A proxy to `path`, replacing `\` with `/` for all results (supports UNC paths) & new methods to normalize & join keeping leading `./` and add, change, default, trim file extensions.
npm install upath
A drop-in replacement / proxy to nodejs's path that:
* Replaces the windows \ with the unix / in all string params & results. This has significant positives - see below.
* Adds filename extensions functions addExt, trimExt, removeExt, changeExt, and defaultExt.
* Add a normalizeSafe function to preserve any meaningful leading ./ & a normalizeTrim which additionally trims any useless ending /.
* Plus a helper toUnix that simply converts \ to / and consolidates duplicates.
Useful note: these docs are actually auto generated from specs, running on Linux.
Notes:
* upath.sep is set to '/' for seamless replacement (as of 1.0.3).
* upath has no runtime dependencies, except built-in path (as of 1.0.4)
* travis-ci tested in node versions 8 to 14 (on linux)
* Also tested on Windows / node@12.18.0 (without CI)
History brief:
* 1.x : Initial release and various features / fixes
* 2.0.0 : Adding UNC paths support - see https://github.com/anodynos/upath/pull/38
Normal path doesn't convert paths to a unified format (ie /) before calculating paths (normalize, join), which can lead to numerous problems.
Also path joining, normalization etc on the two formats is not consistent, depending on where it runs. Running path on Windows yields different results than when it runs on Linux / Mac.
In general, if you code your paths logic while developing on Unix/Mac and it runs on Windows, you may run into problems when using path.
Note that using Unix / on Windows works perfectly inside nodejs (and other languages), so there's no reason to stick to the Windows legacy at all.
##### Examples / specs
Check out the different (improved) behavior to vanilla path:
upath.normalize(path) --returns-->
✓ 'c:/windows/nodejs/path' ---> 'c:/windows/nodejs/path' // equal to path.normalize()
✓ 'c:/windows/../nodejs/path' ---> 'c:/nodejs/path' // equal to path.normalize()
✓ 'c:\\windows\\nodejs\\path' ---> 'c:/windows/nodejs/path' // path.normalize() gives 'c:\windows\nodejs\path'
✓ 'c:\\windows\\..\\nodejs\\path' ---> 'c:/nodejs/path' // path.normalize() gives 'c:\windows\..\nodejs\path'
✓ '/windows\\unix/mixed' ---> '/windows/unix/mixed' // path.normalize() gives '/windows\unix/mixed'
✓ '\\windows//unix/mixed' ---> '/windows/unix/mixed' // path.normalize() gives '\windows/unix/mixed'
✓ '\\windows\\..\\unix/mixed/' ---> '/unix/mixed/' // path.normalize() gives '\windows\..\unix/mixed/'
Joining paths can also be a problem:
upath.join(paths...) --returns-->
✓ 'some/nodejs/deep', '../path' ---> 'some/nodejs/path' // equal to path.join()
✓ 'some/nodejs\\windows', '../path' ---> 'some/nodejs/path' // path.join() gives 'some/path'
✓ 'some\\windows\\only', '..\\path' ---> 'some/windows/path' // path.join() gives 'some\windows\only/..\path'
Parsing with path.parse() should also be consistent across OSes:
upath.parse(path) --returns-->
✓ 'c:\Windows\Directory\somefile.ext' ---> { root: '', dir: 'c:/Windows/Directory', base: 'somefile.ext', ext: '.ext', name: 'somefile'
}
// path.parse() gives '{ root: '', dir: '', base: 'c:\\Windows\\Directory\\somefile.ext', ext: '.ext', name: 'c:\\Windows\\Directory\\somefile'
}'
✓ '/root/of/unix/somefile.ext' ---> { root: '/', dir: '/root/of/unix', base: 'somefile.ext', ext: '.ext', name: 'somefile' // equal to
}path.parse()
#### upath.toUnix(path)
Just converts all ` to / and consolidates duplicates, without performing any normalization.
##### Examples / specs
upath.toUnix(path) --returns-->
✓ './/windows\//unix//mixed////' ---> './windows/unix/mixed/'
✓ ---> '../windows/../unix/mixed'
#### upath.normalizeSafe(path)
Exactly like path.normalize(path), but it keeps the first meaningful ./ or //.
Note that the unix / is returned everywhere, so windows \ is always converted to unix /.
##### Examples / specs & how it differs from vanilla path
upath.normalizeSafe(path) --returns-->
✓ '' ---> '.' // equal to path.normalize()
✓ ---> '.' // equal to path.normalize()
✓ ---> './' // equal to path.normalize()
✓ ---> './' // equal to path.normalize()
✓ ---> './' // path.normalize() gives '.\'
✓ ---> './' // path.normalize() gives '.\/'
✓ ---> '..' // equal to path.normalize()
✓ ---> '..' // equal to path.normalize()
✓ ---> '../' // equal to path.normalize()
✓ ---> '../' // path.normalize() gives '.\..\'
✓ ---> '../dep' // equal to path.normalize()
✓ ---> '../dep' // equal to path.normalize()
✓ ---> '../path/dep' // equal to path.normalize()
✓ ---> '../dep' // equal to path.normalize()
✓ ---> 'dep' // equal to path.normalize()
✓ ---> 'path/dep' // equal to path.normalize()
✓ ---> './dep' // path.normalize() gives 'dep'
✓ ---> './path/dep' // path.normalize() gives 'path/dep'
✓ ---> './dep' // path.normalize() gives 'dep'
✓ ---> './windows/unix/mixed/' // path.normalize() gives 'windows\unix/mixed/'
✓ ---> '../windows/unix/mixed' // path.normalize() gives '../windows\unix/mixed'
✓ ---> 'windows/unix/mixed/' // path.normalize() gives 'windows\unix/mixed/'
✓ ---> '../unix/mixed' // path.normalize() gives '../windows\..\unix/mixed'
✓ ---> '//server/share/file' // path.normalize() gives '\\server\share\file'
✓ ---> '//server/share/file' // path.normalize() gives '/server/share/file'
✓ ---> '//?/UNC/server/share/file' // path.normalize() gives '\\?\UNC\server\share\file'
✓ ---> '//LOCALHOST/c$/temp/file' // path.normalize() gives '\\LOCALHOST\c$\temp\file'
✓ ---> '//?/c:/temp/file' // path.normalize() gives '\\?\c:\temp\file'
✓ ---> '//./c:/temp/file' // path.normalize() gives '\\.\c:\temp\file'
✓ ---> '//./c:/temp/file' // path.normalize() gives '/c:/temp/file'
✓ ---> '//./c:/temp/file' // path.normalize() gives '/\.\c:/temp\/file'
#### upath.normalizeTrim(path)
Exactly like path.normalizeSafe(path), but it trims any useless ending /.
##### Examples / specs
upath.normalizeTrim(path) --returns-->
✓ './' ---> '.' // upath.normalizeSafe() gives './'
✓ ---> '..' // upath.normalizeSafe() gives '../'
✓ ---> '../dep' // upath.normalizeSafe() gives '../dep/'
✓ ---> 'path/dep' // upath.normalizeSafe() gives 'path/dep/'
✓ ---> './windows/unix/mixed' // upath.normalizeSafe() gives './windows/unix/mixed/'
#### upath.joinSafe([path1][, path2][, ...])
Exactly like path.join(), but it keeps the first meaningful ./ or //.
Note that the unix / is returned everywhere, so windows \ is always converted to unix /.
##### Examples / specs & how it differs from vanilla path
upath.joinSafe(path) --returns-->
✓ 'some/nodejs/deep', '../path' ---> 'some/nodejs/path' // equal to path.join()
✓ ---> './some/local/path' // path.join() gives 'some/local/path'
✓ ---> './some/current/path' // path.join() gives 'some\current\mixed/..\path'
✓ ---> '../some/relative/path' // path.join() gives '../some/relative/destination/..\path'
✓ ---> '//server/share/path' // path.join() gives '\\server\share\file/..\path'
✓ ---> '//./c:/temp/path' // path.join() gives '\\.\c:\temp\file/..\path'
✓ ---> '//server/share/path' // path.join() gives '/server/share/path'
✓ ---> '//./c:/temp/path' // path.join() gives '/c:/temp/path'
Happy notes:
In all functions you can:
* use both .ext & ext - the dot . on the extension is always adjusted correctly.
* omit the ext param (pass null/undefined/empty string) and the common sense thing will happen.
* ignore specific extensions from being considered as valid ones (eg .min, .dev .aLongExtIsNotAnExt etc), hence no trimming or replacement takes place on them.
#### upath.addExt(filename, [ext])
Adds .ext to filename, but only if it doesn't already have the exact extension.
##### Examples / specs
upath.addExt(filename, 'js') --returns-->
✓ 'myfile/addExt' ---> 'myfile/addExt.js'
✓ ---> 'myfile/addExt.txt.js'
✓ ---> 'myfile/addExt.js'
✓ ---> 'myfile/addExt.min..js'
It adds nothing if no ext param is passed.
upath.addExt(filename) --returns-->
✓ 'myfile/addExt' ---> 'myfile/addExt'
✓ ---> 'myfile/addExt.txt'
✓ ---> 'myfile/addExt.js'
✓ ---> 'myfile/addExt.min.'
#### upath.trimExt(filename, [ignoreExts], [maxSize=7])
Trims a filename's extension.
* Extensions are considered to be up to maxSize chars long, counting the dot (defaults to 7).
* An Array of ignoreExts (eg ['.min']) prevents these from being considered as extension, thus are not trimmed.
##### Examples / specs
upath.trimExt(filename) --returns-->
✓ 'my/trimedExt.txt' ---> 'my/trimedExt'
✓ ---> 'my/trimedExt'
✓ ---> 'my/trimedExt'
✓ ---> 'my/trimedExt.min'
✓ ---> '../my/trimedExt.longExt'
It is ignoring .min & .dev as extensions, and considers exts with up to 8 chars.
upath.trimExt(filename, ['min', '.dev'], 8) --returns-->
✓ 'my/trimedExt.txt' ---> 'my/trimedExt'
✓ ---> 'my/trimedExt.min'
✓ ---> 'my/trimedExt.dev'
✓ ---> '../my/trimedExt'
✓ ---> '../my/trimedExt.longRExt'
#### upath.removeExt(filename, ext)
Removes the specific ext extension from filename, if it has it. Otherwise it leaves it as is..ext
As in all upath functions, it be or ext.
##### Examples / specs
upath.removeExt(filename, '.js') --returns-->
✓ 'removedExt.js' ---> 'removedExt'
✓ ---> 'removedExt.txt'
✓ ---> 'notRemoved.txt'
It does not care about the length of exts.
upath.removeExt(filename, '.longExt') --returns-->
✓ 'removedExt.longExt' ---> 'removedExt'
✓ ---> 'removedExt.txt'
✓ ---> 'notRemoved.txt'
#### upath.changeExt(filename, [ext], [ignoreExts], [maxSize=7])
Changes a filename's extension to ext. If it has no (valid) extension, it adds it.
* Valid extensions are considered to be up to maxSize chars long, counting the dot (defaults to 7).
* An Array of ignoreExts (eg ['.min']) prevents these from being considered as extension, thus are not changed - the new extension is added instead.
##### Examples / specs
upath.changeExt(filename, '.js') --returns-->
✓ 'my/module.min' ---> 'my/module.js'
✓ ---> 'my/module.js'
✓ ---> 'my/module.js'
✓ ---> 'file/withDot.js'
✓ ---> 'file/change.longExt.js'
If no ext param is given, it trims the current extension (if any).
upath.changeExt(filename) --returns-->
✓ 'my/module.min' ---> 'my/module'
✓ ---> 'my/module'
✓ ---> 'my/module'
✓ ---> 'file/withDot'
✓ ---> 'file/change.longExt'
It is ignoring .min & .dev as extensions, and considers exts with up to 8 chars.
upath.changeExt(filename, 'js', ['min', '.dev'], 8) --returns-->
✓ 'my/module.coffee' ---> 'my/module.js'
✓ ---> 'file/notValidExt.min.js'
✓ ---> 'file/notValidExt.dev.js'
✓ ---> 'file/change.js'
✓ ---> 'file/change.longRExt.js'
#### upath.defaultExt(filename, [ext], [ignoreExts], [maxSize=7])
Adds .ext to filename, only if it doesn't already have _any_ old extension.
* (Old) extensions are considered to be up to maxSize chars long, counting the dot (defaults to 7).
* An Array of ignoreExts (eg ['.min']) will force adding default .ext even if one of these is present.
##### Examples / specs
upath.defaultExt(filename, 'js') --returns-->
✓ 'fileWith/defaultExt' ---> 'fileWith/defaultExt.js'
✓ ---> 'fileWith/defaultExt.js'
✓ ---> 'fileWith/defaultExt.min'
✓ ---> 'fileWith/defaultExt.longExt.js'
If no ext param is passed, it leaves filename intact.
upath.defaultExt(filename) --returns-->
✓ 'fileWith/defaultExt' ---> 'fileWith/defaultExt'
✓ ---> 'fileWith/defaultExt.js'
✓ ---> 'fileWith/defaultExt.min'
✓ ---> 'fileWith/defaultExt.longExt'
It is ignoring .min & .dev as extensions, and considers exts with up to 8 chars.
upath.defaultExt(filename, 'js', ['min', '.dev'], 8) --returns-->
✓ 'fileWith/defaultExt' ---> 'fileWith/defaultExt.js'
✓ ---> 'fileWith/defaultExt.min.js'
✓ ---> 'fileWith/defaultExt.dev.js'
✓ ---> 'fileWith/defaultExt.longExt'
✓ ---> 'fileWith/defaultExt.longRext.js'`
Copyright(c) 2014-2020 Angelos Pikoulas (agelos.pikoulas@gmail.com)
Permission is hereby granted, free of charge, to any person
obtaining a copy of this software and associated documentation
files (the "Software"), to deal in the Software without
restriction, including without limitation the rights to use,
copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the
Software is furnished to do so, subject to the following
conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
OTHER DEALINGS IN THE SOFTWARE.