ccfx 1.0.9__tar.gz → 1.1.0__tar.gz

{ccfx-1.0.9/ccfx.egg-info → ccfx-1.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ccfx
-Version: 1.0.9
+Version: 1.1.0
 Summary: This package simplifies regular common actions for quick prototyping in a user friendly way
 Author-email: Celray James CHAWANDA <celray@chawanda.com>
 License-Expression: MIT
@@ -26,6 +26,9 @@ Requires-Dist: mutagen
 Requires-Dist: requests
 Requires-Dist: tqdm
 Requires-Dist: pillow
+Requires-Dist: scipy
+Requires-Dist: rasterio
+Requires-Dist: matplotlib
 Dynamic: license-file
 
 # ccfx

{ccfx-1.0.9 → ccfx-1.1.0}/ccfx/ccfx.py

@@ -22,7 +22,6 @@ import platform
 import zipfile
 import pickle
 import time
-from shapely.geometry import box, Point
 import geopandas, pandas
 from collections import defaultdict
 import py7zr
@@ -38,6 +37,12 @@ import yt_dlp
 from typing import Optional, Any
 from datetime import datetime, timedelta
 from PIL import Image
+import scipy as scipy
+from shapely.geometry import LineString, Polygon, MultiPolygon, box, Point
+from shapely.ops import polygonize, unary_union
+import rasterio
+from rasterio import features
+from rasterio.transform import from_bounds
 
 # functions
 def listFiles(path: str, ext: Optional[str] = None) -> list:
@@ -568,6 +573,14 @@ def correctFisheye(inputFile: str, outputFile: str = '',
     subprocess.run(cmd, check=True)
     return outputFile
 
+def correctLens(inputFile: str, outputFile: str = '',
+                        k1: float = -0.1, k2: float = 0.05,
+                        cx: float = 0.5, cy: float = 0.5,
+                        crf: int = 20) -> str:
+    """
+    Alias for correctFisheye
+    """
+    return correctFisheye(inputFile, outputFile, k1, k2, cx, cy, crf)
 
 def formatStringBlock(input_str: str, max_chars: int = 70) -> str:
     '''
@@ -680,6 +693,76 @@ def downloadFile(url: str, save_path: str, exists_action: str = 'resume', num_co
 
 
 
+
+def createPolygonFromOuterPoints(pointsGdf: geopandas.GeoDataFrame, alpha: float = 1.6, keepHoles: bool = False) -> geopandas.GeoDataFrame:
+    """
+    Concave hull (alpha-shape) from points.
+    alpha: larger -> tighter/more detailed; too large can fragment.
+    keepHoles: keep interior holes if True, otherwise drop them.
+    """
+
+    pointsGdf = pointsGdf[pointsGdf.geometry.type.eq("Point") & pointsGdf.geometry.notna()]
+    if pointsGdf.empty or pointsGdf.geometry.nunique() < 3:
+        raise ValueError("need at least three distinct points")
+
+    # coordinates (N, 2)
+    coords = numpy.array([(g.x, g.y) for g in pointsGdf.geometry])
+
+    tri = scipy.spatial.Delaunay(coords)
+    simplices = tri.simplices  # (M, 3) indices into coords
+    triPts = coords[simplices] # (M, 3, 2)
+
+    # side lengths
+    a = numpy.linalg.norm(triPts[:, 1] - triPts[:, 2], axis=1)
+    b = numpy.linalg.norm(triPts[:, 0] - triPts[:, 2], axis=1)
+    c = numpy.linalg.norm(triPts[:, 0] - triPts[:, 1], axis=1)
+
+    s = (a + b + c) / 2.0
+    # heron area, guard against tiny/negative due to fp error
+    areaSq = numpy.maximum(s * (s - a) * (s - b) * (s - c), 0.0)
+    area = numpy.sqrt(areaSq)
+    valid = area > 0.0
+    if not numpy.any(valid):
+        hull = pointsGdf.unary_union.convex_hull
+        return geopandas.GeoDataFrame({"name": ["outerBoundary"]}, geometry=[hull], crs=pointsGdf.crs)
+
+    circumradius = (a * b * c) / (4.0 * area)
+    keep = valid & (circumradius < (1.0 / alpha))
+    keptSimplices = simplices[keep]
+    if keptSimplices.size == 0:
+        hull = pointsGdf.unary_union.convex_hull
+        return geopandas.GeoDataFrame({"name": ["outerBoundary"]}, geometry=[hull], crs=pointsGdf.crs)
+
+    # count triangle edges; boundary edges appear exactly once
+    edgeCounts: dict[tuple[int, int], int] = {}
+    for i0, i1, i2 in keptSimplices:
+        for e in ((i0, i1), (i1, i2), (i2, i0)):
+            key = (e[0], e[1]) if e[0] < e[1] else (e[1], e[0])
+            edgeCounts[key] = edgeCounts.get(key, 0) + 1
+
+    boundaryLines = [
+        LineString([coords[i], coords[j]])
+        for (i, j), count in edgeCounts.items()
+        if count == 1
+    ]
+    if not boundaryLines:
+        hull = pointsGdf.unary_union.convex_hull
+        return geopandas.GeoDataFrame({"name": ["outerBoundary"]}, geometry=[hull], crs=pointsGdf.crs)
+
+    polygons = list(polygonize(boundaryLines))
+    if not polygons:
+        hull = pointsGdf.unary_union.convex_hull
+        return geopandas.GeoDataFrame({"name": ["outerBoundary"]}, geometry=[hull], crs=pointsGdf.crs)
+
+    merged = unary_union(polygons)
+    if isinstance(merged, MultiPolygon):
+        merged = max(merged.geoms, key=lambda g: g.area)
+    if not keepHoles and isinstance(merged, Polygon):
+        merged = Polygon(merged.exterior)
+
+    return geopandas.GeoDataFrame({"name": ["outerBoundary"]}, geometry=[merged], crs=pointsGdf.crs)
+
+
 def mergeRasterTiles(tileList:list, outFile:str) -> str:
     '''
     Merge raster tiles into one raster file
@@ -1030,6 +1113,116 @@ def extractCompressedFile(inputFile: str, outputDir: str, v: bool = False) -> No
     """
     uncompress(inputFile, outputDir, v)
 
+
+
+def rasterizeGDF(gdf: geopandas.GeoDataFrame, valueField: str, outRasterFN: str, resolution: float, isCOG: bool = True, allTouched: bool = False, profileOverrides: dict | None = None) -> str:
+    """
+    Rasterize a GeoDataFrame to GeoTIFF/COG.
+
+    Parameters
+    ----------
+    gdf : geopandas.GeoDataFrame
+    valueField : str
+        Column to burn as pixel values.
+    outRasterFN : str
+    resolution : float
+        Pixel size in CRS units (assumes a projected CRS).
+    isCOG : bool, default True
+        If True, writes Cloud-Optimized GeoTIFF. Otherwise plain GeoTIFF.
+    allTouched : bool, default False
+        Pass-through to rasterio.features.rasterize.
+    profileOverrides : dict | None
+        Extra GDAL profile options (e.g., {"compress": "LZW"}). Overrides defaults.
+
+    Returns
+    -------
+    str
+        The path written to (outRasterFN).
+    """
+    # basic checks
+    if gdf is None or len(gdf) == 0 or gdf.geometry.isna().all():
+        raise ValueError("gdf is empty or has no valid geometries.")
+    if gdf.crs is None:
+        raise ValueError("gdf must have a defined CRS.")
+    if resolution <= 0:
+        raise ValueError("resolution must be > 0.")
+    if valueField not in gdf.columns:
+        raise ValueError(f"valueField '{valueField}' not found in gdf.")
+
+    # compute raster shape + transform
+    bounds = gdf.total_bounds  # (minx, miny, maxx, maxy)
+    width = int(numpy.ceil((bounds[2] - bounds[0]) / float(resolution)))
+    height = int(numpy.ceil((bounds[3] - bounds[1]) / float(resolution)))
+    if width < 1 or height < 1:
+        raise ValueError("computed raster dimensions are invalid (check resolution and bounds).")
+
+    transform = from_bounds(bounds[0], bounds[1], bounds[2], bounds[3], width, height)
+
+    # infer dtype + nodata
+    pandasDtype = gdf[valueField].dtype
+    if numpy.issubdtype(pandasDtype, numpy.floating):
+        dtype = numpy.float32
+        nodata = numpy.nan
+        fillValue = numpy.nan
+    elif numpy.issubdtype(pandasDtype, numpy.bool_):
+        dtype = numpy.uint8
+        nodata = 255  # sentinel for bool raster
+        fillValue = nodata
+    else:
+        dtype = numpy.int32
+        nodata = -9999
+        fillValue = nodata
+
+    # prefill target array
+    raster = numpy.full((height, width), fillValue, dtype=dtype)
+
+    # build shapes generator (ensure python scalars)
+    shapes = ((geom, (None if numpy.isnan(val) else val) if isinstance(val, float) else int(val) if numpy.issubdtype(type(val), numpy.integer) else float(val))
+              for geom, val in zip(gdf.geometry, gdf[valueField]))
+
+    # burn
+    features.rasterize(
+        shapes=shapes,
+        out_shape=raster.shape,
+        transform=transform,
+        fill=fillValue,
+        out=raster,
+        all_touched=allTouched,
+        dtype=dtype
+    )
+
+    # default profile settings
+    profile = {
+        "driver": "COG" if isCOG else "GTiff",
+        "height": raster.shape[0],
+        "width": raster.shape[1],
+        "count": 1,
+        "dtype": raster.dtype,
+        "crs": gdf.crs,
+        "transform": transform,
+        "nodata": nodata,
+    }
+    # sane compression defaults
+    if isCOG:
+        profile.setdefault("compress", "LZW")
+        profile.setdefault("blocksize", 512)  # GDAL COG option via rasterio
+        # overviews are handled by the COG driver
+    else:
+        profile.setdefault("compress", "LZW"),
+        profile.setdefault("tiled", True)
+        profile.setdefault("blockxsize", 512)
+        profile.setdefault("blockysize", 512)
+
+    if profileOverrides:
+        profile.update(profileOverrides)
+
+    # write
+    with rasterio.open(outRasterFN, "w", **profile) as dst:
+        dst.write(raster, 1)
+
+    return outRasterFN
+
+
 def moveDirectory(srcDir:str, destDir:str, v:bool = False) -> bool:
     '''
     this function moves all files from srcDir to destDir
@@ -2076,7 +2269,7 @@ def getTimeseriesStats(data:pandas.DataFrame, observed:Optional[str] = None, sim
 
     return stats
 
-def readSWATPlusOutputs(filePath: str, column: Optional[str] = None, unit: Optional[int] = None, gis_id: Optional[int] = None, name: Optional[str] = None) -> Optional[pandas.DataFrame]:
+def readSWATPlusOutputs(filePath: str, column: Optional[str] = None, unit: Optional[int] = None, gis_id: Optional[int] = None, name: Optional[str] = None, coerceNumeric: bool = True) -> Optional[pandas.DataFrame]:
     '''
     Read SWAT+ output files and return a pandas DataFrame with proper date handling
     and optional filtering capabilities.
@@ -2138,8 +2331,15 @@ def readSWATPlusOutputs(filePath: str, column: Optional[str] = None, unit: Optio
     # Convert all columns to numeric except 'name' (which is string)
     for col in df.columns:
         if col != 'name':
-            df[col] = pandas.to_numeric(df[col], errors='coerce')
+            if coerceNumeric:
+                df[col] = pandas.to_numeric(df[col], errors='coerce')
     
+    if not coerceNumeric:
+        # If not coercing to numeric, ensure date columns are numeric
+        for mandatoryCol in ['yr', 'mon', 'day', 'gis_id']:
+            if mandatoryCol in df.columns:
+                df[mandatoryCol] = pandas.to_numeric(df[mandatoryCol], errors='coerce')
+
     # Create date column from yr, mon, day
     try:
         df['date'] = pandas.to_datetime(pandas.DataFrame({'year': df.yr, 'month': df.mon, 'day': df.day}))
@@ -2179,4 +2379,5 @@ def readSWATPlusOutputs(filePath: str, column: Optional[str] = None, unit: Optio
     
     return df
 
+
 ignoreWarnings()

{ccfx-1.0.9 → ccfx-1.1.0/ccfx.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ccfx
-Version: 1.0.9
+Version: 1.1.0
 Summary: This package simplifies regular common actions for quick prototyping in a user friendly way
 Author-email: Celray James CHAWANDA <celray@chawanda.com>
 License-Expression: MIT
@@ -26,6 +26,9 @@ Requires-Dist: mutagen
 Requires-Dist: requests
 Requires-Dist: tqdm
 Requires-Dist: pillow
+Requires-Dist: scipy
+Requires-Dist: rasterio
+Requires-Dist: matplotlib
 Dynamic: license-file
 
 # ccfx

{ccfx-1.0.9 → ccfx-1.1.0}/ccfx.egg-info/requires.txt

@@ -14,3 +14,6 @@ mutagen
 requests
 tqdm
 pillow
+scipy
+rasterio
+matplotlib

{ccfx-1.0.9 → ccfx-1.1.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "ccfx"
-version = "1.0.9"
+version = "1.1.0"
 description = "This package simplifies regular common actions for quick prototyping in a user friendly way"
 readme = "README.md"
 license = "MIT"
@@ -31,6 +31,9 @@ dependencies = [
     "requests",
     "tqdm",
     "pillow",
+    "scipy",
+    "rasterio",
+    "matplotlib",
 ]
 
 [project.urls]