a –½Df=0ã@sÖddlZddlZddlZddlZddlmZddlm Z ddl m Z m Z ddl mZddlmZddlmZmZmZmZmZmZd d „Zd d „Zd d„ZGdd„deƒZdd„Zddd„Zdd„Zdd„Zdd„Z dS)éNé)ÚDataset)Ú BoundingBox)ÚPandasInterfaceÚdefault_datatype)Ú Operation)ÚSlice)Úcartesian_productÚdatetime_typesÚ is_cyclicÚis_nanÚ one_to_oneÚsort_topologicallyc Cs°| dd¡d}| d¡}t dgt t |¡¡dddgg¡}g}t |¡}tt |ƒdƒD]J}||d}t |||dur„dn|dƒ}|j |}t |ƒr`|  |¡q`|S)z^ Split a Path type containing a single NaN separated path into multiple subpaths. réN) ÚsplitÚdimension_valuesÚnpÚ concatenateÚwhereÚisnanrÚ as_dframeÚrangeÚlenÚsliceZilocÚappend) ÚpathÚvaluesÚsplitsZsubpathsÚdataÚiÚendÚslcÚsubpath©r#úc/nfs/NAS7/SABIOD/METHODE/ermites/ermites_venv/lib/python3.9/site-packages/holoviews/element/util.pyÚ split_paths (    r%cCs’|\}}|\}}|j ¡\}}} } |j|j} } d| } d| }t|tƒr`t tt | ƒƒ|j ¡} t|tƒr€t tt |ƒƒ|j ¡}t|t ƒr¾|j dur˜|n t ||j ƒ}|jdur²| n t| |jƒ} t|t ƒrü|j durÖ|n t ||j ƒ}|jdurð| n t| |jƒ} t||f| | ffd}t||ƒ}| |¡ ¡\}}} } t|t ƒsØt|ttfƒsT|g}t|ƒdkrv| tt|ƒƒ|} gg}}|D]:}| ||¡\}}| || d¡| || d¡q„t |¡t  |¡}} n¤t|t ƒs|t|ttfƒsú|g}t|ƒdkr| tt|ƒƒ|} gg}}|D]:}| ||¡\}}| ||d¡| ||d¡q*t |¡t  |¡}} t||f| | ffdS)zµ Given a 2D selection consisting of slices/coordinates, a SheetCoordinateSystem and the shape of the array returns a new BoundingBox representing the sliced region. gð?N)Zpointsrr)ÚboundsZlbrtÚxdensityÚydensityÚ isinstancer rZ timedelta64ÚintÚroundZ _time_unitrÚstartÚmaxÚstopÚminrrZcompute_boundsÚlistÚsetrÚfloatZclosest_cell_centerr)ZslicesZscsÚshapeZxidxZyidxÚysÚxsÚlÚbÚrÚtr'r(ZxunitZyunitr&r!ZlsÚrsÚidxZxcÚ_ÚbsÚtsZycr#r#r$Úcompute_slice_bounds)sP         r?cCs6t|tjƒr|jn|}|D]}t|ƒs|SqtjS)z? Aggregation function to get the first non-zero value. )r)ÚpdZSeriesrr rÚnan)ÚxrÚvr#r#r$Ú reduce_fn]s  rDc@sHeZdZdZejddgddZedd„ƒZdd „Z d d „Z dd d„Z d S)Úcategorical_aggregate2da§ Generates a gridded Dataset of 2D aggregate arrays indexed by the first two dimensions of the passed Element, turning all remaining dimensions into value dimensions. The key dimensions of the gridded array are treated as categorical indices. Useful for data indexed by two independent categorical variables such as a table of population values indexed by country and year. Data that is indexed by continuous dimensions should be binned before aggregation. The aggregation will retain the global sorting order of both dimensions. >> table = Table([('USA', 2000, 282.2), ('UK', 2005, 58.89)], kdims=['Country', 'Year'], vdims=['Population']) >> categorical_aggregate2d(table) Dataset({'Country': ['USA', 'UK'], 'Year': [2000, 2005], 'Population': [[ 282.2 , np.nan], [np.nan, 58.89]]}, kdims=['Country', 'Year'], vdims=['Population']) ZxarrayZgridzO The grid interface types to use when constructing the gridded Dataset.)ÚdefaultÚdoccCsŒ|jdddd…\}}| |d¡}| |d¡}|jjdvrFt |¡}|jjdvr`|t |¡fS|j|ttd  ¡}i}d}|D]œ} |  |d¡} t | ƒdkr°| d g|| d <n4t t | ƒdƒD]"} | | | d…\} } | g|| <qÀ|r€| jjd vr t  | ¡d k  ¡}q€t t | ¡| ¡}q€|s0t||ƒrdnd}| |¡}tjj ||j j d}|  |¡}t |ƒ}t dd „|Dƒƒ} |jD]} ||| jj | ¡jf7}qŠ|j||jj|d S) NcSsg|] }|j‘qSr#)rh©Ú.0Údr#r#r$Ú ÊózEcategorical_aggregate2d._aggregate_dataset_pandas..FreÚuniquez non-unique)Únamescss|]}|jdVqdS)rN)r3rur#r#r$Ú ÑryzDcategorical_aggregate2d._aggregate_dataset_pandas..)rcrI)rarZ set_indexrOrkrr_r@Z MultiIndexZ from_productÚindexr{rjÚtuplerbrhrrnÚTrmrorc) rprVZ index_colsrsrIÚlevelsr}rrrr3rqr#r#r$Ú_aggregate_dataset_pandasÉs   z1categorical_aggregate2d._aggregate_dataset_pandasNcCsZt|tƒr|jjr|S|jdkr*tdƒ‚nt| ¡ƒdkrBtdƒ‚t|dgd}| |¡S)zÇ Generates a categorical 2D aggregate by inserting NaNs at all cross-product locations that do not already have a value assigned. Returns a 2D gridded Dataset object. rz)Cannot aggregate more than two dimensionsézUMust have at two dimensions to aggregate overand one value dimension to aggregate on.Z dataframerd) r)rriZgriddedZndimsÚ ValueErrorrrKr)rprVÚkeyr#r#r$Ú_processÖs  z categorical_aggregate2d._process)N) Ú__name__Ú __module__Ú __qualname__Ú__doc__ÚparamÚListrcÚ classmethodr_rtrr…r#r#r#r$rEhs &# rEcCsLt|ƒ}|sgggfStj|t |¡d}t |¡}t |¡}|||fS)z7 Lay out nodes on a circle and add node index. r)rrÚpiZarangeÚcosÚsin)ÚnodesÚNÚcircrBÚyr#r#r$Úcircular_layoutès   r”©rré2cCsÔt dd|¡}|\}}|\}}|\} } |\} } d|d|dd|d|| dd||d| |d|} d|d|dd|d|| dd||d| |d|}t | |g¡S)zd Compute quadratic bezier spline given start and end coordinate and two control points. rrr‚r)rZlinspaceZ column_stack)r,r Zc0Zc1ZstepsZsxZsyÚexZeyZcx0Zcy0Zcx1Zcy1r5r4r#r#r$Úquadratic_bezierõs &ÿ ÿ&ÿ ÿr˜c Csä| ¡}d|j_| ¡}|j ¡}|j\}}|jjdd…\}}}tj|||jg|jgd}|j|jd|jdid}tj|||jg|jgd}|j|jd|jd id}|  d¡j dgd d }gd ¢} t || j   |jjd d ¡ƒ} | S)zá Given a Graph element containing abstract edges compute edge segments directly connecting the source and target nodes. This operation depends on pandas and is a lot faster than the pure NumPy equivalent. Zgraph_edge_indexNr‚©Zleft_onZright_onÚsrc_xÚsrc_y©ÚcolumnsÚdst_xÚdst_yr©Úaxis)ršr›ržrŸr)Údframer}rhrlrrar@ÚmergeÚrenameÚ sort_valuesÚdropr0rrnÚsize) ÚgraphÚedgesrÚsrcZtgtrBr“r;rsÚcolsZ edge_segmentsr#r#r$Úconnect_edges_pds  r¬c Cs,| ¡ ¡}d|j_|jdddd ¡}|j ¡ ¡}d|j_|jddgddd}|j\}}}|jjdd …\}}}tj |||jg|jgd } | j |jd |jd id } tj | ||jg|jgd } | j |jd|jdid } tj | ||jg|jgd } | j |jd|jdid } |   d¡jdgdd} | gd¢S)zã Given a TriMesh element containing abstract edges compute edge segments directly connecting the source and target nodes. This operation depends on pandas and is a lot faster than the pure NumPy equivalent. Ztrimesh_edge_indexÚcolorÚignorer)Úerrorsr¡Z node_indexÚzNr‚r™Úx0Úy0rœÚx1Úy1Úx2Úy2r )r±r²r³r´rµr¶) r¢r`r}rhr¦rlrrar@r£r¤r¥) Ztrimeshr©rZv1Zv2Zv3rBr“r;rsr#r#r$Úconnect_tri_edges_pd s   r·cCsªg}| |j¡D]”\}}|jdd…dd…|f}|jdd…dd…|f}t|ƒrXt|ƒs`tdƒ‚| |jdd…¡}| |jdd…¡}| t |d|dg¡¡q|S)zù Given a Graph element containing abstract edges compute edge segments directly connecting the source and target nodes. This operation just uses internal HoloViews operations and will be a lot slower than the pandas equivalent. Nz+Could not find node positions for all edgesrr)Úarrayrarrrƒrr)r¨Úpathsr,r Zstart_dsZend_dsr#r#r$Ú connect_edges:srº)r•r•r–)!rSÚnumpyrZpandasr@rŠÚcorerZcore.boundingregionrZ core.datarrZcore.operationrZcore.sheetcoordsrZ core.utilr r r r r rr%r?rDrEr”r˜r¬r·rºr#r#r#r$Ús&     4