print statement is highlighted by pylint

[noonedeadpunk]

Environment data

VS Code version: 1.19.3
Python Extension version: 2018.1
Python Version: 2.7, 3.5
OS and version: OS independent

Actual behavior

pylint marks print statement and print() function are marked as error on both python 2.7 and python 3.5. This behavior appeared just after module upgrade to 2018.1

Expected behavior

pylint shouldn't mark error if using print() function on any python version and print statement on python 2.7

Steps to reproduce:

• create new .py file
• enter print("Hello, world!")
• save file
  

Logs

Output from Python output panel

[pylint]E1601:print statement used (1, 1)

Output from Console window (Help->Developer Tools menu)

console.ts:123 [Extension Host] Python Extension: Failed to get conda info from conda null
t.log @ console.ts:123
console.ts:123 [Extension Host] (node:16735) DeprecationWarning: os.tmpDir() is deprecated. Use os.tmpdir() instead.

[MikhailArkhipov]

This is PyLint message. Extension simply outputs what PyLint produces. In 2.7 the message should appear according to PyLint docs:
https://pylint.readthedocs.io/en/latest/technical_reference/features.html

print-statement (E1601):
  print statement used (print is a function in Python 3)

However, it should not appear in 3+ then. You can suppress the message in pylintrc file.

[noonedeadpunk]

But it seems, that according to pylint, this message should be printed only if print is used as the statement, while it marks as an error while using print as a function.

Moreover, I couldn't reproduce this with system pylint (which is installed on the same machine, as the affected VS Code):

dimar@SVTech-Workstation:~/Documents/Python$ /usr/bin/pylint --version
No config file found, using default configuration
pylint 1.5.2, 
astroid 1.4.4
Python 2.7.12 (default, Dec  4 2017, 14:50:18) 
[GCC 5.4.0 20160609]
dimar@SVTech-Workstation:~/Documents/Python$ /usr/bin/pylint test.py
No config file found, using default configuration
************* Module test
C:  1, 0: Unnecessary parens after 'print' keyword (superfluous-parens)
C:  1, 0: Missing module docstring (missing-docstring)


Report
======
1 statements analysed.

Statistics by type
------------------

+---------+-------+-----------+-----------+------------+---------+
|type     |number |old number |difference |%documented |%badname |
+=========+=======+===========+===========+============+=========+
|module   |1      |1          |=          |0.00        |0.00     |
+---------+-------+-----------+-----------+------------+---------+
|class    |0      |0          |=          |0           |0        |
+---------+-------+-----------+-----------+------------+---------+
|method   |0      |0          |=          |0           |0        |
+---------+-------+-----------+-----------+------------+---------+
|function |0      |0          |=          |0           |0        |
+---------+-------+-----------+-----------+------------+---------+



Raw metrics
-----------

+----------+-------+------+---------+-----------+
|type      |number |%     |previous |difference |
+==========+=======+======+=========+===========+
|code      |1      |50.00 |NC       |NC         |
+----------+-------+------+---------+-----------+
|docstring |0      |0.00  |NC       |NC         |
+----------+-------+------+---------+-----------+
|comment   |0      |0.00  |NC       |NC         |
+----------+-------+------+---------+-----------+
|empty     |1      |50.00 |NC       |NC         |
+----------+-------+------+---------+-----------+



Duplication
-----------

+-------------------------+------+---------+-----------+
|                         |now   |previous |difference |
+=========================+======+=========+===========+
|nb duplicated lines      |0     |NC       |NC         |
+-------------------------+------+---------+-----------+
|percent duplicated lines |0.000 |NC       |NC         |
+-------------------------+------+---------+-----------+



Messages by category
--------------------

+-----------+-------+---------+-----------+
|type       |number |previous |difference |
+===========+=======+=========+===========+
|convention |2      |0        |+2.00      |
+-----------+-------+---------+-----------+
|refactor   |0      |0        |=          |
+-----------+-------+---------+-----------+
|warning    |0      |0        |=          |
+-----------+-------+---------+-----------+
|error      |0      |1        |-1.00      |
+-----------+-------+---------+-----------+



Messages
--------

+-------------------+------------+
|message id         |occurrences |
+===================+============+
|superfluous-parens |1           |
+-------------------+------------+
|missing-docstring  |1           |
+-------------------+------------+



Global evaluation
-----------------
Your code has been rated at -10.00/10

dimar@SVTech-Workstation:~/Documents/Python$

And if I replace print() with a statement, only "Missing module docstring (missing-docstring)" notification will stay. So this is possible only if the module uses it's own pylint, which is different version from the system one.

It seems, that it's my fault about reporting problem for python3 - pylint just hadn't re-read the file after changing Python version. So this problem is referenced only to python2.

[zerkms]

@noonedeadpunk run pylint with these arguments

--disable=all --enable=F,E,unreachable,duplicate-key,unnecessary-semicolon,global-variable-not-assigned,unused-variable,unused-wildcard-import,binary-op-exception,bad-format-string,anomalous-backslash-in-string,bad-open-mode --reports=n --output-format=text --msg-template='{line},{column},{category},{msg_id}:{msg}'

[noonedeadpunk]

@zerkms Yep, you're right, enabling "E" raises "print statement used". Most likely, that vscode-python used other list of enabled modules before update, as this problem appeared only after module update.

And it's pretty irritating, as all of projects are highlighted due to "print statement used". But it seems, that the problem should be addressed to the pylint.

[zerkms]

@noonedeadpunk if you report it to the pylink could you please post a link here

[noonedeadpunk]

@zerkms it seems, that this actual problem with pylint started some time ago, but only now it started to annoy in vscode. Probably you know if --enable=E (or probably some other option) was added by default with the module update?

So, from __future__ import print_function resolves problem, but it looks more like workaround, than the real solution.

[zerkms]

I personally use logger instead of print ;-)

[noonedeadpunk]

Actually, something is really became wrong with pylint after module upgrade. I suppose, that the point is in the default flags for pylint.

So, when I enabled all flags locally for file, which were kindly provided by @zerkms I recieved the following result (which is absolutely correct, except print statement error, which is seemed to be really pylint problem, when -E(--error-mode) flag is used):

But if I will remove this pylint options string, all other notifications will be suppressed. If I will remove print() statement, pylint won't display suppressed problems at all.
And I am 100% sure, that these notifications were shown before the module update

[Omegastick]

In this update, PyLint also stopped highlighting missing docstrings for me. Not sure if it's related or not.

[evanwang22]

I've been experiencing a similar issue since the latest update to 2018.1. All print statements trigger E1601:print statement used despite explicitly disabling print-statement in my .pylintrc at the project level. Additionally, "Convention" messages no longer show in VSCode (which could be similar to @Omegastick's issue, as missing-docstring is a Convention-level rule)

As the screenshots show, pylint within VSCode and running pylint from the CLI give me different outputs. Happy to provide more detail if it's helpful

[MikhailArkhipov]

The message comes straight from from pylint. It flags both statement and the function in 2.7. It does not flag function in 3+ but does flag the statement.

I found that

• changing interpreters did not re-run linting. I am fixing this.
• changing pylintrc and saving it does not re-run linting. This may be non-trivial to fix since we can't watch all the possible pylintrc locations. Alternative is to add command 'run lint' for the manual activation.

So far I see if I disable the warning in pylintrc it does go away. It also does not appear in Python 3+ (in 2018.1 you'd have to reopen the file after changing interpreters or edit and save the document)

Python 2.7

Python 3.6

Disabled:

[noonedeadpunk]

The problem is, that in this release, you've changed flags with which pylint is launched. So the most of messages from pylint just suppressed in VS code.

It stopped displaying at least trailing-whitespace, trailing-newlines, missing-final-newline, line-too-long, missing-docstring, etc. These are pretty good notifications, and I don't see the reason to disable them.

To reproduce it just try to place the following code, from the screenshot

It shows the problems only because of the string:
# pylint: enable=F,E,unreachable,duplicate-key,unnecessary-semicolon,global-variable-not-assigned,unused-variable,unused-wildcard-import,binary-op-exception,bad-format-string,anomalous-backslash-in-string,bad-open-mode

If you remove it, these messages won't be displayed.

[MikhailArkhipov]

This change is by design since many people don't care about whitespace or styling. In many editors these kind of reports are off by default. There is a setting though:

pylintUseMinimalCheckers

You can set it to false and the arguments you mentioned won't be passed to pylint.

[MikhailArkhipov]

18be595 further improves this

[robhaswell]

I'm pretty sure the old behaviour was to use the workspace .pylintrc? If not, is that possible somehow? Personally I don't care what is/isn't highlighted, as long as it is exactly equal to the things which are going to fail the linting step in my CI pipeline.

[patrys]

@robhaswell See #788, it seems to be a bug.

[adgedenkers]

Forgive me, but is this resolved? I see the labels "closed-verified fixed" and "closed-fixed" had been added. Can you explain how you go about resolving this issue?

[noonedeadpunk]

@adgedenkers It's actually pylint bug and is caused by -E flag (pylint-dev/pylint#1869).

But vscode-python has changed the way to launch pylint since the update "by design". In order to get old way of pylint you may set "python.linting.pylintUseMinimalCheckers": false for the vscode user-space settings.

[crazydogen]

why not just disable the E1601 errors in settings?

{
    "python.linting.pylintArgs":
   [
        "–disable-msg=E1601"
   ]
}

[brettcannon]

It's looking like Pylint won't be fixing this before our next release, so we will probably have to add an explicit disablement of that checker.

[brettcannon]

If people want to help us test our fix, please install the insiders build of the extension and make sure you have Pylint 1.8 installed.

[brettcannon]

I'm re-opening the issue as you get an extraneous error from Pylint about --disable=print-statement not being a module ATM.

[brettcannon]

In case we have to list every E error code manually to avoid enabling --py3k checkers:

E0001,E0011,E0012,E0100,E0101,E0102,E0103,E0104,E0105,E0107,E0108,E0110,E0111,E0112,E0113,E0114,E0115,E0116,E0117,E0118,E0202,E0203,E0211,E0213,E0236,E0237,E0238,E0239,E0240,E0241,E0301,E0302,E0303,E0401,E0402,E0601,E0602,E0603,E0604,E0611,E0632,E0633,E0701,E0702,E0703,E0704,E0710,E0711,E0712,E1003,E1101,E1102,E1111,E1120,E1121,E1123,E1124,E1125,E1126,E1127,E1128,E1129,E1130,E1131,E1132,E1133,E1134,E1135,E1136,E1137,E1138,E1139,E1200,E1201,E1205,E1206,E1300,E1301,E1302,E1303,E1304,E1305,E1306,E1310,E1700,E1701

You will notice that E1601 is not on there. This was created using https://gist.github.com/brettcannon/eff7f38a60af48d39814cbb2f33b3d1d .

[MatthieuScarset]

So... is the only solution to have a .pylintrc file in all our project?

I've tried the custom setting but it's not working.
#722 (comment)

[robhaswell]

So... is the only solution to have a .pylintrc file in all our project?

I think this is a good thing @MatthieuScarset. It's no different to other per-project configuration being present (e.g. .gitignore) and it means that the feedback in VS Code will match up with the results of running pylint from the command-line.

The only problem is that this doesn't work as of the latest release.

[brettcannon]

@MatthieuScarset @robhaswell We have some fixes rolling out in the next release to strengthen our detection of custom settings and thus turning off our simpler rules. And as for this specific problem, we are literally meeting about this specific bug in less than an hour to come up with a solution (if we can), so we just ask people be patient as we are actively working on this and aiming to release this week (assuming no more showstopping bugs crop up 😉 ).