Add a way to disable the syntax detector (fixes #27)

Before this commit, there was no way to explicitly disable the
automatic syntax detector. That is useful, for example, if it screws up
and you don't want any syntax in your messages.

This commit adds a special `plain` syntax which forces plain text.

botogram/syntaxes.py

@@ -55,7 +55,9 @@ def guess_syntax(message, provided):
         else:
             return None
 
-    if provided in ("md", "markdown", "Markdown"):
+    if provided in ("plain",):
+        return None
+    elif provided in ("md", "markdown", "Markdown"):
         return "Markdown"
     elif provided in ("html", "HTML"):
         return "HTML"

docs/changelog.rst

@@ -25,8 +25,10 @@ botogram 0.1.2
 
 *Bugfix release, not yet released*
 
+* Add a way to disable the syntax detector (`issue 27`_)
 * Fix automatic syntax detector recognizing markdown in URLs (`issue 28`_)
 
+.. _issue 27: https://github.com/pietroalbini/botogram/issues/27
 .. _issue 28: https://github.com/pietroalbini/botogram/issues/28
 
 .. _changelog-0.1.1:

docs/tricks.rst

@@ -71,6 +71,13 @@ That parameter accepts the following values:
 * ``markdown``, or its aliases ``md`` and ``Markdown``
 * ``html``, or its alias ``HTML``
 
+Also, if you don't want to use any rich formatting but the detector spots
+something, you can disable it providing the special syntax ``plain`` to it:
+
+.. code-block:: python
+
+   chat.send("*I don't want this to be detected*", syntax="plain")
+
 .. note::
 
    Support for rich formatting depends on your users' Telegram client. If

tests/test_syntaxes.py

@@ -41,6 +41,9 @@ def test_is_html():
 
 def test_guess_syntax():
     # Provided syntax name
+    for name in ("plain",):
+        assert botogram.syntaxes.guess_syntax("", name) is None
+
     for name in ("md", "markdown", "Markdown"):
         assert botogram.syntaxes.guess_syntax("", name) == "Markdown"