Autocomplete in the editor appends the completed variable onto the preceding expression if there's a token in between

[spyder-bot]

From jon...@gmail.com on 2014-11-07T19:27:37Z

Spyder Version: 2.4.0dev
Python Version: 2.7.8
Qt Version : 4.8.6, PyQt4 (API v2) 4.10.4 on Windows
pyflakes >=0.5.0: 0.8.1 (OK)
pep8 >=0.6 : 1.5.7 (OK)
IPython >=0.13 : 2.3.0 (OK)
pygments >=1.6 : 1.6 (OK)
pandas >=0.13.1 : 0.15.0 (OK)
sphinx >=0.6.6 : 1.2.3 (OK)
rope >=0.9.2 : 0.9.4 (OK)
matplotlib >=1.0: 1.4.0 (OK)
sympy >=0.7.0 : 0.7.5 (OK)
pylint >=0.25 : None (NOK)

What steps will reproduce the problem?

1. Go to the editor
   1. Type in a new expression, a token such as '+', and the stem of an existing variable
   2. Hit to autocomplete

What is the expected output? What do you see instead?

For example, if "first_instance" already exists in the editor, typing "XYZ + first", then enter yields the text XYZfirst_instance

Please provide any additional information below

.

Original issue: http://code.google.com/p/spyderlib/issues/detail?id=2044

[spyder-bot]

From ccordoba12 on 2014-11-09T15:09:17Z

Yes, I've seen this problem too. Steven, could you take a look at it?

Status: Accepted
Owner: steven.s...@gmail.com
Labels: -Priority-Medium Priority-High Cat-Editor MS-v2.3.2

[spyder-bot]

From steven.s...@gmail.com on 2014-11-09T17:42:39Z

Sure can.

[spyder-bot]

From steven.s...@gmail.com on 2014-11-09T17:49:08Z

So, I can almost replicate this. When I try it, I get varying overlaps on the word itself:

XYZ + first -> XYZ + fifirst_instance
XYZ + first_ -> XYZ + firfirst_instance

However, this behaviour is completely fixed when I try it using the Jedi-0.8 PR branch. I vote we defer this in favour of that PR.

[spyder-bot]

From contrebasse on 2014-11-10T05:37:57Z

Yes, this was corrected in the Jedi branch, I remember reporting it there.

[spyder-bot]

From ccordoba12 on 2014-11-10T10:20:12Z

But the Jedi work is going to be merged for 2.4, and we need a fix for 2.3 in the meantime.

[spyder-bot]

From jon...@gmail.com on 2014-11-10T14:20:56Z

I had updated to the 2.4.0dev to fix another issue, and I can't remember/didn't test if it was this way in 2.3.1 or not.

[spyder-bot]

From steven.s...@gmail.com on 2014-11-10T14:32:25Z

@jonsca, I ran my tests on the current default branch on Bitbucket. @-ccordoba12, I've got a lot of irons in the fire at the moment, I don't think I'll be able to get to this for 2.3.

[spyder-bot]

From ccordoba12 on 2014-11-11T13:36:52Z

Ok, no problem Steven. I'll dig into what's happening because it's really annoying, and we need a fix for 2.3.2.

Owner: ccordoba12
Labels: -Priority-High Priority-Critical

[spyder-bot]

From steven.s...@gmail.com on 2014-11-11T14:34:53Z

Hold on Carlos, I've got it. I was overwhelmed this weekend but I'm caught up.

—
Sent from Mailbox

[spyder-bot]

From ccordoba12 on 2014-11-12T09:30:58Z

Ok, thanks for letting me know. I'll wait for your PR then :-)

Owner: steven.s...@gmail.com

[spyder-bot]

From steven.s...@gmail.com on 2014-11-12T19:29:20Z

Its a pretty simple change, actually:

$ hg diff
diff -r 733f231362e3 spyderlib/widgets/editor.py
--- a/spyderlib/widgets/editor.py Sun Nov 02 15:51:16 2014 -0500
+++ b/spyderlib/widgets/editor.py Wed Nov 12 21:27:29 2014 -0600
@@ -368,7 +368,7 @@
if comp_list:
completion_text = re.findall(r"[\w.]+", text, re.UNICODE)
if completion_text:

•                completion_text = completion_text[0]
  

•                completion_text = completion_text[-1]
           else:
               comp_list = []
           if '.' in completion_text:
  

[spyder-bot]

From steven.s...@gmail.com on 2014-11-12T19:31:22Z

Not worth creating a new branch/PR for.

[spyder-bot]

From ccordoba12 on 2014-11-15T09:27:14Z

Ok, don't worry about it. I'll commit myself. Thanks a lot for your help Steve, you saved me a lot of time :-)

[spyder-bot]

From ccordoba12 on 2014-11-15T10:39:08Z

This issue was closed by revision f7b82b909ba0 .

Status: Fixed