Mercurial support

[frantic]

This will help with Mercurial compatibility – it doesn't output index line, and diff-so-fancy depends on it to draw the first horizontal line.

All bats tests pass.

Fixes #186

[frantic]

Related to #38. Maybe I should add a new test for the Mercurial usecase?

[scottchiefbaker]

This is a pretty crucial part of the git diff logic. It captures the color of the header that we use later. Simply skipping this WILL cause output coloring issues.

[frantic]

But all the automated tests pass, and when trying it out locally with git the coloring was fine. I guess what we could do is we could capture the color here but print the horizontal line when getting to diff --git part. This will require additional checks because mercurial does not print index at all.

[scottchiefbaker]

Are you still willing to work on this? I'd like to discuss options moving forward.

[frantic]

@scottchiefbaker – I've been using this locally for a while now and I'm pretty happy with it. Would be awesome to get this upstreamed. Are you willing to merge this in? If so I'll spend some time rebasing it

[scottchiefbaker]

Ya I'd be willing to merge this pending some tweaks and testing on my side.

I don't use mercurial, so I'll have to rely on your for that. As long as these changes don't break the git pieces I'm ok with that.

[scottchiefbaker]

@frantic where are we at on this? Just going through old issues trying to clean up.

[frantic]

Hey @scottchiefbaker! I've been using this without modifications for 1.5 years now. You mentioned in your last comment that you'll do some testing on your side?

If you are still onboard I'll try rebasing

[scottchiefbaker]

Yeah go ahead and do that and I'll do some testing on my side!

[frantic]

A quick update. Basically git uses this:

diff --git a/diff-so-fancy b/diff-so-fancy
index e5b1ef0..925bf26 100755
--- a/diff-so-fancy
+++ b/diff-so-fancy
@@ -116,7 +116,7 @@ sub do_dsf_stuff {

Mercurial uses this:

diff -r 7bedd0c41e23 -r 120d1b486a5f diff-so-fancy
--- a/diff-so-fancy	Wed Dec 27 16:20:27 2017 -0800
+++ a/diff-so-fancy	Wed Dec 27 18:09:27 2017 -0800
@@ -116,7 +116,7 @@

I guess I could refactor the code a little to not duplicate the logic, but don't have time at the moment.

[scottchiefbaker]

I can look at tweaking the code... can you get me a simple diff out from Mercurial? Something like hg diff HEAD~1 or whatever the equivalent is. I need the complete output to make sure we're compatible.

If it's like you list above it should be super minor changes. I'm pretty confident we can make it work.

[scottchiefbaker]

I found a sample hg repo and used it as a base... I think I have a rough implementation here. Can you test that and let me know?

[frantic]

@scottchiefbaker your version works great with my hg show! However I found my hg diff has the following header

diff --git a/test.js b/test.js
--- a/test.js
+++ b/test.js
@@ -17,3 +17,4 @@

i.e. no "index" and no "-r" either

[scottchiefbaker]

That looks like pretty standard git output, minus the index line. It should already work with d-s-f. Are you saying it doesn't?

[frantic]

Because the index line and "-r" are missing, there's no initial line (first output is hg show, second output is hg diff):

in raw format:

$ hg show
diff -r 000000000000 -r 7d10fc5fa483 test.js
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test.js	Fri Dec 29 12:21:40 2017 -0800
@@ -0,0 +1,1 @@
+console.log({a: 1})

$ hg diff
diff --git a/test.js b/test.js
--- a/test.js
+++ b/test.js
@@ -1,1 +1,2 @@
 console.log({a: 1})
+console.log({b: 1})

[scottchiefbaker]

Try a git pull I think I just fixed this.

[frantic]

Still seeing no line before the file name…

[scottchiefbaker]

Even with 0e73558a35a8d45e7014925242ea398426694128? That's weird... can you post your full output

hg diff --color always > /tmp/out.txt

I need the ANSI codes to be included.

[frantic]

hgdiff.txt

[scottchiefbaker]

Try pulling again... I just landed 4fcb82013f988798454395bbf9ff1ea6a4650606

[frantic]

Yay!

[scottchiefbaker]

Ok great!

Play around with it for a couple of days and make sure it works for you. I'll land it on next next week if we're happy with it. I haven't noticed any side effects on the Git parsing, which is our primary focus.

[frantic]

Awesome, thank you!

[scottchiefbaker]

@frantic have you had a chance to test things on your side? I know it's been a long holiday weekend, so you may not have had time yet.

I'd like to merge this if you think it's good enough. I don't use mercurial, so I don't have an easy way to verify this.

[frantic]

@scottchiefbaker sorry for the delay! I tested a few commands and the results look amazing! I'll give it another try next week when I resume heavy work in Mercurial repo, this should help me test edge cases more.

[scottchiefbaker]

I've merged all my changes in to next. Barring anything crazy this will land in the next official release. That said we don't need this pull request any more so I'm closing it.