Using split-diff 1.6.0
Private, in house, language grammar (sets codepage to ISO-8859-1, while Atom is set to use UTF-8)
Our language uses the '¬' symbol to designate NOT. When executing split-diff on a file that contains this symbol, the last committed version in git is diffed, but the '¬' symbol becomes '�'. So every line with the NOT symbol shows as a changed line in the compare.
The same corruption of the NOT symbol can be seen if I just change the code page on the original file from 8859-1 to utf-8.
So, I do not know how split-diff determines which code page to use when opening a file from git to compare to, but would it be possible to honor the codepage dictated in the language grammar?
Our language grammar defines this in the in the settings folder, in the language.cson file:
'editor':
'atomicSoftTabs': false
'tabType': "soft"
'tabLength': 2
'preferredLineLength': 72
'softWrap': false
'showInvisibles': false
'core':
'fileEncoding': "iso88591"
'whitespace':
'removeTrailingWhitespace': false
Using split-diff 1.6.0 Private, in house, language grammar (sets codepage to ISO-8859-1, while Atom is set to use UTF-8) Our language uses the '¬' symbol to designate NOT. When executing split-diff on a file that contains this symbol, the last committed version in git is diffed, but the '¬' symbol becomes '�'. So every line with the NOT symbol shows as a changed line in the compare.
The same corruption of the NOT symbol can be seen if I just change the code page on the original file from 8859-1 to utf-8.
So, I do not know how split-diff determines which code page to use when opening a file from git to compare to, but would it be possible to honor the codepage dictated in the language grammar?
Our language grammar defines this in the in the settings folder, in the language.cson file: 'editor': 'atomicSoftTabs': false 'tabType': "soft" 'tabLength': 2 'preferredLineLength': 72 'softWrap': false 'showInvisibles': false 'core': 'fileEncoding': "iso88591" 'whitespace': 'removeTrailingWhitespace': false