view tests/test-check-code.t @ 24545:9e0c67e84896

json: implement {tags} template

Tags is pretty easy to implement. Let's start there.

The output is slightly different from `hg tags -Tjson`. For reference,
the CLI has the following output:

  [
   {
    "node": "e2049974f9a23176c2addb61d8f5b86e0d620490",
    "rev": 29880,
    "tag": "tip",
    "type": ""
   },
   ...
  ]

Our output has the format:

  {
    "node": "0aeb19ea57a6d223bacddda3871cb78f24b06510",
    "tags": [
      {
        "node": "e2049974f9a23176c2addb61d8f5b86e0d620490",
        "tag": "tag1",
        "date": [1427775457.0, 25200]
      },
      ...
    ]
  }

"rev" is omitted because it isn't a reliable identifier. We shouldn't
be exposing them in web APIs and giving the impression it remotely
resembles a stable identifier. Perhaps we could one day hide this behind
a config option (it might be useful to expose when running servers
locally).

The "type" of the tag isn't defined because this information isn't yet
exposed to the hgweb templater (it could be in a follow-up) and because
it is questionable whether different types should be exposed at all.
(Should the web interface really be exposing "local" tags?)

We use an object for the outer type instead of Array for a few reasons.
First, it is extensible. If we ever need to throw more global properties
into the output, we can do that without breaking backwards compatibility
(property additions should be backwards compatible). Second, uniformity
in web APIs is nice. Having everything return objects seems much saner than
a mix of array and object. Third, there are security issues with arrays
in older browsers. The JSON web services world almost never uses arrays
as the main type for this reason.

Another possibly controversial part about this patch is how dates are
defined. While JSON has a Date type, it is based on the JavaScript Date
type, which is widely considered a pile of garbage. It is a non-starter
for this reason.

Many of Mercurial's built-in date filters drop seconds resolution. So
that's a non-starter as well, since we want the API to be lossless where
possible. rfc3339date, rfc822date, isodatesec, and date are all lossless.
However, they each require the client to perform string parsing on top of
JSON decoding. While date parsing libraries are pretty ubiquitous, some
languages don't have them out of the box. However, pretty much every
programming language can deal with UNIX timestamps (which are just
integers or floats). So, we choose to use Mercurial's internal date
representation, which in JSON is modeled as float seconds since UNIX
epoch and an integer timezone offset from UTC (keep in mind
JavaScript/JSON models all "Numbers" as double prevision floating point
numbers, so there isn't a difference between ints and floats in JSON).

  $ cat > correct.py <<EOF
  > def toto(arg1, arg2):
  >     del arg2
  >     return (5 + 6, 9)
  > EOF
  $ cat > wrong.py <<EOF
  > def toto( arg1, arg2):
  >     del(arg2)
  >     return ( 5+6, 9)
  > EOF
  $ cat > quote.py <<EOF
  > # let's use quote in comments
  > (''' ( 4x5 )
  > but """\\''' and finally''',
  > """let's fool checkpatch""", '1+2',
  > '"""', 42+1, """and
  > ( 4-1 ) """, "( 1+1 )\" and ")
  > a, '\\\\\\\\', "\\\\\\" x-2", "c-1"
  > EOF
  $ cat > non-py24.py <<EOF
  > # Using builtins that does not exist in Python 2.4
  > if any():
  >     x = all()
  >     y = format(x)
  >     # next(generator) is new in 2.6
  >     z = next(x)
  >     # but generator.next() is okay
  >     x.next()
  >     # and we can make our own next
  >     def next(stuff):
  >         pass
  > 
  > # Do not complain about our own definition
  > def any(x):
  >     pass
  > 
  > # try/except/finally block does not exist in Python 2.4
  >     try:
  >         pass
  >     except StandardError, inst:
  >         pass
  >     finally:
  >         pass
  > 
  > # nested try/finally+try/except is allowed
  >     try:
  >         try:
  >             pass
  >         except StandardError, inst:
  >             pass
  >     finally:
  >         pass
  > 
  > # yield inside a try/finally block is not allowed in Python 2.4
  >     try:
  >         pass
  >         yield 1
  >     finally:
  >         pass
  >     try:
  >         yield
  >         pass
  >     finally:
  >         pass
  > 
  > EOF
  $ cat > classstyle.py <<EOF
  > class newstyle_class(object):
  >     pass
  > 
  > class oldstyle_class:
  >     pass
  > 
  > class empty():
  >     pass
  > 
  > no_class = 1:
  >     pass
  > EOF
  $ check_code="$TESTDIR"/../contrib/check-code.py
  $ "$check_code" ./wrong.py ./correct.py ./quote.py ./non-py24.py ./classstyle.py
  ./wrong.py:1:
   > def toto( arg1, arg2):
   gratuitous whitespace in () or []
  ./wrong.py:2:
   >     del(arg2)
   Python keyword is not a function
  ./wrong.py:3:
   >     return ( 5+6, 9)
   gratuitous whitespace in () or []
   missing whitespace in expression
  ./quote.py:5:
   > '"""', 42+1, """and
   missing whitespace in expression
  ./non-py24.py:2:
   > if any():
   any/all/format not available in Python 2.4
  ./non-py24.py:3:
   >     x = all()
   any/all/format not available in Python 2.4
  ./non-py24.py:4:
   >     y = format(x)
   any/all/format not available in Python 2.4
  ./non-py24.py:6:
   >     z = next(x)
   no next(foo) in Python 2.4 and 2.5, use foo.next() instead
  ./non-py24.py:18:
   >     try:
   no try/except/finally in Python 2.4
  ./non-py24.py:35:
   >     try:
   no yield inside try/finally in Python 2.4
  ./non-py24.py:40:
   >     try:
   no yield inside try/finally in Python 2.4
  ./classstyle.py:4:
   > class oldstyle_class:
   old-style class, use class foo(object)
  ./classstyle.py:7:
   > class empty():
   class foo() not available in Python 2.4, use class foo(object)
  [1]
  $ cat > python3-compat.py << EOF
  > foo <> bar
  > reduce(lambda a, b: a + b, [1, 2, 3, 4])
  > dict(key=value)
  > EOF
  $ "$check_code" python3-compat.py
  python3-compat.py:1:
   > foo <> bar
   <> operator is not available in Python 3+, use !=
  python3-compat.py:2:
   > reduce(lambda a, b: a + b, [1, 2, 3, 4])
   reduce is not available in Python 3+
  python3-compat.py:3:
   > dict(key=value)
   dict() is different in Py2 and 3 and is slower than {}
  [1]

  $ cat > is-op.py <<EOF
  > # is-operator comparing number or string literal
  > x = None
  > y = x is 'foo'
  > y = x is "foo"
  > y = x is 5346
  > y = x is -6
  > y = x is not 'foo'
  > y = x is not "foo"
  > y = x is not 5346
  > y = x is not -6
  > EOF

  $ "$check_code" ./is-op.py
  ./is-op.py:3:
   > y = x is 'foo'
   object comparison with literal
  ./is-op.py:4:
   > y = x is "foo"
   object comparison with literal
  ./is-op.py:5:
   > y = x is 5346
   object comparison with literal
  ./is-op.py:6:
   > y = x is -6
   object comparison with literal
  ./is-op.py:7:
   > y = x is not 'foo'
   object comparison with literal
  ./is-op.py:8:
   > y = x is not "foo"
   object comparison with literal
  ./is-op.py:9:
   > y = x is not 5346
   object comparison with literal
  ./is-op.py:10:
   > y = x is not -6
   object comparison with literal
  [1]

  $ cat > for-nolineno.py <<EOF
  > except:
  > EOF
  $ "$check_code" for-nolineno.py --nolineno
  for-nolineno.py:0:
   > except:
   naked except clause
  [1]

  $ cat > warning.t <<EOF
  >   $ function warnonly {
  >   > }
  >   $ diff -N aaa
  >   $ function onwarn {}
  > EOF
  $ "$check_code" warning.t
  $ "$check_code" --warn warning.t
  warning.t:1:
   >   $ function warnonly {
   warning: don't use 'function', use old style
  warning.t:3:
   >   $ diff -N aaa
   warning: don't use 'diff -N'
  warning.t:4:
   >   $ function onwarn {}
   warning: don't use 'function', use old style
  [1]
  $ cat > raise-format.py <<EOF
  > raise SomeException, message
  > # this next line is okay
  > raise SomeException(arg1, arg2)
  > EOF
  $ "$check_code" not-existing.py raise-format.py
  Skipping*not-existing.py* (glob)
  raise-format.py:1:
   > raise SomeException, message
   don't use old-style two-argument raise, use Exception(message)
  [1]

  $ cat > rst.py <<EOF
  > """problematic rst text
  > 
  > .. note::
  >     wrong
  > """
  > 
  > '''
  > 
  > .. note::
  > 
  >     valid
  > 
  > new text
  > 
  >     .. note::
  > 
  >         also valid
  > '''
  > 
  > """mixed
  > 
  > .. note::
  > 
  >   good
  > 
  >     .. note::
  >         plus bad
  > """
  > EOF
  $ $check_code -w rst.py
  rst.py:3:
   > .. note::
   warning: add two newlines after '.. note::'
  rst.py:26:
   >     .. note::
   warning: add two newlines after '.. note::'
  [1]

  $ cat > ./map-inside-gettext.py <<EOF
  > print _("map inside gettext %s" % v)
  > 
  > print _("concatenating " " by " " space %s" % v)
  > print _("concatenating " + " by " + " '+' %s" % v)
  > 
  > print _("mapping operation in different line %s"
  >         % v)
  > 
  > print _(
  >         "leading spaces inside of '(' %s" % v)
  > EOF
  $ "$check_code" ./map-inside-gettext.py
  ./map-inside-gettext.py:1:
   > print _("map inside gettext %s" % v)
   don't use % inside _()
  ./map-inside-gettext.py:3:
   > print _("concatenating " " by " " space %s" % v)
   don't use % inside _()
  ./map-inside-gettext.py:4:
   > print _("concatenating " + " by " + " '+' %s" % v)
   don't use % inside _()
  ./map-inside-gettext.py:6:
   > print _("mapping operation in different line %s"
   don't use % inside _()
  ./map-inside-gettext.py:9:
   > print _(
   don't use % inside _()
  [1]

web templates

  $ mkdir -p mercurial/templates
  $ cat > mercurial/templates/example.tmpl <<EOF
  > {desc}
  > {desc|escape}
  > {desc|firstline}
  > {desc|websub}
  > EOF

  $ "$check_code" --warnings mercurial/templates/example.tmpl
  mercurial/templates/example.tmpl:2:
   > {desc|escape}
   warning: follow desc keyword with either firstline or websub
  [1]