Re^2: read files one by one in directory and redirct output to a file

Replies are listed 'Best First'.
Re^3: read files one by one in directory and redirct output to a file by Marshall (Canon) on Jul 08, 2018 at 00:24 UTC
On Windows, I believe the equivalent to your Unix command would be: `type *.log > result` [download] On Unix, I think copy is cp. On either O/S it is possible to concatenate a bunch of binary files together into a single result file. copy (cp) works with binary files. type or cat is designed to work with text files.	[reply] [d/l]
Re^4: read files one by one in directory and redirct output to a file by soonix (Chancellor) on Jul 08, 2018 at 07:34 UTC
cat is the abbreviation for 'concatenate' and will do binary files just fine. cp will complain if it's instructed to copy multiple source files and the target is not a directory. Windows 'copy' is different from both.	[reply]
Re^5: read files one by one in directory and redirct output to a file by Marshall (Canon) on Jul 08, 2018 at 08:17 UTC
Well I guess.. `cp [option]... [-T] source destination` [download] Where -T means: Treat destination as a normal file. I don't work much with Unix nowadays.	[reply] [d/l]
Re^6: read files one by one in directory and redirct output to a file by soonix (Chancellor) on Jul 08, 2018 at 13:18 UTC
Re^4: read files one by one in directory and redirct output to a file by haukex (Archbishop) on Jul 08, 2018 at 09:52 UTC
copy (cp) works with binary files. type or cat is designed to work with text files. Note that on NIX/POSIX, there is generally no distinction between "binary" and "text" files. `cp` doesn't concatenate multiple files (`-T`, from your reply below, makes no difference here) - `cat` is "the" NIX tool for concatenating files. If I guess that by "binary" vs. "text" you maybe mean "block-by-block" instead of "line-by-line", then that is course a valid point in regards to performance. But at least GNU Coreutils' `cat` is optimized to read and write files block-by-block, not line-by-line, when it doesn't need to do any line-by-line processing - see its `simple_cat` function.	[reply] [d/l] [select]
Re^5: read files one by one in directory and redirct output to a file by Marshall (Canon) on Jul 08, 2018 at 12:22 UTC
If I guess that by "binary" vs. "text" you maybe mean "block-by-block" instead of "line-by-line", then that is course a valid point in regards to performance. that is what I meant. I defer to your Unix knowledge.	[reply]
Re^4: read files one by one in directory and redirct output to a file by afoken (Chancellor) on Jul 08, 2018 at 18:48 UTC
On Windows, I believe the equivalent to your Unix command would be: `type *.log > result` [download] No, it is not. `type` treats Ctrl-Z (ASCII 26) as end-of-file marker. cat writes its entire input to STDOUT. `cat` can handle binary files, `type` can't. X:\>perl -E "say qq[Hello\cZWorld!]" > foo X:\>type foo Hello X:\>perl -pe 1 < foo Hello→World! X:\> Alexander -- Today I will gladly share my knowledge and experience, for there are no sweeter words than "I told you so". ;-)	[reply] [d/l] [select]