Skip to contents

Version 0.4

pipette 0.4.0 (2020-01-19)

Major changes:

  • Renamed package from brio to pipette, in preparation of CRAN submission.
  • export: Reworked internal methods to use new compress and decompress functions defined in acidbase package.
  • localOrRemoteFile: Reworked to use new decompress defined in acidbase internally, with improved tempfile handling.
  • Migrated coercion methods and other utitilies from the now archived transformer package: atomize, coerceToList, droplevels, decode, encode, factorize, matchRowNameColumn, and metadata2.

Minor changes:

  • Switched to using cli package for improved messages.

pipette 0.4.1 (2020-01-28)

Minor changes:

  • Switched license from MIT to GPL-3.
  • Ensure that coerce method is reexported – thanks @dpryan79 for catching this issue in basejump.

pipette 0.4.2 (2020-04-10)

Minor changes:

  • droplevels: Ensuring S4 generic variant defined in S4Vectors package gets reexported and masks base S3 generic. This helps avoid a C stack usage issue that has popped up in the latest version of R.
  • import: Fix for importing JSON files without extension. Can now declare using the format argument. This fix was needed to import GitHub JSON URLs inside new installGitHub function defined in bb8 package.

pipette 0.4.3 (2020-04-15)

Minor changes:

  • import: Bug fix for format argument erroring on some supported file types.

pipette 0.4.4 (2020-05-12)

Major changes:

  • import and export functions now default to using vroom engine instead of data.table. Internally, these now call vroom and vroom_write. We noticed that the data.table fwrite function in particular can have issues writing many files on AWS EC2 instances, resulting in a stack imbalance. The vroom package seems to be more stable currently.

Minor changes:

  • loadData, saveData: Switched to using cli_alert instead of cli_text internally for status messages.

pipette 0.4.5 (2020-05-18)

Minor changes:

  • import: Now setting delim internally for vroom import call, to handle single column data frame import. Otherwise vroom will warn about failing to detect expected delimiter.

pipette 0.4.6 (2020-06-11)

Minor changes:

  • import: Added support for skip argument, which allows the user to skip a certain number of lines in the input.
  • import and export of source code lines now uses readr package internally (read_lines and write_lines) instead of base readLines and writeLines.

pipette 0.4.7 (2020-07-07)

Minor changes:

  • import Hardened Excel input to intentionally error on any warnings returned by internal read_excel call, which is too liberal in coercing data types, in my opinion.
  • import: Switched XLS parser from gdata (no longer updated) back to readxl, which is more actively developed.
  • import: Added makeNames argument, to override default internal handling. This allows the user to apply snake case and/or camel case formatting automatically with this argument.

pipette 0.4.8 (2020-07-24)

Minor changes:

  • export: Improved messaages to include full output path.
  • Increased minimum R dependency to 4.0.

pipette 0.4.9 (2020-07-27)

Minor changes:

  • Decreased data.table dependency from 1.13.0 back to 1.12.8, so we can build successfully on bioconda.

pipette 0.4.10 (2020-08-03)

Minor changes:

  • Relax name checks for import.

pipette 0.4.11 (2020-08-11)

Minor changes:

  • export: Ensuring that full directory path is always resolved in message.
  • Miscellaneous message improvements, related to internal toString handling.

pipette 0.4.12 (2020-08-13)

Minor changes:

  • import: Improved messages to always resolve full path to import directory.
  • Bug fix for AppVeyor CI config.

pipette 0.4.13 (2020-08-18)

Minor changes:

  • export: Now dropping non-atomic columns (e.g. Entrez ID list column) from data frames automatically prior to export. Previously, the allAreAtomic assert check was called automatically and would error on non-atomics.

pipette 0.4.14 (2020-09-14)

Minor changes:

  • sanitizeNA: Added support for “N/A” string, which is present in some Excel spreadsheets.

pipette 0.4.15 (2020-10-06)

New functions:

  • cacheURL: Utility function for easy package file caching using BiocFileCache package internally.

Minor changes:

  • Updated Acid Genomics package dependencies.

pipette 0.4.16 (2020-10-06)

New functions:

  • getURLDirList: Migrated this function from previous definition in basejump, so we can inherit inside new AcidGenomes package.

Minor changes:

  • cacheURL: Added package argument, so other packages that use this function will automatically inherit the current package, as expected.

pipette 0.4.17 (2020-10-09)

New functions:

  • Migrated transmit here from basejump.

pipette 0.4.18 (2020-10-09)

Minor changes:

  • Bug fix for breaking change in readr v1.4 release. In the write_* functions, including write_lines and write_csv, the path argument has been renamed to file. Now requiring readr v1.4+ in pipette.

pipette 0.4.19 (2020-10-12)

Minor changes:

  • transmit: Added download argument support, to optionally return matching URLs without downloading. This is useful for handing off to cacheURL function for caching files inside of packages with BiocFileCache.

pipette 0.4.20 (2020-12-09)

Minor changes:

  • Updated dependency versions.
  • export: Added append option for character method. Also relaxed checks on character method, allowing for exporting of empty vectors.

pipette 0.4.21 (2020-12-10)

New functions:

  • Added download, which acts as a hardened wrapper for utils::download.file. Annoying, download.file returns status codes but does not intentionally error on any unsuccessful downloads. Our wrapper ensures that R always errors on any file download issue. It also sets a longer timeout internally, to avoid any potential issues with the timeout option being defined in Rprofile.

pipette 0.4.22 (2020-12-11)

Minor changes:

  • export: Added option to intentionally not export column and/or rownames for matrix, data.frame, and DataFrame classes.

pipette 0.4.23 (2020-12-24)

Minor changes:

  • Made some previous imports conditional suggested packages: jsonlite, readr, rtracklayer, yaml.
  • export: Improved internal file name handling for CLI messages.
  • Reworked rtracklayer as a suggested package instead of an explicit import.
  • import: Improved internal bcbio counts importer code to use default TSV method, rather than relying on data.table fread function manually.
  • Removed dependency on readr. Import of lines now uses vroom::vroom_lines internally, and export character method will conditionally switch to using base writeLines if the readr package is not installed.
  • Removed Matrix readMM and writeMM as imports.
  • Removed data.table fread and fwrite as imports.
  • Removed BiocGenerics dependency, in favor of rexports defined in AcidGenerics. This helps keep the number of dependencies declared in the package more compact and manageable.
  • export: Bug fix for handling of GZ file name extension for character method.
  • Bug fix for vroom_lines error import error of bcbio log: Unnamed col_types must have the same length as col_names.
  • Now including additional reexports from data.table and tibble packages.
  • Bug fixes for import/export of tx2gene file handling for pending AcidGenomes package update.
  • import: Hardened against makeNames usage on objects that don’t support names assignment.

pipette 0.4.24 (2021-01-06)

Minor changes:

  • Switched from using cli package internally to AcidCLI.

pipette 0.4.25 (2021-01-08)

Minor changes:

  • getURLDirList now returns sorted.
  • Reexporting url.exists function from RCurl, for convenience.