Unescape link names

See issue #3

lammpsdoc/txt2html.py

@@ -655,7 +655,7 @@ class TxtConverter:
     def get_output_filename(self, path):
         return ""
 
-    def create_converter(self, args):
+    def create_converter(self, args, filename='default.txt'):
         return None
 
     def run(self, args=sys.argv[1:], out=sys.stdout, err=sys.stderr):
@@ -704,7 +704,7 @@ class Txt2HtmlConverter(TxtConverter):
         parser.add_argument('files',  metavar='file', nargs='+', help='one or more files to convert')
         return parser
 
-    def create_converter(self, args):
+    def create_converter(self, args, filename='default.txt'):
         converter = Txt2Html()
         converter.append_page_break = args.breakflag
         converter.create_title = args.create_title

lammpsdoc/txt2rst.py

@@ -23,11 +23,25 @@ import re
 import argparse
 from lammpsdoc import lammps_filters
 from lammpsdoc.txt2html import Markup, Formatting, TxtParser, TxtConverter
+from collections import OrderedDict
+
+
+def simplify_anchor(anchor_url):
+    """ Remove repeating words, but leave repeating numbers """
+    def unique_words(s):
+        parts = s.split('_')
+        last_word = ""
+        for word in parts:
+            if word.isdecimal() or word != last_word:
+                yield word
+                last_word = word
+    return '_'.join(unique_words(anchor_url))
 
 
 class RSTMarkup(Markup):
-    def __init__(self):
+    def __init__(self, filename='default.rst'):
         super().__init__()
+        self.filename = os.path.basename(filename)
 
     def bold_start(self):
         return "**"
@@ -104,11 +118,20 @@ class RSTMarkup(Markup):
 
         anchor_pos = href.find('#')
 
-        if anchor_pos >= 0 and not href.startswith('http'):
-            href = href[anchor_pos+1:]
+        if anchor_pos > 0 and not href.startswith('http'):
+            filename = os.path.splitext(href[0:anchor_pos])[0]
+            href = href[anchor_pos + 1:]
+            href = simplify_anchor(filename + "_" + href)
+            return ":ref:`%s <%s>`" % (content, href)
+        elif anchor_pos == 0 and not href.startswith('http'):
+            filename = os.path.splitext(self.filename)[0]
+            href = href[anchor_pos + 1:]
+            href = simplify_anchor(filename + "_" + href)
             return ":ref:`%s <%s>`" % (content, href)
 
         if href in self.references:
+            filename = os.path.splitext(self.filename)[0]
+            href = simplify_anchor(filename + "_" + href)
             return ":ref:`%s <%s>`" % (content, href)
         elif href in self.aliases:
             href = "%s_" % href
@@ -122,8 +145,9 @@ class RSTMarkup(Markup):
 class RSTFormatting(Formatting):
     RST_HEADER_TYPES = '#*=-^"'
 
-    def __init__(self, markup):
+    def __init__(self, markup, filename = 'default.rst'):
         super().__init__(markup)
+        self.filename = os.path.basename(filename)
         self.indent_level = 0
 
     def paragraph(self, content):
@@ -165,7 +189,9 @@ class RSTFormatting(Formatting):
 
     def named_link(self, paragraph, name):
         self.markup.add_internal_reference(name)
-        return (".. _%s:\n\n" % name) + paragraph
+        filename = os.path.splitext(self.filename)[0]
+        name = simplify_anchor(filename + "_" + name)
+        return (".. _%s:\n\n" % self.markup.unescape_rst_chars(name)) + paragraph
 
     def define_link_alias(self, paragraph, alias, value):
         self.markup.add_link_alias(alias, value)
@@ -347,10 +373,10 @@ class RSTFormatting(Formatting):
 
 
 class Txt2Rst(TxtParser):
-    def __init__(self):
+    def __init__(self, filename='default.rst'):
         super().__init__()
-        self.markup = RSTMarkup()
-        self.format = RSTFormatting(self.markup)
+        self.markup = RSTMarkup(filename=filename)
+        self.format = RSTFormatting(self.markup,filename=filename)
         self.register_filters()
 
     def register_filters(self):
@@ -394,8 +420,8 @@ class Txt2RstConverter(TxtConverter):
         parser.add_argument('files',  metavar='file', nargs='+', help='one or more files to convert')
         return parser
 
-    def create_converter(self, args):
-        return Txt2Rst()
+    def create_converter(self, args, filename='default.txt'):
+        return Txt2Rst(filename=self.get_output_filename(filename))
 
     def get_output_filename(self, path):
         filename, ext = os.path.splitext(path)

tests/test_txt2rst.py

@@ -411,18 +411,26 @@ class TestSpecialCommands(unittest.TestCase):
     def test_internal_reference_link(self):
         s = self.txt2rst.convert("one :link(name)\n"
                                   "a \"link\"_name to above\n")
-        self.assertEqual(".. _name:\n"
+        self.assertEqual(".. _default_name:\n"
                          "\n"
                          "one \n\n"
-                         "a :ref:`link <name>` to above\n\n", s)
+                         "a :ref:`link <default_name>` to above\n\n", s)
 
     def test_local_anchor_link(self):
         s = self.txt2rst.convert("one :link(name)\n"
                                   "a \"link\"_#name to above\n")
-        self.assertEqual(".. _name:\n"
+        self.assertEqual(".. _default_name:\n"
                          "\n"
                          "one \n\n"
-                         "a :ref:`link <name>` to above\n\n", s)
+                         "a :ref:`link <default_name>` to above\n\n", s)
+
+    def test_repeating_words_in_anchor_link(self):
+        s = self.txt2rst.convert("one :link(name_name_2_2)\n"
+                                  "a \"link\"_#name_name_2_2 to above\n")
+        self.assertEqual(".. _default_name_2_2:\n"
+                         "\n"
+                         "one \n\n"
+                         "a :ref:`link <default_name_2_2>` to above\n\n", s)
 
     def test_external_anchor_link(self):
         s = self.txt2rst.convert('some text "containing a\n'