Add an internal Document representation

Doing everything on strings is kinda wonky, so this adds an intermediate
representation. The idea behind this is that the pipeline now goes

    Wikicode [1]-> Document [2]-> Output String

Where step 1 takes care of templates and everything, and step 2 does the
actual output formatting. This has the benefit that we can support
multiple output types, some with more and some with less features (e.g.,
adding a Markdown output which keeps some of the original formatting
intact), and it has the benefit of being less wonky (no hacks with
"<!NUM!>" for numbered lists, more streamlined formatting with newlines,
...).

6 files changed, 30 insertions, 24 deletions

diff --git a/wikimini/templates/cite.py b/wikimini/templates/cite.py
index ac4f597..e342a5f 100644
--- a/wikimini/templates/cite.py
+++ b/wikimini/templates/cite.py
@@ -1,5 +1,6 @@
 """Citation related templates."""
 from . import registry
+from ..document import Plain
 
 
 def tmpl_citation(wikimini, obj):
@@ -23,7 +24,7 @@ def tmpl_citation(wikimini, obj):
             names.append(last)
         elif first:
             names.append(first)
-    return "{} ({})".format(title, "; ".join(names))
+    return [Plain("{} ({})".format(title, "; ".join(names)))]
 
 
 for name in ["cite", "citation", "cite arXiv", "cite AV media", "cite book",
diff --git a/wikimini/templates/convert.py b/wikimini/templates/convert.py
index a7a3f44..8bab782 100644
--- a/wikimini/templates/convert.py
+++ b/wikimini/templates/convert.py
@@ -1,20 +1,21 @@
 """Implementations for the unit conversion templates."""
 from . import registry
+from ..document import Plain
 
 
 def tmpl_convert(wikimini, obj):
     """Renders the ``{{convert|...}}`` template."""
     if str(obj.params[1]) in {"-", "to"}:
-        return "{0}{3} {1} {2}{3}".format(
+        return [Plain("{0}{3} {1} {2}{3}".format(
             obj.params[0].value.strip_code(),
             obj.params[1].value.strip_code(),
             obj.params[2].value.strip_code(),
             obj.params[3].value.strip_code(),
-        )
-    return "{}{}".format(
+        ))]
+    return [Plain("{}{}".format(
         obj.params[0].value.strip_code(),
         obj.params[1].value.strip_code(),
-    )
+    ))]
 
 
 registry.insert("convert", tmpl_convert)
diff --git a/wikimini/templates/language.py b/wikimini/templates/language.py
index e8ab738..22320da 100644
--- a/wikimini/templates/language.py
+++ b/wikimini/templates/language.py
@@ -1,12 +1,15 @@
 """Language related templates."""
 from . import registry
+from ..document import Plain
 
 import pycountry
 
 
 def tmpl_ipa(wikimini, obj):
     """Renders the ``{{IPA|...}}`` template."""
-    return "pronounced [{}]".format(wikimini._convert(obj.params[0].value))
+    return [Plain("pronounced [{}]".format(
+        wikimini.convert(obj.params[0].value).plain()
+    ))]
 
 
 registry.insert("IPA", tmpl_ipa)
@@ -14,7 +17,7 @@ registry.insert("IPA", tmpl_ipa)
 
 def tmpl_lang(wikimini, obj):
     """Renders the ``{{Lang|...}}`` template."""
-    return wikimini._convert(obj.params[1].value)
+    return wikimini.convert(obj.params[1].value).nodes()
 
 
 registry.insert("lang", tmpl_lang)
@@ -24,18 +27,18 @@ registry.insert("script", tmpl_lang)
 def tmpl_lang_code(language_name):
     """Creates a template renderer for a ``{{lang-xx|...}}`` template."""
     def inner(wikimini, obj):
-        return "{}: {}".format(
-            language_name, wikimini._convert(obj.params[0].value)
-        )
+        return [Plain("{}: {}".format(
+            language_name, wikimini.convert(obj.params[0].value).plain()
+        ))]
     return inner
 
 
 def tmpl_ipa_code(language_name):
     """Creates a template renderer for a ``{{IPA-xx|...}}`` template."""
     def inner(wikimini, obj):
-        return "{} pronunciation: [{}]".format(
-            language_name, wikimini._convert(obj.params[0].value)
-        )
+        return [Plain("{} pronunciation: [{}]".format(
+            language_name, wikimini.convert(obj.params[0].value).plain()
+        ))]
     return inner
 
 
@@ -52,7 +55,7 @@ for language in pycountry.languages:
 def tmpl_country_flag(country):
     """Creates a template renderer for ``{{BRA}}`` country flags."""
     def inner(wikimini, obj):
-        return country
+        return [Plain(country)]
     return inner
 
 
diff --git a/wikimini/templates/mainlinks.py b/wikimini/templates/mainlinks.py
index ffcbc5e..8c9abbf 100644
--- a/wikimini/templates/mainlinks.py
+++ b/wikimini/templates/mainlinks.py
@@ -1,14 +1,15 @@
 """Renders templates that link to further articles."""
 from . import registry
 
+from ..document import Paragraph, Plain, BlockLink
 
 def tmpl_main(wikimini, obj):
     """Renders the ``{{main|...}}`` template."""
     links = [
-        "=> {} {}".format(wikimini.page_url(str(t.value)), t.value)
+        BlockLink(wikimini.page_url(str(t.value)), t.value.strip_code())
         for t in obj.params
     ]
-    return "Main articles:\n{}\n".format("\n".join(links))
+    return [Paragraph([Plain("Main articles:")])] + links
 
 
 registry.insert("main", tmpl_main)
diff --git a/wikimini/templates/quotes.py b/wikimini/templates/quotes.py
index b51d92d..39f6fa2 100644
--- a/wikimini/templates/quotes.py
+++ b/wikimini/templates/quotes.py
@@ -1,15 +1,15 @@
 """Renders various quote related templates."""
 from . import registry
 
+from ..document import Blockquote
 
 def tmpl_quote(wikimini, obj):
     """Renders the ``{{blockquote|...}}`` template."""
     text = obj.get("text", None)
     if not text:
         return ""
-    content = wikimini._convert(text.value)
-    lines = content.split("\n")
-    return "\n".join(f"> {line}" for line in lines)
+    content = wikimini.convert(text.value).nodes()
+    return [Blockquote(content)]
 
 
 registry.insert("blockquote", tmpl_quote)
@@ -19,9 +19,8 @@ registry.insert("quote", tmpl_quote)
 def tmpl_cquote(wikimini, obj):
     """Renders the ``{{cquote|...}}`` template."""
     text = obj.params[0]
-    content = wikimini._convert(text.value)
-    lines = content.split("\n")
-    return "\n".join(f"> {line}" for line in lines)
+    content = wikimini.convert(text.value).nodes()
+    return [Blockquote(content)]
 
 
 registry.insert("cquote", tmpl_cquote)
diff --git a/wikimini/templates/various.py b/wikimini/templates/various.py
index 8c6e0d5..6076ac4 100644
--- a/wikimini/templates/various.py
+++ b/wikimini/templates/various.py
@@ -1,16 +1,17 @@
 """Various small templates."""
 from . import registry
+from ..document import Plain
 
 
 def tmpl_reign(wikimini, obj):
     """Renders the ``{{reign|...}}`` template."""
     if not obj.params:
-        return "r. "
+        return [Plain("r. ")]
     first = obj.params[0].value.strip_code().strip() or "?"
     second = ""
     if len(obj.params) > 1:
         second = obj.params[1].value.strip_code().strip()
-    return f"r. {first} – {second}"
+    return [Plain(f"r. {first} – {second}")]
 
 
 registry.insert("reign", tmpl_reign)