From 4c618bcd02cb5db63076e2140fae2b54477cfa4a Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 12:52:26 +0200
Subject: [PATCH 01/15] Test new Zenodo-download procedure (ref #583)

---
 .github/workflows/ci_poetry.yml            |  3 ++
 .github/workflows/download_zenodo_data.yml | 12 ++++++++
 tests/data/download_zenodo_data.sh         | 32 ++++++++++++++++++++++
 3 files changed, 47 insertions(+)
 create mode 100644 .github/workflows/download_zenodo_data.yml
 create mode 100644 tests/data/download_zenodo_data.sh

diff --git a/.github/workflows/ci_poetry.yml b/.github/workflows/ci_poetry.yml
index cf32e7850..b26fc9590 100644
--- a/.github/workflows/ci_poetry.yml
+++ b/.github/workflows/ci_poetry.yml
@@ -42,6 +42,9 @@ jobs:
       - name: Install dependencies (without extras)
         run: poetry install --with dev --without docs --no-interaction
 
+      - name: Download Zenodo data
+        uses: ./.github/workflows/download_zenodo_data.yml
+
       - name: Test core library with pytest
         run: poetry run coverage run -m pytest tests --ignore tests/tasks
 
diff --git a/.github/workflows/download_zenodo_data.yml b/.github/workflows/download_zenodo_data.yml
new file mode 100644
index 000000000..491c86c3f
--- /dev/null
+++ b/.github/workflows/download_zenodo_data.yml
@@ -0,0 +1,12 @@
+name: Download Zenodo data
+on: [workflow_call]
+jobs:
+  zenodo:
+    name: "Download Zenodo data"
+    runs-on: ubuntu-22
+    timeout-minutes: 5
+    steps:
+      - name: Install jq
+        uses: dcarbone/install-jq-action@v2.0.2
+      - name: Test core library with pytest
+        run: bash tests/data/download_zenodo_data.sh
diff --git a/tests/data/download_zenodo_data.sh b/tests/data/download_zenodo_data.sh
new file mode 100644
index 000000000..f5221742e
--- /dev/null
+++ b/tests/data/download_zenodo_data.sh
@@ -0,0 +1,32 @@
+#!/bin/bash
+
+LIST_RECORD_ID="7059515 8091756"
+
+for RECORD_ID in $LIST_RECORD_ID; do
+    echo "****************************"
+    echo "START RECORD_ID=$RECORD_ID"
+    OUTPUT_FOLDER=tests/data/10_5281_zenodo_$RECORD_ID
+    echo "OUTPUT_FOLDER: $OUTPUT_FOLDER"
+
+    if [ -d $OUTPUT_FOLDER ]; then
+        echo "OUTPUT_FOLDER already exists. Exit."
+    else
+        mkdir $OUTPUT_FOLDER
+        FILES=`curl https://zenodo.org/api/records/$RECORD_ID | jq -r ".files[].links.download"`
+        echo "curl exit code: $?"
+        for FILE in $FILES; do
+            FILEPATH=${FILE%"/content"}
+            FILENAME=`basename $FILEPATH`
+            echo "FILE:     $FILE"
+            echo "FILEPATH: $FILEPATH"
+            echo "FILENAME: $FILENAME"
+            echo
+            wget $FILE --output-document=${OUTPUT_FOLDER}/${FILENAME}
+            echo
+        done
+    fi
+
+    echo "END RECORD_ID=$RECORD_ID"
+    echo "****************************"
+    echo
+done

From 8df5a71671da59d4092f56d79ed0b633b84e01a5 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 12:56:06 +0200
Subject: [PATCH 02/15] Update github action

---
 .github/workflows/ci_poetry.yml            |  2 +-
 .github/workflows/download_zenodo_data.yml | 12 ------------
 2 files changed, 1 insertion(+), 13 deletions(-)
 delete mode 100644 .github/workflows/download_zenodo_data.yml

diff --git a/.github/workflows/ci_poetry.yml b/.github/workflows/ci_poetry.yml
index b26fc9590..5653cc57c 100644
--- a/.github/workflows/ci_poetry.yml
+++ b/.github/workflows/ci_poetry.yml
@@ -43,7 +43,7 @@ jobs:
         run: poetry install --with dev --without docs --no-interaction
 
       - name: Download Zenodo data
-        uses: ./.github/workflows/download_zenodo_data.yml
+        run: bash tests/data/download_zenodo_data.sh
 
       - name: Test core library with pytest
         run: poetry run coverage run -m pytest tests --ignore tests/tasks
diff --git a/.github/workflows/download_zenodo_data.yml b/.github/workflows/download_zenodo_data.yml
deleted file mode 100644
index 491c86c3f..000000000
--- a/.github/workflows/download_zenodo_data.yml
+++ /dev/null
@@ -1,12 +0,0 @@
-name: Download Zenodo data
-on: [workflow_call]
-jobs:
-  zenodo:
-    name: "Download Zenodo data"
-    runs-on: ubuntu-22
-    timeout-minutes: 5
-    steps:
-      - name: Install jq
-        uses: dcarbone/install-jq-action@v2.0.2
-      - name: Test core library with pytest
-        run: bash tests/data/download_zenodo_data.sh

From 1aff5092662f1235a8e6edd04571588ff9cd39b1 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 12:57:21 +0200
Subject: [PATCH 03/15] Add wget --no-verbose option

---
 tests/data/download_zenodo_data.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/data/download_zenodo_data.sh b/tests/data/download_zenodo_data.sh
index f5221742e..5513e2ec7 100644
--- a/tests/data/download_zenodo_data.sh
+++ b/tests/data/download_zenodo_data.sh
@@ -21,7 +21,7 @@ for RECORD_ID in $LIST_RECORD_ID; do
             echo "FILEPATH: $FILEPATH"
             echo "FILENAME: $FILENAME"
             echo
-            wget $FILE --output-document=${OUTPUT_FOLDER}/${FILENAME}
+            wget --no-verbose $FILE --output-document=${OUTPUT_FOLDER}/${FILENAME}
             echo
         done
     fi

From fc2c85d8ba8656e43dfe5a6d9acab8e8115f6e76 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 13:00:06 +0200
Subject: [PATCH 04/15] Add caching

---
 .github/workflows/ci_poetry.yml | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/.github/workflows/ci_poetry.yml b/.github/workflows/ci_poetry.yml
index 5653cc57c..4ae2f78d0 100644
--- a/.github/workflows/ci_poetry.yml
+++ b/.github/workflows/ci_poetry.yml
@@ -42,7 +42,15 @@ jobs:
       - name: Install dependencies (without extras)
         run: poetry install --with dev --without docs --no-interaction
 
+      - name: Cache Zenodo data
+        id: cache-zenodo-data
+        uses: actions/cache@v3
+        with:
+          path: tests/data/
+          key: zenodo-data
+
       - name: Download Zenodo data
+        if: steps.cache-zenodo-data.outputs.cache-hit != 'true'
         run: bash tests/data/download_zenodo_data.sh
 
       - name: Test core library with pytest
@@ -91,6 +99,17 @@ jobs:
       - name: Check manifest task metadata
         run: poetry run python fractal_tasks_core/dev/check_manifest.py
 
+      - name: Cache Zenodo data
+        id: cache-zenodo-data
+        uses: actions/cache@v3
+        with:
+          path: tests/data/
+          key: zenodo-data
+
+      - name: Download Zenodo data
+        if: steps.cache-zenodo-data.outputs.cache-hit != 'true'
+        run: bash tests/data/download_zenodo_data.sh
+
       - name: Test tasks with pytest
         run: poetry run coverage run -m pytest tests/tasks
 

From 87a57fe5c25e3d015e67110e436b97436ce4bb7c Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 13:01:47 +0200
Subject: [PATCH 05/15] Trigger ci

---
 x | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 create mode 100644 x

diff --git a/x b/x
new file mode 100644
index 000000000..e69de29bb

From 246567819f9799a40b8ee208edf2cf8f15a4379e Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 13:02:01 +0200
Subject: [PATCH 06/15] Revert last dummy commit

---
 x | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 delete mode 100644 x

diff --git a/x b/x
deleted file mode 100644
index e69de29bb..000000000

From 157867898aac6d970a0e66514a46916d80638ae4 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 13:24:20 +0200
Subject: [PATCH 07/15] Add echo

---
 tests/data/download_zenodo_data.sh | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tests/data/download_zenodo_data.sh b/tests/data/download_zenodo_data.sh
index 5513e2ec7..cecbc56ad 100644
--- a/tests/data/download_zenodo_data.sh
+++ b/tests/data/download_zenodo_data.sh
@@ -14,6 +14,7 @@ for RECORD_ID in $LIST_RECORD_ID; do
         mkdir $OUTPUT_FOLDER
         FILES=`curl https://zenodo.org/api/records/$RECORD_ID | jq -r ".files[].links.download"`
         echo "curl exit code: $?"
+        echo
         for FILE in $FILES; do
             FILEPATH=${FILE%"/content"}
             FILENAME=`basename $FILEPATH`
@@ -24,6 +25,11 @@ for RECORD_ID in $LIST_RECORD_ID; do
             wget --no-verbose $FILE --output-document=${OUTPUT_FOLDER}/${FILENAME}
             echo
         done
+
+        if [ $RECORD_ID == "8091756" ]; then
+            unzip tests/data/10_5281_zenodo_8091756/20200812-CardiomyocyteDifferentiation14-Cycle1.zarr.zip -d tests/data/10_5281_zenodo_8091756
+            unzip tests/data/10_5281_zenodo_8091756/20200812-CardiomyocyteDifferentiation14-Cycle1_mip.zarr.zip -d tests/data/10_5281_zenodo_8091756
+        fi
     fi
 
     echo "END RECORD_ID=$RECORD_ID"

From 26fbf702c3078786274619482a38624d909e58a3 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 13:43:59 +0200
Subject: [PATCH 08/15] Update zenodo fixtures and download script

---
 tests/conftest.py                  | 93 +++++++++++++-----------------
 tests/data/download_zenodo_data.sh |  1 +
 2 files changed, 41 insertions(+), 53 deletions(-)

diff --git a/tests/conftest.py b/tests/conftest.py
index 931acc172..4a2d4f3b4 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -1,13 +1,19 @@
 import json
+import logging
 import os
 import shutil
 import time
 from pathlib import Path
-from urllib.parse import unquote
 
+import anndata as ad
 import pytest
 import requests  # type: ignore
 import wget
+import zarr
+from devtools import debug
+
+from fractal_tasks_core.lib_regions_of_interest import reset_origin
+from fractal_tasks_core.lib_write import write_table
 
 
 @pytest.fixture(scope="session")
@@ -17,48 +23,38 @@ def testdata_path() -> Path:
 
 
 @pytest.fixture(scope="function")
-def zenodo_images(testdata_path, capsys):
+def zenodo_images(testdata_path):
     """
     Inspired by
     https://github.com/dvolgyes/zenodo_get/blob/master/zenodo_get/zget.py
-
-    See https://docs.pytest.org/en/7.4.x/how-to/capture-stdout-stderr.html for
-    the use of capsys
     """
-
     t_start = time.perf_counter()
 
-    url = "10.5281/zenodo.7059515"
-    folder = str(testdata_path / (url.replace(".", "_").replace("/", "_")))
+    # Download images and metadata files
+    recordID = "7059515"
+    url = "10_5281_zenodo_7059515"
+    folder = str(testdata_path / f"10_5281_zenodo_{recordID}")
     if os.path.isdir(folder):
-        print(f"{folder} already exists, skip")
-        return folder
-    os.makedirs(folder)
-    url = "https://doi.org/" + url
-    print(f"I will download {url} files to {folder}")
-
-    r = requests.get(url)
-    recordID = r.url.split("/")[-1]
-    url = "https://zenodo.org/api/records/"
-    r = requests.get(url + recordID)
-
-    js = json.loads(r.text)
-    files = js["files"]
-    for f in files:
-        fname = f["filename"]
-        link = f"https://zenodo.org/record/{recordID}/files/{fname}"
-        print(link)
-        link = unquote(link)
-        wget.download(link, out=folder)
-        print()
+        print(f"{folder} already exists, skip download")
+    else:
+        os.makedirs(folder)
+        url = f"https://zenodo.org/api/records/{recordID}"
+        r = requests.get(url)
+        js = json.loads(r.text)
+        files = js["files"]
+        for f in files:
+            file_url = f["links"]["download"]
+            file_name = file_url.split("/")[-2]
+            wget.download(file_url, out=f"{folder}/{file_name}", bar=False)
 
     # Add an image with invalid name, that should be skipped during parsing
     with open(f"{folder}/invalid_path.png", "w") as f:
         f.write("This file has an invalid filename, which cannot be parsed.")
 
     t_end = time.perf_counter()
-    with capsys.disabled():
-        print(f"\n    Time spent in zenodo_images: {t_end-t_start:.2f} s")
+    logging.warning(
+        f"\n    Time spent in zenodo_images: {t_end-t_start:.2f} s"
+    )
 
     return folder
 
@@ -78,12 +74,8 @@ def zenodo_images_multiplex(testdata_path, zenodo_images):
     return cycle_folders
 
 
-@pytest.fixture(scope="function")
-def zenodo_zarr(testdata_path, tmpdir_factory, capsys):
-    """
-    See https://docs.pytest.org/en/7.4.x/how-to/capture-stdout-stderr.html for
-    the use of capsys
-    """
+@pytest.fixture(scope="session")
+def zenodo_zarr(testdata_path, tmpdir_factory):
     t_start = time.perf_counter()
 
     doi = "10.5281/zenodo.8091756"
@@ -91,33 +83,30 @@ def zenodo_zarr(testdata_path, tmpdir_factory, capsys):
     platenames = ["plate.zarr", "plate_mip.zarr"]
     folders = [rootfolder / plate for plate in platenames]
 
+    # Download dataset
     if rootfolder.exists():
-        print(f"{str(rootfolder)} already exists, skip")
-        folders = [str(f) for f in folders]
-        return folders
+        print(f"{str(rootfolder)} already exists, skip download part")
     else:
-
-        import zarr
-        import anndata as ad
-        import logging
-
-        from fractal_tasks_core.lib_regions_of_interest import reset_origin
-        from fractal_tasks_core.lib_write import write_table
-
         rootfolder.mkdir()
         tmp_path = tmpdir_factory.mktemp("zenodo_zarr")
         zarrnames = [
             "20200812-CardiomyocyteDifferentiation14-Cycle1.zarr",
             "20200812-CardiomyocyteDifferentiation14-Cycle1_mip.zarr",
         ]
-        for zarrname, folder in zip(zarrnames, folders):
+        for zarrname in zarrnames:
             zipname = f"{zarrname}.zip"
             url = f"https://zenodo.org/record/8091756/files/{zipname}"
+            debug(url)
             wget.download(url, out=str(tmp_path / zipname), bar=None)
             shutil.unpack_archive(
-                str(tmp_path / zipname), extract_dir=rootfolder, format="zip"
+                str(tmp_path / zipname),
+                extract_dir=rootfolder,
+                format="zip",
             )
-            shutil.move(str(rootfolder / zarrname), str(folder))
+
+        # Rename and update OME-Zarr
+        for zarrname, folder in zip(zarrnames, folders):
+            shutil.copytree(str(rootfolder / zarrname), str(folder))
 
             # Update well/FOV ROI tables, by shifting their origin to 0
             # TODO: remove this fix, by uploading new zarrs to zenodo (ref
@@ -139,9 +128,7 @@ def zenodo_zarr(testdata_path, tmpdir_factory, capsys):
     folders = [str(f) for f in folders]
 
     t_end = time.perf_counter()
-    with capsys.disabled():
-        print(f"\n    Time spent in zenodo_zarr: {t_end-t_start:.2f} s")
-
+    logging.warning(f"\n    Time spent in zenodo_zarr: {t_end-t_start:.2f} s")
     return folders
 
 
diff --git a/tests/data/download_zenodo_data.sh b/tests/data/download_zenodo_data.sh
index cecbc56ad..c6e868c34 100644
--- a/tests/data/download_zenodo_data.sh
+++ b/tests/data/download_zenodo_data.sh
@@ -1,5 +1,6 @@
 #!/bin/bash
 
+LIST_RECORD_ID="7059515 8091756"
 LIST_RECORD_ID="7059515 8091756"
 
 for RECORD_ID in $LIST_RECORD_ID; do

From 90b94697c17550ee0e28bcfff2fc41781c695a00 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 13:48:27 +0200
Subject: [PATCH 09/15] Add time.sleep in zenodo_zarr fixture

---
 tests/conftest.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/conftest.py b/tests/conftest.py
index 4a2d4f3b4..0866a8f6f 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -98,6 +98,7 @@ def zenodo_zarr(testdata_path, tmpdir_factory):
             url = f"https://zenodo.org/record/8091756/files/{zipname}"
             debug(url)
             wget.download(url, out=str(tmp_path / zipname), bar=None)
+            time.sleep(0.5)
             shutil.unpack_archive(
                 str(tmp_path / zipname),
                 extract_dir=rootfolder,

From a0158d406c83b13e85a5db8f2e2e97543c02d6a0 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 13:51:45 +0200
Subject: [PATCH 10/15] Clean up zenodo fixtures

---
 tests/conftest.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/tests/conftest.py b/tests/conftest.py
index 0866a8f6f..b9b634a2c 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -22,7 +22,7 @@ def testdata_path() -> Path:
     return TEST_DIR / "data/"
 
 
-@pytest.fixture(scope="function")
+@pytest.fixture(scope="session")
 def zenodo_images(testdata_path):
     """
     Inspired by
@@ -59,7 +59,7 @@ def zenodo_images(testdata_path):
     return folder
 
 
-@pytest.fixture(scope="function")
+@pytest.fixture(scope="session")
 def zenodo_images_multiplex(testdata_path, zenodo_images):
     folder = str(testdata_path / "fake_multiplex")
     cycle_folder_1 = str(Path(folder) / "cycle1")
@@ -107,6 +107,7 @@ def zenodo_zarr(testdata_path, tmpdir_factory):
 
         # Rename and update OME-Zarr
         for zarrname, folder in zip(zarrnames, folders):
+            shutil.rmtree(str(folder))
             shutil.copytree(str(rootfolder / zarrname), str(folder))
 
             # Update well/FOV ROI tables, by shifting their origin to 0

From be06f2b54e4522a911bfddf59c92d6fc0fd90d64 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 13:57:07 +0200
Subject: [PATCH 11/15] Clean up zenodo_zarrs fixture

---
 tests/conftest.py | 50 ++++++++++++++++++++++++-----------------------
 1 file changed, 26 insertions(+), 24 deletions(-)

diff --git a/tests/conftest.py b/tests/conftest.py
index b9b634a2c..f2a0eeac6 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -82,6 +82,10 @@ def zenodo_zarr(testdata_path, tmpdir_factory):
     rootfolder = testdata_path / (doi.replace(".", "_").replace("/", "_"))
     platenames = ["plate.zarr", "plate_mip.zarr"]
     folders = [rootfolder / plate for plate in platenames]
+    zarrnames = [
+        "20200812-CardiomyocyteDifferentiation14-Cycle1.zarr",
+        "20200812-CardiomyocyteDifferentiation14-Cycle1_mip.zarr",
+    ]
 
     # Download dataset
     if rootfolder.exists():
@@ -89,10 +93,6 @@ def zenodo_zarr(testdata_path, tmpdir_factory):
     else:
         rootfolder.mkdir()
         tmp_path = tmpdir_factory.mktemp("zenodo_zarr")
-        zarrnames = [
-            "20200812-CardiomyocyteDifferentiation14-Cycle1.zarr",
-            "20200812-CardiomyocyteDifferentiation14-Cycle1_mip.zarr",
-        ]
         for zarrname in zarrnames:
             zipname = f"{zarrname}.zip"
             url = f"https://zenodo.org/record/8091756/files/{zipname}"
@@ -105,27 +105,29 @@ def zenodo_zarr(testdata_path, tmpdir_factory):
                 format="zip",
             )
 
-        # Rename and update OME-Zarr
-        for zarrname, folder in zip(zarrnames, folders):
+    # Based on the Zenodo OME-Zarrs, create the appropriate OME-Zarrs to be
+    # used in tests
+    for zarrname, folder in zip(zarrnames, folders):
+        if os.path.isdir(str(folder)):
             shutil.rmtree(str(folder))
-            shutil.copytree(str(rootfolder / zarrname), str(folder))
-
-            # Update well/FOV ROI tables, by shifting their origin to 0
-            # TODO: remove this fix, by uploading new zarrs to zenodo (ref
-            # issue 526)
-            image_group_path = folder / "B/03/0"
-            group_image = zarr.open_group(str(image_group_path))
-            for table_name in ["FOV_ROI_table", "well_ROI_table"]:
-                table_path = str(image_group_path / "tables" / table_name)
-                old_table = ad.read_zarr(table_path)
-                new_table = reset_origin(old_table)
-                write_table(
-                    group_image,
-                    table_name,
-                    new_table,
-                    overwrite=True,
-                    logger=logging.getLogger(),
-                )
+        shutil.copytree(str(rootfolder / zarrname), str(folder))
+
+        # Update well/FOV ROI tables, by shifting their origin to 0
+        # TODO: remove this fix, by uploading new zarrs to zenodo (ref
+        # issue 526)
+        image_group_path = folder / "B/03/0"
+        group_image = zarr.open_group(str(image_group_path))
+        for table_name in ["FOV_ROI_table", "well_ROI_table"]:
+            table_path = str(image_group_path / "tables" / table_name)
+            old_table = ad.read_zarr(table_path)
+            new_table = reset_origin(old_table)
+            write_table(
+                group_image,
+                table_name,
+                new_table,
+                overwrite=True,
+                logger=logging.getLogger(),
+            )
 
     folders = [str(f) for f in folders]
 

From 82f987f55628c01d37f5853247f73e6becf8462c Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 14:00:22 +0200
Subject: [PATCH 12/15] Also cache Zenodo data in pip CI

---
 .github/workflows/ci_pip.yml | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/.github/workflows/ci_pip.yml b/.github/workflows/ci_pip.yml
index eaf07fb55..f4e41bc4f 100644
--- a/.github/workflows/ci_pip.yml
+++ b/.github/workflows/ci_pip.yml
@@ -35,6 +35,17 @@ jobs:
       - name: Install some testing dependencies (hard-coded)
         run: python -m pip install pytest devtools jsonschema requests wget
 
+      - name: Cache Zenodo data
+        id: cache-zenodo-data
+        uses: actions/cache@v3
+        with:
+          path: tests/data/
+          key: zenodo-data
+
+      - name: Download Zenodo data
+        if: steps.cache-zenodo-data.outputs.cache-hit != 'true'
+        run: bash tests/data/download_zenodo_data.sh
+
       - name: Test core library with pytest
         run: pytest tests --ignore tests/tasks
 
@@ -65,5 +76,16 @@ jobs:
       - name: Install some testing dependencies (hard-coded)
         run: python -m pip install pytest devtools jsonschema requests wget
 
+      - name: Cache Zenodo data
+        id: cache-zenodo-data
+        uses: actions/cache@v3
+        with:
+          path: tests/data/
+          key: zenodo-data
+
+      - name: Download Zenodo data
+        if: steps.cache-zenodo-data.outputs.cache-hit != 'true'
+        run: bash tests/data/download_zenodo_data.sh
+
       - name: Test tasks with pytest
         run: pytest tests tests/tasks

From 4c394eda994065ef654c66981f1c1f5d63932995 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 14:06:02 +0200
Subject: [PATCH 13/15] BROKEN add `raise` to fixture, to verify expected
 behavior

---
 tests/conftest.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/conftest.py b/tests/conftest.py
index f2a0eeac6..8f3ecee60 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -45,6 +45,7 @@ def zenodo_images(testdata_path):
         for f in files:
             file_url = f["links"]["download"]
             file_name = file_url.split("/")[-2]
+            raise ValueError("We are downloading data from zenodo_images")
             wget.download(file_url, out=f"{folder}/{file_name}", bar=False)
 
     # Add an image with invalid name, that should be skipped during parsing
@@ -97,6 +98,7 @@ def zenodo_zarr(testdata_path, tmpdir_factory):
             zipname = f"{zarrname}.zip"
             url = f"https://zenodo.org/record/8091756/files/{zipname}"
             debug(url)
+            raise ValueError("We are downloading data from zenodo_zarr")
             wget.download(url, out=str(tmp_path / zipname), bar=None)
             time.sleep(0.5)
             shutil.unpack_archive(

From 773cace77740f83612cf2e8d81209af2e20a7c31 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 14:07:56 +0200
Subject: [PATCH 14/15] Revert "BROKEN add `raise` to fixture, to verify
 expected behavior"

This reverts commit 4c394eda994065ef654c66981f1c1f5d63932995.
---
 tests/conftest.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/tests/conftest.py b/tests/conftest.py
index 8f3ecee60..f2a0eeac6 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -45,7 +45,6 @@ def zenodo_images(testdata_path):
         for f in files:
             file_url = f["links"]["download"]
             file_name = file_url.split("/")[-2]
-            raise ValueError("We are downloading data from zenodo_images")
             wget.download(file_url, out=f"{folder}/{file_name}", bar=False)
 
     # Add an image with invalid name, that should be skipped during parsing
@@ -98,7 +97,6 @@ def zenodo_zarr(testdata_path, tmpdir_factory):
             zipname = f"{zarrname}.zip"
             url = f"https://zenodo.org/record/8091756/files/{zipname}"
             debug(url)
-            raise ValueError("We are downloading data from zenodo_zarr")
             wget.download(url, out=str(tmp_path / zipname), bar=None)
             time.sleep(0.5)
             shutil.unpack_archive(

From ce0c9d061640de8df3b241961a0c08d5b56e1855 Mon Sep 17 00:00:00 2001
From: Tommaso Comparin <3862206+tcompa@users.noreply.github.com>
Date: Fri, 20 Oct 2023 14:10:03 +0200
Subject: [PATCH 15/15] Update CHANGELOG [skip ci]

---
 CHANGELOG.md | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index c4810eeaa..d6946bb08 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,10 @@
 **Note**: Numbers like (\#123) point to closed Pull Requests on the fractal-tasks-core repository.
 
+# Unreleased
+
+* Testing:
+    * Cache Zenodo data, within GitHub actions (\#585).
+
 # 0.13.0
 
 * Tasks: