Package 'rustycogs'

references as tiles

Description

Add extent information xmin,xmax,ymin,ymax to refs table

Usage

refs_tiles(x)

Arguments

x	file path or url

Value

dataframe with xmin,xmax,ymin,ymax add to the tiffs_refs output

See Also

tiles_to_matrix

Examples

vfile <- system.file("extdata/volcano-cog.tif", package = "rustycogs", mustWork = TRUE)
refs <- refs_tiles(vfile)
dim(refs)
str(refs)

---

Summarise IFD-level metadata from TIFF/COG files.

Description

Returns one row per IFD per file — no tile explosion. Useful for understanding file structure before deciding which tiles to fetch.

Returns one row per IFD per file with no tile explosion. Use this to understand file structure — how many IFDs (overviews), tile grid dimensions, dtype, CRS, geotransform — before deciding which tiles to fetch with tiff_refs() or tiff_read_tiles().

Usage

tiff_ifd_info(paths, region = NULL, anon = FALSE, concurrency = 16L)

tiff_ifd_info(paths, region = NULL, anon = FALSE, concurrency = 16L)

Arguments

paths	Character vector of file paths or URLs.
region	Optional AWS region string (e.g. "ap-southeast-2").
anon	Logical. Use anonymous/unsigned requests. Default FALSE.
concurrency	Integer. Maximum concurrent file scans. Default 16L.

Value

A data.frame with one row per IFD containing image dimensions, tile grid size, dtype, compression, photometric interpretation, predictor, planar configuration, CRS, nodata, and geotransform.

A data frame with one row per IFD and columns: path, ifd, is_tiled, image_w, image_h, tile_w, tile_h, n_tiles_x, n_tiles_y, dtype, compression, bits_per_sample, samples_per_pixel, photometric, predictor, planar_configuration, crs_epsg, gdal_nodata, scale_x, scale_y, origin_x, origin_y.

See Also

tiff_refs() for the full per-tile data frame

Examples

## Not run: 
tiff_ifd_info("scene.tif")
tiff_ifd_info("s3://my-bucket/data.tif", anon = TRUE)

## End(Not run)

---

Extract tile byte-range references from TIFF/COG files.

Description

Scans one or more TIFF or COG files and returns a data frame with one row per tile per IFD, containing the byte-range metadata needed to construct Kerchunk or Zarr virtual stores.

Usage

tiff_refs(paths, region = NULL, anon = FALSE, concurrency = 16L)

tiff_refs(paths, region = NULL, anon = FALSE, concurrency = 16L)

Arguments

paths	Character vector of file paths or URLs. Supported schemes: ⁠s3://⁠, ⁠gs://⁠, ⁠az://⁠, ⁠http://⁠, ⁠https://⁠, or plain local paths.
region	Optional AWS region string (e.g. "ap-southeast-2"). Ignored for non-S3 URLs.
anon	Logical. Use anonymous/unsigned requests. Default FALSE.
concurrency	Integer. Maximum number of files scanned concurrently. Default 16L.

Value

A data.frame with columns path, ifd, tile_col, tile_row, offset, length, image_w, image_h, tile_w, tile_h, valid_w, valid_h, dtype, compression, bits_per_sample, samples_per_pixel, crs_epsg, gdal_nodata, planar_configuration, scale_x, scale_y, origin_x, origin_y.

A data frame with columns: path, ifd, tile_col, tile_row, offset, length, image_w, image_h, tile_w, tile_h, dtype, compression, bits_per_sample, samples_per_pixel, crs_epsg, gdal_nodata, planar_configuration, scale_x, scale_y, origin_x, origin_y.

See Also

tiff_ifd_info() for a compact one-row-per-IFD summary, tiff_read_tiles() to fetch pixel data

Examples

## Not run: 
refs <- tiff_refs("s3://my-bucket/data.tif", anon = TRUE)
refs <- tiff_refs(c("file1.tif", "file2.tif"), concurrency = 4L)

## End(Not run)

---

Fetch and decode a single tile from a TIFF/COG file.

Description

Fetch and decode a single tile from a TIFF/COG file.

Fetch and decode a single tile from a TIFF/COG file

Usage

tiff_tile(path, ifd_index = 0L, col, row, region = NULL, anon = FALSE)

tiff_tile(path, ifd_index = 0L, col, row, region = NULL, anon = FALSE)

Arguments

path	File path or URL to the TIFF.
ifd_index	IFD index (0-based). 0 is the full-resolution image.
col	Tile column index (0-based).
row	Tile row index (0-based).
region	Optional AWS region string.
anon	Logical. Use anonymous requests. Default FALSE.

Value

A named list with data, dim, dtype.

A named list with:

• data: numeric vector of decoded pixel values (row-major)

• dim: integer vector c(height, width, bands)

• dtype: character dtype string (e.g. "<f4", "<u2")

See Also

tile_to_array() to reshape the result, tiff_tiles() for batches

Examples

## Not run: 
tile <- tiff_tile("scene.tif", col = 0L, row = 0L)
m <- tile_to_array(tile)

## End(Not run)

---

Fetch and decode multiple tiles from a TIFF/COG file.

Description

Uses ifd.fetch_tiles() for a single vectorized range request.

Fetches a batch of tiles from a single file in one vectorized request. More efficient than calling tiff_tile() in a loop. For multi-file batches use tiff_read_tiles().

Usage

tiff_tiles(path, ifd_index = 0L, cols, rows, region = NULL, anon = FALSE)

tiff_tiles(path, ifd_index = 0L, cols, rows, region = NULL, anon = FALSE)

Arguments

path	File path or URL to the TIFF.
ifd_index	IFD index (0-based). Default 0L.
cols	Integer vector of tile column indices (0-based).
rows	Integer vector of tile row indices (0-based). Must be the same length as cols.
region	Optional AWS region string.
anon	Logical. Use anonymous requests. Default FALSE.

Value

A list of tile result lists, each with data, dim, dtype.

A list of tile results, each with data, dim, and dtype components as described in tiff_tile().

See Also

tile_to_array() to reshape individual results, tiff_read_tiles() for multi-file batches

Examples

## Not run: 
tiles <- tiff_tiles("scene.tif", cols = 0:3, rows = rep(0L, 4))
arrays <- lapply(tiles, tile_to_array)

## End(Not run)

---

Convert a tile result to a matrix or array

Description

Reshapes the flat numeric vector returned by tiff_tile(), each element of tiff_tiles(), or each element of the data column from tiff_read_tiles() into a matrix (single band) or 3D array (multi-band).

Usage

tile_to_array(tile)

Arguments

tile	A tile result list from tiff_tile(), tiff_tiles(), or an element of the data column from tiff_read_tiles().

Details

The matrix is filled row-by-row (byrow = TRUE), consistent with the row-major (C) order returned by async-tiff and expected by graphics::rasterImage() and ximage(). Note that as.vector(m) on the result does not recover the original input order; use as.vector(t(m)) for a round-trip.

Value

A matrix for single-band tiles, or a 3D array with dimensions ⁠[height, width, bands]⁠ for multi-band tiles.

Examples

## Not run: 
tile <- tiff_tile("scene.tif", col = 0L, row = 0L)
m <- tile_to_array(tile)
dim(m)  # c(256, 256) for a single-band 256x256 tile
ximage::ximage(m)

## End(Not run)

---

Turn already-read data into a full matrix

Description

Turn already-read data into a full matrix

Usage

tiles_to_matrix(refs)

Arguments

refs	data frame, output of tiff_read_tiles

Value

matrix, data shaped from raw read into 'raster order'

See Also

refs_tiles

Examples

vfile <- system.file("extdata/volcano-cog.tif", package = "rustycogs", mustWork = TRUE)
refs <- refs_tiles(vfile)
mat <- refs |> subset(ifd == 4) |>
 tiff_read_tiles() |> tiles_to_matrix()

---