Skip to content

Latest commit

 

History

History
50 lines (44 loc) · 2.98 KB

STATUS.md

File metadata and controls

50 lines (44 loc) · 2.98 KB
Raw

Discussion

Done

  • 2022-01-21:

  • With files named: a*band aab, chars a*b now works correctly

  • Case folding for Windows only is somewhat implemented in case.go

    • This can be disabled by the -s switch; thus allowing for character classes in between a set of []
    • PRs to improve this are most welcomed

  • chars.OutputTextTable() now returns an error when output can not be displayed:

    • Example: chars /dev/null > /dev/full

  • Added to README.md:

    • Use< redirection when possible instead of |
    • Using Get-Content -AsByteStream with newer versions of powershell
    • Under powershell, consider using curl --output

  • When reading file chunks, the block size has been changed from 1024 to 4096 -- a small performance improvement

  • chars.searchForSpecialChars() has been rewritten to be significantly more performant.

  • 2022-01-19:

  • chars is now a package, with a corresponding cmd/chars/cmd.go for the CLI interface

  • This will now report special chars for STDIN

    • cat filename | chars -

  • Running chars by itself with no cmd-line args will also read from STDIN

  • A BytesRead column was added to the output

    • Depending on how data is piped into chars (and which OS), values can differ vs reading the same filename from the cmd-line. See below

  • If a glob doesn't match any files, no output or error will occur

  • Added -j output to create JSON format

  • I think globbing is fixed for cross-platform.

    • I am still globbing because I want this feature for Windows, but since Linux bash, macOS zsh already expand file names beforehand this should not cause a problem.
    • The program is able to handle files on Linux actually containing a * in their name.
    • Please let me know if you can think of any problems that may arise with this method.

  • It reads files in chunks as to not consume too much memory.

To Do

  • On Windows, filenames are case-sensitive, but expected behavior is case-insensitive
  • Checking for errors when writing output has only been implemented for json output and not for table.Render()
  • I am not very happy with searchForSpecialChars(). It works, but seems kludgy -- how could this be function be simplified?
  • You mentioned "paths are a tricky business". I'm OK if my program does not work for unconventional, non-standard file names.
  • Create chars_test.go once everything has been stabilized.
  • Create API examples in README.md

Reading from STDIN on Windows

  • YMMV when piping to STDIN under Windows
  • cmd and powershell will skip BOM characters; these 2 fields will both report a value of 0
  • cmd and powershell will skip NUL characters; this field report a value of 0
  • cmd will convert LF to CRLF for UTF-16 encoded files
  • powershell will convert LF to CRLF
  • Piping from programs such as curl will return LF characters under cmd, but CRLF under powershell