fix for issue #72, dot allowed in anchor

7 files changed, 81 insertions, 15 deletions

diff --git a/README.rst b/README.rst
index 5418ead..01b6985 100644
--- a/README.rst
+++ b/README.rst
@@ -18,8 +18,18 @@ ChangeLog
 
 ::
 
+  0.12.15 (2016-09-21):
+   - allow dot char (and many others) in anchor name
+     Fix issue 72 (reported by Shalon Wood)
+   - Slightly smarter behaviour dumping strings when no style is
+     specified. Single string scalars that start with single quotes
+     or have newlines now are dumped double quoted: "'abc\nklm'" instead of
+     '''abc
+
+        klm'''
+
   0.12.14 (2016-09-21):
-   - preserve round-trip sequences that are mapping keys 
+   - preserve round-trip sequences that are mapping keys
      (prompted by stackoverflow question 39595807 from Nowox)
 
   0.12.13 (2016-09-15):
@@ -27,7 +37,7 @@ ChangeLog
      keys incorrect (reported by Tal Liron)
 
   0.12.11 (2016-09-06):
-   - Fix issue 58 endless loop in scanning tokens (reported by 
+   - Fix issue 58 endless loop in scanning tokens (reported by
      Christopher Lambert)
 
   0.12.10 (2016-09-05):
@@ -45,7 +55,7 @@ ChangeLog
      - fixing issue 46 empty lines between top-level keys were gobbled (but
        not between sequence elements, nor between keys in netsted mappings
        (reported by Alex Harvey)
-   
+
   0.12.5 (2016-08-20):
     - fixing issue 45 preserving datetime formatting (submitted by altuin)
       Several formatting parameters are preserved with some normalisation:
diff --git a/_test/test_anchor.py b/_test/test_anchor.py
index 593bf8a..05c8693 100644
--- a/_test/test_anchor.py
+++ b/_test/test_anchor.py
@@ -17,8 +17,8 @@ def load(s):
     return round_trip_load(dedent(s))
 
 
-def compare(data, s):
-    assert round_trip_dump(data) == dedent(s)
+def compare(d, s):
+    assert round_trip_dump(d) == dedent(s)
 
 
 class TestAnchorsAliases:
@@ -323,3 +323,33 @@ class TestMergeKeysValues:
         if PY3:
             ref -= 1
         assert len(x) == ref
+
+
+class TestFullCharSetAnchors:
+    def test_master_of_orion(self):
+        # https://bitbucket.org/ruamel/yaml/issues/72/not-allowed-in-anchor-names
+        # submitted by Shalon Wood
+        yaml_str = '''
+        - collection: &Backend.Civilizations.RacialPerk
+            items:
+                  - key: perk_population_growth_modifier
+        - *Backend.Civilizations.RacialPerk
+        '''
+        data = load(yaml_str)  # NOQA
+
+    def test_roundtrip_00(self):
+        yaml_str = '''
+        - &dotted.words.here
+          a: 1
+          b: 2
+        - *dotted.words.here
+        '''
+        data = round_trip(yaml_str)  # NOQA
+
+    def test_roundtrip_01(self):
+        yaml_str = '''
+        - &dotted.words.here[a, b]
+        - *dotted.words.here
+        '''
+        data = load(yaml_str)  # NOQA
+        compare(data, yaml_str.replace('[', ' ['))  # an extra space is inserted
diff --git a/compat.py b/compat.py
index 001fbf9..1cbe923 100644
--- a/compat.py
+++ b/compat.py
@@ -120,3 +120,23 @@ def dbg(val=None):
 def nprint(*args, **kw):
     if dbg:
         print(*args, **kw)
+
+# char checkers following production rules
+
+
+def check_namespace_char(ch):
+    if u'\x21' <= ch <= u'\x7E':  # ! to ~
+        return True
+    if u'\xA0' <= ch <= u'\xD7FF':
+        return True
+    if (u'\xE000' <= ch <= u'\xFFFD') and ch != u'\xFEFF':  # excl. byte order mark
+        return True
+    if u'\x10000' <= ch <= u'\x10FFFF':
+        return True
+    return False
+
+
+def check_anchorname_char(ch):
+    if ch in u',[]{}':
+        return False
+    return check_namespace_char(ch)
diff --git a/emitter.py b/emitter.py
index 497f173..51ce3ac 100644
--- a/emitter.py
+++ b/emitter.py
@@ -12,7 +12,8 @@ from __future__ import print_function
 
 from ruamel.yaml.error import YAMLError
 from ruamel.yaml.events import *                                     # NOQA
-from ruamel.yaml.compat import utf8, text_type, PY2, nprint, dbg, DBG_EVENT
+from ruamel.yaml.compat import utf8, text_type, PY2, nprint, dbg, DBG_EVENT, \
+    check_anchorname_char
 
 __all__ = ['Emitter', 'EmitterError']
 
@@ -586,6 +587,9 @@ class Emitter(object):
             if (not self.flow_level and not self.simple_key_context and
                     self.analysis.allow_block):
                 return self.event.style
+        if not self.event.style and self.analysis.allow_double_quoted:
+            if "'" in self.event.value or '\n' in self.event.value:
+                return '"'
         if not self.event.style or self.event.style == '\'':
             if (self.analysis.allow_single_quoted and
                     not (self.simple_key_context and self.analysis.multiline)):
@@ -703,8 +707,7 @@ class Emitter(object):
         if not anchor:
             raise EmitterError("anchor must not be empty")
         for ch in anchor:
-            if not (u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or
-                    u'a' <= ch <= u'z' or ch in u'-_'):
+            if not check_anchorname_char(ch):
                 raise EmitterError("invalid character %r in the anchor: %r"
                                    % (utf8(ch), utf8(anchor)))
         return anchor
diff --git a/nodes.py b/nodes.py
index 03f158f..214284a 100644
--- a/nodes.py
+++ b/nodes.py
@@ -60,7 +60,7 @@ class ScalarNode(Node):
       " -> double quoted
       ' -> single quoted
       | -> literal style
-      > ->
+      > -> folding style
     """
     id = 'scalar'
 
diff --git a/representer.py b/representer.py
index 77a3f59..56890b3 100644
--- a/representer.py
+++ b/representer.py
@@ -97,7 +97,7 @@ class BaseRepresenter(object):
         """
         David Fraser: Extract a method to represent keys in mappings, so that
         a subclass can choose not to quote them (for example)
-        used in repesent_mapping
+        used in represent_mapping
         https://bitbucket.org/davidfraser/pyyaml/commits/d81df6eb95f20cac4a79eed95ae553b5c6f77b8c
         """
         return self.represent_data(data)
diff --git a/scanner.py b/scanner.py
index fbe190c..2a57c5d 100644
--- a/scanner.py
+++ b/scanner.py
@@ -32,7 +32,7 @@ from __future__ import print_function
 #
 from ruamel.yaml.error import MarkedYAMLError
 from ruamel.yaml.tokens import *                # NOQA
-from ruamel.yaml.compat import utf8, unichr, PY3
+from ruamel.yaml.compat import utf8, unichr, PY3, check_anchorname_char
 
 __all__ = ['Scanner', 'RoundTripScanner', 'ScannerError']
 
@@ -937,8 +937,9 @@ class Scanner(object):
         self.forward()
         length = 0
         ch = self.peek(length)
-        while u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z' \
-                or ch in u'-_':
+        # while u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z' \
+        #         or ch in u'-_':
+        while check_anchorname_char(ch):
             length += 1
             ch = self.peek(length)
         if not length:
@@ -948,8 +949,10 @@ class Scanner(object):
                 % utf8(ch), self.get_mark())
         value = self.prefix(length)
         self.forward(length)
-        ch = self.peek()
-        if ch not in u'\0 \t\r\n\x85\u2028\u2029?:,]}%@`':
+        # ch1 = ch
+        # ch = self.peek()   # no need to peek, ch is already set
+        # assert ch1 == ch
+        if ch not in u'\0 \t\r\n\x85\u2028\u2029?:,[]{}%@`':
             raise ScannerError(
                 "while scanning an %s" % name, start_mark,
                 "expected alphabetic or numeric character, but found %r"