fix issue with arrays

Author: giannisdoukas

.gitignore

@@ -246,3 +246,4 @@ tmp.py
 /html/
 cwlbuild
 /tests/repo-like/result.yaml
+/tests/repo-like/messages.txt

ipython2cwl/cwltoolextractor.py

@@ -55,7 +55,10 @@ def visit_AnnAssign(self, node):
         try:
             if (isinstance(node.annotation, ast.Name) and node.annotation.id in self.input_type_mapper) or \
                     (isinstance(node.annotation, ast.Str) and node.annotation.s in self.input_type_mapper):
-                mapper = self.input_type_mapper[node.annotation.id]
+                if hasattr(node.annotation, 'id'):
+                    mapper = self.input_type_mapper[node.annotation.id]
+                else:
+                    mapper = self.input_type_mapper[node.annotation.s]
                 self.extracted_nodes.append(
                     (node, mapper[0], mapper[1], True, True, False)
                 )
@@ -152,6 +155,7 @@ def from_jupyter_notebook_node(cls, node: NotebookNode) -> 'AnnotatedIPython2CWL
 
     @classmethod
     def _wrap_script_to_method(cls, tree, variables) -> str:
+        add_args = cls.__get_add_arguments__([v for v in variables if v.is_input])
         main_template_code = os.linesep.join([
             f"def main({','.join([v.name for v in variables if v.is_input])}):",
             "\tpass",
@@ -160,19 +164,31 @@ def _wrap_script_to_method(cls, tree, variables) -> str:
                 "import argparse",
                 'import pathlib',
                 "parser = argparse.ArgumentParser()",
-                *[f'parser.add_argument("--{variable.name}", '
-                  f'type={variable.argparse_typeof}, '
-                  f'required={variable.required})'
-                  for variable in variables],
+                *add_args,
                 "args = parser.parse_args()",
-                f"main({','.join([f'{v.name}=args.{v.name}' for v in variables if v.is_input])})"
+                f"main({','.join([f'{v.name}=args.{v.name} ' for v in variables if v.is_input])})"
             ]],
         ])
         main_function = ast.parse(main_template_code)
         [node for node in main_function.body if isinstance(node, ast.FunctionDef) and node.name == 'main'][0] \
             .body = tree.body
         return astor.to_source(main_function)
 
+    @classmethod
+    def __get_add_arguments__(cls, variables):
+        args = []
+        for variable in variables:
+            is_array = variable.cwl_typeof.endswith('[]')
+            arg: str = f'parser.add_argument("--{variable.name}", '
+            arg += f'type={variable.argparse_typeof}, '
+            arg += f'required={variable.required}, '
+            if is_array:
+                arg += f'nargs="+", '
+            arg = arg.strip()
+            arg += ')'
+            args.append(arg)
+        return args
+
     def cwl_command_line_tool(self, docker_image_id: str = 'jn2cwl:latest') -> Dict:
         """
         Creates the description of the CWL Command Line Tool.

tests/repo-like/example1.ipynb

@@ -2,30 +2,24 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
-    "from ipython2cwl.iotypes import CWLFilePathInput, CWLFilePathOutput\n",
-    "import yaml"
+    "from ipython2cwl.iotypes import CWLFilePathInput, CWLStringInput, CWLFilePathOutput\n",
+    "from typing import List\n",
+    "import yaml\n",
+    "import os"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "{'entry1': 1, 'entry2': 'foo', 'entry3': 'bar'}\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "datafilename: CWLFilePathInput = 'data.yaml'\n",
-    "\n",
+    "messages: List[CWLStringInput] = ['hello', 'world']\n",
     "with open(datafilename) as fd: \n",
     "    data = yaml.safe_load(fd)\n",
     "print(data)"
@@ -42,13 +36,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
     "results_filename: CWLFilePathOutput = 'result.yaml'\n",
     "with open(results_filename, 'w') as fd:\n",
-    "    yaml.safe_dump(data, fd)"
+    "    yaml.safe_dump(data, fd)\n",
+    "messages_outputs: CWLFilePathOutput = 'messages.txt'\n",
+    "with open(messages_outputs, 'w') as f:\n",
+    "    f.write(' '.join(messages))"
    ]
   }
  ],

tests/repo-like/requirements.txt

@@ -1,2 +1 @@
-PyYAML==5.3.1
-ipython2cwl==0.0.1
+PyYAML==5.3.1

tests/simple.ipynb

@@ -2,27 +2,29 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
     "import pandas as pd\n",
+    "from typing import List\n",
     "import matplotlib\n",
-    "from ipython2cwl.iotypes import CWLFilePathInput, CWLFilePathOutput"
+    "from ipython2cwl.iotypes import CWLFilePathInput, CWLFilePathOutput, CWLStringInput"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
-    "dataset: CWLFilePathInput = 'example.csv'"
+    "dataset: CWLFilePathInput = './data/data.csv'\n",
+    "messages: List[CWLStringInput] = ['hello', 'world']"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -36,17 +38,28 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
     "# transform data\n",
-    "data.sort_values(by='Random B', ascending=False, inplace=True, ignore_index=True)\n",
+    "data.sort_values(by='Y', ascending=False, inplace=True, ignore_index=True)\n",
     "fig = data.plot()\n",
     "\n",
     "after_transform_data: CWLFilePathOutput = 'new_data.png'\n",
     "fig.figure.savefig(after_transform_data)"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "messages_filename = 'messages.txt'\n",
+    "with open(messages_filename, 'w') as f:\n",
+    "    f.write(' '.join(messages))"
+   ]
   }
  ],
  "metadata": {

tests/test_ipython2cwl_from_repo.py

@@ -1,3 +1,4 @@
+import ast
 import os
 import shutil
 import tempfile
@@ -39,6 +40,15 @@ def test_docker_build(self):
         docker_client = docker.from_env()
         script = docker_client.containers.run(dockerfile_image_id, '/app/cwl/bin/simple', entrypoint='/bin/cat')
         self.assertIn('fig.figure.savefig(after_transform_data)', script.decode())
+        messages_array_arg_line = ast.parse([line.strip() for line in script.decode().splitlines() if '--messages' in line][-1])
+        self.assertEqual(
+            '+',  # nargs = '+'
+            [k.value.s for k in messages_array_arg_line.body[0].value.keywords if k.arg == 'nargs'][0]
+        )
+        self.assertEqual(
+            'str',  # type = 'str'
+            [k.value.id for k in messages_array_arg_line.body[0].value.keywords if k.arg == 'type'][0]
+        )
         self.assertDictEqual(
             {
                 'cwlVersion': "v1.1",
@@ -54,6 +64,12 @@ def test_docker_build(self):
                         'inputBinding': {
                             'prefix': '--dataset'
                         }
+                    },
+                    'messages': {
+                        'type': 'string[]',
+                        'inputBinding': {
+                            'prefix': '--messages'
+                        }
                     }
                 },
                 'outputs': {

tests/test_system_tests.py

@@ -43,10 +43,17 @@ def test_repo2cwl(self):
 
         example1_tool = fac.make(os.path.join(output_dir, 'example1.cwl'))
         result = example1_tool(
-            datafilename={'class': 'File', 'location': os.path.join(self.repo_like_dir, 'data.yaml')})
+            datafilename={
+                'class': 'File', 'location': os.path.join(self.repo_like_dir, 'data.yaml')
+            },
+            messages=["hello", "test", "!!!"]
+        )
         with open(result['results_filename']['location'][7:]) as f:
             new_data = yaml.safe_load(f)
         self.assertDictEqual({'entry1': 2, 'entry2': 'foo', 'entry3': 'bar'}, new_data)
+        with open(result['messages_outputs']['location'][7:]) as f:
+            message = f.read()
+        self.assertEqual("hello test !!!", message)
         shutil.rmtree(output_dir)
 
     def test_repo2cwl_output_dir_does_not_exists(self):