Commite00fffc

committed

Shorten comments on _decode steps

This removes the parenthesized examples from the per-step commentsin the WinBashStatus._decode helper.

1 parent3303c74 commite00fffcCopy full SHA for e00fffc

File tree

-8

lines changed

-8

lines changed

Lines changed: 1 addition & 8 deletions

Original file line number	Diff line number	Diff line change
`@@ -139,10 +139,6 @@ def _decode(stdout):`
`139`	`139`	`# uses the ASCII subset, so we can safely guess a wrong code page for it. Errors`
`140`	`140`	`# from such an environment can contain any text, but unlike WSL's own messages,`
`141`	`141`	`# they go to stderr, not stdout. So we can try the system ANSI code page first.`
`142`		`-# (Console programs often use the OEM code page, but the ACP seems more accurate`
`143`		`-# here. For example, on en-US Windows with the original system code page but the`
`144`		`-# display language set to fr-FR, the message, if not UTF-16LE, is windows-1252,`
`145`		`-# same as the ACP, while the OEMCP is 437, which can't decode its accents.)`
`146`	`142`	`acp=_get_windows_ansi_encoding()`
`147`	`143`	`try:`
`148`	`144`	`returnstdout.decode(acp)`
`@@ -151,10 +147,7 @@ def _decode(stdout):`
`151`	`147`	`exceptLookupErroraserror:`
`152`	`148`	`log.warning("%s",str(error))# Message already says "Unknown encoding:".`
`153`	`149`
`154`		`-# Assume UTF-8. If invalid, substitute Unicode replacement characters. (For`
`155`		`-# example, on zh-CN Windows set to display fr-FR, errors from WSL itself, if not`
`156`		`-# UTF-16LE, are in windows-1252, even though the ANSI and OEM code pages both`
`157`		`-# default to 936, and decoding as code page 936 or as UTF-8 both have errors.)`
	`150`	`+# Assume UTF-8. If invalid, substitute Unicode replacement characters.`
`158`	`151`	`returnstdout.decode("utf-8",errors="replace")`
`159`	`152`
`160`	`153`

Comments

(0)