Walk

Summary

Returns data names in directory and database structures by moving through the tree from the top down or the bottom up. Each directory or workspace yields a tuple of three: directory path, directory names, and file names.

Discussion

The os module includes an os.walk function that can be used to move through a directory tree and find data. The os.walk function is file based and does not recognize database content such as geodatabase feature classes, tables, or rasters. For better performance, it is recommended that you use os.walk for file-based formats. The arcpy.da.Walk function can be used to catalog data.

Syntax

Walk (top, {topdown}, {onerror}, {followlinks}, {datatype}, {type})

Parameter	Explanation	Data Type
top	The top-level workspace that will be used.	String
topdown	If topdown is True or not specified, the tuple for a directory is generated before the tuple for any of its workspaces (workspaces are generated trom the top down). If topdown is False, the tuple for a workspace is generated after the tuple for all of its subworkspaces (workspaces are generated from the bottom up). When topdown is True, the dirnames list can be modified in place, and Walk will only recurse into the subworkspaces whose names remain in dirnames. This can be used to limit the search, impose a specific order of visiting, or inform Walk about directories the caller creates or renames before it resumes Walk again. Modifying dirnames when topdown is False is ineffective, because in bottom-up mode, the workspaces in dirnames are generated before dirpath is generated. (The default value is True)	Boolean
onerror	Errors are ignored by default. The onerror function will be called with an OSError instance. This function can be used to report the error and continue with Walk or raise an exception to cancel. Note: The file name is available as the filename attribute of the exception object. (The default value is None)	Function
followlinks	By default, Walk does not visit connection files. Set followlinks to True to visit connection files. (The default value is False)	Boolean
datatype	Specifies the data type that will be used to limit the results. Any—All data types are returned. This is equivalent to using None or skipping the argument. CadDrawing—CAD files (.dwg, .dxf, and .dgn) are returned. FeatureClass—Feature classes in a geodatabase and shapefiles (.shp) are returned. FeatureDataset—Feature datasets in a geodatabase are returned. GeometricNetwork—Geometric networks in a geodatabase are returned. LasDataset—LAS dataset files (.las and .lasd) are returned. Layer—Layer files (.lyrx) are returned. Locator—Locator files (.loc) are returned. Map— ArcGIS Pro projects (.aprx files), ArcGIS Pro map documents (.mapx files), and ArcMap documents (.mxd files) are returned. MosaicDataset—Mosaic datasets in a geodatabase are returned. NetworkDataset—Network datasets in a geodatabase are returned. ParcelDataset—Parcel fabric datasets are returned. RasterCatalog—Raster catalogs in a geodatabase are returned. RasterDataset—Raster datasets in a geodatabase or in a folder are returned. RelationshipClass—Relationship classes in a geodatabase are returned. RepresentationClass—Representation classes in a geodatabase are returned. Table—Tables in a geodatabase or folder and sheets in Excel workbooks are returned. Terrain—Terrain datasets are returned. Tin—TIN surfaces in a geodatabase are returned. Tool—Tools in a toolbox are returned. Topology—Topologies in a geodatabase are returned. UtilityNetwork—Utility networks in a geodatabase are returned. Multiple data types are supported if they're entered as a list or tuple. for dirpath, dirnames, filenames in arcpy.da.Walk(workspace, datatype=['MosaicDataset', 'RasterDataset']): (The default value is None)	String
type	Specifies whether feature and raster data types will be further limited by type. All—All types are returned. This is equivalent to using None or skipping the argument. Any—All types are returned. This is equivalent to using None or skipping the argument. Valid feature types are the following: Multipatch—Multipatch feature classes are returned. Multipoint—Multipoint feature classes are returned. Point—Point feature classes are returned. Polygon—Polygon feature classes are returned. Polyline—Polyline feature classes are returned. Valid raster types are the following: BIL— Esri Band Interleaved by Line file BIP— Esri Band Interleaved by Pixel file BMP— Bitmap graphic raster dataset format BSQ— Esri Band Sequential file DAT— ENVI DAT file GIF— Graphic Interchange Format for raster datasets GRID— Esri Grid raster dataset format IMG— ERDAS IMAGINE raster data format JP2— JPEG 2000 raster dataset format JPG— Joint Photographic Experts Group raster dataset format PNG— Portable Network Graphic raster dataset format TIF— Tag Image File Format for raster datasets Multiple data types are supported if they're entered as a list or tuple. for dirpath, dirnames, filenames in arcpy.da.Walk(workspace, datatype='FeatureClass', type=['Polygon', 'Polyline']): (The default value is None)	String

Return Value

Data Type

Explanation

Generator

Yields a tuple of three that includes the workspace, directory names, and file names. dirpath is the path to the workspace as a string. dirnames is a list of names of subdirectories and other workspaces in dirpath. filenames is a list of names of nonworkspace contents in dirpath. Note: Names in the lists include only the base name; no path components are included. For a full path (which begins with top) to a file or directory in dirpath, use os.path.join(dirpath, name).

Code sample

Walk example 1

Use the Walk function to catalog polygon feature classes.

import arcpy
import os

workspace = "c:/data"
feature_classes = []

walk = arcpy.da.Walk(workspace, datatype="FeatureClass", type="Polygon")

for dirpath, dirnames, filenames in walk:
    for filename in filenames:
        feature_classes.append(os.path.join(dirpath, filename))

Walk example 2

Use the Walk function to catalog raster data. Any rasters in a folder named back_up will be ignored.

import arcpy
import os

workspace = "c:/data"
rasters = []

walk = arcpy.da.Walk(workspace, topdown=True, datatype="RasterDataset")

for dirpath, dirnames, filenames in walk:
    # Disregard any folder named 'back_up' in creating list of rasters
    if "back_up" in dirnames:
        dirnames.remove('back_up')
    for filename in filenames:
        rasters.append(os.path.join(dirpath, filename))