Added unittests for util.mimeparse and made the module more robust and somewhat faster.

This also closes tickets #290, #330, #370 and #381.

3 files changed, 256 insertions, 51 deletions

diff --git a/.hgignore b/.hgignore
index dda7f55..09b0959 100644
--- a/.hgignore
+++ b/.hgignore
@@ -1,5 +1,13 @@
 syntax: glob
-docs/_build
-Paste.egg-info
+
+.project
+.pydevproject
+.settings
+*.pyc
+*.pyo
+*.log
+*.tmp
+*.egg-info
 dist/
 build/
+docs/_build
diff --git a/paste/util/mimeparse.py b/paste/util/mimeparse.py
index a8793b2..7a68772 100644
--- a/paste/util/mimeparse.py
+++ b/paste/util/mimeparse.py
@@ -28,15 +28,23 @@ def parse_mime_type(mime_type):
 
        ('application', 'xhtml', {'q', '0.5'})
        """
-    parts = mime_type.split(";")
-    params = dict([tuple([s.strip() for s in param.split("=")])\
-            for param in parts[1:] ])
-    full_type = parts[0].strip()
-    # Java URLConnection class sends an Accept header that includes a single "*"
-    # Turn it into a legal wildcard.
-    if full_type == '*': full_type = '*/*'
-    (type, subtype) = full_type.split("/")
-    return (type.strip(), subtype.strip(), params)
+    type = mime_type.split(';')
+    type, plist = type[0], type[1:]
+    try:
+        type, subtype = type.split('/', 1)
+    except ValueError:
+        type, subtype = type.strip() or '*', '*'
+    else:
+        type = type.strip() or '*'
+        subtype = subtype.strip() or '*'
+    params = {}
+    for param in plist:
+        param = param.split('=', 1)
+        if len(param) == 2:
+            key, value = param[0].strip(), param[1].strip()
+            if key and value:
+                params[key] = value
+    return type, subtype, params
 
 def parse_media_range(range):
     """Carves up a media range and returns a tuple of the
@@ -51,12 +59,17 @@ def parse_media_range(range):
        is a value for 'q' in the params dictionary, filling it
        in with a proper default if necessary.
        """
-    (type, subtype, params) = parse_mime_type(range)
-    if not params.has_key('q') or not params['q'] or \
-            not float(params['q']) or float(params['q']) > 1\
-            or float(params['q']) < 0:
+    type, subtype, params = parse_mime_type(range)
+    if 'q' in params:
+        try:
+            q = float(params['q'])
+            if not 0 <= q <= 1:
+                raise ValueError
+        except ValueError:
+            params['q'] = '1'
+    else:
         params['q'] = '1'
-    return (type, subtype, params)
+    return type, subtype, params
 
 def fitness_and_quality_parsed(mime_type, parsed_ranges):
     """Find the best match for a given mime-type against
@@ -65,24 +78,26 @@ def fitness_and_quality_parsed(mime_type, parsed_ranges):
        the fitness value and the value of the 'q' quality
        parameter of the best match, or (-1, 0) if no match
        was found. Just as for quality_parsed(), 'parsed_ranges'
-       must be a list of parsed media ranges. """
-    best_fitness = -1
-    best_fit_q = 0
-    (target_type, target_subtype, target_params) =\
-            parse_media_range(mime_type)
-    for (type, subtype, params) in parsed_ranges:
-        if (type == target_type or type == '*' or target_type == '*') and \
-                (subtype == target_subtype or subtype == '*' or target_subtype == '*'):
-            param_matches = reduce(lambda x, y: x+y, [1 for (key, value) in \
-                    target_params.iteritems() if key != 'q' and \
-                    params.has_key(key) and value == params[key]], 0)
-            fitness = (type == target_type) and 100 or 0
-            fitness += (subtype == target_subtype) and 10 or 0
-            fitness += param_matches
+       must be a list of parsed media ranges."""
+    best_fitness, best_fit_q = -1, 0
+    target_type, target_subtype, target_params = parse_media_range(mime_type)
+    for type, subtype, params in parsed_ranges:
+        if (type == target_type
+                    or type == '*' or target_type == '*') and (
+                subtype == target_subtype
+                    or subtype == '*' or target_subtype == '*'):
+            fitness = 0
+            if type == target_type:
+                fitness += 100
+            if subtype == target_subtype:
+                fitness += 10
+            for key in target_params:
+                if key != 'q' and key in params:
+                    if params[key] == target_params[key]:
+                        fitness += 1
             if fitness > best_fitness:
                 best_fitness = fitness
                 best_fit_q = params['q']
-
     return best_fitness, float(best_fit_q)
 
 def quality_parsed(mime_type, parsed_ranges):
@@ -90,9 +105,9 @@ def quality_parsed(mime_type, parsed_ranges):
     a list of media_ranges that have already been
     parsed by parse_media_range(). Returns the
     'q' quality parameter of the best match, 0 if no
-    match was found. This function bahaves the same as quality()
+    match was found. This function behaves the same as quality()
     except that 'parsed_ranges' must be a list of
-    parsed media ranges. """
+    parsed media ranges."""
     return fitness_and_quality_parsed(mime_type, parsed_ranges)[1]
 
 def quality(mime_type, ranges):
@@ -103,24 +118,27 @@ def quality(mime_type, ranges):
     0.7
 
     """
-    parsed_ranges = [parse_media_range(r) for r in ranges.split(",")]
+    parsed_ranges = map(parse_media_range, ranges.split(','))
     return quality_parsed(mime_type, parsed_ranges)
 
 def best_match(supported, header):
     """Takes a list of supported mime-types and finds the best
-    match for all the media-ranges listed in header. The value of
-    header must be a string that conforms to the format of the
-    HTTP Accept: header. The value of 'supported' is a list of
-    mime-types.
+    match for all the media-ranges listed in header. In case of
+    ambiguity, whatever comes first in the list will be chosen.
+    The value of header must be a string that conforms to the format
+    of the HTTP Accept: header. The value of 'supported' is a list
+    of mime-types.
 
     >>> best_match(['application/xbel+xml', 'text/xml'], 'text/*;q=0.5,*/*; q=0.1')
     'text/xml'
     """
-    parsed_header = [parse_media_range(r) for r in header.split(",")]
-    weighted_matches = [(fitness_and_quality_parsed(mime_type, parsed_header), mime_type)\
-            for mime_type in supported]
-    weighted_matches.sort()
-    return weighted_matches[-1][0][1] and weighted_matches[-1][1] or ''
+    if not supported:
+        return ''
+    parsed_header = map(parse_media_range, header.split(','))
+    best_type = max([
+            (fitness_and_quality_parsed(mime_type, parsed_header), -n)
+            for n, mime_type in enumerate(supported)])
+    return best_type[0][1] and supported[-best_type[1]] or ''
 
 def desired_matches(desired, header):
     """Takes a list of desired mime-types in the order the server prefers to
@@ -132,15 +150,13 @@ def desired_matches(desired, header):
     the desired list in the same order should each one be Accepted by the
     browser.
 
-    >>> sorted_match(['text/html', 'application/xml'], \
+    >>> desired_matches(['text/html', 'application/xml'], \
     ...     'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png')
     ['text/html', 'application/xml']
-    >>> sorted_match(['text/html', 'application/xml'], 'application/xml,application/json')
+    >>> desired_matches(['text/html', 'application/xml'], 'application/xml,application/json')
     ['application/xml']
     """
-    matches = []
-    parsed_ranges = [parse_media_range(r) for r in header.split(",")]
-    for mimetype in desired:
-        if quality_parsed(mimetype, parsed_ranges):
-            matches.append(mimetype)
-    return matches
+    parsed_ranges = map(parse_media_range, header.split(','))
+    return [mimetype for mimetype in desired
+        if quality_parsed(mimetype, parsed_ranges)]
+
diff --git a/tests/test_util/test_mimeparse.py b/tests/test_util/test_mimeparse.py
new file mode 100644
index 0000000..c24b5e5
--- /dev/null
+++ b/tests/test_util/test_mimeparse.py
@@ -0,0 +1,181 @@
+# (c) 2010 Ch. Zwerschke and contributors

+# This module is part of the Python Paste Project and is released under

+# the MIT License: http://www.opensource.org/licenses/mit-license.php

+

+from time import localtime

+from datetime import date

+from paste.util.mimeparse import *

+

+def test_parse_mime_type():

+    parse = parse_mime_type

+    assert parse('*/*') == ('*', '*', {})

+    assert parse('text/html') == ('text', 'html', {})

+    assert parse('audio/*; q=0.2') == ('audio', '*', {'q': '0.2'})

+    assert parse('text/x-dvi;level=1') == ('text', 'x-dvi', {'level': '1'})

+    assert parse('image/gif; level=2; q=0.4') == (

+        'image', 'gif', {'level': '2', 'q': '0.4'})

+    assert parse('application/xhtml;level=3;q=0.5') == (

+        'application', 'xhtml', {'level': '3', 'q': '0.5'})

+

+def test_parse_illformed_mime_type():

+    parse = parse_mime_type

+    assert parse('*') == ('*', '*', {})

+    assert parse('text') == ('text', '*', {})

+    assert parse('text/') == ('text', '*', {})

+    assert parse('/plain') == ('*', 'plain', {})

+    assert parse('/') == ('*', '*', {})

+    assert parse('text/plain;') == ('text', 'plain', {})

+    assert parse(';q=0.5') == ('*', '*', {'q': '0.5'})

+    assert parse('*; q=.2') == ('*', '*', {'q': '.2'})

+    assert parse('image; q=.7; level=3') == (

+        'image', '*', {'q': '.7', 'level': '3'})

+    assert parse('*;q=1') == ('*', '*', {'q': '1'})

+    assert parse('*;q=') == ('*', '*', {})

+    assert parse('*;=0.5') == ('*', '*', {})

+    assert parse('*;q=foobar') == ('*', '*', {'q': 'foobar'})

+    assert parse('image/gif; level=2; q=2') == (

+        'image', 'gif', {'level': '2', 'q': '2'})

+

+def test_parse_media_range():

+    parse = parse_media_range

+    assert parse('application/*;q=0.5') == ('application', '*', {'q': '0.5'})

+    assert parse('text/plain') == ('text', 'plain', {'q': '1'})

+    assert parse('*') == ('*', '*', {'q': '1'})

+    assert parse(';q=0.5') == ('*', '*', {'q': '0.5'})

+    assert parse('*;q=0.5') == ('*', '*', {'q': '0.5'})

+    assert parse('*;q=1') == ('*', '*', {'q': '1'})

+    assert parse('*;q=') == ('*', '*', {'q': '1'})

+    assert parse('*;q=-1') == ('*', '*', {'q': '1'})

+    assert parse('*;q=foobar') == ('*', '*', {'q': '1'})

+    assert parse('*;q=0.0001') == ('*', '*', {'q': '0.0001'})

+    assert parse('*;q=1000.0') == ('*', '*', {'q': '1'})

+    assert parse('*;q=0') == ('*', '*', {'q': '0'})

+    assert parse('*;q=0.0000') == ('*', '*', {'q': '0.0000'})

+    assert parse('*;q=1.0001') == ('*', '*', {'q': '1'})

+    assert parse('*;q=2') == ('*', '*', {'q': '1'})

+    assert parse('*;q=1e3') == ('*', '*', {'q': '1'})

+    assert parse('image/gif; level=2') == (

+        'image', 'gif', {'level': '2', 'q': '1'})

+    assert parse('image/gif; level=2; q=0.5') == (

+        'image', 'gif', {'level': '2', 'q': '0.5'})

+    assert parse('image/gif; level=2; q=2') == (

+        'image', 'gif', {'level': '2', 'q': '1'})

+

+def test_fitness_and_quality_parsed():

+    faq = fitness_and_quality_parsed

+    assert faq('*/*;q=0.7', [

+        ('foo', 'bar', {'q': '0.5'})]) == (0, 0.5)

+    assert faq('foo/*;q=0.7', [

+        ('foo', 'bar', {'q': '0.5'})]) == (100, 0.5)

+    assert faq('*/bar;q=0.7', [

+        ('foo', 'bar', {'q': '0.5'})]) == (10, 0.5)

+    assert faq('foo/bar;q=0.7', [

+        ('foo', 'bar', {'q': '0.5'})]) == (110, 0.5)

+    assert faq('text/html;q=0.7', [

+        ('foo', 'bar', {'q': '0.5'})]) == (-1, 0)

+    assert faq('text/html;q=0.7', [

+        ('text', 'bar', {'q': '0.5'})]) == (-1, 0)

+    assert faq('text/html;q=0.7', [

+        ('foo', 'html', {'q': '0.5'})]) == (-1, 0)

+    assert faq('text/html;q=0.7', [

+        ('text', '*', {'q': '0.5'})]) == (100, 0.5)

+    assert faq('text/html;q=0.7', [

+        ('*', 'html', {'q': '0.5'})]) == (10, 0.5)

+    assert faq('text/html;q=0.7', [

+        ('*', '*', {'q': '0'}), ('text', 'html', {'q': '0.5'})]) == (110, 0.5)

+    assert faq('text/html;q=0.7', [

+        ('*', '*', {'q': '0.5'}), ('audio', '*', {'q': '0'})]) == (0, 0.5)

+    assert faq('audio/mp3;q=0.7', [

+        ('*', '*', {'q': '0'}), ('audio', '*', {'q': '0.5'})]) == (100, 0.5)

+    assert faq('*/mp3;q=0.7', [

+        ('foo', 'mp3', {'q': '0.5'}), ('audio', '*', {'q': '0'})]) == (10, 0.5)

+    assert faq('audio/mp3;q=0.7', [

+        ('audio', 'ogg', {'q': '0'}), ('*', 'mp3', {'q': '0.5'})]) == (10, 0.5)

+    assert faq('audio/mp3;q=0.7', [

+        ('*', 'ogg', {'q': '0'}), ('*', 'mp3', {'q': '0.5'})]) == (10, 0.5)

+    assert faq('text/html;q=0.7', [

+        ('text', 'plain', {'q': '0'}),

+        ('plain', 'html', {'q': '0'}),

+        ('text', 'html', {'q': '0.5'}),

+        ('html', 'text', {'q': '0'})]) == (110, 0.5)

+    assert faq('text/html;q=0.7;level=2', [

+        ('plain', 'html', {'q': '0', 'level': '2'}),

+        ('text', '*', {'q': '0.5', 'level': '3'}),

+        ('*', 'html', {'q': '0.5', 'level': '2'}),

+        ('image', 'gif', {'q': '0.5', 'level': '2'})]) == (100, 0.5)

+    assert faq('text/html;q=0.7;level=2', [

+        ('text', 'plain', {'q': '0'}), ('text', 'html', {'q': '0'}),

+        ('text', 'plain', {'q': '0', 'level': '2'}),

+        ('text', 'html', {'q': '0.5', 'level': '2'}),

+        ('*', '*', {'q': '0', 'level': '2'}),

+        ('text', 'html', {'q': '0', 'level': '3'})]) == (111, 0.5)

+    assert faq('text/html;q=0.7;level=2;opt=3', [

+        ('text', 'html', {'q': '0'}),

+        ('text', 'html', {'q': '0', 'level': '2'}),

+        ('text', 'html', {'q': '0', 'opt': '3'}),

+        ('*', '*', {'q': '0', 'level': '2', 'opt': '3'}),

+        ('text', 'html', {'q': '0', 'level': '3', 'opt': '3'}),

+        ('text', 'html', {'q': '0.5', 'level': '2', 'opt': '3'}),

+        ('*', '*', {'q': '0', 'level': '3', 'opt': '3'})]) == (112, 0.5)

+

+def test_quality_parsed():

+    qp = quality_parsed

+    assert qp('image/gif;q=0.7', [('image', 'jpg', {'q': '0.5'})]) == 0

+    assert qp('image/gif;q=0.7', [('image', '*', {'q': '0.5'})]) == 0.5

+    assert qp('audio/mp3;q=0.7;quality=100', [

+        ('*', '*', {'q': '0', 'quality': '100'}),

+        ('audio', '*', {'q': '0', 'quality': '100'}),

+        ('*', 'mp3', {'q': '0', 'quality': '100'}),

+        ('audio', 'mp3', {'q': '0', 'quality': '50'}),

+        ('audio', 'mp3', {'q': '0.5', 'quality': '100'}),

+        ('audio', 'mp3', {'q': '0.5'})]) == 0.5

+

+def test_quality():

+    assert quality('text/html',

+        'text/*;q=0.3, text/html;q=0.7, text/html;level=1,'

+        ' text/html;level=2;q=0.4, */*;q=0.5') == 0.7

+    assert quality('text/html;level=2',

+        'text/*;q=0.3, text/html;q=0.7, text/html;level=1,'

+        ' text/html;level=2;q=0.4, */*;q=0.5') == 0.4

+    assert quality('text/plain',

+        'text/*;q=0.3, text/html;q=0.7, text/html;level=1,'

+        ' text/html;level=2;q=0.4, */*;q=0.5') == 0.3

+    assert quality('plain/text',

+        'text/*;q=0.3, text/html;q=0.7, text/html;level=1,'

+        ' text/html;level=2;q=0.4, */*;q=0.5') == 0.5

+

+def test_best_match():

+    bm = best_match

+    assert bm([], '*/*') == ''

+    assert bm(['application/xbel+xml', 'text/xml'],

+        'text/*;q=0.5,*/*; q=0.1') == 'text/xml'

+    assert bm(['application/xbel+xml', 'audio/mp3'],

+        'text/*;q=0.5,*/*; q=0.1') == 'application/xbel+xml'

+    assert bm(['application/xbel+xml', 'audio/mp3'],

+        'text/*;q=0.5,*/mp3; q=0.1') == 'audio/mp3'

+    assert bm(['application/xbel+xml', 'text/plain', 'text/html'],

+        'text/*;q=0.5,*/plain; q=0.1') == 'text/plain'

+    assert bm(['application/xbel+xml', 'text/html', 'text/xhtml'],

+        'text/*;q=0.1,*/xhtml; q=0.5') == 'text/html'

+    assert bm(['application/xbel+xml', 'text/html', 'text/xhtml'],

+        '*/html;q=0.1,*/xhtml; q=0.5') == 'text/xhtml'

+

+def test_illformed_best_match():

+    bm = best_match

+    assert bm(['image/png', 'image/jpeg', 'image/gif', 'text/html'],

+        'text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2') == 'image/jpeg'

+    assert bm(['image/png', 'image/jpg', 'image/tif', 'text/html'],

+        'text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2') == 'text/html'

+    assert bm(['image/png', 'image/jpg', 'image/tif', 'audio/mp3'],

+        'text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2') == 'image/png'

+

+def test_sorted_match():

+    dm = desired_matches

+    assert dm(['text/html', 'application/xml'],

+        'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,'

+        'text/plain;q=0.8,image/png') == ['text/html', 'application/xml']

+    assert dm(['text/html', 'application/xml'],

+        'application/xml,application/json') == ['application/xml']

+    assert dm(['text/xhtml', 'text/plain', 'application/xhtml'],

+        'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,'

+        'text/plain;q=0.8,image/png') == ['text/plain']