String>>#expandMacros new line macro rationale
Hi, yesterday I was playing with this method and found something awkward. <n> macro is expanded to CR. I was expecting that expands to the line ending of the platform, or at least to LF. Anybody knows if this have some explanation? I'm tempted to change the <n> expansion to use OSPlatform current lineEnding , and add <r> as a valid macro expanding to CR. I think it's more consistent with the general use. What do you think? If there's consensus I would create an issue and provide a Slice with the changes and some test cases. Gabriel
Hi gabriel
<n> macro is expanded to CR. I was expecting that expands to the line ending of the platform, or at least to LF.
Anybody knows if this have some explanation?
I'm tempted to change the <n> expansion to use OSPlatform current lineEnding , and add <r> as a valid macro expanding to CR. I think it's more consistent with the general use.
Yes
What do you think? If there's consensus I would create an issue and provide a Slice with the changes and some test cases. Yes I would love that someone or a group of Pharo lover fix the issues with the encoding. I remember a discussion a while ago.
Gabriel
Slice in Inbox https://pharo.fogbugz.com/f/cases/16863/Make-expandMacros-more-consistent-wi... On Sat, Oct 24, 2015 at 4:19 AM, stepharo <stepharo@free.fr> wrote:
Hi gabriel
<n> macro is expanded to CR. I was expecting that expands to the line ending of the platform, or at least to LF.
Anybody knows if this have some explanation?
I'm tempted to change the <n> expansion to use OSPlatform current lineEnding , and add <r> as a valid macro expanding to CR. I think it's more consistent with the general use.
Yes
What do you think? If there's consensus I would create an issue and provide a Slice with the changes and some test cases.
Yes I would love that someone or a group of Pharo lover fix the issues with the encoding. I remember a discussion a while ago.
Gabriel
As there is a macro for OSPlatform current lineEnding (<n>) and one for CR (<r>) shouldnât there also be one for LF? E.g. in the Java String.format method %n is used for OSPlatform current lineEnding, \n for LF and \r for CR to cover all three cases. Udo On 24.10.2015, at 22:22, Gabriel Cotelli <g.cotelli@gmail.com> wrote:
Slice in Inbox https://pharo.fogbugz.com/f/cases/16863/Make-expandMacros-more-consistent-wi...
On Sat, Oct 24, 2015 at 4:19 AM, stepharo <stepharo@free.fr> wrote: Hi gabriel
<n> macro is expanded to CR. I was expecting that expands to the line ending of the platform, or at least to LF.
Anybody knows if this have some explanation?
I'm tempted to change the <n> expansion to use OSPlatform current lineEnding , and add <r> as a valid macro expanding to CR. I think it's more consistent with the general use.
Yes What do you think? If there's consensus I would create an issue and provide a Slice with the changes and some test cases. Yes I would love that someone or a group of Pharo lover fix the issues with the encoding. I remember a discussion a while ago.
Gabriel
participants (3)
-
Gabriel Cotelli -
stepharo -
Udo Borkowski