patch: handle 0 context lines (diff.unified=0) when parsing patches

Previously, if there were no context lines, we would just keep updating the
ranges and the hunk, but not actually storing the hunk (just overwriting it each
time).  Thus a diff like this:

    $ hg diff --config diff.unified=0
    diff --git a/bar b/bar
    --- a/bar
    +++ b/bar
    @@ -1,0 +2,1 @@ 1
    +change1
    @@ -3,0 +5,1 @@ 3
    +change2

would come out of the parser like this (change1 is lost):

    bar:
    @@ -3,0 +5,1 @@ 3
    +change2

This had some really weird side effects for things like commit --interactive,
split, etc.

Differential Revision: https://phab.mercurial-scm.org/D5743

mercurial/patch.py

@@ -1609,6 +1609,7 b' def parsepatch(originalchunks, maxcontex'
             self.headers = []
 
         def addrange(self, limits):
+            self.addcontext([])
             fromstart, fromend, tostart, toend, proc = limits
             self.fromline = int(fromstart)
             self.toline = int(tostart)
@@ -1629,6 +1630,8 b' def parsepatch(originalchunks, maxcontex'
             if self.context:
                 self.before = self.context
                 self.context = []
+            if self.hunk:
+                self.addcontext([])
             self.hunk = hunk
 
         def newfile(self, hdr):

tests/test-commit-interactive.t

@@ -1807,3 +1807,38 b' even if none of mode, size and timestamp'
   n   0         -1 unset               subdir/f1
   $ hg status -A subdir/f1
   M subdir/f1
+
+Test diff.unified=0
+
+  $ hg init $TESTTMP/b
+  $ cd $TESTTMP/b
+  $ cat > foo <<EOF
+  > 1
+  > 2
+  > 3
+  > 4
+  > 5
+  > EOF
+  $ hg ci -qAm initial
+  $ cat > foo <<EOF
+  > 1
+  > change1
+  > 2
+  > 3
+  > change2
+  > 4
+  > 5
+  > EOF
+  $ printf 'y\ny\ny\n' | hg ci -im initial --config diff.unified=0
+  diff --git a/foo b/foo
+  2 hunks, 2 lines changed
+  examine changes to 'foo'? [Ynesfdaq?] y
+  
+  @@ -1,0 +2,1 @@ 1
+  +change1
+  record change 1/2 to 'foo'? [Ynesfdaq?] y
+  
+  @@ -3,0 +5,1 @@ 3
+  +change2
+  record change 2/2 to 'foo'? [Ynesfdaq?] y
+  

tests/test-split.t

@@ -103,6 +103,12 b' Split a head'
   abort: cannot split multiple revisions
   [255]
 
+This function splits a bit strangely primarily to avoid changing the behavior of
+the test after a bug was fixed with how split/commit --interactive handled
+`diff.unified=0`: when there were no context lines, it kept only the last diff
+hunk. When running split, this meant that runsplit was always recording three commits,
+one for each diff hunk, in reverse order (the base commit was the last diff hunk
+in the file).
   $ runsplit() {
   > cat > $TESTTMP/messages <<EOF
   > split 1
@@ -113,8 +119,11 b' Split a head'
   > EOF
   > cat <<EOF | hg split "$@"
   > y
+  > n
+  > n
   > y
   > y
+  > n
   > y
   > y
   > y
@@ -123,13 +132,23 b' Split a head'
 
   $ HGEDITOR=false runsplit
   diff --git a/a b/a
-  1 hunks, 1 lines changed
+  3 hunks, 3 lines changed
   examine changes to 'a'? [Ynesfdaq?] y
   
+  @@ -1,1 +1,1 @@
+  -1
+  +11
+  record change 1/3 to 'a'? [Ynesfdaq?] n
+  
+  @@ -3,1 +3,1 @@ 2
+  -3
+  +33
+  record change 2/3 to 'a'? [Ynesfdaq?] n
+  
   @@ -5,1 +5,1 @@ 4
   -5
   +55
-  record this change to 'a'? [Ynesfdaq?] y
+  record change 3/3 to 'a'? [Ynesfdaq?] y
   
   transaction abort!
   rollback completed
@@ -140,13 +159,23 b' Split a head'
   $ HGEDITOR="\"$PYTHON\" $TESTTMP/editor.py"
   $ runsplit
   diff --git a/a b/a
-  1 hunks, 1 lines changed
+  3 hunks, 3 lines changed
   examine changes to 'a'? [Ynesfdaq?] y
   
+  @@ -1,1 +1,1 @@
+  -1
+  +11
+  record change 1/3 to 'a'? [Ynesfdaq?] n
+  
+  @@ -3,1 +3,1 @@ 2
+  -3
+  +33
+  record change 2/3 to 'a'? [Ynesfdaq?] n
+  
   @@ -5,1 +5,1 @@ 4
   -5
   +55
-  record this change to 'a'? [Ynesfdaq?] y
+  record change 3/3 to 'a'? [Ynesfdaq?] y
   
   EDITOR: HG: Splitting 1df0d5c5a3ab. Write commit message for the first split changeset.
   EDITOR: a2
@@ -160,13 +189,18 b' Split a head'
   EDITOR: HG: changed a
   created new head
   diff --git a/a b/a
-  1 hunks, 1 lines changed
+  2 hunks, 2 lines changed
   examine changes to 'a'? [Ynesfdaq?] y
   
+  @@ -1,1 +1,1 @@
+  -1
+  +11
+  record change 1/2 to 'a'? [Ynesfdaq?] n
+  
   @@ -3,1 +3,1 @@ 2
   -3
   +33
-  record this change to 'a'? [Ynesfdaq?] y
+  record change 2/2 to 'a'? [Ynesfdaq?] y
   
   EDITOR: HG: Splitting 1df0d5c5a3ab. So far it has been split into:
   EDITOR: HG: - e704349bd21b: split 1