[READY] Calculate the start codepoint when we don't resolve items in LSC

[puremourning]

Resolving lots of completion items is very slow, so we only do that if
there are 100 or feweer completions.

Most unfortunately Visual Studio Code does some heuristics on completion
items, so that any overlap with existing code is replaced, rather than
appended. This happens to be how jdt.ls returns import completions.

It's common for there to be more than 100 suggestions when typing,
say 'import com.'. This leads to bad behaviour where the insertion ends
up being 'com.com.youcompletme...' which is clearly wrong.

To combad this, when we don't have a textEdit to apply, we detect any
overlap between the insertion_text and the existing text (up to the
completion location) and offset the start_codepoint by that much.

The overlap detection algorithm is efficient (see the link) and simpler
than a dynamic programming one.

Incidentally, we add the 'prefix' key to request_wrap to avoid
recalculating it for each element.

---

This change is 

[codecov-io]

Codecov Report

Merging #905 into master will increase coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #905      +/-   ##
==========================================
+ Coverage   96.18%   96.19%   +0.01%     
==========================================
  Files          83       83              
  Lines        6468     6496      +28     
==========================================
+ Hits         6221     6249      +28     
  Misses        247      247

[micbou]

Reviewed 5 of 5 files at r1, 2 of 2 files at r2.
Review status: all files reviewed at latest revision, 6 unresolved discussions.

---

ycmd/completers/language_server/language_server_completer.py, line 786 at r1 (raw file):

    #
    # ycmd protocol only supports a single start column, so we must post-process
    # the completion items to work out a single start colum to use, as follows:

column.

---

ycmd/completers/language_server/language_server_completer.py, line 1483 at r1 (raw file):

      raise IncompatibleCompletionException( insertion_text )
  else:
    # Calculate the start codepoint based on the overlappoing text in the

overlapping

---

ycmd/completers/language_server/language_server_completer.py, line 1531 at r1 (raw file):

  #      return 4 (from previous iteration)

  # More complex examle: 'Some CoCo' vs 'CoCo Bean'

example.

---

ycmd/completers/language_server/language_server_completer.py, line 1532 at r1 (raw file):

  # More complex examle: 'Some CoCo' vs 'CoCo Bean'
  #   No trunation

truncation.

---

ycmd/completers/language_server/language_server_completer.py, line 1540 at r1 (raw file):

  #   Iter 1 (overlap = 1): p('b') = 3, overlap = 4, aaab!=caab, return 0

One blank line is enough.

---

ycmd/tests/request_wrap_test.py, line 63 at r1 (raw file):

         RequestWrap( PrepareJson( line_num = 1,
                                   contents = line,
                                   column_num = col ) ) [ 'prefix' ] )

Space before [ can be removed.

---

Comments from Reviewable

[puremourning]

Review status: 3 of 5 files reviewed at latest revision, 6 unresolved discussions, some commit checks pending.

---

ycmd/completers/language_server/language_server_completer.py, line 786 at r1 (raw file):

Previously, micbou wrote…

column.

Done.

---

ycmd/completers/language_server/language_server_completer.py, line 1483 at r1 (raw file):

Previously, micbou wrote…

overlapping

Done.

---

ycmd/completers/language_server/language_server_completer.py, line 1531 at r1 (raw file):

Previously, micbou wrote…

example.

Done.

---

ycmd/completers/language_server/language_server_completer.py, line 1532 at r1 (raw file):

Previously, micbou wrote…

truncation.

Done.

---

ycmd/completers/language_server/language_server_completer.py, line 1540 at r1 (raw file):

Previously, micbou wrote…

One blank line is enough.

Done.

---

ycmd/tests/request_wrap_test.py, line 63 at r1 (raw file):

Previously, micbou wrote…

Space before [ can be removed.

Done.

---

Comments from Reviewable

[micbou]

This works great.

---

Reviewed 2 of 2 files at r3.
Review status: all files reviewed at latest revision, 1 unresolved discussion.

---

ycmd/request_wrap.py, line 193 at r3 (raw file):

  def _Prefix( self ):
    return self[ 'line_value' ][ : ( self[ 'start_codepoint' ] - 1 ) ]

Parentheses are not needed (see the _Query function just above).

---

Comments from Reviewable

[puremourning]

Review status: all files reviewed at latest revision, 1 unresolved discussion.

---

ycmd/request_wrap.py, line 193 at r3 (raw file):

Previously, micbou wrote…

Parentheses are not needed (see the _Query function just above).

I kinda like them for the clarity. The -1 is quite subtle IMO.

---

Comments from Reviewable

[micbou]

Review status: all files reviewed at latest revision, 1 unresolved discussion.

---

ycmd/request_wrap.py, line 193 at r3 (raw file):

Previously, puremourning (Ben Jackson) wrote…

I kinda like them for the clarity. The -1 is quite subtle IMO.

If you think this improves readability then we should keep them.

---

Comments from Reviewable

[bstaletic]

@zzbot r=micbou

---

Reviewed 2 of 5 files at r1, 1 of 2 files at r2, 2 of 2 files at r3.
Review status: all files reviewed at latest revision, all discussions resolved.

---

Comments from Reviewable

[zzbot]

📌 Commit 3209927 has been approved by micbou

[zzbot]

⌛ Testing commit 3209927 with merge 5b1615a...

[zzbot]

☀️ Test successful - status-appveyor, status-travis
Approved by: micbou
Pushing 5b1615a to master...