new proposal for doctests

website/Makefile

@@ -1,12 +1,13 @@
-all: dir site
+all: src/code site
 
-dir:
+src/code:
-	mkdir -p site
+	mkdir -p src/code/
+	./create_code_samples tests/ src/code/
 
 site: site/index.html site/blog/ site/docs/ site/license/ site/blog/introducing-spacy/ site/blog/parsing-english-in-python/ site/blog/part-of-speech-POS-tagger-in-python/ site/tutorials/twitter-filter/ site/tutorials/syntax-search/ site/tutorials/mark-adverbs/ site/blog/writing-c-in-cython/ site/blog/how-spacy-works/
 
 site/index.html: src/jade/header.jade src/jade/*.jade
-	./run_jade src/jade/home/index.jade $@
+	jade -P src/jade/home/index.jade --out site/
 
 site/docs/: src/jade/docs/*.jade src/jade/header.jade
 	jade -P src/jade/docs/index.jade --out $@

website/create_code_samples

@@ -0,0 +1,54 @@
+#!/usr/bin/env python
+import sys
+import re
+import os
+import ast
+
+
+src_dirname = sys.argv[1]
+dst_dirname = sys.argv[2]
+prefix = "test_"
+
+
+for filename in os.listdir(src_dirname):
+    match = re.match(re.escape(prefix) + r"(.+)\.py", filename)
+    if not match:
+        continue
+
+    name = match.group(1)
+    source = open(os.path.join(src_dirname, filename)).readlines()
+    tree = ast.parse("".join(source))
+
+    for item in tree.body:
+        if isinstance(item, ast.FunctionDef) and item.name.startswith(prefix):
+
+            # only ast.expr and ast.stmt have line numbers, see:
+            # https://docs.python.org/2/library/ast.html#ast.AST.lineno
+            line_numbers = [x.lineno for x in ast.iter_child_nodes(item)
+                            if isinstance(x, ast.expr) or
+                               isinstance(x, ast.stmt)]
+
+            body = source[min(line_numbers)-1:max(line_numbers)]
+
+            # make sure we are inside an indented function body
+            assert all([re.match(r"\s", l[0]) for l in body])
+
+            offset = 0
+            for line in body:
+                match = re.search(r"[^\s]", line)
+                if match:
+                    offset = match.start(0)
+
+            # remove indentation
+            assert offset > 0
+
+            for i in range(len(body)):
+                body[i] = body[i][offset:] if len(body[i]) > offset else "\n"
+
+            # make sure empty lines contain a newline
+            assert all([l[-1] == "\n" for l in body])
+
+            code_filename = "%s.%s" % (name, item.name[len(prefix):])
+
+            with open(os.path.join(dst_dirname, code_filename), "w") as f:
+                f.write("".join(body))

website/run_jade

@@ -1,59 +0,0 @@
-#!/usr/bin/env node
-'use strict';
-
-var fs = require('fs');
-var jade = require('jade');
-
-// returns all: code and return value (default)
-jade.filters.doctest_all = function (html, _, use_rv) {
-    use_rv = use_rv === undefined ? true : use_rv;
-
-    var lines = html.trim().split(/\n/),
-        block = [],
-        res = '';
-
-    lines.forEach(function (line) {
-        if (line.indexOf('>>> ') === 0) {
-            // we use ### to suppress lines
-            if (line.indexOf("###") === -1) {
-                block.push(line.replace(/^>>> /gm, ''));
-            }
-        } else if (block.length > 0) {
-            res += '<pre class="language-python"><code>' + block.join('\n') + '</code></pre>';
-            block = [];
-
-            if (use_rv) {
-                res += '<p>Which produces:</p>';
-                res += '<pre class="language-python"><code>' + line + '</code></pre>';
-            }
-        }
-    });
-
-    if (block.length > 0) {
-        res += '<pre class="language-python"><code>' + block.join('\n') + '</code></pre>';
-    }
-
-    return res;
-};
-
-// returns only code
-jade.filters.doctest = function (html) {
-    return jade.filters.doctest_all(html, null, false);
-};
-
-if (process.argv[0] === "node") {
-    process.argv.shift();
-}
-
-var in_file = process.argv[1];
-var out_file = process.argv[2];
-
-var html = jade.renderFile(in_file, {
-    pretty: true
-});
-
-fs.writeFile(out_file, html, function (err) {
-    if (err) {
-        throw err;
-    }
-});

website/tests/test_home.py

@@ -0,0 +1,40 @@
+from __future__ import unicode_literals
+import pytest
+
+
+@pytest.fixture(scope="session")
+def nlp():
+    from spacy.en import English
+    return English()
+
+
+@pytest.fixture()
+def doc(nlp):
+    return nlp('Hello, world. Here are two sentences.')
+
+
+@pytest.fixture()
+def token(doc):
+    return doc[0]
+
+
+def test_load_resources_and_process_text():
+    from spacy.en import English
+    nlp = English()
+    doc = nlp('Hello, world. Here are two sentences.')
+
+
+def test_get_tokens_and_sentences(doc):
+    token = doc[0]
+    sentence = doc.sents.next()
+
+    assert token is sentence[0]
+    assert sentence.text == 'Hello, world.'
+
+
+def test_use_integer_ids_for_any_strings(nlp, token):
+    hello_id = nlp.vocab.strings['Hello']
+    hello_str = nlp.vocab.strings[hello_id]
+
+    assert token.orth  == hello_id  == 3404
+    assert token.orth_ == hello_str == 'Hello'