Gdc client galaxy tool

.gitignore

@@ -1,3 +1,7 @@
+# Planemo-generated test outputs
+tool_test_output.html
+tool_test_output.json
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[codz]

tools/gdc-downloader/dependencies/Dockerfile

@@ -0,0 +1,12 @@
+FROM ubuntu:20.04 AS builder
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends wget unzip ca-certificates && \
+    wget https://gdc.cancer.gov/system/files/public/file/gdc-client_2.3_Ubuntu_x64-py3.8-ubuntu-20.04.zip && \
+    unzip gdc-client_2.3_Ubuntu_x64-py3.8-ubuntu-20.04.zip && \
+    unzip gdc-client_2.3_Ubuntu_x64.zip && \
+    chmod 755 gdc-client
+
+FROM ubuntu:20.04
+
+COPY --from=builder /gdc-client /usr/local/bin/gdc-client

tools/gdc-downloader/gdc-downloader.xml

@@ -0,0 +1,86 @@
+<?xml version="1.0"?>
+<tool
+    id="gdc_downloader"
+    name="GDC Data Downloader"
+    version="0.0.1"
+    hidden="false"
+    profile="25.1"
+>
+    <description>Acquire data from the NCI Genomic Data Commons</description>
+    <command><![CDATA[
+        mkdir ./downloads &&
+        mkfifo -m 600 token.pipe &&
+        (printf '%s' "\$GDC_AUTH_TOKEN" > token.pipe &) &&
+        gdc-client download -m "$manifest_file" -t token.pipe -d ./downloads &&
+        rm -f token.pipe &&
+        rm -rf ./downloads/*/logs
+    ]]></command>
+    <requirements>
+        <container type="docker">quay.io/goeckslab/gdc-client:2.3</container>
+        <credentials
+            name="gdc_auth"
+            version="1.0"
+            optional="true"
+            label="GDC credentials"
+            description="Used when acquiring protected datasets you are authorized to access"
+        >
+            <secret
+                name="gdc_token"
+                inject_as_env="GDC_AUTH_TOKEN"
+                label="Authorization token"
+                description="Acquired from the GDC portal"
+            />
+        </credentials>
+    </requirements>
+    <inputs>
+        <param
+            name="manifest_file"
+            type="data"
+            format="txt"
+            label="Manifest file downloaded from GDC"
+        />
+    </inputs>
+    <outputs>
+        <collection name="gdc_files" type="list" label="GDC files from manifest">
+            <discover_datasets pattern="__name_and_ext__" directory="downloads" recurse="true" />
+        </collection>
+    </outputs>
+    <tests>
+        <test expect_num_outputs="1">
+            <param name="manifest_file" value="test-manifest.txt" ftype="txt" />
+            <output_collection name="gdc_files" type="list" count="1">
+                <element name="TCGA-E2-A15K-11A-01-TS1.28656e31-f16f-431e-95d4-915892ecdff8">
+                    <assert_contents>
+                        <has_size value="43659635" />
+                    </assert_contents>
+                </element>
+            </output_collection>
+        </test>
+    </tests>
+    <help><![CDATA[
+**What it does**
+
+Downloads data files from the `NCI Genomic Data Commons (GDC) <https://portal.gdc.cancer.gov/>`_ using
+``gdc-client``, based on a GDC manifest file.
+
+**Manifest**
+
+The input is a manifest file listing the files to download, generated from the
+`GDC Data Portal <https://portal.gdc.cancer.gov/>`_ (e.g. via the cart download option).
+
+**Authorization token (optional)**
+
+A GDC authorization token is only required to download **controlled-access** datasets; open-access
+data can be downloaded without one. Tokens are also obtained from the
+`GDC Data Portal <https://portal.gdc.cancer.gov/>`_, under your account profile.
+
+To use your authorization token, store it using Galaxy's credentials
+system, which keeps it encrypted and injects it securely at runtime. See
+`Galaxy's credentials documentation <https://galaxyproject.org/news/2026-01-12-tool-credentials-system/#how-to-use-it-for-galaxy-users>`_
+for instructions on adding your GDC token as a credential.
+
+**Output**
+
+A collection containing the files listed in the manifest, as downloaded from the GDC.
+    ]]></help>
+</tool>

tools/gdc-downloader/test-data/test-manifest.txt

@@ -0,0 +1,2 @@
+id	filename	md5	size	state
+360367d5-4aa4-4c32-aaf3-7d8528aec089	TCGA-E2-A15K-11A-01-TS1.28656e31-f16f-431e-95d4-915892ecdff8.svs	7c401ac82c290ff845f8262fa8e41798	43659635	released