[FIX] CREATE TASK wrongly identified as query (#42)

This fixes #41

README.md

@@ -36,7 +36,7 @@ repos:
 
 To exemplify the formatting let's say you have a SQL query like this
 
-```python
+```
 example_sql = """
 create or replace table mytable as -- mytable example
 seLecT a.asdf, b.qwer, -- some comment here
@@ -55,7 +55,7 @@ groUp by a.asdf
 
 Then you can use this package to format it so that it is better readable
 
-```python
+```
 from sql_formatter.core import format_sql
 print(format_sql(example_sql))
 ```
@@ -79,7 +79,7 @@ print(format_sql(example_sql))
 
 It can even deal with subqueries and it will correct my favourite simple careless mistake (comma at the end of SELECT statement before of FROM) for you on the flow :-)
 
-```python
+```
 print(format_sql("""
 select asdf, cast(qwer as numeric), -- some comment
 qwer1
@@ -109,7 +109,7 @@ where qwer1 >= 0
 
 The formatter is also robust against nested subqueries
 
-```python
+```
 print(format_sql("""
 select field1, field2 from (select field1, 
 field2 from (select field1, field2, 
@@ -131,7 +131,7 @@ field3 from table1 where a=1 and b>=100))
 
 If you do not want to get some query formatted in your SQL file then you can use the marker `/*skip-formatter*/` in your query to disable formatting for just the corresponding query
 
-```python
+```
 from sql_formatter.format_file import format_sql_commands
 print(format_sql_commands(
 """

docs/core.html

@@ -59,6 +59,8 @@ <h2 id="General-formatting">General formatting<a class="anchor-link" href="#Gene
     <div class="input_area">
 <div class=" highlight hl-ipython3"><pre><span></span><span class="n">example_sql</span> <span class="o">=</span> <span class="s2">&quot;&quot;&quot;</span>
 <span class="s2">create or replace table mytable as -- Mytable example</span>
+<span class="s2">/* multi line</span>
+<span class="s2">   comment */</span>
 <span class="s2">seLecT a.asdf, b.qwer, -- some comment here</span>
 <span class="s2">/* and here is a line comment inside select */</span>
 <span class="s2">substr(c.asdf, 1, 2) as substr_asdf, </span>
@@ -99,6 +101,8 @@ <h2 id="General-formatting">General formatting<a class="anchor-link" href="#Gene
 <div class="inner_cell">
     <div class="input_area">
 <div class=" highlight hl-ipython3"><pre><span></span><span class="n">expected_sql</span> <span class="o">=</span> <span class="s2">&quot;&quot;&quot;CREATE OR REPLACE TABLE mytable AS -- Mytable example</span>
+<span class="s2">/* multi line</span>
+<span class="s2">comment */</span>
 <span class="s2">SELECT a.asdf,</span>
 <span class="s2">       b.qwer, -- some comment here</span>
 <span class="s2">       /* and here is a line comment inside select */</span>
@@ -820,7 +824,8 @@ <h4 id="preformat_statements" class="doc_header"><code>preformat_statements</cod
 <div class="output_subarea output_stream output_stdout output_text">
 <pre>
 CREATE OR REPLACE TABLE mytable AS -- Mytable example[C]
-SELECT a.asdf, b.qwer, -- some comment here[C][CS]/* and here is a line comment inside select */[C]substr(c.asdf, 1, 2) as substr_asdf,[CS]/* some commenT there */[C]case when a.asdf= 1 then &#39;b&#39; /* here a case comment */[C]when b.qwer =2 then &#39;c&#39; else &#39;d&#39; end as new_field, -- Some comment[C]b.asdf2
+SELECT a.asdf, b.qwer, -- some comment here[C][CS]/* and here is a line comment inside select */[C]substr(c.asdf, 1, 2) as substr_asdf,[CS]/* some multi line
+commenT there */[C]case when a.asdf= 1 then &#39;b&#39; /* here a case comment */[C]when b.qwer =2 then &#39;c&#39; else &#39;d&#39; end as new_field, -- Some comment[C]b.asdf2
 FROM table1 as a
 LEFT JOIN table2 as b -- and here a comment[C]
 ON a.asdf = b.asdf /* joiN this way */[C]
@@ -2557,6 +2562,8 @@ <h4 id="format_simple_sql" class="doc_header"><code>format_simple_sql</code><a h
 
 <div class="output_subarea output_stream output_stdout output_text">
 <pre>CREATE OR REPLACE TABLE mytable AS -- Mytable example
+/* multi line
+comment */
 SELECT a.asdf,
        b.qwer, -- some comment here
        /* and here is a line comment inside select */
@@ -2732,7 +2739,7 @@ <h3 id="Main-function-handling-queries-with-subqueries">Main function handling q
 
 
 <div class="output_markdown rendered_html output_subarea ">
-<h4 id="format_sql" class="doc_header"><code>format_sql</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/core.py#L259" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>format_sql</code>(<strong><code>s</code></strong>, <strong><code>add_semicolon</code></strong>=<em><code>True</code></em>)</p>
+<h4 id="format_sql" class="doc_header"><code>format_sql</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/core.py#L260" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>format_sql</code>(<strong><code>s</code></strong>, <strong><code>add_semicolon</code></strong>=<em><code>True</code></em>)</p>
 </blockquote>
 <p>Format SQL query with subqueries. If <code>add_semicolon</code> is True then add a semicolon at the end</p>
 

docs/utils.html

@@ -692,7 +692,7 @@ <h4 id="mark_comments" class="doc_header"><code>mark_comments</code><a href="htt
 
 
 <div class="output_markdown rendered_html output_subarea ">
-<h4 id="identify_queries" class="doc_header"><code>identify_queries</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/utils.py#L254" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>identify_queries</code>(<strong><code>s</code></strong>)</p>
+<h4 id="identify_queries" class="doc_header"><code>identify_queries</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/utils.py#L255" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>identify_queries</code>(<strong><code>s</code></strong>)</p>
 </blockquote>
 <p>Split by queries in string <code>s</code></p>
 
@@ -1725,6 +1725,29 @@ <h4 id="check_sql_query" class="doc_header"><code>check_sql_query</code><a href=
 <div class="cell border-box-sizing code_cell rendered">
 <div class="input">
 
+<div class="inner_cell">
+    <div class="input_area">
+<div class=" highlight hl-ipython3"><pre><span></span><span class="k">assert</span> <span class="n">check_sql_query</span><span class="p">(</span><span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">--- Table 1---</span>
+<span class="s2">creaTe or replace view my_table as</span>
+<span class="s2">select asdf</span>
+<span class="s2">from table</span>
+<span class="s2">where asdf = 1</span>
+<span class="s2">&quot;&quot;&quot;</span><span class="o">.</span><span class="n">strip</span><span class="p">())</span>
+</pre></div>
+
+    </div>
+</div>
+</div>
+
+</div>
+    {% endraw %}
+
+    {% raw %}
+
+<div class="cell border-box-sizing code_cell rendered">
+<div class="input">
+
 <div class="inner_cell">
     <div class="input_area">
 <div class=" highlight hl-ipython3"><pre><span></span><span class="k">assert</span> <span class="n">check_sql_query</span><span class="p">(</span><span class="s2">&quot;&quot;&quot;</span>
@@ -1761,6 +1784,46 @@ <h4 id="check_sql_query" class="doc_header"><code>check_sql_query</code><a href=
 <div class="cell border-box-sizing code_cell rendered">
 <div class="input">
 
+<div class="inner_cell">
+    <div class="input_area">
+<div class=" highlight hl-ipython3"><pre><span></span><span class="k">assert</span> <span class="ow">not</span> <span class="n">check_sql_query</span><span class="p">(</span><span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">create or replace task my_task as</span>
+<span class="s2">&quot;&quot;&quot;</span><span class="p">)</span>
+</pre></div>
+
+    </div>
+</div>
+</div>
+
+</div>
+    {% endraw %}
+
+    {% raw %}
+
+<div class="cell border-box-sizing code_cell rendered">
+<div class="input">
+
+<div class="inner_cell">
+    <div class="input_area">
+<div class=" highlight hl-ipython3"><pre><span></span><span class="k">assert</span> <span class="ow">not</span> <span class="n">check_sql_query</span><span class="p">(</span><span class="s2">&quot;&quot;&quot;</span>
+<span class="s2">create or replace task my_task as</span>
+<span class="s2">create or replace table as</span>
+<span class="s2">select asdf</span>
+<span class="s2">&quot;&quot;&quot;</span><span class="p">)</span>
+</pre></div>
+
+    </div>
+</div>
+</div>
+
+</div>
+    {% endraw %}
+
+    {% raw %}
+
+<div class="cell border-box-sizing code_cell rendered">
+<div class="input">
+
 <div class="inner_cell">
     <div class="input_area">
 <div class=" highlight hl-ipython3"><pre><span></span><span class="k">assert</span> <span class="ow">not</span> <span class="n">check_sql_query</span><span class="p">(</span><span class="s2">&quot;&quot;&quot;</span>
@@ -1794,7 +1857,7 @@ <h4 id="check_sql_query" class="doc_header"><code>check_sql_query</code><a href=
 
 
 <div class="output_markdown rendered_html output_subarea ">
-<h4 id="check_skip_marker" class="doc_header"><code>check_skip_marker</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/utils.py#L249" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>check_skip_marker</code>(<strong><code>s</code></strong>)</p>
+<h4 id="check_skip_marker" class="doc_header"><code>check_skip_marker</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/utils.py#L250" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>check_skip_marker</code>(<strong><code>s</code></strong>)</p>
 </blockquote>
 <p>Checks whether user set marker /<em>skip-formatter</em>/ to not format query</p>
 
@@ -1878,7 +1941,7 @@ <h4 id="check_skip_marker" class="doc_header"><code>check_skip_marker</code><a h
 
 
 <div class="output_markdown rendered_html output_subarea ">
-<h4 id="identify_queries" class="doc_header"><code>identify_queries</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/utils.py#L254" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>identify_queries</code>(<strong><code>s</code></strong>)</p>
+<h4 id="identify_queries" class="doc_header"><code>identify_queries</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/utils.py#L255" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>identify_queries</code>(<strong><code>s</code></strong>)</p>
 </blockquote>
 <p>Split by queries in string <code>s</code></p>
 
@@ -1959,7 +2022,7 @@ <h4 id="identify_queries" class="doc_header"><code>identify_queries</code><a hre
 
 
 <div class="output_markdown rendered_html output_subarea ">
-<h4 id="split_by_query" class="doc_header"><code>split_by_query</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/utils.py#L283" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>split_by_query</code>(<strong><code>s</code></strong>)</p>
+<h4 id="split_by_query" class="doc_header"><code>split_by_query</code><a href="https://github.com/PabloRMira/sql_formatter/tree/master/sql_formatter/utils.py#L284" class="source_link" style="float:right">[source]</a></h4><blockquote><p><code>split_by_query</code>(<strong><code>s</code></strong>)</p>
 </blockquote>
 <p>Split string <code>s</code> by its queries</p>
 

nbs/00_core.ipynb

@@ -4,7 +4,16 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The autoreload extension is already loaded. To reload it, use:\n",
+      "  %reload_ext autoreload\n"
+     ]
+    }
+   ],
    "source": [
     "#hide\n",
     "%load_ext autoreload\n",
@@ -75,6 +84,8 @@
    "source": [
     "example_sql = \"\"\"\n",
     "create or replace table mytable as -- Mytable example\n",
+    "/* multi line\n",
+    "   comment */\n",
     "seLecT a.asdf, b.qwer, -- some comment here\n",
     "/* and here is a line comment inside select */\n",
     "substr(c.asdf, 1, 2) as substr_asdf, \n",
@@ -107,6 +118,8 @@
    "outputs": [],
    "source": [
     "expected_sql = \"\"\"CREATE OR REPLACE TABLE mytable AS -- Mytable example\n",
+    "/* multi line\n",
+    "comment */\n",
     "SELECT a.asdf,\n",
     "       b.qwer, -- some comment here\n",
     "       /* and here is a line comment inside select */\n",
@@ -660,7 +673,8 @@
      "text": [
       "\n",
       "CREATE OR REPLACE TABLE mytable AS -- Mytable example[C]\n",
-      "SELECT a.asdf, b.qwer, -- some comment here[C][CS]/* and here is a line comment inside select */[C]substr(c.asdf, 1, 2) as substr_asdf,[CS]/* some commenT there */[C]case when a.asdf= 1 then 'b' /* here a case comment */[C]when b.qwer =2 then 'c' else 'd' end as new_field, -- Some comment[C]b.asdf2\n",
+      "SELECT a.asdf, b.qwer, -- some comment here[C][CS]/* and here is a line comment inside select */[C]substr(c.asdf, 1, 2) as substr_asdf,[CS]/* some multi line\n",
+      "commenT there */[C]case when a.asdf= 1 then 'b' /* here a case comment */[C]when b.qwer =2 then 'c' else 'd' end as new_field, -- Some comment[C]b.asdf2\n",
       "FROM table1 as a\n",
       "LEFT JOIN table2 as b -- and here a comment[C]\n",
       "ON a.asdf = b.asdf /* joiN this way */[C]\n",
@@ -1846,7 +1860,8 @@
     "    s = lowercase_query(s)  # everything lowercased but not the comments\n",
     "    s = preformat_statements(s)  # add breaklines for the main statements\n",
     "    s = format_statements(s)  # format statements\n",
-    "    s = re.sub(r\"\\[C\\]\", \"\", s)  # replace remaining [EOC]\n",
+    "    s = re.sub(r\"\\[C\\]\", \"\", s)  # replace remaining [C]\n",
+    "    s = re.sub(r\"\\[CS\\]\", \"\\n\", s)  # replace remainig [CS]\n",
     "    if add_semicolon:\n",
     "        s = add_ending_semicolon(s)  # add ending semicolon if not there yet\n",
     "    return s"
@@ -1862,6 +1877,8 @@
      "output_type": "stream",
      "text": [
       "CREATE OR REPLACE TABLE mytable AS -- Mytable example\n",
+      "/* multi line\n",
+      "comment */\n",
       "SELECT a.asdf,\n",
       "       b.qwer, -- some comment here\n",
       "       /* and here is a line comment inside select */\n",

nbs/02_utils.ipynb

@@ -1191,7 +1191,8 @@
     "#export\n",
     "def check_sql_query(s):\n",
     "    \"Checks whether `s` is a SQL query\"\n",
-    "    return bool(re.match(pattern=r\".*(?:select|create).*\", string=s, flags=re.I | re.DOTALL))"
+    "    return (bool(re.search(pattern=r\".*(?:select|create.{0,10}(?:table|view)).*\", string=s, flags=re.I)) and\n",
+    "            not bool(re.search(pattern=r\"create(?!.*(?:table|view))\", string=s, flags=re.I)))"
    ]
   },
   {
@@ -1209,6 +1210,21 @@
     "\"\"\".strip())"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "assert check_sql_query(\"\"\"\n",
+    "--- Table 1---\n",
+    "creaTe or replace view my_table as\n",
+    "select asdf\n",
+    "from table\n",
+    "where asdf = 1\n",
+    "\"\"\".strip())"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -1229,6 +1245,30 @@
     "assert not check_sql_query(\"use database my_database;\")"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "assert not check_sql_query(\"\"\"\n",
+    "create or replace task my_task as\n",
+    "\"\"\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "assert not check_sql_query(\"\"\"\n",
+    "create or replace task my_task as\n",
+    "create or replace table as\n",
+    "select asdf\n",
+    "\"\"\")"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,

sql_formatter/core.py

@@ -250,7 +250,8 @@ def format_simple_sql(s, add_semicolon=True):
     s = lowercase_query(s)  # everything lowercased but not the comments
     s = preformat_statements(s)  # add breaklines for the main statements
     s = format_statements(s)  # format statements
-    s = re.sub(r"\[C\]", "", s)  # replace remaining [EOC]
+    s = re.sub(r"\[C\]", "", s)  # replace remaining [C]
+    s = re.sub(r"\[CS\]", "\n", s)  # replace remainig [CS]
     if add_semicolon:
         s = add_ending_semicolon(s)  # add ending semicolon if not there yet
     return s

sql_formatter/utils.py

@@ -243,7 +243,8 @@ def format_subquery(s, previous_s):
 # Cell
 def check_sql_query(s):
     "Checks whether `s` is a SQL query"
-    return bool(re.match(pattern=r".*(?:select|create).*", string=s, flags=re.I | re.DOTALL))
+    return (bool(re.search(pattern=r".*(?:select|create.{0,10}(?:table|view)).*", string=s, flags=re.I)) and
+            not bool(re.search(pattern=r"create(?!.*(?:table|view))", string=s, flags=re.I)))
 
 # Cell
 def check_skip_marker(s):