jsfmt diff

[brettlangdon]

jsfmt diff ignores space changes git diff -b and when running jsfmt --diff=true --format=true ./rewrite.js I missed a bunch of continuation changes that I didn't actually intend to make:

diff --git a/rewrite.js b/rewrite.js
index 3675a93..9a638b0 100644
--- a/rewrite.js
+++ b/rewrite.js
@@ -84,13 +84,13 @@ function match(wildcards, pattern, node) {
         return false;
       }
       return match(wildcards, pattern.key, node.key)
-          && match(wildcards, pattern.value, node.value);
+      && match(wildcards, pattern.value, node.value);
     case 'MemberExpression':
       if (pattern.computed != node.computed) {
         return false;
       }
       return match(wildcards, pattern.object, node.object)
-          && match(wildcards, pattern.property, node.property);
+      && match(wildcards, pattern.property, node.property);
     case 'ArrayExpression':
       return partial(wildcards, pattern.elements, node.elements);
     case 'ObjectExpression':
@@ -100,12 +100,12 @@ function match(wildcards, pattern, node) {
         return false;
       }
       return match(wildcards, pattern.left, node.left)
-          && match(wildcards, pattern.right, node.right);
+      && match(wildcards, pattern.right, node.right);
     case 'ForStatement':
       return match(wildcards, pattern.init, node.init)
-          && match(wildcards, pattern.test, node.test)
-          && match(wildcards, pattern.update, node.update)
-          && match(wildcards, pattern.body, node.body);
+      && match(wildcards, pattern.test, node.test)
+      && match(wildcards, pattern.update, node.update)
+      && match(wildcards, pattern.body, node.body);
     case 'VariableDeclaration':

is this the desired behavior? if not, I might suggest removing the -b option from jsfmt diff.

[jimfleming]

Nope, not desired, good catch.

[jimfleming]

I think this is the last known issue. Going to resolve it real quick then bump the npm version. I'll post it as a PR so its all out in the open now that jsfmt is public.

[brettlangdon]

One last thing I wanted to ask, should it be --format=true by default to match the behavior of running gofmt without any parameters?

[jimfleming]

Yah, I can get behind that. One reason I went with no default action is discoverability of other options but I think familiarity with gofmt is part of the draw towards jsfmt.

[brettlangdon]

Adding --help might... "Help"
On May 12, 2014 1:01 PM, "Jim Fleming" [email protected] wrote:

Yah, I can get behind that. One reason I went with no default action is
discoverability of other options but I think familiarity with gofmt is part
of the draw towards jsfmt.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/13#issuecomment-42858730
.

[jimfleming]

Haha, well, --help works out of the box because any unrecognized flag prints the help.

[brettlangdon]

haha, nice.

On the topic of help text/etc, I am a complete http://docopt.org/ fanboy... so I have to at least mention it once ;)

[jimfleming]

Interesting, so you just describe the help doc and it figures out the params for you? I like that.

This issue is of concern but assuming it doesn't interfere with jsfmt's functionality I don't mind switching.

[brettlangdon]

ooo, that seems like a fun issue.

but yeah, just give it the help text and it does the rest. I'll see if I can do a port later today and do a PR so you can see what it would look like.

[jimfleming]

#14