o jk*@s@dZddlZddlZddlmZddlmZGdddeZdS)aObject with a bounding box, e.g. Block, Line, Span. Based on ``PyMuPDF``, the coordinates (e.g. bbox of ``page.get_text('rawdict')``) are generally provided relative to the un-rotated page; while this ``pdf2docx`` library works under real page coordinate system, i.e. with rotation considered. So, any instances created by this Class are always applied a rotation matrix automatically. Therefore, the bbox parameter used to create ``Element`` instance MUST be relative to un-rotated CS. If final coordinates are provided, should update it after creating an empty object:: Element().update_bbox(final_bbox) .. note:: An exception is ``page.get_drawings()``, the coordinates are converted to real page CS already. N)IText) constantsc @s eZdZdZedZeddZeddZ d6de fd d Z d d Z d dZ eddZejddZddZdefddZddZddZd7dddefddZd8defd!d"Zd9d$ed%efd&d'Zd9d$ed%efd(d)Zd*d+Zd,d-Zd:d0ed1ed2ed3efd4d5ZdS);Elementz.Boundary box with attribute in fitz.Rect type.cCs"|r t|tjr||_dSdSdS)zlSet global rotation matrix. Args: Rotation_matrix (fitz.Matrix): target matrix N) isinstancefitzMatrixROTATION_MATRIX)clsZrotation_matrixr X/var/www/html/Deteccion_Ine/venv/lib/python3.10/site-packages/pdf2docx/common/Element.pyset_rotation_matrixs zElement.set_rotation_matrixcCs&|j\}}}}}}t||||ddS)zHPure rotation matrix used for calculating text direction after rotation.r)r rr )r abcdefr r r pure_rotation_matrix*szElement.pure_rotation_matrixNrawcCsBt|_||_d|p ivrt|dtj}||dSdS)zJ Initialize Element and convert to the real (rotation considered) page CS.bboxN)rRectr_parentrr update_bbox)selfrparentrectr r r __init__1s  zElement.__init__cCs t|jS)z!Real object when bbox is defined.)boolrrr r r __bool__=s zElement.__bool__cCs|jjdt|jdS)N()) __class____name__tuplerr r r r __repr__IszElement.__repr__cCs|jSNrr r r r rOszElement.parentcCs ||_dSr(r))rrr r r rRs cCs"|jd}|_t|}||_|S)zmake a deep copy.N)rrcopydeepcopy)rrobjr r r r*Ys z Element.copydtcCs|j| | ||fS)aGet expanded bbox with margin in both x- and y- direction. Args: dt (float): Expanding margin. Returns: fitz.Rect: Expanded bbox. .. note:: This method creates a new bbox, rather than changing the bbox of itself. )r)rr-r r r get_expand_bboxcs zElement.get_expand_bboxcCstdd|D|_|S)zUpdate current bbox to specified ``rect``. Args: rect (fitz.Rect or list): bbox-like ``(x0, y0, x1, y1)``, in real page CS (with rotation considered). cSsg|]}t|dqS)r)round.0xr r r ysz'Element.update_bbox..)rrr)rrr r r rrszElement.update_bboxcCs||j|jBS)zUpdate current bbox to the union with specified Element. Args: e (Element): The target to get union Returns: Element: self )rr)rrr r r union_bbox}s zElement.union_bbox?r thresholdcCsx|j}|s dS|j|j@}t||d}||krdS|jj|jjkr1|jjtj|jjkS|jjtj|jjkS)a5Whether given element is contained in this instance, with margin considered. Args: e (Element): Target element threshold (float, optional): Intersection rate. Defaults to 1.0. The larger, the stricter. Returns: bool: [description] F)rget_arear/widthheightrZ MINOR_DIST)rrr6S intersectionfactorr r r containss   zElement.containsffffff?c Csx|j}t|dr |jnt|}||@}|jrdS|||}}}|r0|t||nd} | |kr:||BSdS)aCIf the intersection with ``e`` exceeds the threshold, return the union of these two elements; else return None. Args: e (Element): Target element. threshold (float, optional): Intersection rate. Defaults to 0.95. Returns: fitz.Rect: Union bbox or None. rNgư>)rhasattrrrZis_emptyr8min) rrr6Zbbox_1Zbbox_2rZa1Za2rr=r r r get_main_bboxs  zElement.get_main_bboxTr=text_directionc C|rt|sdS|r|jrdnd}|j|d|j|}|j|d|j|}t|j|d|j|dt|j||j|}d}|||||t||kS)aCheck whether two Element instances have enough intersection in vertical direction, i.e. perpendicular to reading direction. Args: e (Element): Object to check with factor (float, optional): Threshold of overlap ratio, the larger it is, the higher probability the two bbox-es are aligned. text_direction (bool, optional): Consider text direction or not. True by default. Returns: bool: [description] Examples:: +--------------+ | | +--------------+ L1 +-------------------+ | | +-------------------+ L2 An enough intersection is defined based on the minimum width of two boxes:: L1+L2-L>factor*min(L1,L2) Frrr7MbP?rZis_vertical_textrmaxrA rrr=rCidxZL1ZL2Lepsr r r vertically_align_withs4zElement.vertically_align_withc CrD)aCheck whether two Element instances have enough intersection in horizontal direction, i.e. along the reading direction. Args: e (Element): Element to check with factor (float, optional): threshold of overlap ratio, the larger it is, the higher probability the two bbox-es are aligned. text_direction (bool, optional): consider text direction or not. True by default. Examples:: +--------------+ | | L1 +--------------------+ +--------------+ | | L2 +--------------------+ An enough intersection is defined based on the minimum width of two boxes:: L1+L2-L>factor*min(L1,L2) Frrr7rErFrHr r r horizontally_align_withs4zElement.horizontally_align_withcCs|r|j|jkr dS|jrdnd}|j||j|dd}|j||j|dd}||j|dko>||j|dk}|S)aCheck whether in same row/line with specified Element instance. With text direction considered. Taking horizontal text as an example: * yes: the bottom edge of each box is lower than the centerline of the other one; * otherwise, not in same row. Args: e (Element): Target object. .. note:: The difference to method ``horizontally_align_with``: they may not in same line, though aligned horizontally. Frrr7g@)Zis_horizontal_textr)rrrIc1c2resr r r in_same_row s$zElement.in_same_rowcCsdtdd|jDiS)zStore properties in raw dict.rcss|]}|VqdSr(r r0r r r -sz Element.store..)r&rr r r r store+sz Element.storerrr?stroker9filldashesc Cs|j|j||||ddddS)z(Plot bbox in PDF page for debug purpose.FrU)colorrWr9rXoverlayZ fill_opacityN)Z draw_rectr)rpagerVr9rWrXr r r plot0s z Element.plot)NN)r5)r?)rT)rTrUNN) r% __module__ __qualname____doc__rr r classmethodrrdictrr!r'propertyrsetterr*floatr.rr4r>rBrrLrMrQrSr&strr\r r r r rs2         )" r)r_r*rZsharerrrr r r r s