + def getEditActions(self): + return self._actions.copy() + + def takeTextInputActionAt(self, pointf): + '''根据鼠标位置查找已保存的文本输入结果，找到后取出''' + for i in range(len(self._actions)): + action = self._actions[i] + if action[0] == 'text' and action[3].contains(pointf): + return self._actions.pop(i) + return None + + def undoEditAction(self): + reply = False + if self._actions: + reply = self._actions.pop() + if not self._actions: # 所有编辑行为都被撤销后退出编辑模式 + self.god.exitEditMode() + else: + self.god.exitEditMode() + return reply + + def clearEditActions(self): + self._actions.clear() + + def setBeginEditPoint(self, pointf): + '''在截图区域上绘制矩形、椭圆时鼠标左键按下的位置（topLeft）''' + self._pt_startEdit = pointf + self.god.isDrawing = True + + def setEndEditPoint(self, pointf): + '''在截图区域上绘制矩形、椭圆时鼠标左键松开的位置（bottomRight）''' + self._pt_endEdit = pointf + + def saveRectangleAction(self): + rectf = self.normalizeRectF(self._pt_startEdit, self._pt_endEdit) + self._actions.append(('rectangle', self.god.toolbar.curColor(), self.god.toolbar.curLineWidth(), + rectf.topLeft(), rectf.bottomRight())) + self._pt_startEdit = QPointF() + self._pt_endEdit = QPointF() + self.god.isDrawing = False + + def saveEllipseleAction(self): + rectf = self.normalizeRectF(self._pt_startEdit, self._pt_endEdit) + self._actions.append(('ellipse', self.god.toolbar.curColor(), self.god.toolbar.curLineWidth(), + rectf.topLeft(), rectf.bottomRight())) + self._pt_startEdit = QPointF() + self._pt_endEdit = QPointF() + self.god.isDrawing = False + + def saveGraffitiPointF(self, pointf, first=False): + self._pointfs.append(pointf) + if first: + self.god.isDrawing = True + + def getGraffitiPointFs(self): + return self._pointfs.copy() + + def saveGraffitiAction(self): + if self._pointfs: + self._actions.append(('graffiti', self.god.toolbar.curColor(), self.god.toolbar.curLineWidth(), self._pointfs.copy())) + self._pointfs.clear() + self.god.isDrawing = False + + def setBeginInputTextPoint(self, pointf): + '''在截图区域上输入文字时鼠标左键按下的位置（topLeft）''' + self.god.isDrawing = True + self.god.textInputWg.beginNewInput(pointf, self._pt_end) + + def saveTextInputAction(self): + txt = self.god.textInputWg.toPlainText() + if txt: + rectf = self.god.textInputWg._maxRect # 取最大矩形的topLeft + rectf.setSize(QRectF(self.god.textInputWg.rect()).size()) # 取实际矩形的宽高 + self._actions.append(('text', self.god.toolbar.curColor(), self.god.toolbar.curFont(), + rectf, txt)) + self.god.textInputWg.clear() + self.god.textInputWg.hide() # 不管保存成功与否都取消编辑 + self.god.isDrawing = False + + def saveNightAreaImg(self): + '''将九宫格区域保存为本地图片，仅用于开发测试''' + screenPixmap = self.screenPhysicalPixmapCopy() + self._painter.begin(screenPixmap) + self._painter.setPen(self.pen_SolidLine_lightBlue) + self._painter.setFont(self.god.font_normal) + self._painter.drawRect(self._rt_center) + for area in self.aroundAreaIn8Direction(): + self._painter.drawRect(area) + for pointf in [self._rt_center.topLeft(), self._rt_center.topRight(), + self._rt_center.bottomLeft(), self._rt_center.bottomRight(), + self._pt_centerTopMid, self._pt_centerBottomMid, + self._pt_centerLeftMid, self._pt_centerRightMid]: + self._painter.drawText(pointf + QPointF(5, -5), '(%s, %s)' % (pointf.x(), pointf.y())) + self._painter.end() + screenPixmap.save('1.jpg', quality=100) + self.centerPhysicalPixmap().save('2.jpg', quality=100) + + +class ScreenShotWidget(QWidget): + + fileType_all = '所有文件 (*);;Excel文件 (*.xls *.xlsx);;图片文件 (*.jpg *.jpeg *.gif *.png *.bmp)' + fileType_img = '图片文件 (*.jpg *.jpeg *.gif *.png *.bmp)' + dir_lastAccess = os.getcwd() # 最后访问目录 + + def __init__(self): + super().__init__() + self.setMouseTracking(True) + self.setWindowFlags(QtCore.Qt.WindowType.FramelessWindowHint | QtCore.Qt.WindowType.WindowStaysOnTopHint) + self.initPainterTool() + self.initFunctionalFlag() + self.screenArea = ScreenArea(self) + self.toolbar = ScreenShotToolBar(self) + self.textInputWg = TextInputWidget(self) + + self.captureImage = None + # 设置 screenPixmap 为窗口背景 + # palette = QtGui.QPalette() + # palette.setBrush(QtGui.QPalette.ColorRole.Window, QtGui.QBrush(self.screenArea.screenPhysicalPixmapCopy())) + # self.setPalette(palette) + + def rejectSlot(self): + self.captureImage = None + self.close() + + def acceptSlot(self): + self.captureImage = self.screenArea.centerPhysicalPixmap().toImage() + self.close() + + def start(self): + self.screenArea.captureScreen() + self.setGeometry(self.screenArea.screenPhysicalRectF().toRect()) + self.clearScreenShotArea() + self.showFullScreen() + + def initPainterTool(self): + self.painter = QPainter() + self.color_transparent = QtCore.Qt.GlobalColor.transparent + self.color_black = QColor(0, 0, 0, 64) # 黑色背景 + self.color_lightBlue = QColor(30, 120, 255) # 浅蓝色。深蓝色QtCore.Qt.GlobalColor.blue + self.font_normal = QtGui.QFont('Times New Roman', 11, QtGui.QFont.Weight.Normal) + self.font_textInput = QtGui.QFont('微软雅黑', 16, QtGui.QFont.Weight.Normal) # 工具条文字工具默认字体 + self.pen_transparent = QPen(QtCore.Qt.PenStyle.NoPen) # 没有笔迹，画不出线条 + self.pen_white = QPen(QtCore.Qt.GlobalColor.white) + self.pen_SolidLine_lightBlue = QPen(self.color_lightBlue) # 实线，浅蓝色 + self.pen_SolidLine_lightBlue.setStyle(QtCore.Qt.PenStyle.DashLine) # 实线SolidLine，虚线DashLine，点线DotLine + self.pen_SolidLine_lightBlue.setWidthF(0) # 0表示线宽为1 + self.pen_DashLine_lightBlue = QPen(self.color_lightBlue) # 虚线，浅蓝色 + self.pen_DashLine_lightBlue.setStyle(QtCore.Qt.PenStyle.DashLine) + + def initFunctionalFlag(self): + self.hasScreenShot = False # 是否已通过拖动鼠标左键划定截图区域 + self.isCapturing = False # 正在拖动鼠标左键选定截图区域时 + self.isMoving = False # 在截图区域内拖动时 + self.isAdjusting = False # 在截图区域的边框按住鼠标左键调整大小时 + self.isDrawing = False # 是否已在截图区域内开始绘制 + self.isDrawRectangle = False # 正在截图区域内画矩形 + self.isDrawEllipse = False # 正在截图区域内画椭圆 + self.isDrawGraffiti = False # 正在截图区域内进行涂鸦 + self.isDrawText = False # 正在截图区域内画文字 + self.setCursor(QtCore.Qt.CursorShape.CrossCursor) # 设置鼠标样式 十字 + + def paintEvent(self, event): + centerRectF = self.screenArea.centerLogicalRectF() + screenSizeF = self.screenArea.screenLogicalSizeF() + canvasPixmap = self.screenArea.screenPhysicalPixmapCopy() + # canvasPixmap = QPixmap(screenSizeF.toSize()) + # canvasPixmap.fill(self.color_transparent) + # 在屏幕截图的副本上绘制已选定的截图区域 + self.painter.begin(canvasPixmap) + if self.hasScreenShot: + self.paintCenterArea(centerRectF) # 绘制中央截图区域 + self.paintMaskLayer(screenSizeF, fullScreen=False) # 绘制截图区域的周边区域遮罩层 + else: + self.paintMaskLayer(screenSizeF) + self.paintMagnifyingGlass(screenSizeF) # 在鼠标光标右下角显示放大镜 + self.paintToolbar(centerRectF, screenSizeF) # 在截图区域右下角显示工具条 + self.paintEditActions() # 在截图区域绘制编辑行为结果 + self.painter.end() + # 把画好的绘制结果显示到窗口上 + self.painter.begin(self) + self.painter.drawPixmap(0, 0, canvasPixmap) # 从坐标(0, 0)开始绘制 + self.painter.end() + + def paintCenterArea(self, centerRectF): + '''绘制已选定的截图区域''' + self.painter.setRenderHint(QPainter.RenderHint.Antialiasing, True) # 反走样 + # 1.绘制矩形线框 + self.painter.setPen(self.pen_DashLine_lightBlue) + self.painter.drawRect(centerRectF) + # 2.绘制矩形线框4个端点和4条边框的中间点 + if centerRectF.width() >= 100 and centerRectF.height() >= 100: + points = [ # 点坐标 + centerRectF.topLeft(), centerRectF.topRight(), centerRectF.bottomLeft(), centerRectF.bottomRight(), + self.screenArea.centerLeftMid(), self.screenArea.centerRightMid(), + self.screenArea.centerTopMid(), self.screenArea.centerBottomMid() + ] + blueDotRadius = QPointF(2, 2) # 椭圆蓝点 + self.painter.setBrush(self.color_lightBlue) + for point in points: + self.painter.drawEllipse(QRectF(point - blueDotRadius, point + blueDotRadius)) + # 3.在截图区域左上角显示截图区域宽高 + if centerRectF.topLeft().y() > 20: + labelPos = centerRectF.topLeft() + QPointF(5, -5) + else: # 拖拽截图区域到贴近屏幕上边缘时“宽x高”移动到截图区域左上角的下侧 + labelPos = centerRectF.topLeft() + QPointF(5, 15) + centerPhysicalRect = self.screenArea.centerPhysicalRectF().toRect() + self.painter.setPen(self.pen_white) + self.painter.setFont(self.font_normal) + self.painter.drawText(labelPos, '%s x %s' % (centerPhysicalRect.width(), centerPhysicalRect.height())) + # 4.在屏幕左上角预览截图结果 + # self.painter.drawPixmap(0, 0, self.screenArea.centerPhysicalPixmap()) # 从坐标(0, 0)开始绘制 + + def paintMaskLayer(self, screenSizeF, fullScreen=True): + if fullScreen: # 全屏遮罩层 + maskPixmap = QPixmap(screenSizeF.toSize()) + maskPixmap.fill(self.color_black) + self.painter.drawPixmap(0, 0, maskPixmap) + else: # 绘制截图区域的周边区域遮罩层，以凸显截图区域 + # 方法一：截图区域以外的8个方向区域 + # for area in self.screenArea.aroundAreaIn8Direction(): + # area = area.normalized() + # maskPixmap = QPixmap(area.size().toSize()) # 由于float转int的精度问题，可能会存在黑线条缝隙 + # maskPixmap.fill(self.color_black) + # self.painter.drawPixmap(area.topLeft(), maskPixmap) + # 方法二：截图区域以外的上下左右区域（有交集，交集部分颜色加深，有明显的纵横效果） + # for area in self.screenArea.aroundAreaIn4Direction(): + # maskPixmap = QPixmap(area.size().toSize()) + # maskPixmap.fill(self.color_black) + # self.painter.drawPixmap(area.topLeft(), maskPixmap) + # 方法三：截图区域以外的上下左右区域（无交集） + for area in self.screenArea.aroundAreaWithoutIntersection(): + maskPixmap = QPixmap(area.size().toSize()) + maskPixmap.fill(self.color_black) + self.painter.drawPixmap(area.topLeft(), maskPixmap) + + def paintMagnifyingGlass(self, screenSizeF, glassSize=150, offset=30, labelHeight=30): + '''未划定截图区域模式时、正在划定截取区域时、调整截取区域大小时在鼠标光标右下角显示放大镜 + glassSize:放大镜正方形边长 + offset:放大镜任意一个端点距离鼠标光标位置的最近距离 + labelHeight:pos和rgb两行文字的高度''' + if self.hasScreenShot and (not self.isCapturing) and (not self.isAdjusting): + return + pos = QtGui.QCursor.pos() + glassPixmap = self.screenArea.paintMagnifyingGlassPixmap(pos, glassSize) # 画好纵横十字线后的放大镜内QPixmap + # 限制放大镜显示不超出屏幕外 + glassRect = glassPixmap.rect() + if (pos.x() + glassSize + offset) < screenSizeF.width(): + if (pos.y() + offset + glassSize + labelHeight) < screenSizeF.height(): + glassRect.moveTo(pos + QPoint(offset, offset)) + else: + glassRect.moveBottomLeft(pos + QPoint(offset, -offset)) + else: + if (pos.y() + offset + glassSize + labelHeight) < screenSizeF.height(): + glassRect.moveTopRight(pos + QPoint(-offset, offset)) + else: + glassRect.moveBottomRight(pos + QPoint(-offset, -offset)) + self.painter.drawPixmap(glassRect.topLeft(), glassPixmap) + # 显示pos:(x, y)、rgb:(255,255,255) + qrgb = QtGui.QRgba64.fromArgb32(glassPixmap.toImage().pixel(glassPixmap.rect().center())) + labelRectF = QRectF(glassRect.bottomLeft().x(), glassRect.bottomLeft().y(), glassSize, labelHeight) + self.painter.setPen(self.pen_transparent) + self.painter.setBrush(self.color_black) # 黑底 + self.painter.drawRect(labelRectF) + self.painter.setPen(self.pen_white) + self.painter.setFont(self.font_normal) + self.painter.drawText(labelRectF, + QtCore.Qt.AlignmentFlag.AlignLeft | QtCore.Qt.AlignmentFlag.AlignVCenter, + 'pos:(%s, %s)\nrgb:(%s, %s, %s)' % (pos.x(), pos.y(), qrgb.red8(), qrgb.green8(), qrgb.blue8())) + + def paintToolbar(self, centerRectF, screenSizeF): + '''在截图区域右下角显示工具条''' + if self.hasScreenShot: + if self.isCapturing or self.isAdjusting: + self.toolbar.hide() # 正在划定截取区域时、调整截图区域大小时不显示工具条 + else: + self.toolbar.adjustSize() + toolbarRectF = QRectF(self.toolbar.rect()) + # 工具条位置优先顺序：右下角下侧，右上角上侧，右下角上侧 + if (screenSizeF.height() - centerRectF.bottomRight().y()) > toolbarRectF.height(): + toolbarRectF.moveTopRight(centerRectF.bottomRight() + QPointF(-5, 5)) + elif centerRectF.topRight().y() > toolbarRectF.height(): + toolbarRectF.moveBottomRight(centerRectF.topRight() + QPointF(-5, -5)) + else: + toolbarRectF.moveBottomRight(centerRectF.bottomRight() + QPointF(-5, -5)) + # 限制工具条的x坐标不为负数，不能移出屏幕外 + if toolbarRectF.x() < 0: + pos = toolbarRectF.topLeft() + pos.setX(centerRectF.x() + 5) + toolbarRectF.moveTo(pos) + self.toolbar.move(toolbarRectF.topLeft().toPoint()) + self.toolbar.show() + else: + self.toolbar.hide() + + def paintEditActions(self): + '''在截图区域绘制编辑行为结果。编辑行为超出截图区域也无所谓，保存图像时只截取截图区域内''' + # 1.绘制正在拖拽编辑中的矩形、椭圆、涂鸦 + if self.isDrawRectangle: + self.screenArea.paintRectangle(self.painter, self.toolbar.curColor(), self.toolbar.curLineWidth()) + elif self.isDrawEllipse: + self.screenArea.paintEllipse(self.painter, self.toolbar.curColor(), self.toolbar.curLineWidth()) + elif self.isDrawGraffiti: + self.screenArea.paintGraffiti(self.painter, self.toolbar.curColor(), self.toolbar.curLineWidth()) + # 2.绘制所有已保存的编辑行为 + self.screenArea.paintEachEditAction(self.painter) + + def clearEditFlags(self): + self.isDrawing = False + self.isDrawRectangle = False + self.isDrawEllipse = False + self.isDrawGraffiti = False + self.isDrawText = False + + def exitEditMode(self): + '''退出编辑模式''' + self.clearEditFlags() + # self.toolbar.onActionTriggered(None) # 清空工具条工具按钮选中状态 + self.textInputWg.hide() + + def clearScreenShotArea(self): + '''清空已划定的截取区域''' + self.screenArea.clearEditActions() # 清除已保存的编辑行为 + self.exitEditMode() + self.hasScreenShot = False + self.isCapturing = False + pos = QPointF() + self.screenArea.setCenterArea(pos, pos) + self.update() + self.setCursor(QtCore.Qt.CursorShape.CrossCursor) # 设置鼠标样式 十字 + + def mousePressEvent(self, event): + if event.button() == QtCore.Qt.MouseButton.LeftButton: + pos = event.pos() + if self.hasScreenShot: + if self.isDrawRectangle or self.isDrawEllipse: + self.screenArea.setBeginEditPoint(pos) + elif self.isDrawGraffiti: # 保存涂鸦经过的每一个点 + self.screenArea.saveGraffitiPointF(pos, first=True) + elif self.isDrawText: + if self.isDrawing: + if QRectF(self.textInputWg.rect()).contains(pos): + pass # 在输入框内调整光标位置，忽略 + else: # 鼠标点到输入框之外，完成编辑 + self.screenArea.saveTextInputAction() + else: # 未开始编辑时（暂不支持文本拖拽） + action = self.screenArea.takeTextInputActionAt(pos) + if action: # 鼠标点到输入框之内，修改旧的文本输入 + self.textInputWg.loadTextInputBy(action) + else: # 鼠标点到输入框之外，开始新的文本输入 + self.screenArea.setBeginInputTextPoint(pos) + elif self.screenArea.isMousePosInCenterRectF(pos): + self.isMoving = True # 进入拖拽移动模式 + self.screenArea.setBeginDragPoint(pos) + else: + self.isAdjusting = True # 进入调整大小模式 + self.screenArea.setBeginAdjustPoint(pos) + else: + self.screenArea.setCenterArea(pos, pos) + self.isCapturing = True # 进入划定截图区域模式 + if event.button() == QtCore.Qt.MouseButton.RightButton: + if self.hasScreenShot or self.isCapturing: # 清空已划定的的截图区域 + self.clearScreenShotArea() + else: + self.close() + + def mouseReleaseEvent(self, event): + if event.button() == QtCore.Qt.MouseButton.LeftButton: + if self.isDrawRectangle: + self.screenArea.saveRectangleAction() + elif self.isDrawEllipse: + self.screenArea.saveEllipseleAction() + elif self.isDrawGraffiti: + self.screenArea.saveGraffitiAction() + self.isCapturing = False + self.isMoving = False + self.isAdjusting = False + self.toolbar.show() + + def mouseMoveEvent(self, event): + pos = event.pos() + if self.isDrawing: + if self.isDrawRectangle or self.isDrawEllipse: + self.screenArea.setEndEditPoint(pos) + elif self.isDrawGraffiti: + self.screenArea.saveGraffitiPointF(pos) + elif self.isCapturing: + self.hasScreenShot = True + self.screenArea.setEndPoint(pos, remake=True) + elif self.isMoving: + self.screenArea.moveCenterAreaTo(pos) + elif self.isAdjusting: + self.screenArea.adjustCenterAreaBy(pos) + self.update() + if self.hasScreenShot: + self.setCursor(self.screenArea.getMouseShapeBy(pos)) + else: + self.setCursor(QtCore.Qt.CursorShape.CrossCursor) # 设置鼠标样式 十字 + + def mouseDoubleClickEvent(self, event): + if event.button() == QtCore.Qt.MouseButton.LeftButton: + if self.screenArea.isMousePosInCenterRectF(event.pos()): + self.save2Clipboard() + self.close() + + def keyPressEvent(self, QKeyEvent): + if QKeyEvent.key() == QtCore.Qt.Key.Key_Escape: + self.close() + if QKeyEvent.key() in (QtCore.Qt.Key.Key_Return, QtCore.Qt.Key.Key_Enter): # 大键盘、小键盘回车 + self.save2Clipboard() + self.close() + + def save2Clipboard(self): + '''将截图区域复制到剪贴板''' + if self.hasScreenShot: + mimData = QtCore.QMimeData() + mimData.setImageData(self.screenArea.centerPhysicalPixmap().toImage()) + QApplication.clipboard().setMimeData(mimData) + # self.screenArea.saveNightAreaImg() + self.close() + + def save2Local(self): + fileType = self.fileType_img + filePath, fileFormat = self.sys_selectSaveFilePath(self, fileType=fileType) + if filePath: + self.screenArea.centerPhysicalPixmap().save(filePath, quality=100) + self.close() + + def sys_getCurTime(self, fmt='%Y-%m-%d %H:%M:%S'): + '''获取字符串格式的当前时间''' + # return QtCore.QDateTime.currentDateTime().toString('yyyy-MM-dd hh:mm:ss') + return datetime.now().strftime(fmt) + + def sys_selectSaveFilePath(self, widget, title='选择文件保存路径', saveFileDir=None, + saveFileName='', defaultFileFmt='%Y%m%d%H%M%S', fileType=None): + '''选择文件保存路径 + title:选择窗口标题 + saveFileDir:指定保存目录 + saveFileName:默认保存文件名 + defaultFileFmt:不指定saveFileName时，自动以此格式的时间字符串命名文件 + fileType:可以选择的文件类型 + return:(所选的文件保存路径, 文件的类型) + ''' + options = QFileDialog.Option.ReadOnly + if saveFileName == '': + saveFileName = self.sys_getCurTime(defaultFileFmt) + if not saveFileDir: + saveFileDir = self.dir_lastAccess + saveFilePath = os.path.join(saveFileDir, saveFileName) + if not fileType: + fileType = self.fileType_all + filePath, fileFormat = QFileDialog.getSaveFileName(widget, title, saveFilePath, fileType, options=options) + if filePath: + self.dir_lastAccess = os.path.dirname(filePath) + return (filePath, fileFormat) + + diff --git a/icons/chinese.png b/icons/chinese.png new file mode 100644 index 0000000000000000000000000000000000000000..328e2fff73bd75188fa888aae45c8cb4ca844f57 Binary files /dev/null and b/icons/chinese.png differ diff --git a/icons/english.png b/icons/english.png new file mode 100644 index 0000000000000000000000000000000000000000..536c4a910ae6fc05040f4958477a34aeae891ea0 Binary files /dev/null and b/icons/english.png differ diff --git a/icons/folder-open.png b/icons/folder-open.png new file mode 100644 index 0000000000000000000000000000000000000000..ab5f55f5a4819add116113b55f717f7a21aeafdd Binary files /dev/null and b/icons/folder-open.png differ diff --git a/icons/folder-plus.png b/icons/folder-plus.png new file mode 100644 index 0000000000000000000000000000000000000000..01ce6c10b0ed3e3975edbebbc8e886f846fabe8d Binary files /dev/null and b/icons/folder-plus.png differ diff --git a/image2doc.py b/image2doc.py new file mode 100644 index 0000000000000000000000000000000000000000..700126e0249cdbf7df415f7e8a07931b1241c4bf --- /dev/null +++ b/image2doc.py @@ -0,0 +1,349 @@ +import sys +import tarfile +import os +import time +import functools +import cv2 +import platform +import numpy as np +from qtpy import QtWidgets +from qtpy.QtGui import QImage, QPixmap, QIcon + +from ppstructure.predict_system import StructureSystem, save_structure_res +from ppstructure.utility import parse_args, draw_structure_result +from ppocr.utils.network import download_with_progressbar +from ppstructure.recovery.recovery_to_doc import sorted_layout_boxes, convert_info_docx +from ScreenShotWidget import ScreenShotWidget + +__APPNAME__ = "Image2Doc" +__VERSION__ = "0.0.2" +here = os.path.dirname(os.path.abspath(__file__)) +URLs_EN = { + # 下载超英文轻量级PP-OCRv3模型的检测模型并解压 + "en_PP-OCRv3_det_infer": "https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_infer.tar", + # 下载英文轻量级PP-OCRv3模型的识别模型并解压 + "en_PP-OCRv3_rec_infer": "https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_infer.tar", + # 下载超轻量级英文表格英文模型并解压 + "en_ppstructure_mobile_v2.0_SLANet_infer": "https://paddleocr.bj.bcebos.com/ppstructure/models/slanet/en_ppstructure_mobile_v2.0_SLANet_infer.tar", + # 英文版面分析模型 + "picodet_lcnet_x1_0_fgd_layout_infer": "https://paddleocr.bj.bcebos.com/ppstructure/models/layout/picodet_lcnet_x1_0_fgd_layout_infer.tar", +} +DICT_EN = { + "rec_char_dict_path": "en_dict.txt", + "layout_dict_path": "layout_publaynet_dict.txt", +} + +URLs_CN = { + # 下载超中文轻量级PP-OCRv3模型的检测模型并解压 + "cn_PP-OCRv3_det_infer": "https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar", + # 下载中文轻量级PP-OCRv3模型的识别模型并解压 + "cn_PP-OCRv3_rec_infer": "https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_infer.tar", + # 下载超轻量级英文表格英文模型并解压 + "cn_ppstructure_mobile_v2.0_SLANet_infer": "https://paddleocr.bj.bcebos.com/ppstructure/models/slanet/en_ppstructure_mobile_v2.0_SLANet_infer.tar", + # 中文版面分析模型 + "picodet_lcnet_x1_0_fgd_layout_cdla_infer": "https://paddleocr.bj.bcebos.com/ppstructure/models/layout/picodet_lcnet_x1_0_fgd_layout_cdla_infer.tar", +} +DICT_CN = { + "rec_char_dict_path": "ppocr_keys_v1.txt", + "layout_dict_path": "layout_cdla_dict.txt", +} + + +def QImageToCvMat(incomingImage): + ''' + Converts a QImage into an opencv MAT format + ''' + + incomingImage = incomingImage.convertToFormat(QImage.Format.Format_RGBA8888) + + width = incomingImage.width() + height = incomingImage.height() + + ptr = incomingImage.bits() + ptr.setsize(height * width * 4) + arr = np.frombuffer(ptr, np.uint8).reshape((height, width, 4)) + return arr + + +class APP_Image2Doc(QtWidgets.QWidget): + def __init__(self): + super(QtWidgets.QWidget, self).__init__() + self.pb = None # 进度条 + self.pb_text = "已载入: {} / 已转换: {}" + self.imagePaths = [] + # self.resultPath = os.path.join(here, "output") + self.screenShotWg = ScreenShotWidget() + self.screenShot = None + self.save_pdf = False + + self.vis_font_path = os.path.join(here, + "doc", "fonts", "simfang.ttf") + + # 初始化界面 + self.setupUi() + + # 下载模型 + self.downloadModels(URLs_EN) + self.downloadModels(URLs_CN) + + self.structure_sys_en = self.initPredictor('EN') + self.structure_sys_cn = self.initPredictor('CN') + + def setupUi(self): + self.setObjectName("MainWindow") + self.setWindowTitle(__APPNAME__ + " " + __VERSION__) + + layout = QtWidgets.QGridLayout() + + openFileButton = QtWidgets.QPushButton("打开文件") + openFileButton.setIcon(QIcon(QPixmap("./icons/folder-plus.png"))) + layout.addWidget(openFileButton, 0, 0, 1, 1) + openFileButton.clicked.connect(self.openFileSlot) + + # screenShotButton = QtWidgets.QPushButton("截图识别") + # layout.addWidget(screenShotButton, 0, 1, 1, 1) + # screenShotButton.clicked.connect(self.screenShotSlot) + # screenShotButton.setEnabled(False) # temporarily disenble + + startCNShotButton = QtWidgets.QPushButton("中文转换") + startCNShotButton.setIcon(QIcon(QPixmap("./icons/chinese.png"))) + layout.addWidget(startCNShotButton, 0, 1, 1, 1) + startCNShotButton.clicked.connect( + functools.partial(self.startSlot, 'CN')) + + startENButton = QtWidgets.QPushButton("英文转换") + startENButton.setIcon(QIcon(QPixmap("./icons/english.png"))) + layout.addWidget(startENButton, 0, 2, 1, 1) + startENButton.clicked.connect( + functools.partial(self.startSlot, 'EN')) + + showResultButton = QtWidgets.QPushButton("显示结果") + showResultButton.setIcon(QIcon(QPixmap("./icons/folder-open.png"))) + layout.addWidget(showResultButton, 0, 3, 1, 1) + showResultButton.clicked.connect(self.showResultSlot) + + self.pb = QtWidgets.QLabel( + self.pb_text.format(0, 0)) + layout.addWidget(self.pb, 1, 0, 1, 4) + + self.setLayout(layout) + + def downloadModels(self, URLs): + # using custom model + tar_file_name_list = [ + 'inference.pdiparams', + 'inference.pdiparams.info', + 'inference.pdmodel', + 'model.pdiparams', + 'model.pdiparams.info', + 'model.pdmodel' + ] + model_path = os.path.join(here, 'inference') + os.makedirs(model_path, exist_ok=True) + + # download and unzip models + for name in URLs.keys(): + url = URLs[name] + print("Try downloading file: {}".format(url)) + tarname = url.split('/')[-1] + tarpath = os.path.join(model_path, tarname) + if os.path.exists(tarpath): + print("File have already exist. skip") + else: + try: + download_with_progressbar(url, tarpath) + except Exception as e: + print("Error occurred when downloading file, error message:") + print(e) + + # unzip model tar + try: + with tarfile.open(tarpath, 'r') as tarObj: + storage_dir = os.path.join(model_path, name) + os.makedirs(storage_dir, exist_ok=True) + for member in tarObj.getmembers(): + filename = None + for tar_file_name in tar_file_name_list: + if tar_file_name in member.name: + filename = tar_file_name + if filename is None: + continue + file = tarObj.extractfile(member) + with open( + os.path.join(storage_dir, filename), + 'wb') as f: + f.write(file.read()) + except Exception as e: + print("Error occurred when unziping file, error message:") + print(e) + + def initPredictor(self, lang='EN'): + # init predictor args + args = parse_args() + args.table_max_len = 488 + args.ocr = True + args.recovery = True + args.save_pdf = self.save_pdf + args.table_char_dict_path = os.path.join(here, + "ppocr", "utils", "dict", "table_structure_dict.txt") + if lang == 'EN': + args.det_model_dir = os.path.join(here, # 此处从这里找到模型存放位置 + "inference", "en_PP-OCRv3_det_infer") + args.rec_model_dir = os.path.join(here, + "inference", "en_PP-OCRv3_rec_infer") + args.table_model_dir = os.path.join(here, + "inference", "en_ppstructure_mobile_v2.0_SLANet_infer") + args.output = os.path.join(here, "output") # 结果保存路径 + args.layout_model_dir = os.path.join(here, + "inference", "picodet_lcnet_x1_0_fgd_layout_infer") + lang_dict = DICT_EN + elif lang == 'CN': + args.det_model_dir = os.path.join(here, # 此处从这里找到模型存放位置 + "inference", "cn_PP-OCRv3_det_infer") + args.rec_model_dir = os.path.join(here, + "inference", "cn_PP-OCRv3_rec_infer") + args.table_model_dir = os.path.join(here, + "inference", "cn_ppstructure_mobile_v2.0_SLANet_infer") + args.output = os.path.join(here, "output") # 结果保存路径 + args.layout_model_dir = os.path.join(here, + "inference", "picodet_lcnet_x1_0_fgd_layout_cdla_infer") + lang_dict = DICT_CN + else: + raise ValueError("Unsupported language") + args.rec_char_dict_path = os.path.join(here, + "ppocr", "utils", + lang_dict['rec_char_dict_path']) + args.layout_dict_path = os.path.join(here, + "ppocr", "utils", "dict", "layout_dict", + lang_dict['layout_dict_path']) + # init predictor + return StructureSystem(args) + + def openFileSlot(self): + ''' + 可以多选图像文件 + ''' + selectedFiles = QtWidgets.QFileDialog.getOpenFileNames(self, + "多文件选择", "/", "图片文件 (*.png *.jpeg *.jpg *.bmp *.pdf)")[0] + if len(selectedFiles) > 0: + self.imagePaths = selectedFiles + self.screenShot = None # discard screenshot temp image + self.updateProgressBar(len(selectedFiles), 0) + + def screenShotSlot(self): + ''' + 选定图像文件和截图的转换过程只能同时进行一个 + 截图只能同时转换一个 + ''' + self.screenShotWg.start() + if self.screenShotWg.captureImage: + self.screenShot = self.screenShotWg.captureImage + self.imagePaths.clear() # discard openfile temp list + self.updateProgressBar(1, 0) + + def startSlot(self, lang): + if self.screenShot: # for screenShot + img_name = 'screenshot_' + time.strftime("%Y%m%d%H%M%S", time.localtime()) + image = QImageToCvMat(self.screenShot) + self.predictAndSave(image, img_name, lang) + # update Progress Bar + self.updateProgressBar(1, 1) + QtWidgets.QMessageBox.information(self, + u'Information', "文档提取完成") + elif len(self.imagePaths) > 0 : # for image file selection + self.output_dir = os.path.join( + os.path.dirname(self.imagePaths[0]), "output") # output_dir shold be same as imagepath + os.makedirs(self.output_dir, exist_ok=True) + for i, image_file in enumerate(self.imagePaths): + if os.path.basename(image_file)[-3:] in ['pdf']: + import fitz + from PIL import Image + imgs = [] + with fitz.open(image_file) as pdf: + for pg in range(0, pdf.pageCount): + page = pdf[pg] + mat = fitz.Matrix(2, 2) + pm = page.getPixmap(matrix=mat, alpha=False) + + # if width or height > 2000 pixels, don't enlarge the image + if pm.width > 2000 or pm.height > 2000: + pm = page.getPixmap(matrix=fitz.Matrix(1, 1), alpha=False) + + img = Image.frombytes("RGB", [pm.width, pm.height], pm.samples) + img = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR) + imgs.append(img) + + else: + img = cv2.imread(image_file) + if img is None: + print("error in loading image:{}".format(image_file)) + continue + imgs = [img] + + img_name = os.path.basename(image_file).split('.')[0] + os.makedirs(os.path.join(self.output_dir, img_name), exist_ok=True) + self.predictAndSave(imgs, img_name, lang) + + # update Progress Bar + self.updateProgressBar(len(self.imagePaths), i+1) + QtWidgets.QMessageBox.information(self, + u'Information', "文档提取完成") + else: + print('empty input') + + def predictAndSave(self, imgs, img_name, lang): + all_res = [] + for index, img in enumerate(imgs): + if lang == 'EN': + res, time_dict = self.structure_sys_en(img) + elif lang == 'CN': + res, time_dict = self.structure_sys_cn(img) + + # save output + save_structure_res(res, self.output_dir, img_name) + draw_img = draw_structure_result(img, res, self.vis_font_path) + img_save_path = os.path.join(self.output_dir, img_name, 'show_{}.jpg'.format(index)) + if res != []: + cv2.imwrite(img_save_path, draw_img) + + # recovery + h, w, _ = img.shape + res = sorted_layout_boxes(res, w) + all_res += res + + try: + convert_info_docx(img, all_res, self.output_dir, img_name, self.save_pdf) + except Exception as ex: + QtWidgets.QMessageBox.information(self, + u'Information', "error in layout recovery image:{}, err msg: {}".format( + img_name, ex)) + + print('result save to {}'.format(self.output_dir)) + + def showResultSlot(self): + if os.path.exists(self.output_dir): + if platform.system() == 'Windows': + os.startfile(self.output_dir) + else: + os.system('open ' + os.path.normpath(self.lastOpenDir)) + else: + QtWidgets.QMessageBox.information(self, + u'Information', "输出文件不存在") + + def updateProgressBar(self, loaded, finished): + self.pb.setText( + self.pb_text.format(loaded, finished)) + + +def main(): + app = QtWidgets.QApplication(sys.argv) + + window = APP_Image2Doc() # 创建对象 + window.show() # 全屏显示窗口 + + QtWidgets.QApplication.processEvents() + sys.exit(app.exec()) + + +if __name__ == "__main__": + main() \ No newline at end of file