Merge branch 'wb/t32' into wb/t10

Author: peterbjohnson

in2lambda/api/question.py

@@ -178,3 +178,33 @@ def add_part_text(self, elem: Any) -> None:
             self.parts[self._last_part["text"]].text = elem_text
 
         self._last_part["text"] += 1
+
+    def to_json(self, output_dir: str, number: int = 0) -> None:
+        """Turns this question alone into Lambda Feedback JSON/ZIP files.
+
+        This is what Lambda Feedback takes when importing a single question into a set
+        that already exists: the question's JSON and its images under ``media``, with
+        no set file. Images keep the names the JSON refers to them by.
+
+        Files of the same name are overwritten; nothing else in the directory is
+        touched.
+
+        Args:
+            output_dir: Where to output the final Lambda Feedback JSON/ZIP files.
+            number: The question's order number, which also prefixes its file names.
+
+        Examples:
+            >>> import os
+            >>> import tempfile
+            >>> from in2lambda.api.question import Question
+            >>> question = Question(title="Q", main_text="Some text")
+            >>> with tempfile.TemporaryDirectory() as temp_dir:
+            ...     question.to_json(temp_dir)
+            ...     sorted(os.listdir(temp_dir))
+            ...     sorted(os.listdir(f"{temp_dir}/question_000_Q"))
+            ['question_000_Q', 'question_000_Q.zip']
+            ['question_000_Q.json']
+        """
+        from in2lambda.json_convert import json_convert
+
+        json_convert.write_question(self, output_dir, number)

in2lambda/api/set.py

@@ -105,7 +105,8 @@ def increment_current_question(self) -> None:
     def to_json(self, output_dir: str) -> None:
         """Turns this set into Lambda Feedback JSON/ZIP files.
 
-        WARNING: This will overwrite any existing files in the directory.
+        Files of the same name are overwritten; nothing else in the directory is
+        touched, and only what was written goes into the zip.
 
         Args:
             output_dir: Where to output the final Lambda Feedback JSON/ZIP files.

in2lambda/json_convert/json_convert.py

@@ -20,20 +20,40 @@
 MINIMAL_SET_TEMPLATE = "minimal_template_set.json"
 
 
-def _zip_sorted_folder(folder_path, zip_path):
-    """Zips the contents of a folder, preserving the directory structure.
+def _templates() -> tuple[dict[str, Any], dict[str, Any]]:
+    """Loads the minimal question and set templates that the writer fills in.
+
+    Returns:
+        The question template and the set template.
+    """
+    # Use path so minimal template can be found regardless of where the user is running python from.
+    with open(Path(__file__).with_name(MINIMAL_QUESTION_TEMPLATE), "r") as file:
+        question_template = json.load(file)
+
+    with open(Path(__file__).with_name(MINIMAL_SET_TEMPLATE), "r") as file:
+        set_template = json.load(file)
+
+    return question_template, set_template
+
+
+def _zip(files: list[Path], root: Path, zip_path: str) -> None:
+    """Zips the given files, keeping where they sit relative to a folder.
+
+    Only what this run wrote is listed, so whatever else the folder holds is neither
+    uploaded nor removed.
 
     Args:
-        folder_path: The path to the folder to zip.
+        files: The files to include, all inside root.
+        root: The folder the archive names are relative to.
         zip_path: The path where the zip file will be created.
     """
+    # Sort by archive name for deterministic, alphabetical order. A file can be
+    # written more than once — an image used by both a question and its worked
+    # solution — and is still one file on disk, so name it once here too.
+    names = sorted({str(file.relative_to(root)): file for file in files}.items())
     with zipfile.ZipFile(zip_path, "w") as zf:
-        for root, dirs, files in os.walk(folder_path):
-            # Sort files for deterministic, alphabetical order
-            for file in sorted(files):
-                abs_path = os.path.join(root, file)
-                rel_path = os.path.relpath(abs_path, folder_path)
-                zf.write(abs_path, arcname=rel_path)
+        for name, file in names:
+            zf.write(file, arcname=name)
 
 
 def _response_area_to_json(area: ResponseArea, order: int) -> dict[str, Any]:
@@ -156,13 +176,29 @@ def _part_to_json(
     return output
 
 
+def _question_title(question: Question, i: int) -> str:
+    return question.title if question.title != "" else f"Question {i + 1}"
+
+
+def _question_stem(i: int, title: str) -> str:
+    # Lambda Feedback names the file after the title with only spaces made
+    # underscores. Path separators go too, so a title cannot leave the set folder,
+    # and so do the characters Windows forbids in file names.
+    return (
+        "question_"
+        + str(i).zfill(3)
+        + "_"
+        + re.sub(r'[\s/\\<>:"|?*]', "_", title.strip())
+    )
+
+
 def _question_json(
     question: Question, i: int, template: dict[str, Any]
 ) -> dict[str, Any]:
     output = deepcopy(template)
 
     output["orderNumber"] = i  # order number starts at 0
-    output["title"] = question.title if question.title != "" else f"Question {i + 1}"
+    output["title"] = _question_title(question, i)
     output["masterContent"] = question.main_text
 
     output["publish"] = question.publish
@@ -190,6 +226,59 @@ def _question_json(
     return output
 
 
+def _write_question(
+    question: Question, i: int, template: dict[str, Any], folder: Path
+) -> list[Path]:
+    """Writes one question's JSON, and any images it uses, into an existing folder.
+
+    Args:
+        question: The question to write.
+        i: Its order number, which also prefixes the file name.
+        template: The loaded JSON from the minimal question template.
+        folder: The folder to write into.
+
+    Returns:
+        The files written.
+    """
+    output = _question_json(question, i, template)
+
+    json_file = folder / f"{_question_stem(i, output['title'])}.json"
+    with open(json_file, "w") as file:
+        json.dump(output, file)
+    written = [json_file]
+
+    for image in question.images:
+        # If images exist, create a media directory
+        media = folder / "media"
+        media.mkdir(exist_ok=True)
+        # The JSON refers to an image by its file name, so copying keeps that name.
+        written.append(Path(shutil.copy(os.path.abspath(image), media)))
+
+    return written
+
+
+def write_question(question: Question, output_dir: str, number: int = 0) -> None:
+    """Writes a single question as its own Lambda Feedback import.
+
+    The question gets a folder named after it, holding its JSON and its images under
+    ``media``, and a zip of that folder. There is no set file: this is what Lambda
+    Feedback takes when importing one question into a set that already exists.
+
+    Args:
+        question: The question to write.
+        output_dir: Where to put the question's folder and zip.
+        number: The question's order number, which also prefixes its file names.
+    """
+    question_template, _ = _templates()
+
+    folder = Path(output_dir) / _question_stem(
+        number, _question_title(question, number)
+    )
+    folder.mkdir(parents=True, exist_ok=True)
+    written = _write_question(question, number, question_template, folder)
+    _zip(written, folder, f"{folder}.zip")
+
+
 def converter(
     question_template: dict[str, Any],
     set_template: dict[str, Any],
@@ -225,62 +314,29 @@ def converter(
         SetQuestions._structuredTutorialVisibility.status
     )
     # create the set file
-    with open(f"{output_question}/set_{set_name}.json", "w") as file:
+    folder = Path(output_question)
+    set_file = folder / f"set_{set_name}.json"
+    with open(set_file, "w") as file:
         json.dump(set_template, file)
 
-    for i in range(len(ListQuestions)):
-        output = _question_json(ListQuestions[i], i, question_template)
-
-        # Lambda Feedback names the file after the title with only spaces made
-        # underscores. Path separators go too, so a title cannot leave the set folder,
-        # and so do the characters Windows forbids in file names.
-        filename = (
-            "question_"
-            + str(i).zfill(3)
-            + "_"
-            + re.sub(r'[\s/\\<>:"|?*]', "_", output["title"].strip())
-        )
-
-        # write questions into directory
-        with open(f"{output_question}/{filename}.json", "w") as file:
-            json.dump(output, file)
-
-        # write image into directory
-        for k in range(len(ListQuestions[i].images)):
-            image_path = os.path.abspath(
-                ListQuestions[i].images[k]
-            )  # converts computer path into python path
-            # If images exist, create a media directory
-            output_image = os.path.join(output_question, "media")
-            os.makedirs(output_image, exist_ok=True)
-            shutil.copy(image_path, output_image)  # copies image into the directory
+    written = [set_file]
+    for i, question in enumerate(ListQuestions):
+        written += _write_question(question, i, question_template, folder)
 
     # output zip file in destination folder
-    _zip_sorted_folder(output_question, output_question + ".zip")
+    _zip(written, folder, output_question + ".zip")
 
 
 def main(set_questions: Set, output_dir: str) -> None:
-    """Preliminary defensive programming before calling the main converter function.
+    """Loads the templates and calls the main converter function.
 
     This ultimately then produces the Lambda Feedback JSON/ZIP files.
 
     Args:
         set_questions: A Set object containing questions.
         output_dir: Where to output the final Lambda Feedback JSON/ZIP files.
     """
-    # Use path so minimal template can be found regardless of where the user is running python from.
-    with open(Path(__file__).with_name(MINIMAL_QUESTION_TEMPLATE), "r") as file:
-        question_template = json.load(file)
-
-    with open(Path(__file__).with_name(MINIMAL_SET_TEMPLATE), "r") as file:
-        set_template = json.load(file)
-
-    # check if directory exists in file
-    if os.path.isdir(output_dir):
-        try:
-            shutil.rmtree(output_dir)
-        except OSError as e:
-            print("Error: %s : %s" % (output_dir, e.strerror))
+    question_template, set_template = _templates()
     converter(question_template, set_template, set_questions, output_dir)
 
 

tests/test_exports.py

@@ -10,6 +10,7 @@
 import json
 import re
 import uuid
+import zipfile
 from dataclasses import replace
 from pathlib import Path
 
@@ -134,6 +135,85 @@ def test_written_keys_exist_in_export(export_dir: Path, tmp_path: Path) -> None:
     assert not missing, missing
 
 
+@each_export
+def test_question_exports_alone(export_dir: Path, tmp_path: Path) -> None:
+    """Each question writes on its own as the set writes it, with the images it uses."""
+    loaded = Set.from_json(str(export_dir))
+    from_set = _write_back(loaded, tmp_path)
+    exported_media = _relative_files(export_dir / "media")
+
+    for i, question in enumerate(loaded.questions):
+        question.to_json(str(tmp_path / "single"), number=i)
+
+        # The question writes under the name the set gives it, so the files the set
+        # wrote for the same number say what to expect.
+        (set_file,) = from_set.glob(f"question_{i:03}_*.json")
+        folder = tmp_path / "single" / set_file.stem
+        expected = sorted(
+            [set_file.name]
+            + [
+                f"media/{name}"
+                for name in exported_media
+                if name.startswith(f"{set_file.stem}_")
+            ]
+        )
+
+        assert _relative_files(folder) == expected
+        with zipfile.ZipFile(f"{folder}.zip") as zf:
+            assert sorted(zf.namelist()) == expected
+
+        written = (folder / set_file.name).read_text()
+        assert json.loads(written) == json.loads(set_file.read_text())
+
+        # Every image the JSON points at must be beside it, or it will not resolve
+        # once the question is imported.
+        references = re.findall(r"!\[[^\]]*\]\(([^)]+)\)", written)
+        assert all(
+            (folder / "media" / reference).is_file() for reference in references
+        ), references
+
+
+def test_writing_leaves_other_files(tmp_path: Path) -> None:
+    """Writing over an export keeps files it did not write and leaves them out of the zip."""
+    image = tmp_path / "diagram.png"
+    image.write_bytes(b"not really a png")
+    question_set = Set(questions=[Question(title="Q", images=[str(image)])])
+
+    written = _write_back(question_set, tmp_path)
+    strays = [
+        tmp_path / "out" / "notes.txt",
+        written / "notes.txt",
+        written / "media" / "notes.txt",
+    ]
+    for stray in strays:
+        stray.write_text("someone else's work")
+
+    _write_back(question_set, tmp_path)
+
+    assert [stray.read_text() for stray in strays] == ["someone else's work"] * 3
+    with zipfile.ZipFile(f"{written}.zip") as zf:
+        assert sorted(zf.namelist()) == [
+            "media/diagram.png",
+            "question_000_Q.json",
+            "set_set.json",
+        ]
+
+
+def test_repeated_image_zipped_once(tmp_path: Path) -> None:
+    """An image listed twice, as one used in both a question and its solution, is one file."""
+    image = tmp_path / "diagram.png"
+    image.write_bytes(b"not really a png")
+    question_set = Set(questions=[Question(title="Q", images=[str(image), str(image)])])
+
+    written = _write_back(question_set, tmp_path)
+
+    assert _relative_files(written / "media") == ["diagram.png"]
+    with zipfile.ZipFile(f"{written}.zip") as zf:
+        assert [name for name in zf.namelist() if name.startswith("media/")] == [
+            "media/diagram.png"
+        ]
+
+
 def _area_shape(area: dict) -> frozenset[str]:
     # Without indices, an area's shape is the keys it has, not how many tests, cases
     # or symbols it lists.